Can literacy change brain anatomy by feggy ostrosky solis

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 1–4

Can literacy change brain anatomy?

F.Ostrosky-Solís Universidad Nacional Autónoma de México, Lomas de Reforma, Mexico

Several studies have postulated that education and/or literacy may not only protect against the effects of biological ageing (Albert et al, 1995; Christensen & Henderson, 1991; Orrell & Sahakian, 1995), but also against the clinical manifestation of cerebral neuropathology (Katzman, 1993; Stern, 2002; Stern, Gurland, Tatemichi, Tang, Wilder, & Mayeux, 1994; Zhang et al., 1990). In clinical neuropsychology, much debate has centred on whether the brain is more likely to degenerate as a result of overuse or underuse; while some epidemiological studies have suggested that active engagement in intellectual, social, and physical activities may delay the cognitive deterioration associated with normal ageing (Scarmeas, Levy, Tang, Manly, & Stern, 2001), other studies have emphasized that the protective effect of education is not always observed but depends upon the specific cognitive ability that is measured (Ardila, Ostrosky-Solís, Rosselli, & Gómez, 2000; Ostrosky-Solís, Ardila, Rosselli, López, & Mendoza, 1998). Ostrosky-Solís (2002) points out that protection or age-related decline attenuation in well-educated subjects is highly related to verbal abilities; thus, education and verbal advantage could serve as a means of compensatory strategies, such as using verbal cues to aid recall or encoding visuospatial tasks with language. These are the strategies provided by formal education. The use of these strategies could mask otherwise similar rates of biological ageing among different educational groups, and this advantage, coupled with the effects of several important variables such as good health, appropriate occupation, and active engagement with the surrounding environment, could explain why cognitive stimulation can provide some moderating influence on the complex changes in cognitive performance associated with ageing.

It has also been reported that Alzheimer’s disease not only has a later onset but that it is less severe in highly educated people (Katzman, 1993; Stern et al., 1994). This association of high education with late age of onset of dementia has been considered as an evidence of cognitive and/or brain reserve (Katzman, 1993; Mortimer, 1988; Satz, 1993; Stern, 2002).

The articles presented in this Special Issue analyse the impact of literacy on the anatomic and functional organization of the adult brain. Cognitive neuroimaging studies, event-related potentials, neuropsychological data of literate and illiterate subjects, and discussion regarding the origins and evolution of reading and writing are presented.

Stern, Scarmeas, and Habeck point out that the cognitive reserve model suggests that variables such as education and IQ are associated with cognitive reserve (CR) and may mediate differential susceptibility to age-related memory changes. They propose two complementary facets to CR: reserve—individual differences in the capacity to perform task—and compensation—the use of alternate brain networks or cognitive processes to cope with brain pathology. However, up to now the neurophysiologic substrate of CR has not been established. Therefore in order to explore the anatomical basis for CR in healthy young and old individuals, they used H215O PET to analyse the relationship between CR and task-related activation

2

OSTROSKY-SOLÍS

during the performance of a nonverbal recognition memory test. The first two studies focused on young subjects, and found either brain areas or brain networks where the amount of increased activation correlated with CR. The third study compared activation patterns of young and elderly individuals and found locations where the relation between activation and CR differed across the two groups. They suggest that these may represent areas where compensation for the effects of ageing has caused a functional reorganization of the neural substrate for task performance. For the young subjects this brain network may be used to cope with increasing task demands, and may represent a neural manifestation of reserve. These exploratory analyses suggest that it is possible to identify the neural substrate for these two aspects of cognitive reserve.

In a series of pioneering research dating back to 1976, Castro-Caldas and colleagues have used illiteracy as a tool to understand the way the brain adapts to information. Using behavioural and neuroimaging studies, they have shown that learning how to read and write during childhood affects the functional organization of the adult brain. In his article, Castro-Caldas describes two types of effects that can be related to the exposure of the stimulation that is involved in the complex process of schooling: a diffuse effect and a focal effect. He defines the diffuse effect as the one related to the adaptation to a rich environment, such as that present in school, and which introduces several changes in brain function, like the increase of abstract thinking and the development of parallel processing of information. The focal effect is related to the learning of specific skills and operations that constitute the mastery of reading and writing and which may change particular areas of the brain involved in these operations. Castro-Caldas reviews, both from the functional and from the anatomical point of view, how the knowledge of reading and writing has effects on several cognitive process including visual processing, cross-modal operations (audio-visual and visuotactile), and interhemispheric crossing of information. Using the results of neuroimaging studies, he reports differences between groups of literate and illiterate subjects in several areas: the corpus callosum is thinner in the illiterate group in the segment where the parietal lobe fibres cross; the parietal lobe processing of both hemispheres is different between groups; and the occipital lobe processes information more slowly in individuals who learned to read as adults compared to those who learned at the usual age. While dealing with phonology, a complex pattern of brain activation was only present in literate subjects.

The paper by Ostrosky-Solís, Arellano Garcia, and Perez explores the issue of language lateralization in illiterates. Although some studies reported that lesions in the right hemisphere resulted in a greater incidence of language difficulties in illiterate stroke victims than in their literate counterparts, and that aphasia was less severe in left-stroke illiterate patients, other studies did not replicate these findings and found left-hemisphere dominance in illiterates. Up to now, the question remains as to whether the functional balance between the two cerebral hemispheres while processing oral language could be modified by the knowledge of orthography; thus literacy could play a significant role in language lateralization. Using neurophysiological techniques the authors recorded cortical evoked potentials to a probe click stimulus to assess the extent of activation of the two cerebral hemispheres during a verbal memory task in literate and illiterate subjects. They found a left-hemisphere attenuation during the experimental condition in both groups.

Correspondence should be sent to Dr Feggy Ostrosky-Solís, Laboratory of Neuropsychology and Psychophysiology, Faculty of Psychology, Universidad Nacional Autónoma de Mexico. Rivera de Cupia 110–71, Lomas de Reforma, Mexico, DF, 11930 (E-mail: feggy@prodigy.net.mx). © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000231

CAN LITERACY CHANGE BRAIN ANATOMY?

3

However, during the verbal memory task, significant intrahemispheric differences between groups were found at parietotemporal areas. Results seem to indicate that learning how to read and write demands an intrahemispheric specialization with an important activation of parietotemporal areas. These data support the view that the brains of illiterate subjects show patterns of activation that are different to those of literate subjects, thus suggesting that environmental conditions can influence brain organization.

A large number of studies have found that level of education has been proven to have an important impact on the cerebral organization of cognitive skills and on performance in neuropsychological tests, and it has been suggested that the development and organization of psychological processes such as abstraction, inference, and memory depends on the type of symbols (i.e., written system) used by the individuals in their environment. As suggested by Vygotsky (1978) many years ago, reading and writing are cognitive tools that, once acquired, change the way in which stimuli are memorized and conceptualized, thus stimulating abstract thinking. Manly, Byrd, Touradji, Sanchez, and Stern analysed the effects of literacy on neuropsychological test performance among ethnically diverse elders from Northern Manhattan, NY. Instead of years of education, literacy was assessed by using a reading measure; they concluded that reading level is a more sensitive predictor of baseline test performance, and also that literacy skills are protective against memory decline. They point out that differences in organization of visuospatial information, lack of previous exposure to stimuli, and difficulties with interpretation of the logical functions of language are possible factors that affect test performance of elders with low levels of literacy.

Although culture and education are factors that significantly affect cognitive performance, it is often difficult to distinguish between the effects of education and the effects of culture, since educational level influences the sociocultural status of an individual. Ostrosky-SolĂ­s, RamĂ­rez, Lozano, Picasso, and VĂŠlez analysed the influence of education and of culture on the neuropsychological profile of an indigenous and a nonindigenous population. They studied the Maya group, who live in the state of Yucatan in the Mexican Republic. Results showed that indigenous subjects showed higher scores in visuospatial tasks and that level of education had significant effects on working and verbal memory. No significant differences were found in other cognitive processes (orientation, comprehension, and some executive functions). They concluded that culture dictates what is important for survival and that education could be considered as a type of subculture that facilitates the development of certain skills over others. They emphasized that culture and education affects cognitive skills, so that accurate assessment of cognitive dysfunction is dependent on both education and cultural skills.

In a thought-provoking article, Ardila has analysed the origins of reading and writing in human history; he points out that the origins of writing can be traced to early cave paintings. Writing (or pre-writing) was initially a visuoconstructive ability, later involving some stereotyped movements to represent pictograms, and finally involving spoken language. Writing has had a long evolution since the cave painting of Palaeolithic times. Different strategies have been used to visually represent spoken language (ideograms, alphabets, etc.). Writing, however, has continued to evolve since its invention. Evolution has continued with the development of different technical instruments for writing: the feather, the pencil, the typewriter, and the computer. Brain representation of written language has necessarily changed in some ways, too. Ardila suggests that in the future new neuropsychological syndromes resulting from new living conditions will be described.

In human history, writing only dates back some 5000 to 6000 years, and just a few centuries ago, reading and writing abilities were uncommon among the general population. The acquisition of reading and writing skills has changed the brain organization of cognitive activity in general, as well as specific abilities. The papers reported in this issue provide pioneering hard evidence on how culture changes the brain and how the environment can influence brain development; however, many questions remain to be answered

4

OSTROSKY-SOLÍS

regarding idiosyncratic adaptations to diverse cultural influences, the identification of variables that promote cognitive reserve, and the future evolution of the human brain.

REFERENCES Albert, M.S., Jones, K., Savage, C.R., Berkman, L., Seeman, T., Blazer, D., & Rowe, J.W. (1995). Predictors of cognitive change in older persons: MacArthur studies of successful aging. Psychology and Aging, 10, 578–589.

Ardila, A., Ostrosky-Solís, F., Rosselli, M., & Gómez, C. (2000). Age related cognitive decline during normal aging: The complex effect of education. Archives of Clinical Neuropsychology, 15, 495–514.

Christensen, H., & Henderson, A.S. (1991). Is age kinder to the initially more able? A study of eminent scientists and academics. Psychological Medicine, 21, 935–946.

Katzman, R. (1993). Education and the prevalence of dementia and Alzheimer’s disease. Neurology, 43, 13–20.

Mortimer, J.A. (1988). Do psychosocial risk factors contribute to Alzheimer’s disease? In A.S. Hendersen & J.H.Hendersen (Eds.), Etiology of dementia of Alzheimer’s type (pp. 39–52). Chichester, UK: John Wiley.

Orell, M., &Sahakian, B. (1995). Education and dementia. Research evidence supports the concept “use it or lose it”.

British Medical Journal, 310, 951−952.

Ostrosky-Solís, F. (2002). Education effects on cognitive function: Cognitive reserve, compensation or testing bias.

Journal of the International Neuropsychological Association, 8, 290–291.

Ostrosky-Solís, F., Ardila, A., Rosselli, M., López, G., & Mendoza, V. (1998). Neuropsychological test performance in illiterates. Archives of Clinical Neuropsychology, 13, 645–660.

Satz, P. (1993). Brain reserve capacity on symptom onset after brain injury: A formulation and review of evidence for threshold theory. Neuropsychology, 7, 273–295.

Scarmeas, N., Levy, G., Tang, M.X., Manly, J., & Stern, Y. (2001). Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology, 57, 2236−2242.

Stern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve concept. Journal of the International Neuropsychological Society, 8, 448−460.

Stern, Y., Gurland, B., Tatemichi, T.K., Tang, M.X., Wilder, D., & Mayeux, R. (1994). Influence of education and occupation in the incidence of Alzheimer’s desease. Journal of the American Medical Association, 271, 1004–1010.

Vygotsky, L.S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Zhang, M., Katzman, R., Salmon, D., Jin, H., Cai, G., Wang, Z., Qu, G., Grant, I Yu, E., Levy, P., Klauber, M.R., & Liu, W.T. (1990). The prevalence of dementia in Alzheimer’s disease in Shanghai, China: Impact of age, gender and education. Annals of Neurology, 27, 428−437.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 5–17

Targeting regions of interest for the study of the illiterate brain Alexandre Castro-Caldas Centro de Estudos Egas Moniz, Lisboa, Portugal

This paper reviews a work project that uses illiteracy as a tool to understand the way the brain adapts to information. The project follows the exploration of certain targets that can be identified with the functions of reading and writing, both from the functional and from the anatomical points of view. Results concerning visual processing, cross-modal operations (audiovisual and visuotactile), and interhemispheric crossing of information are reported.

Studies with magnetoencephalography, with positron emission tomography, and with functional magnetic resonance provided evidence that the absence of school attendance at the usual age constitutes a handicap for the development of certain biological processes that serve behavioural functioning. Differences between groups of literate and illiterate subjects were found in several areas: while dealing with phonology a complex pattern of brain activation was only present in literate subjects; the corpus callosum in the segment where the parietal lobe fibres cross was thinner in the illiterate group; the parietal lobe processing of both hemispheres was different between groups; and the occipital lobe processed information more slowly in cases that learned to read as adults compared to those that learned at the usual age. Some behavioural studies suggest that there are other operations that can be explored from the image point of view.

Cet article fait la revue d’un projet d’étude qui utilise l’analphabétisme comme outil pour comprendre la façon dont le cerveau s’adapte à l’information. Le projet poursuit l’exploration de certaines cibles qui peuvent être identifiées avec les fonctions de lecture et d’écriture, à la fois du point de vue fonctionnel et anatomique. Les résultats concernant le traitement visuel, les operations trans-modales (audiovisuelles et visuotactiles) et le passage inter-hémisphérique de l’information sont rapportés. Les études utilisant la magnéto-encéphalographie, la tomographie avec emission de positrons et la resonance magnétique fonctionnelle ont démontré que l’absence de scolarisation à l’âge adéquat constitue un handicap pour le développement de certains processus biologiques servant au fonctionnement comportemental. Des differences entre les groupes de participants alphabètes et analphabètes furent soulevées: lorsque les participants font des tâches de phonologie, un patron complexe d’activation cérébrale est present seulement chez les alphabètes; chez le groupe analphabète, le corps calleux est plus mince dans le segment où traversent les fibres du lobe pariétal; le traitement du lobe pariétal des deux hémisphères est different entre les groupes; et le lobe occipital traite l’information plus lentement dans les cas de l’apprentissage de la lecture a l’âge adulte comparativement a

6

CASTRO-CALDAS

l’apprentissage de la lecture a l’âge adéquat. Des études comportementales suggèrent que d’autres operations pourraient être explorées du point de vue de l’image.

Este estudio revisa una línea de investigación que ha utilizado al analfabetismo como una herramienta para entender la forma en la que el cerebro se adapta a la información. Se exploran regiones cerebrales que se han asociado con la lectura y la escritura. Se relacionan cambios anatómicos y funcionales asociados al procesamiento visual, a operaciones intermodales (auditivo-visual y viso-táctil), así como a operaciones que demandan intercruce interhemisférico de la información. El resultado de estudios con diversas técnicas de neuroimagen. las que incluyen el magnetoencefalograma, la tomografía por emisión de positrones y la resonancia magnética han aportando datos sobre el efecto de la falta de asistencia a la escuela durante la niñez en el desarrollo de ciertos procesos biológicos que subyacen a tareas conductuales. Las diferencias entre grupos de sujetos alfabetos y analfabetos se encontraron en diversas areas: la población alfabetizada mostró patrones complejos en la activación cerebral al realizar tareas fonológicas. Se detectó que el cuerpo calloso es mal delgado en el grupo de analfabetos en el segmento donde cruzan las fibras del lóbulo parietal; el procesamiento del lóbulo parietal de ambos hemisferios es diferente entre grupos; y el lóbulo occipital procesa lentamente la información en sujetos que aprendieron a leer de adultos comparados con aquellos que lo hicieron a la edad apropiada. Estudios conductuales sugieren que existen otras operaciones que pueden explorarse mediante estudios de neuroimagen.

There is now a robust body of evidence documenting differences between literate and illiterate subjects in performing a variety of tasks (Ardila, Ardila, Bryden, Ostrosky, Rosselli, & Steenhuis, 1989a; Ardila, Rosselli, & Ostrosky, 1992; Ardila, Rosselli, & Puente, 1994; Bornstein & Suga, 1988; Coppens, Parente & Lecours, 1998; Finlayson, Johnson, & Reitan, 1977; Heaton, Grant, & Mathews, 1986; Leckliter & Matarazzo, 1989; Lecours, Mehler, & Parente, 1987; Matute, 1986; Matute, Leal, Zarabozo, Robles, & Cedillo, 2000; Ostrosky, Ardila, Rosselli, Lopez-Arango, & Uriel-Mendoza, 1998; Ostrosky, Canseco, Quintanar, Navarro, & Ardila, 1985; Ostrosky, Quintanar, Canseco, Meneses, Navarro, & Ardila, 1986; Reis & Castro-Caldas, 1997a; Rosseli, 1993; Rosseli, Ardila, & Rosas, 1990). These behavioural differences between groups reflected certain biological differences that have also been documented (CastroCaldas et al., 1999; CastroCaldas, Peterson, Reis, Stone-Elander, & Ingvar, 1998b; Petersson, Reis, Askelöf, Castro-Caldas, & Ingvar, 2000). The first concern of those researching in this field should be the selection of the populations to be compared. Indeed, illiteracy can be the result of learning difficulties like developmental dyslexia; the result of lack of practice in adulthood after successful learning at the proper age in school; or the reflection of social problems in poorly developed regions. This last is the one that interests us in the project that we are pursuing for the following reasons: illiteracy is very common in the world (more than half of the world population is illiterate); the brains of these subjects have developed until adult life without the experience of connecting the visual and motor system to oral language; and reading and writing, although learned more easily in childhood, can also be learned later in life.

Correspondence should be sent to Professor A.Castro-Caldas, Centro de Estudos Egas Moniz, 1649–028 Lisboa, Portugal (E-mail: ccneurol@esoterica.pt). © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000240

THE STUDY OF THE ILLITERATE BRAIN

7

As Ostrosky et al. (1998) pointed out, the educational effect on neuropsychological test performance is not linear. Differences between 0 and 3 years of education are usually highly significant; differences between 3 and 6 years of education can be lower; between 6 and 9 are even lower; and so forth. Education effect represents a kind of negatively accelerated curve, tending to a plateau. Even if the reason proposed by the authors for this type of curve is true—i.e., the ceiling effect in neuropsychological tests is low—it is also true that a qualitative change occurs in the brain following the simple fact of learning to connect graphemes to phonemes. This is a mental operation that gives a new tool to the brain: A tool that helps to understand new words, helps to organize information, and gives access to less concrete information stimulating abstract thinking, as Vygotsky (1978) pointed out a few years ago. It is important to note that this knowledge does not change basic mental functions necessary for everyday life, as is sometimes argued. Cornelius and Caspi (1987) found that educational level had a substantial relationship with performance on some tests but was not systematically related to everyday problem solving. However, it certainly influences the way in which people deal with the information necessary for modern life, such as a medical prescription, instructions for use of machines, and safety information, just to mention a few. On the other hand it increases the emergence of idiosyncratic strategies in some talented subjects. Figure 1 is an example of a method for representing money and prices created by one of our subjects who had a small business selling vegetables. This system is similar to an abacus of the Eastern cultures. As we will see later, calculation by illiterate subjects is based on strategies that are different from those learned at school.

We can describe two types of effects that can be related to the exposure of the stimulation that is involved in the complex process of schooling: a diffuse effect and a focal effect. The diffuse effect can be defined as being related to the adaptation to a rich environment, such as that present in school, and which introduces several changes in brain function like the increasing of abstract thinking and the development of parallel processing of information. The focal effect is related to the learning of specific skills and operations that constitute the mastery of reading and writing and that may change particular areas of the brain involved in these operations (Castro-Caldas, 2002). We have some animal evidence for both the diffuse and the focal effects. When the brains of animals raised in rich environments were compared to the brains of those that were raised in poor environments, a thinner cerebral cortex and fewer connections in the brains of the animals that were less stimulated became evident. If the experiment involved particular functions like vision or audition, the biological effect could be found in the brain areas that usually convey that type of information, respectively the occipital lobe and the temporal lobe (for revision see Kolb & Whishaw, 1998).

There is also human evidence of similar effects. Jacobs, Schall, and Scheibel (1993) studied the brains of 20 neurologically normal right-handers; variables included gender, side of the brain, and education. They evaluated the total dendritic length, mean dendritic length, and dendritic segment count in Wernicke’s area.

A distinction was made between proximal (1st, 2nd, and 3rd order) and ontogenetically developing later distal (4th order and above) dendritic branches. Besides other interesting results related to individual backgrounds, education had a consistent and substantial effect such that dendritic measures increased along with educational level. As could be expected, distal dendritic branches appeared to exhibit greater epigenetic flexibility than proximal dendrites.

In our programme we are much more interested in the so-called focal effects. Therefore we designed a prospective project looking for targets to be explored from both the behavioural and the biological points of view, related to their knowledge of orthography.

From the evolutionist perspective, reading and writing are very recent acquisitions of human behaviour.

They can be considered to be the result of‘the evolution of previous independent skills that, at a certain moment, connected with oral language. Reading is the result of the capacity to identify visual patterns for which we can give a particular meaning. It is probably the result of the capacity to look for food; identifying

8

CASTRO-CALDAS

Figure 1. An example of a system created by an illiterate subject to represent prices and make simple calculation.

the good foodstuffs or looking for signs left by animals and interpreting these. These basic skills evolved progressively for symbolic and abstract thinking. An important component, therefore, is to explore visual analysis and symbolic and abstract thinking. This relates to occipital cortex and the visual system and to the whole brain. Abstract thinking is indeed a cultural skill that is difficult to localize in the brain and that most probably relates to diffuse integrative networks. It may correspond to the concept of brain reserve proposed

by Stern (Stern, Tang, Denaro, & Mayeux, 1995) as an explanation for the frequently reported fact of a higher incidence of dementia in populations with a low cultural level. This brain reserve delays the emergence of signs of dementia in subjects with higher levels of intellectual achievement. This is still a hypothesis because the finding is not consistent in all series, but it is a concept that can be considered in our model (Castro-Caldas & Guerreiro, 2001).

Writing skills evolved from motor acts. Motor acts became gestures with symbolic meaning and started to be engraved as drawings. The drawings had meanings and were recognized and copied by elements of the society. They became visual clues, as mentioned earlier, and were connected to the visual system. At this point we have to consider a cross-modal operation between the proprioceptive system and the visual system. This occurs in the parietal lobe, where information coming from the dorsal route of the visual system converges with the tactile information framed by the motor system (Cohan & Anderson, 2002). Writing, which was previously logographic, at a certain point became alphabetic. This is the point when a conscious sublexical analysis of oral language became necessary to perform the operation of matching graphemes and phonemes.

Phonological processing is a procedural mechanism and therefore should be supported by structures involved in automatic motor processing. Nevertheless this is also a cross-modal operation between the visual and the auditory systems conveyed through the ventral route, which links occipital and temporal cortices.

Finally, such complex processing concerns both cerebral hemispheres, and the functional distribution of activities can be explored. Indeed, the connection to the language system shifts the information towards the

THE STUDY OF THE ILLITERATE BRAIN

9

left side of the brain while the spatial distribution and the visuoconstructive component shifts it to the right hemisphere.

In the following sections we will review some of our findings related to these topics, taking into consideration the functional and the biologic aspects.

PROBLEMS OF CEREBRAL DOMINANCE Fifty years ago, Critchley (1956) suggested that although the clinical experience did not support the belief that aphasia was rare in illiterate persons, aphasia patients with a premorbid superior mastery of language would be disturbed more severely and for longer. Thus it was suggested that the exercise of functions related to oral language, like reading and writing, would shift the information to the left hemisphere, which resulted in a stronger left hemisphere dominance for language. Nevertheless, after Broca, aphasia was reported and studied in many subjects, independent of their degree of schooling. The authors who contributed to this field for more than a century never raised this issue, although it is likely that hundreds of cases studied by these pioneers were illiterate because literacy was rare in the 19th century (Castro-Caldas, 2002).

In recent work, Fonseca (Fonseca & Castro-Caldas, 2002) from our laboratory compared the evolution of the Aphasia Quotient (AQ) (obtained by combining the scores of four measures: fluency, naming, word repetition and phrase comprehension) of a group of 24 illiterate aphasics with that of 42 literate controls matched for type of aphasia, type of stroke, sex, and age. They were evaluated in the acute phase and again 6 months later. The difference found in the AQ between groups in the first evaluation was the same at the end of the study, and was due to the lower values scored by illiterate patients in the subtests of the AQ. The slope of the curve was similar for both groups (see Figure 2). These findings contradict the idea of less severe form of aphasia in illiterate subjects, suggested by Critchley (1956). However, there were qualitative differences, which are still being studied, that suggested a more modular organization of the different dimensions of language in the illiterate group.

In 1971, Cameron, Currier, and Haerer questioned the proneness of illiterate subjects to became aphasic following left-hemisphere lesions. They studied subjects with brain lesions from three educational groups: a more educated group (10.5 years of schooling), an intermediate group (6.5 years of schooling), and a poor group (2.5 years of schooling). They considered the association of right-sided weakness and language disturbance of any type and found that the association was present in 78% of subjects of the first group, 64% of the second, and only 36% of the third. These results were not reproduced in our series (Damรกsio, CastroCaldas, Grosso, & Ferro, 1976a; Damรกsio, Hamsher, Castro-Caldas, Ferro, & Grosso, 1976b). Among 182 subjects with cerebral lesions who were literate, 157 had lesions in the left hemisphere and 115 (63%) were aphasic. Among the 43 illiterate subjects, 29 (67%) of the 34 that had lesions on the left hemisphere were aphasic. In this series there were no cases of aphasia with lesions on the right hemisphere.

Later on we had the opportunity of studying cases of crossed aphasia and we never found an illiterate subject with this syndrome (Castro-Caldas & Confraria, 1984; Castro-Caldas, Poppe, & Confraria, 1987).

However, A.F.Wechsler (1976) reported a case of crossed aphasia, suggesting that illiteracy was an important cause for the unusual hemispheric representation of functions in this patient. The study of Lecours et al. (1987; Lecours, Mehler, & Parente, 1988), in which normal subjects and left- and righthemisphere lesioned patients, literate and illiterate, were studied, also suggested that the right hemisphere of illiterate subjects played a more important role in language processing of illiterate subjects.

Studies performed in subjects without brain lesions also contributed to the discussion of this question.

Studies with dichotic listening were inconclusive. The first was performed with Portuguese subjects, in our

10

CASTRO-CALDAS

Figure 2. Evolution of the Aphasia Quotient of literate and illiterate aphasics for 6 months.

laboratory. Three different tests were used: one with pairs of different words, another with pairs of digits, and a third with pairs of words differing in the initial phoneme. The results showed similar performances of literate and illiterate subjects in the first two tests and a reversal of the channel effect in the last test in the illiterate group. When the words of the pair were phonetically similar, illiterate subjects preferred the words presented to the left channel while literate ones preferred those presented to the right (Damรกsio, Damรกsio, Castro-Caldas, & Hamsher, 1979). This study suggested a different way of processing formal similitude by each group. A somewhat concordant study was done with Greek illiterate subjects (Tzavaras, Kaprinis, & Gatzoyas, 1981). These authors studied literate and illiterate volunteers on dichotic listening tasks in two different situations. In the first one the subjects were asked to repeat what they had heard without any constraint; in the second a forced choice of channel was required. The authors computed a score of asymmetry and concluded that asymmetry was more evident in the more educated group. Using more carefully designed paradigms, Castro and Morais (1987) were unable to demonstrate differences between literate and illiterate subjects in dichotic listening tasks.

In a positron emission tomography (PET) study performed on volunteers that showed differences, which has been discussed elsewhere (Castro-Caldas et al., 1998b), we were also able to demonstrate that the balance of the activation of homologous areas of each cerebral hemisphere, within the parietal lobe, was different in literate compared to illiterate subjects while they were repeating both words and pseudowords.

The superior parts of the parietal lobe were more active on the left than on the right hemisphere in illiterates compared to literates, while the reverse was the case for the inferior parts of the parietal cortex and the precuneus (Castro-Caldas, Peterson, Reis, Askelof, & Ingvar, 1998a). This was interpreted as the result of a different arrangement of the neural network that supports that particular function of language, depending on whether you are literate or illiterate. The parietal lobe plays a particular role in these functions.

This finding was corroborated by the fact that the region of the corpus callosum where the interparietal fibres cross is thinner in illiterate subjects compared to the same region in literate controls (Castro-Caldas et al., 1999). So far, it is possible to conclude that the knowledge of orthography, acquired at the correct age in childhood, may change the pattern of involvement of each hemisphere in the processing oral language. This involvement does not determine the severity and presence of aphasia but may have certain qualitative aspects that deserve further investigation.

The connection between the cerebral hemisphere through the corpus callosum still deserves more attention. We examined visually guided hand motor behaviour in a test paradigm in which literate and illiterate subjects were asked to manipulate a computer mouse. They had to direct the cursor towards a target

THE STUDY OF THE ILLITERATE BRAIN

11

randomly presented on the screen of a portable computer. All participants were naïve on the use of computers. Time to reach the target was measured and the test was carried out with both hands. Results showed differences in the crossed situation of reaching the left side of the screen with the mouse in the right hand; literate subjects performed better than illiterate subjects (Reis & Castro-Caldas, 1997b). This bias is probably the result of writing being trained with the right hand starting on the left side of the paper.

Very recent results obtained in subjects who learned to read and to write as adults show that there is another handicap related to the transfer of information from one hemisphere to the other. We used a test paradigm in which subjects were asked to write words with the right hand, and the same words in mirror writing with the left hand (Fernandes, Simoes, Nunes, Gonçalves, & Castro-Caldas, 2003). The main result was that mirror writing was more difficult for ex-illiterates than for those who learned at the correct age, suggesting that illiterates were not using a strategy of crossing the motor programme through the corpus callosum but rather a strategy involving the reversal of visual images.

A group of subjects was recently studied by means of magnetoencephalography (MEG) at the Universidade Compultense in Madrid (Castro-Caldas et al., 2003). We compared recently literate women with controls who went to school at the appropriate time. They were submitted to two different tasks. In the first they had to attend to a sequence of words presented through loudspeakers at comfortable sound level.

The second was similarly designed, except that the words were presented in written form on a computer screen. Preliminary results showed that in both tasks there were more fonts of activity in right temporoparietal regions of recently literate subjects than of those who learned to read in childhood. This suggested a more important contribution of the right hemisphere for dealing with oral and written language in recently literate subjects. This also represents the handicap of missing school at a young age and has to be understood when teaching adults, because it reflects different strategies for learning that need to be optimised.

PROBLEMS OF VISUAL PROCESSING As pointed out by Borod, Goodglass, and Kaplan (1980) and also by Kremin et al. (1991), the performance of non-brain-damaged subjects on visual naming tasks is influenced by their level of education. In the first case, drawings of the Boston naming test were used and in the second the Snodgrass and Vanderwart figures (1980) were shown to the subjects. Naming of drawings by illiterates was also poor in the study by Lecours et al. (1987). Those interested in finding norms to study illiterate subjects have discussed this aspect extensively (Ardila et al., 1989a; Rosseli et al, 1990). We also find important differences while standardizing the naming test of the Multilingual Aphasia Examination Battery (Benton, Hamsher, & Siven, 1994).

As is shown in Table 1, the distribution of illiterate and literate subjects on the score range was very different. There were almost no errors in the literate group and there were very few illiterates who reached the maximum score of 84 Several authors have studied this difficulty: Parente (1984) studied the perception and drawing of a 3D cube; Kolinski, Morais, Content, and TABLE 1 Distribution of the individual scores of literates and illiterates in a test of naming drawings Score range

0−30

31–50

51–70

71–80

81–84

Literate (%) Illiterate (%)

0 18

0 21

0 46

32 11

68 4

12

CASTRO-CALDAS

Cary (1987) concluded that the understanding of the drawings was dependent from learning at school and did not emerge naturally; Matute et al. (2000) studied the perception and construction of models made by sticks; and we compared the capacity for naming drawings, photographs, and real objects. In this study we showed that literates and illiterates were similar on naming real objects but that illiterates were much worse at naming photographs and drawings (Reis, Guerreiro, & Castro-Caldas, 1994; Reiss, Petersson, CastroCaldas, & Ingvar, 2001). It is still difficult to understand exactly which level of processing is influenced by schooling. Some observations on individual cases suggest that illiterates have difficulty integrating the details of the drawing in a meaningful whole, which would make them similar to the cases of integrative agnosia (Humphreys & Riddoch, 1998). These authors reported a patient who “copies in a slow and slavish manner because his perceptual representations of the stimuli are not correctly organized and integrated” (p.

69). Figure 3 represents the drawing by copy of a woman who did not attend school in childhood and was attending classes as an adult; her behaviour while performing the task was similar to the description of that of the patient (Gonçalves, 2003). This is probably not exactly related to visual perception but to abstract thinking and integrating information from both hemispheres. Therefore it may be considered a diffuse effect of schooling.

Going back to the study mentioned above, using MEG (Castro-Caldas, 2002), it is important to note that while reading words, the activation of occipital regions in recently literate subjects took much longer than in subjects who learned to read in childhood. Time of information processing was different between groups in some of the involved areas and the route followed in time by the brain activity was also different.

We can conclude that the process of reading following school attendance has a significant training effect on the complex process of seeing. There are certainly many reasons behind the simple fact of learning orthography. These are the result of exposure to the school system, where we also learn to draw and to copy.

PROBLEMS OF AUDIOVISUAL CROSS-MODAL PROCESSING Developmental studies have shown that learning to read and write requires the awareness of the phonological contents of the words. In order to perform the operation of matching graphemes and phonemes one needs to be able to perform sublexical segmentation at the phonological level. This ability develops in children while they are learning at school (Gathercole, Willis, Baddeley, & Emslie, 1994). For those who remain illiterate there is no awareness of the phonological structure, as was demonstrated by Morais (Morais, Cary, Alegria, & Bertelson, 1979).

We addressed this problem by asking illiterate subjects to repeat pseudowords (Reis & Castro-Caldas, 1997a). Without the awareness of the phonological constitution of the words, illiterates use an analogical approach supported by semantics, and have great difficulty performing this task accurately. They do learn new words but it takes longer and requires several attempts. We were asking them to repeat pseudowords for the first time. Compared to their controls (selected from the same cultural environment but having attended school in childhood), they repeated pseudowords much worse but repeated real words as well as controls.

This corresponds to a cross-modal operation in which a sequence of phones is analysed, paced by visual imagery of letter forms. On the other hand, articulatory rehearsal is part of phonological working memory (Baddeley, 2000), and therefore this system has to be more active in a successful operation.

With this orientation in mind we performed a PET study in which six illiterate women were compared to six literate ones, from the same cultural environment, while repeating series of words and series of pseudowords. As expected, illiterate women were worse than literates on pseudoword (Castro-Caldas et al.,

THE STUDY OF THE ILLITERATE BRAIN

13

Figure 3. Example of the copy of one of Bender’s models by an illiterate woman. Numbers indicate the drawing sequence following the direction of the arrows.

1998b) repetition but similar on real word repetition. The pattern of brain activation was, however, different in the two conditions (see Figure 4).

While repeating real words, literate women activated the left posterior parietal cortex (BA 40) more, and while repeating pseudowords they activated a complex pattern of cortical and subcortical structures more.

The areas involved were bilateral anterior insular (BA 14 and 15) and right frontal opercular cortices (BA 44, 45, 47, and 49), left perigenual anterior cingulated gyrus (BA 24 and 32), left basal ganglia (putamen, globus pallidus, and head of the caudate nucleus), anterior thalamus/hypothalamus, and midline cerebellum.

According to Binder and Price (2001), there appear to be two distinct regions of the inferior frontal cortex involved in phonemic processing: the left frontal operculum (around the anterior insula and BA 45) and a more posterior region (BA 44/6). When reading words and pseudowords the supramarginal gyrus was activated in some studies but not in all. These authors call attention to the characteristics of the experimental paradigm, which are very important to the areas involved in each operation. Other functions that require brain activation may be involved simultaneously and may mask the activation related to phonological processing. There is indeed a certain variability in the results according to the type of task required, and there are no other studies with a paradigm similar to ours.

These differences were later confirmed by performing a network analysis on the results. There were no significant differences between the literate and the illiterate group during real word repetition. In contrast, the network interactions differed between the literate and the illiterate group during pseudoword repetition.

In addition to differences similar to those observed in the illiterate group between word and pseudoword repetition, there were differences related to the interactions of the phonological loop between the groups. In particular, the differences related to the interaction between Broca’s area and the inferior parietal cortex as well as the posterior mid-insula bridge between Wernicke’s and Broca’s area (Petersson et al., 2000).

14

CASTRO-CALDAS

Figure 4. Patterns of activation of literate-illiterate while repeating real words (A) and pseudowords (B). (Adapted from Castro-Caldas et al., 1998a.)

It is well accepted that the phonogical loop plays an important role in the process of learning new words and learning to read. After learning, the loop plays an important role in the process of recoding written material (Baddeley, 2000). Therefore, it is important to understand if the loop shows differences in literate and illiterate subjects. The paradigm that Nunes (2002) selected to study this topic was a test of word span performed with and without articulatory suppression. Articulatory suppression is a procedure whereby the subject is required to utter some repeated redundant sound while performing another task such as memory span. Murray (1968) showed that suppression reduces performance in reading tasks but not in the auditory presentation, which relates this task to the reading activity that is naturally absent in illiterate subjects.

Results showed that illiterates had a lower word span than literates but that they were equally affected by articulatory suppression. The difference was on the type of errors made by the two groups of subjects.

While suppressing articulatory rehearsal, literate subjects significantly changed their pattern of errors: they made the same type of errors as illiterate subjects. Illiterates did not change qualitatively between one situation and the other.

These results suggest that some basic operations at the phonological level interfere in different ways with the normal processing of oral language in illiterates compared to literates. A good example of this is the impact that the word form has in several operations. The first example was demonstrated in a word pair association test. Although it is known that this task depends principally on semantic coding (Papagano & Vallar, 1992), we developed a variant of the word-pair association test of the Wechsler Memory Scale (D. Wechsler, 1945). Two sequences of 10 pairs of words were presented to 30 women (20 illiterate and 10 literate from the same cultural environment). In each sequence five words were semantically related and the other five were phonologically related but semantically unrelated. Results showed a clear difference of performance between cultural groups. Literate subjects recalled words associated by semantics and by phonology similarly. Illiterate subjects were much worse in recalling words associated phonologically than those related by semantics (Reis & Castro-Caldas, 1997a). Although these are interesting results, it was difficult to develop a hypothesis that allowed us to look for the difference in a specific region in the brain.

We could think that phonological analogy would be more related to Broca’s area and that semantic analogy

THE STUDY OF THE ILLITERATE BRAIN

15

Figure 5. Performance of literate and illiterate in tasks of verbal fluency. Mean values are represented in the curves (Reis & Castro-Caldas, 1997a).

would be more related to the parietal lobe (Binder & Price, 2001); however, the more important process going on in the brain while performing this task was related to memory, and therefore the activation should show the areas involved in memory processing. This same paradigm was adapted for a PET study and what we found in a group of literate and illiterate women was indeed the activation of left prefrontal and medial temporal lobes during the process of encoding (Petersson, Reis, Castro-Caldas, & Ingvar, 1999). There was no group effect and no test effect, suggesting that the influence of the type of material was irrelevant at the level of processing that was detected.

The impact of illiteracy on cognition at the phonological level can still be found in tasks of verbal fluency.

Several authors found this (Ardila et al., 1992; Ostrosky et al., 1985; Rosseli et al, 1990), both when semantic categories were proposed and when initial phonemes were used. Using semantic cueing, there seem to have been differences in performance according to the topic proposed and the strategy used, but these need to be investigated further. Phonemic cueing has no meaning for illiterate subjects and reflects their lack of awareness for phonology. Figure 5 shows the results obtained from 23 illiterate women and 16 literate ones on tasks of verbal fluency cued by semantic and phonological criteria. As can be seen, the performance is similar for both criteria in the literate group and there is a marked difference in the illiterate group.

Verbal fluency is a good target for studying illiterate subjects from the behavioural point of view as it may reflect some epigenetic components of semantic knowledge; however, it will probably be difficult to find consistent differences in image studies.

16

CASTRO-CALDAS

PROBLEMS OF TACTILE-VISUAL CROSS-MODAL PROCESSING Studies on acquired alexia since Dejerine (1892) have demonstrated the importance of parietal regions in reading and writing skills. Interestingly enough, a recent study (Fukatsu, Fugi, & Yamadori, 1998) reported a case of a Japanese subject who had agraphesthesia without sensory defects in either hand and without visual alexia. This case suggests an independence of the somesthetic images of graphic symbols from their visual counterparts.

Animal studies have showed that the posterior parietal cortex is responsible for integrating information coming from different sensory sources in a similar reference frame. This is important to inform the motor system to perform propositional movements (for a general reference see Cohen & Anderson, 2002). Writing is a skilled propositional movement that requires sophisticated information from vision and somesthetic sources.

As was mentioned above, in our first PET study there was a difference in the pattern of activation between literates and illiterates while repeating real words (see Figure 4). Therefore we raised the possibility that the activation of that area could correspond to a particular development of processes involved in visuotactile cross-modal operations (Simões, Fernandes, Nunes, Gonçalves, & Castro-Caldas, 2003). In this study, three groups of subjects were studied: a group of women who had learned to read in adult life; a group of women who had learned at the usual age but, because they had no training during life, were very poor writers; and a group of women who had learned in childhood and who had current habits of reading and writing. All subjects performed similarly in tests for stereognosia and two-point discrimination.

There were marginal differences in graphesthesia and significant differences in tactile reading of small words written with three-dimensional block letters. This shows that the training of writing generates a new way of having access to graphemes. Subjects that had no writing practice could not identify the tactile features that allow the formation of a visual image of the letters, although they could identify those that are required for object identification. The complexity of this cross-modal operation may explain why it is more difficult for adults (and even for children) to learn to write.

PROBLEMS RELATED TO DEALING WITH NUMBERS AND MENTAL CALCULATION In school we also learn digits, numbers, and calculation. This is also the acquisition of a skill that has fixed rules and that cannot be learned spontaneously. Calculation as a mental operation, however, emerges spontaneously as a tool to solve problems of daily life. Animals can assess quantities and very young babies can perform some forms of mental calculation.

When we start learning the graphic elements of digits and arithmetical signs, we are introducing a symbolic representation in a semantic complex of mental activity. Therefore, as suggested by Grafman and Boller (1987), we can deal with digits and numbers in a semantic way, i.e., when we count, or measure quantities, or we can deal with them in an asemantic way, when we perform abstract calculation. This second aspect is learned at school, therefore those who missed school do not use an asemantic process for their computations. As was shown in Figure 1 (see also Figure 1 in Castro-Caldas, Reis, & Guerreiro, 1997), the magnitude component is expressed by illiterates by the number of graphic elements. For a literate subject, the word «five» is preferentially processed as «5», whereas for an illiterate subject it is processed as «/////». Therefore this is another target for investigation.

Previous publications suggest that performance in digit span tasks is worse in illiterate compared to literate subjects. The hypothesis of an influence of the magnitude component in these results was tested through the comparison of the performance in repeating series composed of digits larger than five and series

THE STUDY OF THE ILLITERATE BRAIN

17

composed of digits smaller than five for both illiterate and literate subjects. It was found that the digit type (>5 or <5) was important for the performance of illiterate subjects. The magnitude component plus a poorly trained phonological memory may account for the bad performance of illiterate subjects in repeating series of digits. This hypothesis was tested in clinical populations with transcortical sensorial and motor aphasia and Alzheimer dementia, which revealed important performance dissociations between illiterate and literate subjects. The lack of access to the magnitude component of the digit (semantic) was reflected in the performance of the illiterate subjects with transcortical sensorial aphasia and Alzheimer dementia, whose defects concerned language comprehension and semantic processing, respectively (Castro-Caldas & Guerreiro, 2001; Reis, Guerreiro, & Castro-Caldas, 1996; Reis, Guerreiro, Garcia, & Castro-Caldas, 1995).

Evidence obtained from the observation of subjects with acquired cerebral lesions early in life is important to orient us in studying this topic. Mathematical achievement in adult life is very poor in subjects who sustained right-hemisphere lesions before the age of school attendance. This is not the case for those whose lesions were in the left hemisphere in the same period of life (Aram & Ekelman, 1988). The reverse is found when the cerebral lesion occurs later in life, after learning the visual images of the digits and the rules for calculation in school. In cases of brain injury acquired in later in life, acalculia is associated with lesions of the left cerebral hemisphere (Martins, Ferreira, & Borges, 1999a; Martins, Parreira, Albuquerque, & Ferro, 1999b) and lesions on the right side of the brain may disrupt some aspects of spatial arrangements of the written operations but not mental calculation itself. This suggests that the neural support for these processes changes. We have to understand if this is due to school attendance or to age.

The estimation of magnitude in the framework of semantics is one of the subcomponents of the EC301 test battery (Deloche, Souza, Braga, & Dellatolas, 1999a), designed to assess calculation shown to be independent of schooling (Deloche, Souza, Braga, & Dellatolas, 1999b). Therefore this was a good choice to study the hypothesis of having more than one strategy to solve problems that could be related to learning the rules and the symbols. Forty-nine subjects, divided into two groups, were studied in Brazil (Braga, Castro-Caldas, Deloche, Dellatolas, & Campos da Paz, 2003): one consisted of 19 illiterates and the other of 30 college graduates. Brain activation was studied by fMRI while they were solving problems of contextual magnitude (for instance: do 12 persons fit inside an automobile?; are 50 bricks enough to built a house?). The pattern of activation differed in the two groups. College graduates activated areas exclusively on the left hemisphere while illiterates activated both hemispheres. On Table 2 are represented the main areas that were more activated in each group after subtraction.

The qualitative analysis of the verbal responses TABLE 2 Contextual magnitude judgments: fMRI activation in each group Illiterate subjects Temporal Lobe Left BA 20 Right BA 22 Occipital Lobe Right BA 37 Left BA 9 Left BA 8

Literate subjects Right BA 20 Right BA 39 Left BA 21 Right BA 19 Frontal Lobe

Temporal Lobe Left BA 22 Left BA 42 Parietal Lobe Left BA 40

18

CASTRO-CALDAS

Illiterate subjects

Literate subjects

Left BA 6

given during post-fMRI interviews about strategies for solving contextual magnitude judgments also revealed findings that differentiated the two groups. In the verbal response analysis, 95% of the illiterates referred to the use of visual images in problem solving, for example: “It’s something that’s hard to explain, but I imagined a school full of classrooms and only 9 students, it looked like so little”; “I saw the bus that I take to go home, the seats, the people sitting, one, two, three seats and then some people standing and I figured that 45 would be a good number of people for that bus if some stayed standing”; “I can’t really explain it, but I can see 50 bricks by just thinking about it because I’m a construction worker, and just by looking at 50 bricks you know straight away that you can’t build no house.”

The college graduate group presented a very different pattern of responses, for example: “I thought abstractly, comparing quantity, volume and space for almost all the answers”; “I compared the proportion taking into account the space and the reality”; “I compared the quantity that was given with what would be the ideal for each situation presented”; “I evaluated according to the quantity and space; these were questions of immediate abstraction, you quickly come to a decision.”

The accounts from both groups showed that illiterates used visualization strategy, while college graduates used “abstraction” for solving the contextual magnitude judgments. Each of these strategies has its corresponding pattern of activation.

CONCLUSION This revision illustrates that it is possible to target behavioural processes that are dependent from school achievement and also regions of the brain that change in accordance with these processes. There is not always a behavioural process related to a particular pattern of brain activation. There are idiosyncratic adaptations that sometimes increase interindividual differences and there are processes done with less efficacy that use the same areas of the brain in both literate and illiterate subjects.

REFERENCES Aram, D.M., & Ekelman, B.L. (1988). Scholastic aptitude and achievement among children with unilateral brain lesions. Neuropsychologia, 26, 903–916.

Ardila, A., Ardila, O., Bryden, M.P., Ostrosky, F., Rosselli, M., & Steenhuis, R. (1989a). Effects of cultural background and education on handedness. Neuropsychologia, 27, 893–898.

Ardila, A., Rosselli, M., & Ostrosky, F. (1992). Sociocultural factors in neuropsychological assessment. In A.E.Puent & R.J.McCaffrey (Eds.), Handbook of neuropsychological assessment: A biopsychosocial perspective (pp. 181–192).

New York: Academic Press.

Ardila, A., Rosselli, M., & Puente, P. (1994). Neuropsychological evaluation of the Spanish speaker. New York: Plenum Press.

Ardila, A., Rosselli, M., & Rosas, P. (1989b). Neuropsychological assessment in illiterates: Visuospatial and memory abilities. Brain and Cognition, 11, 147–166.

Baddeley, A. (2000). Short-term and working memory. In E.Tulving & F.I.M.Craik (Eds.), The Oxford handbook of memory (pp. 77–92). Oxford: Oxford University Press.

Benton, A.L., Hamsher, K.S., & Siven, A.B. (1994). Multilingual Aphasia Examination. Iowa City, IA: AJA Associates.

THE STUDY OF THE ILLITERATE BRAIN

19

Binder, J., & Price, C. (2001). Functional neuroimaging of language. In R.Cabeza & A.Kingstone (Eds.), Handbook functional neuroimaging of cognition (pp. 187–251). Cambridge, MA: MIT Press.

Bornstein, R.A., & Suga, L.J. (1988). Educational level and neuropsychological performance in healthy elderly subjects. Developmental Neuropsychology, 4, 17–22.

Borod, J.C, Goodglass, H., & Kaplan, E. (1980). Normative data on the Boston Diagnostic Aphasia Examination and the Boston Naming Test. Journal of Clinical Neuropsychology, 2, 209–215.

Braga, L.W., Castro-Caldas, A., Deloche, G., Dellatolas, G., & Campos da Paz, A.A. (2003). Culture and education generate alternative pathways in the functional anatomy of the brain. Manuscript submitted for publication.

Cameron, R.F., Currier, R.D., & Haerer, A.F. (1971). Aphasia and literacy. British Journal of Disorders Communication, 6, 161−163.

Castro, S.L., & Morais, J. (1987). Ear differences in illiterates. Neuropsychologia, 25, 409−417.

Castro-Caldas, A. (1999). Book reviews (Patrick Coppens, Yvan Lebrun, & Anna Basso: Aphasia in atypical populations.). Brain, 722, 1200–1202.

Castro-Caldas, A. (2002). Um modelo para o estudo da influência da estimulação cultural na organização biofuncional do cérebro humano. Psicologia, 16, 13–25.

Castro-Caldas, A., & Confraria, A. (1984). Age and type of crossed aphasia in dextrals due to stroke. Brain and Language, 23, 126–133.

Castro-Caldas, A., & Guerreiro, M. (2001). Cultural background as a risk factor for dementia. In F.Boller & S.F.Cappa (Eds.), Handbook of neuropsychology (2nd ed.). Amsterdam: Elsevier Science.

Castro-Caldas, A., Miranda, P., Carmo, I., Reis, A.I., Leote, F., Ribeiro, C, & Ducla-Soares, E. (1999). Influence of learning to read and write on the morphology of the Corpus Callosum. European Journal of Neurology, 6, 23–28.

Castro-Caldas, A., Nunes, V.S., Simões, R., Fernandes, R., Maestú, F., Ortiz, T., Campo, P., Fernandez, A., & Amo, C. (2003). Mapeamemto do cortex receptivo da linguagem em adultos letrados na infância e em adultos letrados na vida adulta utilizando Magnetoencefalografia. To be presented at the SLAN Conference, Canada.

Castro-Caldas, A., Peterson, K.M., Reis, A., Askelof, S., & Ingvar, M. (1998a). Differences in inter-hemispheric interactions related to literacy, assessed by PET [PO1.104]. Neurology, 50, A43.

Castro-Caldas, A., Peterson, K.M., Reis, A., Stone-Elander, S., & Ingvar, M. (1998b). The illiterate brain: Learning to read and write during childhood influences the functional organization of the adult brain. Brain, 121, 1053–1063.

Castro-Caldas, A., Poppe, P., & Confraria, A. (1987). Non-verbal disturbances in crossed aphasia. Aphasiology, 1, 403–413.

Castro-Caldas, A., Reis, A., & Guerreiro, M. (1997). Neuropsychological aspects of illiteracy. Neuropsychological Rehabilitation, 7, 327–338.

Cohen, Y.E., & Anderson, R.A. (2002). A common reference frame for movement plans in the posterior parietal cortex.

Nature Reviews—Neuroscience, 3, 553–562.

Coppens, P., Parente, M.A., & Lecours, A.R. (1998). Aphasia in illiterate individuals. In P.Coopens, Y. Lebrun, & A.Basso (Eds.), Aphasia in atypical populations. Mahwah, NJ: Lawrence Erlbaum Associates Inc. Cornelius, S.W., & Caspi, A. (1987). Everyday problem solving in adulthood and old age. Psychology of Aging, 2, 144–153.

Critchley, M. (1956). Premorbid literacy, and the pattern of subsequent aphasia. Proceedings of the Royal Society of Medicine, 49, 335–336.

Damasio, A., Castro-Caldas, A., Grosso, J.T., & Ferro, J.M. (1976a). Brain specialization for language does not depend on literacy. Archives of Neurology, 33, 300–301.

Damasio, H., Damásio, A.R., Castro-Caldas, A., & Hamsher, K. (1979). Reversal of ear advantage for phonetically different words in illiterates. Journal of Clinical Neuropsychology, 1, 331–338.

Damasio, A.R., Hamsher, K.S., Castro-Caldas, A., Ferro, J.M., & Grosso, J.T. (1976b). Brain specialization for language: Not dependent on literacy. Archives of Neurology, 33, 662.

Déjérine, J. (1892). Contribution a l’étude anatomopathologique et clinique des differentes variétés de cécité verbale.

Memoires-Société Biologie, 4, 61–90.

20

CASTRO-CALDAS

Deloche, G., Souza, L., Braga, L.W., & Dellatolas, G. (1999a). Assessment of calculation and number processing by adults: Cognitive and neuropsychological issues. Perception Motor Skills, 89, 707−738.

Deloche, G., Souza, L., Braga, L.W., & Dellatolas, G. (1999b). A calculation and number processing battery for clinical application in illiterates and semi-illiterates. Cortex, 35, 503–521.

Fernandes, R., Simões, R., Nunes, V., Gonçalves, M., & Castro-Caldas, A. (2003). Interhemispheric connectivity and literacy. Manuscript in preparation.

Finlayson, N.A., Johnson, K.A., & Reitan, R.M. (1977). Relation of level of education to neuropsychological measures in brain damaged and non-brain damaged adults. Journal of Consulting and Clinical Psychology, 45, 536–542.

Fonseca, J., & Castro-Caldas, A. (2002). The knowledge of reading and writing rules influences aphasia outcome.

International Journal of Neuropsychology, 8, 506.

Fukatsu, R., Fugi, T., & Yamadori, A. (1998). Pure somesthetic alexia—somesthetic verbal disconnection for letters.

Brain, 121, 843−850.

Gathercole, S.E., Willis, C.S., Baddeley, A.D., & Emslie, H. (1994). The Children’s Test of Phonological Working Memory. Memory, 65, 103–127.

Gonçalves, M.A. (2003). Reconhecimento e nomeação visual de desenhos a contornos por analfabetos e letrados.

Masters thesis, Faculdade de Medicina de Lisboa.

Grafman, J., & Boller, F. (1987). Cross-cultural approaches to the study of calculation processes. In G. Deloche & X. Seron (Eds.), Mathematical disabilities: A cognitive neuropsychological perspective (pp. 257−271). Hove, UK: Lawrence Erlbaum Associates Ltd. Heaton, R.H., Grant, I., & Mathews, C. (1986). Differences in neuropsychological test performance associated with age, education and sex. In I. Grant & K.M. Adams (Eds.), Neuropsychological assessment in neuropsychiatric disorders (pp. 108–120). New York: Oxford University Press.

Humphreys, G.W., & Riddoch, M.J. (1998). To see but not to see: A case study of visual agnosia (6th ed.) Hove, UK: Psychology Press.

Jacobs, B., Schall, M., & Scheibel, A.B. (1993). A quantitative dendritic analysis of Wernicke’s area in humans. II. Gender, hemispheric, and environmental factors. Journal Comparative Neurology, 327, 97–111.

Kolb, B., & Whishaw, Q. (1998). Brain plasticity and behaviour. Annual Review of Psychology, 49, 43–64.

Kolinski, R., Morais, J., Content, A., & Cary, L. (1987). Finding parts within figures: A developmental study.

Perception, 16, 399–407.

Kremin, H., Deloche, G., Metz-Lutz, M., Hannequin, D., Dordai, M., Perrier, D., Cardebat, D., Ferrand, I Larroque, C, Naud, E., Pichard, B., & Bunel, G. (1991). Short report: The effects of age, educational background and sex on confrontation naming in normals. Principles for testing naming ability. Aphasiology, 5, 579–582.

Leckliter, I.N., & Matarazzo, J.D. (1989). Age, education, IQ, sex, alcohol abuse. Halstead-Reitan Test Battery scores.

Literature review. Journal of Clinical Psychology, 45, 484–512.

Lecours, R.L., Mehler, J., & Parente, M.A. (1987). Illiteracy and brain damage 1: Aphasia testing in culturally contrasted populations (control subjects). Neuropsychologia, 25, 231–245.

Lecours, A.R., Mehler, J., & Parente, M.A. (1988). Illiteracy and brain damage: 3. A contribution to the study of speech and language disorders in illiterates with unilateral brain damage (initial testing). Neuropsychologia, 26, 575–589.

Martins, I.P., Ferreira, J., & Borges, L. (1999a). Acquired procedural dyscalculia associated to a left parietal lesion in a child. Child Neuropsychology, 5, 265–273.

Martins, I.P., Parreira, E., Albuquerque, L., & Ferro, J.M. (1999b). Capacités de calcul chez des enfants scolarisés avec des lesions cérébrales acquises. Approche Neuropsychologique des Apprentissages chez L’Enfant, 11, 6–12.

Matute, E. (1986). Aphasia in illiterates. Journal of Neurolinguistics, 1, 115–130.

Matute, E., Leal, F., Zarabozo, D., Robles, A., & Cedillo, C. (2000). Does literacy have an effect on stick construction tasks? International Journal of Neuropsychology, 6, 668−672.

Morais, J., Cary, L., Alegria, J., & Bertelson, P. (1979). Does awareness of speech as a sequence of phones arise spontaneously? Cognition, 7, 323−331.

Murray, D.J. (1968). Articulatory and acoustic confusability in short term memory. Journal of Experimental Psychology, 78, 679−684.

THE STUDY OF THE ILLITERATE BRAIN

21

Nunes, V.A. (2002). A memória de trabalho e a aprendizagem da leitura—o loop fonológico. Masters thesis, Faculdade de Medicina de Lisboa.

Ostrosky, F., Ardila, A., Rosselli, M., Lopez-Arango, G., & Uriel-Mendoza, V. (1998). Neuropsychological test performance in illiterate subjects. Archives of Clinical Neuropsychology, 13, 645–660.

Ostrosky, F., Canseco, E., Quintanar, L., Navarro, E., & Ardila, A. (1985). Sociocultural effects in neuropsychological assessment. International Journal of Neuroscience, 27, 53−66.

Ostrosky, F., Quintanar, L., Canseco, E., Meneses, S., Navarro, E., & Ardila, A. (1986). Habilidades cognoscitivas y nível sociocultural (Cognitive abilities and sociocultural level). Revista de Investigación Clínica (Mexico), 38, 37−42.

Papagano, C, & Vallar, G. (1992). Phonological short-term memory and the learning of novel words: The effect of phonological similarity and item length. Quarterly Journal of Experimental Psychology, 44A, 47–67.

Parente, M.A. (1984). Habilidades construtivas em analfabetos: Um estudo através de desenho e construção do cubo.

Unpublished dissertation, Pontifícia Universidade Católica de São Paulo, Brasil.

Petersson, K.M., Reis, A., Askelöf, S., Castro-Caldas, A., & Ingvar, M. (2000). Language processing modulated by literacy: A network analysis of verbal repetition in literate and illiterate subjects. Journal of Cognitive Neuroscience, 12, 364−382.

Petersson, K.M., Reis, A.I., Castro-Caldas, A., & Ingvar, M. (1999). Effective auditory-verbal encoding activates the left prefrontal and the medial temporal lobes: A generalization to illiterate subjects. NeuroImage, 10, 45−54.

Reis, A., & Castro-Caldas, A. (1997a). Illiteracy: A cause for biased cognitive development. International Journal of Neuropsychology, 3, 444−450.

Reis, A., & Castro-Caldas, A. (1997b). Learning to read and write increases the efficacy of reaching a target in twodimensional space. International Journal of Neuropsychology, 3, 213.

Reis, A.I., Guerreiro, M., & Castro-Caldas, A. (1994). The influence of educational level of non-braindamaged subjects on visual naming capacities. Journal of Clinical and Experimental Neuropsychology, 16, 939–942.

Reis, A., Guerreiro, M., & Castro-Caldas, A. (1996). The illiterate brain: The influence of an untrained phonological input buffer in oral repetition and digit span of transcortical aphasics. Journal of the International Neuropsychological Society, 2, 187.

Reis, A., Guerreiro, M., Garcia, C., & Castro-Caldas, A. (1995). How does an illiterate subject process the lexical component of arithmetics? Journal of the International Neuropsychological Society, 1, 206.

Reis, A., Petersson, K.M., Castro-Caldas, A., & Ingvar, M. (2001). Formal schooling influences two-but not threedimensional naming skills. Brain and Cognition, 47, 397−411.

Rosseli, M. (1993). Neuropsychology of illiteracy. Behavioral Neurology, 6, 107−112.

Rosseli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in illiterates II: Language and praxic abilities. Brain and Cognition, 12, 281–296.

Simões, R., Fernandes, R., Nunes, V., Gonçalves, M., & Castro-Caldas, A. (2003). Inability of tactile reading in poor writers and late schooled subjects. Manuscript submitted for publication.

Snodgrass, J.G., & Vanderwart, M.A. (1980). A standardized set of 260 pictures: Norms for name agreement, familiarity and visual complexity. Journal of Experimental Psychology: Human Learning and Memory, 6, 174–215.

Stern, Y., Tang, M.X., Denaro, J., & Mayeux, R. (1995). Increased risk of mortality in Alzheimer’s disease patients with more advanced educational and occupational attainment. Annals of Neurology, 37, 590–595.

Tzavaras, A., Kaprinis, G., & Gatzoyas, A. (1981). Literacy and hemispheric specialization for language: Digit dichotic listening in illiterates. Neuropsychologia, 19, 565−570.

Vygotsky, L.S. (1978). Mind in society. The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Wechsler, A.F. (1976). Crossed aphasia in an illiterate dextral. Brain and Language, 3, 164–172.

Wechsler, D. (1945). A standardized memory scale for clinical use. Journal of Psychology, 19, 87–95.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 18–26

Imaging cognitive reserve Yaakov Stern, Nikolaos Scarmeas, and Christian Habeck Taub Institute for Research in Alzheimer’s Disease and the Aging Brain and the College of Physicians and Surgeons of Columbia University, New York, USA

We review three studies that attempt to explore the neural substrates of cognitive reserve (CR).

All studies utilize two conditions of a visual recognition task: a low demand condition, and another where difficulty was titrated such that all subjects performed at 75% recognition accuracy. We hypothesized that there would be different patterns of task-related activation as a function of a proxy measure of CR. The first two studies focused on young subjects, and found either brain areas or brain networks where the amount of increased activation from low to titrated demand correlated with CR. The third study compared activation patterns of young and elderly individuals. As in the previous studies, brain areas were found in both groups where there was differential activation as a function of CR. Most notable were locations where the relation between activation and CR differed across the two groups. These findings provide a basis for a preliminary neural implementation of cognitive reserve. They also suggest that there may be a reorganization of the neural implementation of reserve in normal ageing, which may constitute compensation for the neural effects of ageing.

Nous présentons trois études qui visaient à explorer les substrats neuraux de la réserve cognitive (RC). Ces études utilisaient deux conditions d’une tâche de reconnaissance visuelle, une condition peu demandante et une autre plus difficile pour laquelle tous les participants avaient une performance avec une exactitude de reconnaissance de 75%. L’hypothèse proposée était qu’il y aurait différents patrons d’activation reliée à la tâche comme fonction d’une mesure intermédiaire de la RC. Les deux premières études visaient de jeunes participants. Elles ont montré qu’à la fois les zones cérébrales et les réseaux cérébraux pour lesquels l’importance d’activation augmentée à partir de demandes faibles à élevées étaient corrélés avec la RC. La troisième étude comparait les patrons d’activation d’individus jeunes et ages. Comme dans les études précédentes, les zones cérébrales se sont révélées être une fonction de la RC dans les deux groupes où il y avait une activation différentielle. Les locations où la relation entre l’activation et la RC différait entre les deux groupes est particulierèment remarquable. Ces résultats fournissent une base pour l’accomplissement neural préliminaire de la RC. Ils suggèrent également qu’il peut y avoir une reorganisation de ces accomplissements neuraux de reserve dans le vieillissement normal, lequel peut constituer une compensation pour les effets neuraux associés au vieillissement.

Se revisan tres estudios que exploran los substratos neurales de la reserva cognitiva (RC).

Todos los estudios utilizan dos condiciones en una tarea de reconocimiento visual, una condición de baja demanda y otra condición de alta demanda. Todos sujetos tuvieron un nivel

STERN, SCARMEAS, HABECK

23

de 75% de aciertos en el reconocimiento. Se postularon hipótesis que proponen que habría diferentes patrones de activación en las tareas como una función de una medida proxy de RC. Los dos primeros estudios se enfocaron en sujetos jóvenes, y se encontraron tanto areas como redes cerebrales donde la cantidad de incremento en la activación, de la condición baja a la de alta demanda se correlacionó con la RC. El tercer estudio comparó los patrones de activación de individuos jóvenes y mayores. Como en los estudios previos, se encontraron areas cerebrales en ambos grupos, donde hubo una activación diferencial en función de la RC. Se detectaron regiones donde la relación entre la activación y la RC fueron diferentes entre los grupos. Estos resultados sugieren datos para poder explorar las bases neuronales de la reserva cognitiva.

También sugieren que durante el envejecimiento normal puede existir una reorganización neuronal de la RC, el cual forma parte de una compensación por las consecuencias neurales del envejecimiento.

The idea of cognitive reserve (CR) stems from the repeated observation that there does not appear to be a direct relationship between the degree of brain pathology or brain damage and the clinical manifestation of that damage. There are two concepts that we have proposed as an account for the neurophysiological substrate of CR. In the past, we have tried to differentiate between these two ideas, calling one reserve and the other compensation (Stern, 2002). The concept of reserve is based on the variability that naturally exists across individuals in the ability to recruit brain networks in response to challenging tasks. This variability might be translated into differential susceptibility to factors that disrupt performance, i.e., differential reserve. A related idea is that once pathology disrupts the brain networks that normally underlie performance, people may adapt new, compensatory networks. We call this compensation. Both reserve and compensation fall under the rubric of cognitive reserve.

Figure 1 attempts to illustrate our differential conception of reserve and compensation. In each graph in the figure, the X-axis indicates increasing task load and the Y-axis indicates expression of some brain area or brain network during task performance. Some possible sets of relationships between these variables are illustrated. In a young individual where there is no brain pathology, we would expect increased network expression as task load increases. At some point, expression of this network may plateau, and perhaps an additional network may become active at this higher difficulty level. In the nomenclature we propose, both of these networks would be subsumed under the term “reserve.” These are networks that underlie task performance and are invoked as an individual is challenged by a task. We would hypothesize that individuals with more efficient networks, or those who are better able to activate additional networks as difficulty increases, would have more flexibility in coping with challenges. Furthermore, if there were brain pathology, these individuals would be more able to cope with this pathology and continue to operate effectively. On the right of the figure are two examples of what might happen when pathology occurs. In the top example, an older individual, or one with pathology, continues to use the same networks, albeit with

Correspondence should be sent to Yaakov Stern, 630 West 168th Street, P&S Box 16, New York, NY, 10032, USA (Email: ys11@columbia.edu).

This work was supported by NIH grant AG 14671. © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000259

24

IMAGING COGNITIVE RESERVE

Figure 1. Theoretical representation of reserve and compensation networks.

less efficiency. There are some differences in how the networks are activated as a function of task difficulty. The first network plateaus much sooner. We could say that it is less efficient in that it needs to be activated at a lower level of load. The second network is expressed to a lesser degree than in young subjects. This could be an indication of lowered capacity in that the person with brain damage is less able to invoke the network. However, in both cases, the individual with brain damage is still using the same network as the person without brain damage. We would say that they are using reserve networks.

The lower right-hand figure illustrates an alternate scenario, where the individuals with brain damage express an entirely different network to the young subjects. This network meets our operational definition for compensation in that it is not expressed in the young subjects. The term â&#x20AC;&#x153;compensationâ&#x20AC;? in our nomenclature does not indicate whether the compensatory network actually allows the individual to work at the same level as someone without brain pathology. This is something that can be assessed separately from whether a compensatory network is present or not.

The distinction between reserve and compensatory networks emerges from consideration of findings and functional imaging studies that compare task-related activation in impaired and unimpaired groups. Often, imaging studies comparing an impaired and an unimpaired group will find greater activation in some brain area in the impaired group (Backman, Andersson, Nyberg, Winblad, Nordberg, & Almkvist, 1999; Becker, Mintun, Aleva, Wiseman, Nichols, & DeKosky, 1996; Cabeza, Anderson, Houle, Mangels, & Nyberg, 2000; Grady et al., 1993, 1994, 1996; Logan, Sanders, Snyder, Morris, & Buckner, 2002; Madden et al., 1999; McIntosh et al., 1999; Reuter-Lorenz, 2002; Reuter-Lorenz et al., 2000; Stern et al., 2000). This is often interpreted as the impaired group compensating for pathology by engaging alternate brain areas during task performance. If compensation truly represents a change that is induced by the brain damage, then it is important to distinguish between compensation and reserve. Without careful consideration, it is possible to interpret increased activation in an impaired group as compensation when it is not. As our graphs suggest,

STERN, SCARMEAS, HABECK

25

there may be increased network expression with increased task difficulty. This has been noted often in imaging studies (e.g., Glahn et al., 2002; Grady et al., 1996; Jansma, Ramsey, Coppola, & Kahn, 2000; Jha & McCarthy, 2000; Rypma & Dâ&#x20AC;&#x2122;Esposito, 1999). It is common for a task to be more difficult in an impaired group than in an unimpaired group. Therefore, the differential activation that is seen in an impaired group may simply be a function of increases in task difficulty. That is, after matching for task difficulty across the two groups, one might find that the two groups are actually using the same network, not different networks.

The support for the concept of CR in ageing and Alzheimerâ&#x20AC;&#x2122;s disease (AD) comes from both epidemiologic and imaging studies. Epidemiological data suggests that high education, occupation (Stern, Gurland, Tatemichi, Tang, Wilder, & Mayeux, 1994), or more active engagement in intellectual, social, and physical activities (Fabrigoule, Letenneur, Dartigues, Zarrouk, Commenges, & Barberger-Gateau, 1995; Friedland, Fritsch, & Smyth, 2001; Kondo, Niino, & Shido, 1994; Scarmeas, Levy, Tang, Manly, & Stern, 2001; Wilson et al., 2002) are associated with decreased risk for incident dementia. Similar epidemiological data suggests that increased CR might also reduce the risk of the cognitive changes that occur in normal ageing.

It has been reported that individuals with higher levels of intellectual ability, education, and socioeconomic status are more likely to develop an engaged lifestyle, which in turn contributes to the maintenance of verbal intelligence in later life (Gold, Andres, Etezadi, Arbuckle, Schwartzman, & Chaikelson, 1995). In another study, among other factors, education was related to maintenance of intellectual performance in a sample of Second World War veterans tested twice over a 40-year period (Arbuckle, Gold, Andres, Schwartzman, & Chaikelson, 1992). Low education has been associated with poor health and function in older adults (Snowdon, Ostwald, & Kane, 1989a; Snowdon, Ostwald, Kane, & Keenan, 1989b) as well as with a faster rate of cognitive decline (Butler, Ashford, & Snowdon, 1996; Nguyen, Black, Ray, Espino, & Markides, 2002).

Supporting the concept of CR, PET studies in AD subjects matched for clinical severity have reported negative correlations between resting cerebral blood flow (with CBF; taken as a surrogate for AD pathology) (DeCarli et al., 1992; Friedland, Brun, & Bundinger, 1985; McGeer, McGeer, Harrop, Akiyama, & Kamo, 1990) and education, IQ, occupation, and leisure (Alexander, Furey, Grady, Pietrini, Mentis, & Schapiro, 1997; Stern, Alexander, Prohovnik, & Mayeux, 1992; Stern et al., 1995). The negative correlations are consistent with the prediction that at any given level of behavioural symptomatology, a subject with a higher level of CR should have greater AD pathology (i.e., lower CBF).

While these studies and others provide support for the concept of CR, its neurophysiological substrate has not been established. Since we have proposed that differential CR is associated with different ways of processing task demands, this would predict that differential task-related neural processing in individuals of any given age would be a function of CR. These activation differences across subjects should be present not only in individuals affected by brain pathology, but even in healthy young individuals.

In this review, we will describe three studies that have attempted to identify the neurophysiological substrate underlying a cognitive reserve. The first two focus on young, healthy individuals. Testing for a relationship between a cognitive reserve index such as IQ and differential activation in young subjects has the advantage that one need not be concerned about the confound of a cross-subject variability in agerelated brain pathology. These two studies focus on reserve. The third study compares activation in old and young subjects. This study comparison allows us to search for compensatory networks in ageing.

All three studies used a continuous nonverbal recognition task. The basic task consists of the serial presentation of one or more single unnameable shapes, followed by a series of the same number of recognition probes. For each probe, the subject used a button press to indicate whether or not they had just seen the item. There were two task conditions. In the low demand condition, each study item was followed by a recognition probe. In the titrated demand condition, subjects studied a longer list of items, and then

26

IMAGING COGNITIVE RESERVE

Figure 2. Left: Area where titrated minus low demand activation during study of shape stimuli correlated positively with NART score in young subjects. Right: The scatter graph illustrates this correlation at a selected voxel.

responded to an equally long set of recognition probes. Prior to scanning, the study list size of the titrated demand condition was adjusted for each subject, such as recognition accuracy with 75%. This procedure was intended to match task difficulty (as operationalized by performance) across subjects.

Our intention was to explore how individual differences in cognitive reserve are related to changes in neural activity as the subjects move from the low to the titrated demand task. Our prediction was that certain aspects of task-related activation would be related to cognitive reserve. This simplifies to a prediction that there will be a correlation between fMRI activation and a measure of cognitive reserve.

The first study (Stern, Zarahn, Hilton, Delapaz, & Rakitin, 2003) used 19 healthy young adults between the ages of 18 and 30 years. We used the raw score of the National Adult Reading Test (NART; Nelson, 1982; Grober & Sliwinski, 1991) as a proxy measure for cognitive reserve. This test is a good estimate of verbal IQ, and reading measures have been used effectively as measures of reserve in the past. The data analytic approach in this study was to look voxel by voxel to find brain areas where there was a correlation between individual subjectâ&#x20AC;&#x2122;s NART scores and the degree to which event-related fMRI response amplitude changed from the low to the titrated demand. During the study phase of the task (i.e., when subjects were viewing the shapes to remember them later), positive correlations between titrated minus low demand activation and NART were seen in left middle frontal gyrus and negative correlations were seen at right superior frontal gyrus, middle frontal gyrus, precentral gyrus, medial frontal gyrus, and insular. Figure 2 demonstrates this correlation between task-related activation and NART for one brain area, left middle frontal gyrus. Note the relationship individual by individual between task-related activation and NART scores. We also found brain areas that showed similar relationships between task activation and NART scores during the recognition phase of the task.

The primary finding of this study was that, both during the study and during subsequent retrieval, brain areas were noted where there was a systematic relationship between CR and brain activation. These correlations support our hypothesis that neural processing differs as a function of CR. This differential processing may help explain individual differences in capacity or efficiency, and may underlie reserve against age-related or other pathologic changes.

In a subsequent study, we evaluated 17 healthy young adults, using the same activation methodology (Habeck, Hilton, Zarahn, Flynn, Moeller, & Stern, 2003). The major difference was in how we analysed the functional imaging data. In this second study we used ordinal trend canonical variates analysis (OrT CVA)

STERN, SCARMEAS, HABECK

27

(Habeck, Stern, Posner, & Moeller, 2002). This analysis is methodologically similar to other regional covariance analysis techniques, and is designed to identify a covariance pattern (or â&#x20AC;&#x153;brain networkâ&#x20AC;?) whose expression increases from the low to titrated demand condition for as many subjects as possible. The ordinal trend analysis was first performed on imaging data from the study phase of the task. Once such a brain network, whose subject expression systematically increased in expression from low to titrated demand, was identified, we examined the relationship between individual subjectsâ&#x20AC;&#x2122; change in network expression across the two conditions and their NART scores.

During the study phase of the task, we identified a covariance pattern, or brain network, whose change in expression from the low to the titrated demand condition increased for 15 of the 17 subjects. This is demonstrated in Figure 3, where each line represents subject expression across the two conditions. This brain network reflects activity in all areas of the brain; however, some areas are more involved in this network than others. Further, because this network described a covariance pattern, it implies that change in activation in one area is related to change in another area. This relationship can be positive or negative.

Thus, as the brain network expression increases, some brain areas show increased activation and some areas show decreased activation. Areas that were associated with increases in activation from the low to the titrated condition for the majority of subjects were found in cerebellar locations. Areas associated with decreased activity from the low to the titrated demand condition were attained in precuneus, anterior singular gyrus, bilateral thalamus, right insular, right middle temporal gyrus, and bilateral inferior frontal gyrus.

The key finding for our purposes was that the larger the increase in activation from the low to the titrated demand condition in a subject, the lower their NART IQ. That is, subjects with CR showed the greatest changes in expression of this network across the two difficulty conditions. These findings were independent of any differences in the study list size across subjects. This correlation is illustrated in Figure 3.

Once a brain network is identified in one task condition, one can apply it prospectively to data from another task condition and investigate the association of network expression with other experimental variables. In this case, we determined whether the network that we identified during study also changed its subject expression from the low to the titrated demand condition during the recognition phase of the task.

Indeed, we found that network expression increased from low to titrated demand during recognition in the same manner as it did during encoding. Further, across subjects, the change in expression from the low to the titrated demand condition in the test phase again correlated significantly with NART IQ. Thus, in the second study, a brain network was found that showed increased expression as the load associated with the task increased. That is, as we moved from a low demand task to a titrated demand task, increased expression of this network was noted. Further, this change in activation across tasks was associated with CR: Individuals with lower CR showed greater levels change in network expression. These results complement those of the first study. We have now found a brain network that appears to be differentially expressed as a function of CR. The third study (Scarmeas et al, 2003) used the same activation task. In this study we used PET as the imaging modality, and both old and young subjects were included. Seventeen young adults and 19 healthy elderly adults participated. The cognitive reserve variable that we used in this study was a factor score that summarized years of education and scores on two IQ indices, the NART and WAIS-R vocabulary score. As in the other studies, subjects were scanned while performing the low and titrated demand tasks.

This study used a similar analytic approach to the first study. We began by searching for voxels where there was a correlation between the CR measure and the change in activation from the low to the titrated condition. As in our initial study, areas where such a correlation is found would be likely candidates to be mediating CR. Indeed, we found such areas both in the young and the old subjects. The more crucial

28

IMAGING COGNITIVE RESERVE

Figure 3. Expression of a covariance pattern found during study of shape stimuli. Top: expression of the network increased from low to titrated demand in 15 of 17 subjects. Bottom: Subjectâ&#x20AC;&#x2122;s change in expression of this network across conditions correlated negatively with NART IQ.

theoretical point in this study, however, was to search for areas where the relationship between the taskrelated activation and cognitive reserve differed in young and old subjects. For example, Figure 4 illustrates a voxel in the right inferior temporal gyrus, where the relationship between task-related activation and CR was positive in the young, and negative in the old. If we assume that people with more CR are doing a task

STERN, SCARMEAS, HABECK

29

Figure 4. In right inferior temporal gyrus, change in activation from the low to the titrated demand condition correlated positively with an index of CR in young subjects, and negatively in old subjects.

in a more optimal manner, then the positive relationship in the young would suggest that it is more adaptive to show increased activation at this brain location as a task gets more difficult. However, the older subjects with more reserve are doing exactly the opposite. This finding suggests that there has been some reorganization of the networks underlying task performance in the old subjects versus the young subjects.

This reorganization has resulted in the optimal activation of this brain area in old subjects with high reserve being in the opposite direction to that in young subjects with high reserve. The older subjects are using a brain area differently to the younger subjects in order to compensate for the effects of ageing. We hypothesize that the source of this change in brain utilization is the age-related neural changes in the older group. This change in network utilization would be a candidate for what we have termed â&#x20AC;&#x153;compensation.â&#x20AC;?

Thus, this third study points to a set of brain areas where there may be reorganization of brain responses as a response to ageing. We see this as a first step in understanding the neural substrates of compensation.

In summary, the three studies reviewed represent first steps in identifying the neural implementation of the concepts of reserve and compensation. We have begun with a nonverbal memory task, but plan to generalize our application of these concepts to other tasks as well. If cognitive reserve, as we conceive of it, allows people to cope with brain pathology and maintain effective functioning, then its implementation should not be specific to any particular task.

However, there are certain aspects of our study design that we view as key to exploring reserve and compensation. The first one is careful control of task difficulty. Without this, we cannot be sure that between-subject differences and activation are not simply a function of differences in the difficulty of the task for each subject. This is the reason why we attempted to use a titration procedure to match task

30

IMAGING COGNITIVE RESERVE

difficulty for each subject. A second one involves analytic techniques that allow us to compare the relationship between task-related activation and reserve across groups. We feel that both the voxel-based and network-based approaches will yield valuable information in the future.

REFERENCES Alexander, G.E., Furey, M.L., Grady, C.L., Pietrini, P., Mentis, M.J., & Schapiro, M.B. (1997). Association of premorbid function with cerebral metabolism in Alzheimer’s disease: Implications for the reserve hypothesis.

American Journal of Psychiatry, 154, 165−172.

Arbuckle, T.Y., Gold, D.P., Andres, D., Schwartzman, A., & Chaikelson, J. (1992). The role of psychosocial context, age, and intelligence in memory performance of older men. Psychology and Aging, 7, 25–36.

Backman, L., Andersson, J.L., Nyberg, L., Winblad, B., Nordberg, A., & Almkvist, O. (1999). Brain regions associated with episodic retrieval in normal aging and Alzheimer’s disease. Neurology, 52, 1861–1870.

Becker, J.T., Mintun, M.A., Aleva, K., Wiseman, M.B., Nichols, T., & DeKosky, S.T. (1996). Compensatory reallocation of brain resources supporting verbal episodic memory in Alzheimer’s disease. Neurology, 46, 692–700.

Butler, S.M., Ashford, J.W., & Snowdon, D.A. (1996). Age, education, and changes in the Mini-Mental State Exam scores of older women: Findings from the Nun Study. Journal of the American Geriatrics Society, 44, 675–681.

Cabeza, R., Anderson, N.D., Houle, S., Mangels, J.A., & Nyberg, L. (2000). Age-related differences in neural activity during item and temporal-order memory retrieval: A positron emission tomography study. Journal of Cognitive Neurosciences, 12, 197−206.

DeCarli, C., Atack, J.R., Ball, M.J., Kay, J.A., Grady, C.L., Fewster, P., Pettigrew, K.D., Rapoport, S.I & Schapiro, M.B. (1992). Postmortem regional neurofibrillary tangle densities but not senile plaque densities are related to regional cerebral metabolic rates for glucose during life in Alzheimer’s disease patients. Neurodegeneration, 1, 113–121.

Fabrigoule, C., Letenneur, L., Dartigues, J.F., Zarrouk, M., Commenges, D., & Barberger-Gateau, P. (1995). Social and leisure activities and risk of dementia: A prospective longitudinal study. Journal American Geriatrics Society, 43, 485−490.

Friedland, R.P., Brun, A., & Bundinger, T.F. (1985). Pathological and positron emission tomographic correlations in Alzheimer’s disease. Lancet, 1–228.

Friedland, R.P., Fritsch, T., & Smyth, K.A. (2001). Patients with Alzheimer’s disease have reduced activities in midlife compared with healthy control-group members. Proceedings of the National Academy of Sciences, USA, 98, 3440–3445.

Glahn, D.C, Kim, J., Cohen, M.S., Poutanen, V.P., Therman, S., Bava, S., Van Erp, T.G. M., Manninen, M., Huttunen, M., Lonnqvist, J., Standertskjold-Nordenstam, C.G., & Cannon, T.D. (2002). Maintenance and manipulation in spatial working memory: Dissociations in the prefrontal cortex. Neuroimage, 17, 201–213.

Gold, D.P., Andres, D., Etezadi, J., Arbuckle, T., Schwartzman, A., & Chaikelson, J. (1995). Structural equation model of intellectual change and continuity and predictors of intelligence in older men. Psychology and Aging, 10, 294−303.

Grady, C.L., Haxby, J.V., Horwitz, B., Gillette, J., Salerno, J.A., Gonzalez-Aviles, A., Carson, R.E., Herscovitch, P., Schapiro, M.B., & Rapoport, S.I. (1993). Activation of cerebral blood flow during a visuoperceptual task in patients with Alzheimer-type dementia. Neurobiology of Aging, 14, 35–44.

Grady, C.L., Horwitz, B., Pietrini, P., Mentis, M.J., Ungerleider, L.G., Rapoport, S.I & Haxby, J.V. (1996). Effect of task difficulty on cerebral blood flow during perceptual matching for faces. Human Brain Mapping, 4, 227–239 .

Grady, C.L., Maisog, J.M., Horwitz, B., Ungerleider, L.G., Mentis, M.J., Salerno, J.A., Pietrini, P., Wagner, E., & Haxby, J.V. (1994). Age-related changes in cortical blood flow activation during visual processing of faces and location. Journal of Neurosciences, 14, 1450–1462 .

STERN, SCARMEAS, HABECK

31

Grober, E., & Sliwinski, M. (1991). Development and validation of a model for estimating premorbid verbal intelligence in the elderly. Journal of Clinical and Experimental Neuropsychology, 13, 933–949.

Habeck, C, Hilton, H.J., Zarahn, E., Flynn, J., Moeller, J.R., & Stern, Y. (2003). Relation of cognitive reserve and task performance to expression of regional covariance networks in an event-related fMRI study of non-verbal memory.

Neuroimage, 20, 1723–1733 .

Habeck, C.G., Stern, Y., Posner, H.B., & Moeller, J.R. (2002). Regional covariance analysis of event-related fMRI, SFN Abstracts.

Jansma, J.M., Ramsey, N.F., Coppola, R., & Kahn, R.S. (2000). Specific versus nonspecific brain activity in a parametric n-back task. Neuroimage, 12, 688–697.

Jha, A.P., & McCarthy, G. (2000). The influence of memory load upon delay-interval activity in a working-memory task: An event-related functional MRI study. Journal of Cognitive Neuroscience, 12, 90−105.

Logan, J.M., Sanders, A.L., Snyder, A.Z., Morris, J. C, & Buckner, R.L. (2002). Under-recruitment and nonselective recruitment: Dissociable neural mechanisms associated with aging. Neuron, 33, 827–840.

Madden, D.J., Turkington, T.G., Provenzale, J.M., Denny, L.L., Hawk, T.C., Gottlob, L.R., & Coleman, R.E. (1999).

Adult age differences in the functional neuroanatomy of verbal recognition memory. Human Brain Mapping, 7, 115–135.

McGeer, E.G., McGeer, P.L., Harrop, R., Akiyama, H., & Kamo, H. (1990). Correlations of regional postmortem enzyme activities with premortem local glucose metabolic rates in Alzheimer’s disease. Journal of Neurosciences Research, 27, 612–619.

McIntosh, A.R., Sekuler, A.B., Penpeci, C, Rajah, M.N., Grady, C.L., Sekuler, R., & Bennett, P.J. (1999). Recruitment of unique neural systems to support visual memory in normal aging. Current Biology, 9, 1275–1278.

Nelson, H.E. (1982). The National Adult Reading Test (NART): Test manual. Bury St Edmunds, UK: Thames Valley Test Company.

Nguyen, H.T., Black, S.A., Ray, L.A., Espino, D.V., & Markides, K.S. (2002). Predictors of decline in MMSE scores among older Mexican Americans. Journal of the Gerontological Association, 57, M181.

Reuter-Lorenz, P. (2002). New visions of the aging mind and brain. Trends in Cognitive Science, 6, 394.

Reuter-Lorenz, P.A., Jonides, J., Smith, E.E., Hartley, A., Miller, A., Marshuetz, C., & Koeppe, R.A. (2000). Age differences in the frontal lateralization of verbal and spatial working memory revealed by PET. Journal of Cognitive Neuroscience, 12, 174–187.

Rypma, B., & D’Esposito, M. (1999). The roles of prefrontal brain regions in components of working memory: Effects of memory load and individual differences. Proceedings of the National Academy of Sciences of the United States of America, 96, 6558–6563.

Scarmeas, N., Levy, G., Tang, M.X., Manly, J., & Stern, Y. (2001). Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology, 57, 2236–2242.

Scarmeas, N., Zarahn, E., Anderson, K.E., Hilton, H.J., Flynn, J., Van Heertum, R.L., Sackeim, H.A., & Stern, Y. (2003). Cognitive reserve modulates functional brain responses during memory tasks. A PET study in healthy young and elderly subjects. Neuroimage, 19, 1215–1227.

Snowdon, D.A., Ostwald, S.K., & Kane, R.L. (1989a). Education, survival and independence in elderly Catholic sisters, 1936–1988. American Journal of Epidemiology, 130, 999–1012.

Snowdon, D.A., Ostwald, S.K., Kane, R.L., & Keenan, N.L. (1989b). Years of life with good and poor mental and physical function in the elderly. Journal of Clinical Epidemiology, 42, 1055–1066.

Stern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve concept. Journal of the International Neuropsychological Society, 8, 448−460.

Stern, Y., Alexander, G.E., Prohovnik, I., & Mayeux, R. (1992). Inverse relationship between education and parietotemporal perfusion deficit in Alzheimer’s disease. Annals of Neurology, 32, 371−375.

Stern, Y., Alexander, G.E., Prohovnik, I., Stricks, L., Link, B., Lennon, M.C., & Mayeux, R. (1995). Relationship between lifetime occupation and parietal flow: Implications for a reserve against Alzheimer’s disease pathology.

Neurology, 45, 55–60.

32

IMAGING COGNITIVE RESERVE

Stern, Y., Gurland, B., Tatemichi, T.K., Tang, M.X., Wilder, D., & Mayeux, R. (1994). Influence of education and occupation on the incidence of Alzheimer’s disease. Journal of the American Medical Association, 277, 1004–1010.

Stern, Y., Moeller, J.R., Anderson, K.E., Luber, B., Zubin, N.R., DiMauro, A.A., Park, A., Campbell, C.E., Marder, K., Bell, K., Van Heertum, R., & Sackeim, H.A. (2000). Different brain networks mediate task performance in normal aging and AD: Defining compensation. Neurology, 55, 1291–1297.

Stern, Y., Zarahn, E., Hilton, H.J., Delapaz, R., & Rakitin, B. (2003). Exploring the neural basis of cognitive reserve.

Journal of Clinical and Experimental Neuropsychology, 5, 691–701.

Wilson, R.S., Mendes De Leon, C.F., Barnes, L.L., Schneider, J.A., Bienias, J.L., Evans, D.A., & Bennett, D.A. (2002).

Participation in cognitively stimulating activities and risk of incident Alzheimer disease. Journal of the American Medical Association, 287, 742−748.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 27–35

Can learning to read and write change the brain organization? An electrophysiological study F.Ostrosky-Solís, Miguel Arellano García, and Martha Pérez Universidad Nacional Autónoma de Mexico, Lomas de Reforma, Mexico

It has been suggested that learning how to read and write during childhood influences the functional organization of the adult human brain. In the present study, cortical evoked potentials (ERPs) to a probe click stimulus were used to assess the extent of activation of the two cerebral hemispheres during a verbal memory task in literate and illiterate subjects. Left-hemisphere attenuation during the experimental condition was found in both groups. These findings suggest that for the illiterate subjects, left hemisphere predominantly mediates language processing.

However, during the verbal memory task, significant intrahemispheric differences between groups were found at parieto-temporal areas. Results seem to indicate that learning how to read and write demands an intrahemispheric specialization with an important activation of parietotemporal areas. These data support the view that the brains of illiterate subjects show patterns of activation that are different from those of literate subjects, thus reflecting that environmental conditions can influence brain organization.

Il fut suggéré que l’apprentissage de la lecture et de l’écriture durant l’enfance influence l’organisation fonctionnelle du cerveau humain adulte. Dans la présente étude, les potentiels évoqués corticaux (ERPs) en réponse à des déclics entendus par le participant furent utilises pour évaluer l’étendue de l’activation des deux hémisphères cérébraux durant une tâche de memorisation verbale chez des participants alphabètes et analphabètes. Une attenuation de l’hémisphzère gauche durant la condition expérimentale fut trouvée chez les deux groupes. Ces résultats suggèrent que pour les participants analphabètes, l’hémisphzère gauche gère principalement le traitement du langage. Cependant, durant la tâche de memorisation verbale, des differences intra-hémisphériques significatives furent soulevées entre les groupes dans les zones pariéto-temporales. Les résultats semblent indiquer que l’apprentissage de la lecture et de l’écriture demande une specialisation intra-hémisphérique avec une activation importante des zones pariéto-temporales. Ces données appuient l’idée que le cerveau des analphabètes montre des patrons d’activation qui sont différents de ceux des alphabètes, indiquant que les conditions environnementales peuvent influencer l’organisation cérébrale.

Se ha sugerido que aprender a leer y escribir durante la niñez influye en la organización funcional del S cerebro humano adulto. En el presente estudio se usaron los potenciales relacionados a eventos (PRE) para evaluar la magnitud de activación de los dos hemisferios cerebrales durante una tarea de memoria verbal en sujetos alfabetos y analfabetos. En ambos grupos se encontró una atenuación significativa del hemisferio izquierdo durante la condición

34

OSTROSKY-SOLÍS, GARCÍA, PÉREZ

experimental. Estos resultados sugieren que tanto en los sujetos analfabetos como en los alfabetizados, el hemisferio izquierdo media el proceso del lenguaje. Sin embargo durante la tarea de memoria verbal, se encontraron diferencias significativas intrahemisféricas entre grupos en las areas parieto-temporales. Los resultados sugieren que aprender a leer y escribir demanda una especialización intrahemisférica con una activación importante de areas parietotemporales. Estos datos apoyan la vision de que el cerebro de los sujetos analfabetos muestra redes de activación que son diferentes que aquéllos sujetos instruidos, reflejando que condiciones medioambientales pueden influir en la organización del cerebro.

In the study of brain-behaviour relationships, one of neuropsychology’s most consistent and robust finding has been the lateralization of language functions to the left cerebral hemisphere in most humans, particularly right-handers (Geshwind & Levitsky, 1968; Luria, 1980).

The capacity for language has been associated with a predisposition for the left hemisphere to acquire speech and language. In Western cultures, a significant left-sided lesion will almost invariably result in impairments to most aspects of language, including written production and comprehension. Broadly defined speech areas in the left hemisphere have been identified and delineated, each of which is postulated to perform relatively specific functions. For example, a localized lesion in the posterior region of the left hemisphere near the angular gyrus in the adult brain may result in pure alexia, defined as the specific impairment of reading, while a lesion in the anterior region of the left hemisphere at the third frontal circumvolution will affect oral expression (Benson & Ardila, 1996).

However, as Kennepohl (1999) points out, there is accumulating evidence that left-sided dominance for language may be modified by identifiable environmental variables. For example, differences in language lateralization have been found in reading studies in Japan. Both phonetic (kana) and logographic (kanji) symbols are used in everyday written Japanese. Sasanuma (1975) reported that many aphasics with leftsided lesions exhibit a preserved ability to read kanji but not kana script. Tachistoscopic studies performed with neurologically intact individuals have produced left visual field (i.e., right hemisphere) advantages for kanji words (Elman, Takahashi, & Tohsaku, 1981). Thus, it is not only language but also how language is represented that influences the recruitment of the left or right hemisphere strategies.

Another line of studies has found that cultural factors, as reflected in orthographic systems, can also shape neurophysiological systems. Paulesu et al. (2000) showed that English-speaking subjects and Italian subjects used different areas of their brain while reading in their native languages. Thus, reading English or Italian words required different strategies, which means different patterns of neural involvement. Paulesu et al. pointed out that in English there are 1120 ways of representing 40 sounds (phonemes) by different letter

Correspondence should be sent to Feggy Ostrosky-Solís, PhD, Universidad Nacional Autónoma de Mexico, Rivera de Cupia 110–71, Lomas de Reforma, Mexico, DF 11930 (E-mail: feggy@prodigy.net.mx).

This research was partially supported by a grant from Consejo Nacional de Ciencia y Tecnología (CONACYT) y Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica Universidad Nacional Autónoma de Mexico (PAPITT). Partial data from this paper were presented at the 30th Annual Meeting of the International Neuropsychological Society in Honolulu, Hawaii. © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000268

BRAIN IN ILLITERACY

35

combinations (graphemes). The mapping between graphemes, phonemes, and whole-word sounds are essentially ambiguous in the English language; by contrast, in Italian, 33 graphemes are sufficient to represent the 25 phonemes of language, and the mappings from graphemes to phonemes are unequivocal. In order to find out how theses differences affect brain activation patterns, the authors used positron emission tomography (PET) studies while their subjects read words and nonwords aloud, and found that Italians showed a greater activation in left superior temporal regions associated with phoneme processing, while English readers showed greater activation in left posterior inferior temporal gyrus and anterior frontal gyrus.

A third line of evidence on how environmental variables can alter brain organization comes from studies in illiterate populations. In a series of behavioural as well as neuroimaging studies, Castro-Caldas and Reis (2000) found that learning how to read and write during childhood influences the functional organization of the adult brain. In a series of behavioural studies, they found out that illiterates had difficulties repeating pseudowords, and in a brain activation study using PET (Castro-Caldas, Petersson, Stone-Elander, & Ingvar, 1998) they reported that the pattern of brain activation was very different for each educational group.

According to these authors, learning the written form of language (orthography) interacts with the functions of oral language; thus learning how to read changes the way in which speech is processed.

Moreover, brain lesion data pertaining to illiterate subjects have revealed conflicting results. Lecours et al. (1988) studied the effects of unilateral stroke in large groups of either educated or illiterate Portuguese speakers. They reported that lesions in the right hemisphere resulted in a greater incidence of language difficulties in illiterate stroke victims than in their literate counterparts and that aphasia was less severe in left-sided stroke illiterate patients. However, other studies did not replicate these findings and found lefthemisphere dominance in illiterates (Damasio, Castro-Caldas, Grosso, & Ferro, 1976; Wechsler, 1976).

Matute (1988) reported that although aphasia in illiterates was associated with left-hemisphere lesions, it was less severe and the distribution per aphasia type was different, that is, in the illiterate subjects Brocaâ&#x20AC;&#x2122;s type aphasia was more frequent, even with posterior lesions.

We can say that, up to date, right hemisphere language representation is less well known than the left hemisphere organization. The question remains as to whether the functional balance between the two cerebral hemispheres while processing oral language could be modified by the knowledge of orthography; i.e., could literacy play a significant role in language lateralization?

Another way to address these questions is through the use of neurophysiological techniques such as the event-related potential (ERP). These techniques have provided the researcher with a complementary opportunity to understand more fully what is going on in the brain during the activation of different functional systems. Cognitive processes occur in a matter of milliseconds; up to two and a half correct decisions can be made per second and reaction time can be as fast as 150 milliseconds. Since ERP can measure neural activity occurring within a millisecond time range, it offers the possibility of revealing the sequence and timing of neural events occurring during the activation of specific cognitive tasks. ERPs are transient voltage fluctuations generated in the brain in conjunction with sensory, motor, or cognitive events.

They are considered to represent the summation of electric fields of a large number of neurons activated in synchrony. The basic assumption when working with ERPs is that, as a result of an event, an ensemble of neurons functionally related to that event comes to exhibit a particular spatiotemporal organization; that is, patterns of coherent or coordinate synaptic events will occur in different areas of the brain in a temporal sequence determined by anatomical connections, transmission times, and similar parameters of the ensemble of neurons being stimulated (Harmony, 1984).

It has been found that there are at least two sensory systems involved in creating the electrical responses recorded at the scalp. These systems interact to form the various components of the ERP. One is a specific,

36

OSTROSKY-SOLÍS, GARCÍA, PÉREZ

direct projection to the primary sensory areas that evokes a short latency positive-negative wave complex; the other is a nonspecific or indirect system that primarily affects the later components. Sutton, Braren, Zubin, and John (1965) have reported that the early or specific components result from exogenous influences related to the character of the stimulus such as intensity, frequency, and patterning. These early components show marked cross-modality variations. The later nonspecific components are described as endogenous and vary according to the state of the subject, the meaning of the stimulus, and the informationprocessing demands of the task (Hillyard, Simpson, Woods, Van Voorhis, & Münle, 1984). The use of neurophysiological techniques such as ERPs offer the possibility of identifying the spatial and temporal flow of information that takes place during the activation of different functional systems and might suggest ways in which this flow of information is altered during deficient cognitive processing.

Over the past few years, there has been a great deal of research on the relationship between the ERP and information processing in the brain. These studies have attempted to correlate ERP components with complex psychological variables such as selective attention (the NI00 component), active discrimination of stimulus features (the N200), delivery of task-relevant information (the P300), and expectancy (the CNV).

Although not as reliable as these four components, other endogenous brain potentials have been related to several levels of language processing such as phonetics (Molfese, 1978), semantics (Brown, Marsh, & Smith, 1976; Chapman, McCrary, Chapman, & Martin, 1980), and syntax (Kutas & Hillyard, 1982).

Recently, ERPs have also been applied in the study of reading. A negative wave with a peak latency of 200– 600 ms (N400) occurs when subjects notice semantically deviant words in sentences (Kutas & Hillyard, 1980a, 1980b, 1980c), read unpredictable words (Ostrosky, Canseco, Meneses, Próspero, & Ardila, 1987a; Ostrosky, Canseco, Meneses, Próspero, Ardila, & Zarabozo, 1987b), or name words (Stuss, Sarazin, Leech, & Picton, 1983).

Several studies have reported that ERPs, especially in the context of a probe paradigm, provide a reliable and noninvasive means of studying language lateralization both in normal subjects and in recovering aphasic patients. (Papanicolaou, Levin, & Eisenberg, 1984; Papanicolaou, Moore, Deutsch, Levin, & Eisenberg, 1988).

The probe-evoked potentials paradigm consists of recording ERPs to a repetitive probe stimulus (e.g., click or strobe flash) presented during a control condition when subjects attend to this stimulus exclusively, and during experimental conditions while, at the same time, subjects are engaged in language or other cognitive tasks. The dependent variable is the measure of the amplitude of the ERPs elicited to the probe stimulus exclusively during both control and experimental conditions.

In a series of studies, Papanicolaou et al. (1984, 1988) have reported that normal adults show greater lefthemisphere (LH) attenuation in their auditory probe ERPs when presented with language tasks and greater right-hemisphere (RH) attenuation during visuospatial tasks. The degree of relative attenuation is interpreted as the degree of hemispheric engagement in the task.

The probe paradigm has also been used with adult aphasic subjects, revealing that aphasic adults who recovered language showed greater RH attenuation to language tasks than did dysarthric, RH damage, or normal subjects (Papanicolaou et al., 1988).

In the present study, we used an auditory probe ERP with literate and illiterate neurological intact subjects to study if the functional architecture of language is more bilaterally organized in either group.

BRAIN IN ILLITERACY

37

METHODS Participants Fourteen neurological healthy subjects were tested. Seven subjects were illiterates who, for social reasons, had never attended school or had any notion of how to write even their own name, and nine subjects were literates, with more than 12 years of school attendance and with a regular habit of reading and writing. The mean age of subjects were 40.8 years (SD=6.4) and 41.2 years (SD=6.2), respectively. A neurological and psychiatric screening questionnaire was used to rule out previous neurological and psychiatric conditions such as brain injury, cerebrovascular disease, epilepsy, Parkinson’s disease, psychiatric hospitalizations, etc.

Measurement A handedness questionnaire was also presented; only right-handed subjects were included in the sample population.

In order to evaluate cognitive functioning, the NEUROPSI (Ostrosky, Ardila, & Rosselli, 1999) test was administered. This is a neuropsychological battery developed especially for Spanish-speaking subjects, standardized and validated according to age, from 16 to 85 years, and according to educational level, 0 to 24 years. Only subjects who scored within the normal range were included.

The criteria for inclusion in the illiterate sample were: (a) zero school attendance as a result of economical restrictions and/or long distances between home and school during childhood; (b) inability to write their own name; for this purpose, all the subjects were requested to write their names, and only those unable to do so were included in the illiterate sample; and (c) normal performance in daily life activities (i.e., normal functional intelligence according to the subject’s sociocultural environment); (d) right-handedness; and (e) normal neuropsychological profile.

Electrophysiological recording Probe ERPs were recorded in the Neuropsychology and Psychophysiology Laboratory of the National University of Mexico. Electrophysiological data acquisition and analysis were carried out on a Neuroscan system. Scalp electrical activity was recorded from 32 monopolar derivations according to the 10–20 international system. An electrode cap (Electro-Cap International) was used. All electrodes were referenced to linked ear lobes and an additional EOG bipolar derivation was recorded with electrodes placed at the inner and outer canthi of the right eye. Electrode impedance was always below 5 kOhms. The signal was filtered between 0.1−30 Hz (3 dB down). Each trial comprised of 256 digitized EEG points (analogue to digital converter of 12-bit resolution) acquired at a sampling rate of 256 Hz, total epoch time 1 s. A pre-stimulus baseline of 100 ms was obtained in each trial and data acquisition continued 900ms after stimulus onset. In addition, during the auditory tasks (control and experimental), the subjects fixated their gaze on a target point placed on the centre of the monitor screen to reduce the occurrence of significant contamination by eye movements. Each continuous recording was visually inspected to eliminate epochs contaminated with muscle activation, movement artifacts, and electrocardiographic signals. After that initial rejection, an individual blink rejection routine was applied to reject epochs contaminated with blinks exceeding peak-topeak amplitude of +50 to –50mv. Epochs on which one or more channels of the analogue to digital converter were saturated were also excluded (about 15% of trials were lost due to such artifacts). For every subject,

38

OSTROSKY-SOLĂ?S, GARCĂ?A, PĂ&#x2030;REZ

averaged ERPs (n=30 trials) for each recording site were obtained for each stimulus condition (control and experimental) for both groups. These ERPs were submitted to a low pass-digital (zero phase distortion) filtering with an upper cut-off of 10 Hz. All data points were corrected (prior to plotting or measurement) by subtracting the average prestimulus amplitude value. In other words, all amplitude values were measured with respect to the average amplitude of the prestimulus value corresponding to each ERP. The auditory probe stimulus was a 500-Hz click (10-ms ramp; 50-ms plateau) with an intensity of 70 dB and a rate of 1/s delivered binaurally through headphones. Subjects were asked to report any interaural loudness asymmetry; if present, minor adjustments (in 2-dB steps) were made until a subjective loudness match was found.

During the control condition, the task was simply to attend to the probe stimulus. In the experimental condition (verbal encoding), the participants were presented with a list of 30 low-imagery, high-frequency common nouns (i.e., love, liberty, happiness, peace) at a rate of approximately 0.5 words/s and were instructed to attempt to memorize the words. The list was repeated three times without interruption. In addition to the words, the probe stimulus was also presented over the headphones as before, but was irrelevant to the task (i.e., subjects were instructed to ignore the tones and concentrate on memorizing the words). To ensure compliance with the task subjects were tested immediately after presentation of the word list by a recognition test from a total of 50 words. Testing was accomplished in a single session ranging from 1 to 2 hours.

Statistical analysis For the ERP measures, 11 electrodes sites where the attenuation coefficient was largest were selected (Fz, Cz, Pz, Fc3, Cp3, T3, Tp7, Fc4, Cp4, T4, Tp8). For some analysis, homologous left and right electrodes were collapsed as anterior to posterior sites (frontal, central, parietal) and coronal sites (left, midline, right).

Amplitude and latency measures were subjected to repeated measures ANOVA using group (illiterate and literate subjects) as the between-subject variable and anterior to posterior (frontal central, parietal) or coronal (left, midline, right) as within-subject variables.

RESULTS No significant differences in the behavioural performance of the literate and illiterate subjects were found.

On average, literate subjects recalled 26.7 out of 30 words and illiterate subjects 26.2 out of 30 words.

As to the relative hemispheric activation underlying such performances, the probe ERP data, reported below, also revealed basic similarities in the two groups. Auditory probe EP waveforms recorded from the left and right areas were characterized by two major peaks, an early negative (N1) and a later positive one (P2), with average latencies of 101.5 (SD 29.4) and 210.0 (SD 30.0) ms, respectively. To assess relative hemispheric engagement during each task, the N1-P2 amplitude of the ERPs waveforms were used, following the procedure reported by Papanicolaou et al. (1988). The relative, task-specific attenuation of the probe response in each hemisphere was expressed as the ratio of the amplitude of the probe ERP obtained during processing divided by that obtained during the control condition. Thus, ratio values less than 1.00 indicate different amounts of task-specific probe response attenuation, and consequently different degrees of hemispheric engagement in the task, independently of variability in the raw ERPs. Figure 1 presents the ERPs to clicks and to the words and how the attenuation coefficient was calculated.

A ratio value of 1.0 reflects no amplitude changes across conditions whereas values below 1.0 indicate the degree of attenuation of the probe ERP in each hemisphere during the task.

BRAIN IN ILLITERACY

39

Figure 1. Peak to peak amplitudes of ERPs to clicks and to words and the formula for obtaining the attenuation coefficient. The relative, task-specific attenuation of the probe response in each hemisphere was expressed as the ratio of the amplitude of the probe EP obtained during processing (verbal memory) over that obtained during the control condition (clicks). Thus, ratio values less than 1.00 indicate different amounts of task-specific probe response attenuation, and consequently different degrees of hemispheric engagement in the task, independently of variability in the raw ERPs. A ratio value of 1.0 reflects no amplitude changes across conditions whereas value below 1.0 indicates the degree of attenuation of the probe ERPs in each hemisphere during the task.

To ensure that sensory probe ERPs, independently of their sensitivity to the cognitive task, were of comparable amplitude in the two groups and the two hemispheres, N1-P2 amplitudes of the ERPs obtained during the control condition were submitted to 2 (groups) by 2 (hemispheres) mixed design analysis of variance. As expected, both analyses resulted in null effects (p>.05), indicating that during the control condition the auditory ERPs were of comparable amplitude bilaterally, and did not differ between the literate and illiterate groups. Figure 2 illustrates the mean evoked potential (ERP) amplitude values for each hemisphere and group during the control condition.

Subsequently, to evaluate the relative degree of probe ERP attenuation in each hemisphere during the language task, amplitude ratio scores (task/ control condition) of ERPs recorded in each hemisphere were computed. These ratio scores were submitted to two separate 2 (group) by 2 (hemispheres) mixed design analyses of variance. The analysis of the experimental data resulted in a significant main effect of hemispheres, F (1, 14)=14.19, p=.002, and no significant group or interaction effects (p>.05). These data indicate that, in both groups, engagement in the verbal memory task resulted in a significantly greater left hemisphere attenuation, suggesting that the left hemisphere predominantly mediates the linguistic operations required by this task for both literate and illiterate subjects. Figure 3 illustrates the mean evoked potential (ERP) amplitude ratio scores for each hemisphere and group during the verbal memory task.

To further evaluate if there were differenceswithin hemispheres, ratio scores were submitted to a 2 (group) by 5 (sites: frontal, fronto-temporal, temporo-central, and parieto-temporal) mixed designed analysis of variance. Separate analysis was carried out for the left and right hemispheres. In both hemispheres, a significant

40

OSTROSKY-SOLĂ?S, GARCĂ?A, PĂ&#x2030;REZ

Figure 3. Mean evoked potential ERP amplitude ratio scores for each hemisphere and group during the verbal memory task. Solid lines indicate literates and broken lines indicate illiterates. As can be appreciated in both groups, engagement in the verbal memory task resulted in significantly greater left hemisphere attenuation.

Figure 2. Mean ERP amplitude values for each hemisphere and group during the control condition. During this condition the auditory ERPs were of comparable amplitude bilaterally, and did not differ between the literate and illiterate groups. Solid lines with circles indicate literates and broken lines with squares indicate illiterates.

interaction between group and site was found, F(4, 56)=3.38, p=.01. A posteriori analysis revealed significant differences between groups only in parieto-temporal areas (p<.05). These data indicate that, in the illiterate group, engagement in the verbal memory task resulted in significantly less activation of parietotemporal areas. Figure 4 illustrates the mean ERP amplitude ratio scores for frontal, fronto-temporal, temporo-central, and parieto-temporal sites. Solid lines indicate literates and broken lines, illiterates.

BRAIN IN ILLITERACY

41

Figure 4. Mean ERP amplitude ratio scores for frontal, fronto-temporal, temporo-central, and parieto-temporal sites. Solid lines with circles indicate literates and broken lines with squares indicate illiterates. Significant differences between groups were found in parieto-temporal areas (p<.05).

DISCUSSION The present probe ERPs data demonstrated that during the control condition the auditory ERPs were of comparable amplitude bilaterally, and did not differ between the literate and illiterate subjects. On the other hand, left hemisphere attenuation during the experimental condition was similar in both groups. These findings suggest that for the illiterate subjects, left hemisphere predominantly mediates language processing. However, during the verbal memory task, significant intrahemispheric differences between groups were found in parieto-temporal areas. The literate group shows a very similar attenuation across the five areas whereas the illiterate subjects did not show attenuation in this specific area. These data seem to indicate that learning how to read and write demand an intrahemispheric specialization with an important activation of parieto-temporal areas. These brain regions have been associated with object and word naming and semantic processing in several studies (Paulesu et al., 2000; Price, Moore, Humphreys, Frackowiak, & Friston, 1996; Vandenberghe, Price, Wise, Josephs, & Frackowiak, 1996). Thus it seems that learning how to read and write not only changes the areas involved during reading but also the ones recruited during verbal memory. It is important to point out that the differences between groups were not due to abnormalities in these areas since all of our illiterate subjects had an intact neuropsychological profile and intact functional skills, and their illiteracy was due to social reasons. Furthermore, no significant differences in the behavioural performance of the literate and illiterate subjects were found.

As suggested by Castro-Caldas et al. (1998), learning to read and write adds a visuographic dimension, based on operation of matching phonemes and graphemes, to the internal representational system for spoken language, so learning a specific skill during childhood may partially determine the functional organization of the adult brain.

Our results are also in accordance with other neuroimaging studies, such as that reported by Paulesu et al.

(2000) in English and Italian word reading and Castro-Caldas et al. (1998) in illiterate subjects, since both studies reported different brain activation patterns associated with cultural factors.

42

OSTROSKY-SOLÍS, GARCÍA, PÉREZ

In conclusion, our electrophysiological data supports the notion that cultural factors, as reflected by learning how to read, can powerfully shape adult neuropsychological systems.

Illiteracy still represents a significant social phenomenon. Currently, about 876 million people are illiterate (UNESCO, 2001). In human history, writing only dates back some 5000 to 6000 years, and just a few centuries ago, reading and writing abilities were uncommon among the general population. It might be assumed that the acquisition of reading and writing skills has changed the brain organization of cognitive activity in general. Very important cognitive consequences of learning to read and to write have been suggested, not only in language but also in visual perception, logical reasoning, remembering strategies, and formal operational thinking (Ardila, Rosselli, & Rosas, 1989; Castro-Caldas & Reis, 2000; Manly et al., 1999; Ostrosky-Solís, Ardila, Rosselli, López, & Mendoza, 1988; Ostrosky-Solís, López, & Ardila, 2000).

The analysis of illiteracy can help, in consequence, not only to discern the influence of educational background on cognitive performance, but also to contribute to a better understanding of the cerebral organization for cognitive activities.

Undoubtedly, as Castro-Caldas and Reis point out (2000), the brain can be understood as an organ that adapts to several types of internal and external influences. The interaction of these complex concurrent stimuli through life shapes the highly differentiated biological arrangement of the brain and its consequent physiology. Thus, the analysis of illiteracy performances on cognitive tasks can significantly increase the understanding about brain organization of cognition under normal and abnormal conditions.

REFERENCES Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in illiterates: Visuospatial and memory abilities. Brain and Cognition, 11, 147–1666.

Benson, D.F., & Ardila, A. (1996). Aphasia: A clinical perspective. New York: Oxford University Press.

Brown, W.S., Marsh, J.T., & Smith, J.C. (1976). Evoked potential waveform differences produced by, the perception of different meanings of an ambiguous phrase. Electroencephalography and Clinical Neurophysiology, 41, 113–123.

Castro-Caldas, A., Petersson, K.M., Stone-Elander, S., & Ingvar, M. (1998). The illiterate brain. Learning to read and write during childhood influences the functional organization of the adult brain. Brain, 121, 1053–1063.

Castro-Caldas, A., & Reis, A. (2000). Neurobiological substrates of illiteracy. The Neuroscientist, 6, 475–482.

Chapman, R.M., McCrary, J.W., Chapman, J.A., & Martin, J.K. (1980). Behavioral and neural analyses of connotative meaning: Word classes and rating scales. Brain and Language, 11, 319−339.

Damasio, A.R., Castro-Caldas, A., Grosso, J.T., & Ferro, J.M. (1976). Brain specialisation for language does not depend on literacy. Archives of Neurology, 33, 300–3001.

Elman, J.L., Takahashi, K., & Tohsaku, Y.H. (1981). Lateral asymmetries for the identification of concrete and abstract Kanji. Brain and Language, 13, 290–300.

Geshwind, N., & Levitsky, W. (1968). Human brain: Left-right asymmetries in temporal speech region. Science, 161, 186–187.

Harmony, T. (1984). Event-related potentials and hemispheric specialization. In A.Ardila & F. ostrosky-Solís (Eds.), The right hemisphere: Neurology and neuropsychology. New York: Gordon & Breach.

Hillyard, S.A., Simpson, G.V., Woods, D.L., Van Voorhis, S., & Münle, F.T. (1984). Event-related potentials and selective attention to different modalities. In F.Reinoso-Suarez & C.Ajmone-Marson (Eds.), Cortical integration.

New York: Raven Press.

Kennepohl, S. (1999). Toward a cultural neuropsychology: An alternative view and a preliminary model. Brain and Cognition, 41, 365–380.

Kutas, M., & Hillyard, S.A. (1980a). Reading between the lines: Event-related brain potentials during natural sentence processing. Brain and Language, 11, 354–373.

BRAIN IN ILLITERACY

43

Kutas, M., & Hillyard, S.A. (1980b). Reading senseless sentences: Brain potentials reflect semantic incongruity.

Science, 207, 203–205.

Kutas, M., & Hillyard, S.A. (1980c). Event-related brain potentials to semantically inappropriate and surprisingly large words. Biological Psychology, 11, 99–116.

Kutas, M., & Hillyard, S.A. (1982). Event-related brain potentials and cognitive science. In M. Gazzaniga (Ed.), Cognitive neuroscience. New York: Plenum Press.

Lecours, A.R., Mehler, J., Parente, M.A., Beltrami, M.C, Canossa de Tolipan, L., Castro, M.J., Carrono, V., Chagastelles, L., Dehaut, F., Delgado, R., Evangelista, A., Fajgenbaum, S., Fontoura, C., de Fraga Karmann, D., Gurd, J., Hierro Torne, C, Jakubovicz, R., Kac, R., Lefevre, B., Lima, C., Maciel, J., Mansur, L., Martinez, R., Nobrega, M.C., Osorio, Z., Paciornik, J., Papaterra, F., Jourdan Penedo, M.A., Saboya, B., Scheuer, C., Batista da Silva, A., Spinardi, M., & Texeira, M. (1988). Illiteracy and brain damage 3: A contribution to the study of speech and language disorders in illiterates with unilateral brain damage (initial testing). Neuropsychologia, 26, 575–589.

Luria, A.R. (1980). Higher cortical functions in man. New York: Basic Books.

Manly, J., Jacobs, D., Sano, M., Bell, K., Merchant, C, Small, S., & Stern, Y. (1999). Effect of literacy on neuropsychological test performance in non-demented, education-matched elders. Journal of the International Neuropsychological Society, 5, 191–202.

Matute, E. (1988). El aprendizaje de la lecto-escritura y la especialización hemisférica para el lenguaje. En F.Ostrosky & A.Ardila (Eds.), El lenguaje oral y escrito: Investigación en Latinoamérica (pp. 137–154). Mexico: Trillas.

Molfese, D.L. (1978). Left and right hemisphere involvement in speech perception: Electrophysio logical correlates.

Perceptual Psychophysics, 23, 237–243.

Ostrosky, F., Ardila, A., & Rosselli, M. (1999). NEUROPSI: A brief neuropsychological test battery in Spanish. Journal of the International Neuropsychological Society, 5, 413−433.

Ostrosky, F., Canseco, E., Meneses, S., Próspero, O., & Ardila, A. (1987a). Neuroelectric correlates of a neuropsychological model of word decoding and semantic processing in normal children. International Journal of Neuroscience, 34, 97–113.

Ostrosky, F., Canseco, E., Meneses, S., Próspero, O., Ardila, A., & Zarabozo, D. (1987b). Neuroelectric correlates of a neuropsychological model of word decoding and semantic processing in reading disabled children. International Journal of Neuroscience, 35, 1–10.

Ostrosky-Solís, F., Ardila, A., Rosselli, M., López, G., & Mendoza, V. (1988). Neuropsychological test performance in illiterates. Archives of Clinical Neuropsychology, 13, 645–660.

Ostrosky-Solís, F., López, G., & Ardila, A. (2000). Sensitivity and specificity of the Mini-Mental State Examination in a Spanish-speaking population. Applied Neuropsychology, 7, 25–31.

Papanicolaou, A.C, Levin, H.S., & Eisenberg, H.M. (1984). Evoked potential correlates of recovery from aphasia after focal left hemisphere injury adults. Neurosurgery, 14, 412−415.

Papanicolaou, A.C, Moore, B.D., Deutsch, G., Levin, H.S., & Eisenberg, H.M. (1988). Evidence for right-hemisphere involvement in recovery from aphasia. Archives of Neurology, 45, 1025–1029.

Paulesu, E., McCrory, E., Fazio, F., Menocello, L., Brunswick, N., Cappa, S.F., Cotelli, M., Cossu, G., Corte, F., Lorusso, M., Pesenti, S., Galagher, A., Perani, D., Price, C, Frith, C.D., & Frith, U. (2000). Cultural effect on brain function. Nature Neurocience, 3, 91.

Price, C., Moore, C., Humphreys, G., Frackowiak, R., & Friston, K. (1996). The neural regions sustaining object recognition and naming. Proceedings of the Royal Society of London Biologial Science, 263, 1501–1507.

Sasanuma, S. (1975). Kana and Kanji processing in Japanese aphasics. Brain and Language, 2, 369– 383.

Stuss, D.T., Sarazin, F.F., Leech, E.E., & Picton, T.W. (1983). Event-related potentials during naming and mental rotation. Electroencephalography and Clinical Neurophysiology, 56, 133–146.

Sutton, S., Braren, M., Zubin, J., & John, E.R. (1965). Evoked potentials correlates of stimulus uncertainty. Science, 150, 1187–1188.

UNESCO. (2001). Disponible. Retrieved September 7, 2001, from http://unescostat.unesno.org/en/stats/ stats0.htm.

44

OSTROSKY-SOLÍS, GARCÍA, PÉREZ

Vandenberghe, R., Price, C., Wise, R., Josephs, O., & Frackowiak, R.S. (1996). Functional anatomy of a common semantic system for word and pictures. Nature, 383, 254–256.

Wechsler, A.F. (1976). Crossed aphasia in illiterate dextral. Brain and Language, 3, 164–172.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 36–46

Culture or education? Neuropsychological test performance of a Maya indigenous population F.Ostrosky-Solís, M.Ramírez, A.Lozano, H.Picasso, and A.Vélez Universidad Nacional Autónoma de Mexico, Lomas de Reforma, Mexico

Although culture and education are factors that significantly affect cognitive performance, it is often difficult to distinguish between the effects of education and the effects of culture, since the educational level influences the sociocultural status of an individual. Therefore, although it is common to attribute the differences between the performance in neuropsychological tests to both the level of education and the culture, frequently the effects of the two variables are confounded. In the present study we analysed the influence of education and of culture on the neuropsychological profile of indigenous and a nonindigenous population. We studied a total sample of 44 individuals divided into 4 groups: (1) 7 illiterate indigenous subjects; (2) 7 control subjects with no education; (3) 15 indigenous subjects with 1–4 years of education; and (4) 15 control individuals with 1–4 years of education. Subjects were paired by age and educational level. The indigenous population was Maya, who live in the state of Yucatan in the Mexican Republic. The NEUROPSI, a brief neuropsychological test battery developed and standardized in Mexico (Ostrosky-Solís, Ardila, & Rosselli, 1997, 1999), was individually administered.

Results showed differential effects for both variables. Indigenous subjects showed higher scores in visuospatial tasks, and their level of education had significant effects on working and verbal memory. No significant differences were found in other cognitive processes (orientation, comprehension, and some executive functions). Our data showed that culture dictates what it is important for survival and that education could be considered as a type of subculture that facilitates the development of certain skills instead of others. However, the influences of both variables on cognitive skills are different, which should be considered when assessing subjects of different cultures. The interpretation of neuropsychological tests, leading to accurate assessment of cognitive dysfunction, is dependent on both education and cultural skills.

Quoique la culture et l’éducation soient des facteurs qui affectent de façon significative la performance cognitive, il est souvent difficile de distinguer entre les effets de l’éducation et ceux de la culture, notamment en raison du fait que le niveau d’éducation influence le statut socioculturel des individus. Par consequent, bien qu’il soit courant d’attribuer les differences entre la performance dans les tests neuropsychologiques à la fois au niveau d’éducation et a la culture, fréquemment, les effets des deux variables sont confondus. Dans la présente étude, nous avons analyse l’influence de l’éducation et de la culture sur le profil neuropsychologique de populations indigènes ou non indigènes. Nous avons étudié un total de 44 individus divisés en quatre groupes: (1) 7 participants indigènes analphabètes; (2) 7 participants contrôle sans education; (3) 15 participants indigènes ayant entre 1 et 4 ans de scolarité; (4) 15 participants

46

EFFECTS OF CULTURE AND EDUCATION

contrôle ayant entre 1 et 4 ans de scolarité. Les participants furent pairés selon l’âge et le niveau d’éducation. La population indigène était constituée de Mayas vivant dans l’état du Yucatan en République mexicaine. Le NEUROPSI, une brève batterie de tests neuropsychologiques développée et standardisée a Mexico (Ostrosky-Solís, Ardila, & Rosselli, 1997, 1999) fut administrée individuellement. Les résultats indiquent des effets différents pour les deux variables. Les participants indigènes ont montré des scores plus élevés dans les tâches visuospatiales et le niveau d’éducation avait des effets significatifs sur le travail et la mémoire verbale. Aucune difference significative ne fut trouvée quant aux autres processus cognitifs (orientation, comprehension et fonctions executives). Nos données montrent que la culture dicte ce qui est important pour la survie et que l’éducation doit être considérée comme un type de sousculture qui facilite le développement de certaines habiletés plutôt que d’autres. Cependant, les influences des deux variables sur les habiletés cognitives sont différentes. L’interprétation des tests neuropsychologiques et, par consequent, l’évaluation exacte de dysfonctions cognitives, dependent a la fois de l’éducation et des habiletés culturelles.

A unque la cultura y la educación son factores que afectan significativamente el desempeño cognoscitivo, a menudo es difícil distinguir entre los efectos de la educación y los efectos de la cultura, ya que el nivel educativo tiene influencia sobre el estado socio-cultural de un individuo.

Por consiguiente, aunque es común atribuir las diferencias del desempeño en una prueba neuropsicológica a ambas, el nivel educativo y la cultura, frecuentemente los efectos de las dos variables son confundidos. En el presente estudio se analizó la influencia de la educación y de la cultura en el perfil neuropsicológico de una población indígena y una población no indígena. Se estudió una muestra total de 44 individuos dividida en 4 grupos: (1) 7 sujetos indígenas analfabetos; (2) 7 sujetos control sin educación; (3) 15 sujetos indígenas con 1–4 años de educación; y (4) 15 individuos control con 1–4 años de educación. Los sujetos fueron pareados por edad y el nivel educativo. La población indígena era población Maya que vive en el estado de Yucatan en la República Mexicana. Se administró individualmente la Batería Neuropsicológica Breve en Español NEUROPSI, desarrollada y estandarizada en población hispano-hablante (Ostrosky-Solís, Ardila, & Rosselli, 1997, 1999). Los resultados mostraron efectos diferenciales para ambas variables. Los sujetos indígenas mostraron las puntuaciones más altas en las tareas visoespaciales, y el nivel de educación tuvo efectos significativos en la memoria de trabajo y en la memoria verbal. No se encontraron diferencias significativas en otros procesos cognoscitivos (orientación, comprensión, y algunas funciones ejecutivas). Estos datos sugieren que cultura dicta lo que es importante para la supervivencia y que la educación podría ser considerada como un tipo de subcultura que facilita el desarrollo de ciertas habilidades en lugar de otras, sin embargo las influencias de ambas variables en las habilidades

Correspondence should be sent to Feggy Ostrosky-Solís, PhD, Universidad Nacional Autónoma de Mexico, Rivera de Cupia 110–71, Lomas de Reforma, Mexico, DF 11930 (E-mail: feggy@prodigy.net.mx).

This work was partially supported by a grant from Consejo Nacional de Ciencia y Tecnología (CONACYT) and Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica Universidad Nacional Autónoma de Mexico (PAPITT). © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000277

OSTROSKY-SOLÍS ET AL.

47

cognoscitivas son diferentes, por consiguiente esto los datos deben ser considerados cuando se evalué a sujetos de culturas diferentes. La interpretación de pruebas neuropsicológicas y por consiguiente la valoración de los trastornos cognoscitivos depende tanto de habilidades educativas y culturales.

Level of education has been proven to have an important impact on the cerebral organization of cognitive skills and on performance in neuropsychological tests (Ardila, Ostrosky-Solís, Rosselli, & Gómez, 2000; Ardila, Rosselli, & Rosas, 1989; Castro-Caldas, Petersson, Stone-Elander, & Ingvar, 1998; Castro-Caldas & Reis, 2000; Manly et al., 1999; Matute, Leal, Zarabozo, Robles, & Cedillo, 2000; Ostrosky, Canseco, Quintanar, Navarro, & Ardila, 1985; Ostrosky, Quintanar, Canseco, Memeses, Navarro, & Ardila, 1986; Ostrosky-Solís, 2002; OstroskySolís, Ardila, Rosselli, López-Arango, & Uriel Mendoza, 1998; Rosselli, Ardila, & Rosas, 1990). It has been suggested that illiterate people solved cognitive problems functionally and specifically, and responded better to the perceptual and functional attributes of stimuli, whereas educated subjects responded to abstract concepts and to logic relationships between stimuli (Luria, 1976).

Although level of education has a significant influence on the nature of performance on traditional neuropsychological measures of verbal and nonverbal skills, it is often difficult to distinguish between education and culture, since the educational level influences the sociocultural status of an individual.

Therefore, although it is common to attribute the differences between performances in neuropsychological tests to both level of education and culture, the effects of the two variables are frequently confounded. As Ardila (1996) pointed out, the differences found in the performance on tests between “Anglos” and “Hispanics” in the United States are frequently attributed to cultural variables, without taking into account that a great part of these differences is simply the result of different educational levels.

Culture has been defined as “the way of living of a human group”; it involves everything we learn as members of a society, whether this is within social, political, economic, religious, and/or linguistic institutions. This learning includes not only the knowledge of skills to survive physically or socially, but even how to express emotions, appreciate music, or experience pain (Chinoy, 1992). Although culture is an important variable involved in the development and use of specific cognitive and behavioural skills, currently there are very few studies that have analysed how culture influences neuropsychological performance. Recently, Ardila and Moreno (2001) evaluated a group of Arauco indigenous in Colombia, using a neuropsychological test battery. Twenty indigenous were selected, 12 male and 8 female; the age range was between 8 and 30 years, and education level between 0 and 6 years. The adults were monolingual (indigenous language) and illiterate; the minors were bilingual and educated. The battery with which they were assessed included copying a cube, copying and recalling the Rey-Osterrieth figure, the Spanish version of the WISC-R block design, identification of overlapped figures, identification of multiple-choice figures, ideomotor praxis, drawing a map, spatial memory, verbal fluency, modified Wisconsin card, and a laterality questionnaire. The authors report that in some of the tests, the performance of the indigenous group was almost perfect (identification of overlapped figures and ideomotor praxis skills), whereas performance in other tests was impossible (cubes design, map drawing, Rey-Osterrieth complex figure copying, and spatial memory, modified Wisconsin). They concluded that three variables affected the performance of the subjects. (1) Educational level: A significant correlation between the scoring in the test and this level was found. (2) Cultural relevance: Some tests were significant and important while others did not make sense and were impossible to understand. (3) Age: A significant association was found between performance in the tests and this variable. One of the limitations of this study is that it included a small sample of subjects (n=20) with different levels of education (0–6 years) and a wide range of ages (8–30 years), and therefore it

48

EFFECTS OF CULTURE AND EDUCATION

is not clear whether the results are due to effects of culture, age, or differences in the educational level of the subjects.

The limitation of the studies performed to date is that the effects of culture and education are not separated. Therefore, the purpose of this research was to analyse the influence of each one of these variables (culture and education) while administering a neuropsychological test to an indigenous population.

METHOD Subjects The total sample included 44 individuals: 22 were indigenous and 22 were controls, with an average age of 50.98 years (range 16 to 73 years) and average schooling of 20 months (range 0 years to 4 years). As it has been demonstrated that learning how to read and write influences the functional organization of cognitive processes, the sample was divided into four groups: (1) 7 illiterate indigenous subjects; (2) 7 control subjects with no education; (3) 15 indigenous subjects with 1–4 years of education; and (4) 15 control individuals with 1–4 years of education. The individuals were paired by age and educational level. The descriptive characteristics of the sample are shown in Table 1.

The indigenous population under study lives in the state of Yucatan in the Mexican Republic. The Mayan language is spoken by 800,291 people (INEGI, 2000), and they are the majority population in the state of Yucatan. Yucatan is the state with more indigenous language speakers than anywhere else in Mexico.

Mayan is part of the Maya-Totonaco group; this language is spoken by peninsular indigenous and by a great number of mestizos or persons of mixed race, who use it as an interaction element in their social relationships. Women use the Mayan language more than men, and the new generations speak Spanish more often than Mayan, since Mayan is used only at home (INI, 2002). The traditional Mayan houses have walls made of interwoven branches, with guano, palm leaves, or hay, on top of a soil base. The furnishings are very simple; they generally consist of wooden chairs with leather seats, tree trunk benches, a table, hammocks made of henequen or cotton thread. The social organization of the Maya is made up of municipal authorities that, together with the nojoch tata, the (holy) escribientes or clerks, and the rezadores, or people who pray, administer justice and solve the problems of the community.

The control population was selected from Mexico City; this population was made up of individuals born in the city, who did not speak any indigenous language, and who are merchants, work at different trades, or are domestic employees. The subjects of this population were monolingual in Spanish.

TABLE 1 Descriptive characteristics of the sample Sex Indigenous illiterates Control illiterates Indigenous 1–4 Controls 1–4 Total

n

Age in years (SD) Range of age in years M

F

Level of education in months (SD)

7 7 15 15 44

58.43 (8.88) 57.71 (9.06) 47.73 (17.85) 47.60 (17.85) 50.98 (15.94)

2 5 7 8 22

0 (0.00) 0 (0.00) 30.40 (11.88) 30.40 (13.50) 20.77 (17.56)

16–73 43–69 16–73 16−73 16–73

5 2 8 7 22

OSTROSKY-SOLÍS ET AL.

49

Instruments The following battery was administered for the assessment of the subjects.

1. Clinical history. A neurologic and psychiatric screening questionnaire was used to rule out previous neurological and psychiatric conditions, such as brain injury, cerebrovascular disease, epilepsy, Parkinson’s disease, psychiatric hospitalizations, etc.

2. Guide for the Exploration, Comprehension and Expression of Basic Spanish (Ostrosky-Solís, 2002).

This was applied in order to select subjects who were completely bilingual with an adequate comprehension and expression of Spanish. A score of 70 and above is equivalent to completely bilingual (Maya-Spanish).

3. The NEUROPSI. This neuropsychological test battery is a brief battery developed and standardized in Mexico (Ostrosky-Solís et al., 1997; 1999).

The NEUROPSI test battery This battery includes the following sections.

1. Orientation. Time (day, month, and year), Place (city and specific place), and Person (how old are you?). Max score=6 pts.

2. Attention and concentration. Max score= 27pts.

2.1. Digits backward, up to six digits. Max score=6 pts.

2.2. Visual detection. On a sheet that includes 16 different figures, each one repeated 16 times, the subjects are requested to cross out those figures equal to the one presented as a model. The 16 matching figures are equally distributed at the right and the left visual fields. The test is suspended after 1 minute.

Two scores are obtained: number of correct responses (max score=16), and number of errors.

2.3. 20 minus 3, five consecutive times. Max score=5.

3. Coding. Max score=18.

3.1. Verbal memory. Six common nouns corresponding to three different semantic categories (animals, fruits, and body parts) are presented three times. After each presentation, the subject repeats those words that he or she remembers. The score is the average number of words repeated in the three trials (max score=6). In addition, intrusions, perseverations, and recency and primacy effects are noted.

3.2. Copy of a semi-complex figure. A figure similar to the Rey-Osterrieth Complex Figure, but simpler, is presented to the subject. The subject is instructed to copy the figure as best they can. A special scoring system is used, with a max score of 12pts.

4. Language. Max score=26.

4.1. Naming. Eight different line drawing figures are presented for naming. They correspond to animals, musical instruments, body parts, and objects. If the subject presents with visual difficulties, an alternative procedure is used: the patient is required to name small objects placed in the hand, and body parts. Max score=8.

4.2. Repetition. The subject is asked to repeat one monosyllabic word, one three-syllabic word, one phrase with three words, and one seven-word sentence. Successful repetition in each one is scored 1. Max score=4.

4.3. Comprehension. On a sheet of paper two circles (small and large) and two squares (small and large) are drawn. Six consecutive commands, similar to those used in the Token Test, are given to the subject. The easiest one is “point to the small square,” and the hardest one is “in addition to the circles, point to the small square.” Max score=6.

50

EFFECTS OF CULTURE AND EDUCATION

4.4. Semantic verbal fluency (animals). Two scoring systems were used: (a) the total number of correct words, and (b) a 4-point scale. One point was given for 0–5 words; two points to 6–8 words; three points to 9–14 words; and four points to 15 or more words in a minute. Intrusions and perseverations were noted.

4.5. Phonological verbal fluency (words beginning with the letter F). Two scoring systems were used: (a) the total number of correct words, and (b) a 4-point scale. One point was given to 0–3 words; two points to 4–6 words; three points to 7–9 words; and four points to 10 or more words in a minute.Intrusions and perseverations were noted.

5. Reading. The subject is asked to read aloud a short paragraph (109 words). Three questions about the paragraph are presented. Max score=3.

6. Writing. To write a six-word sentence under dictation; and to copy a different six-word sentence. Max score=2.

7. Conceptual functions. Max score=10.

7.1. Similarities. Three pairs of words (e.g., orange-pear) are presented to find the similarity. An example is provided. Each one is scored as 0 (physical similarity: both are round), 1 (functional similarity: both can be eaten), or 2 (the answer corresponds to the supraordinate word: fruits). Max score=6.

7.2. Calculation abilities. Three simple arithmetical problems are presented. Max score=3.

7.3. Sequences. The subject is asked to continue a sequence of figures drawn on a sheet of paper (what figure continues?). Max score=1.

8. Motor functions. Max score=8.

8.1. Changing the position of the hand. To repeat three positions with the hand (right and left). The model is presented up to three times by the examiner. A max score of 2 is used for the left and for the right hand.

Max score=4.

8.2. Alternating the movements of the hands. To alternate the position of the hands (right hand close, left hand open, and to switch). Max score=2.

8.3. Opposite reactions. If the examiner shows a finger, the subject must show a fist; if the examiner shows a fist, the subject must show a finger. Max score=2.

9. Delay recall. Max score=30.

9.1. Recall of verbal information: (a) spontaneous recall; max score=6; (b) cueing recall—recall by categories (animals, fruits, and body parts); max score=6; (c) recognition—the examiner reads 14 different words, and the subject must tell which ones were previously presented; max score=6.

9.2. Recall of the semi-complex figure. Max score=12.

In total, 26 different scores are obtained. Maximum total score is 130. Administration time is 25 to 30 minutes. Normative scores were obtained in a 1640-subject sample, corresponding to four age ranges (16– 30, 31–50, 51−65, and 66−85 years) and four educational levels (illiterates, 1–4, 5–9, and more than 10 years of formal education) (Ostrosky-Solís et al., 1999). The NEUROPSI manual distinguishes four levels of performance by age and by educational level: normal (within 1 SD), mildly abnormal (between 1 and 2 SDs), moderately abnormal (between 2 and 3 SDs), and severely abnormal (over 3 SDs with regard to the mean scores in that age and education group). Subjects were compared with the norms corresponding to their educational level (illiterates or 1–4 years of formal education). The NEUROPSI is sensible to cognitive alterations associated with several clinical groups. An index of 83.53% of sensitivity and 82.07% of specificity has been reported in patients with mild and moderate dementia (Mejia, Gutierrez, & OstroskySolís, in press; Ostrosky-Solís et al., 1997).

OSTROSKY-SOLÍS ET AL.

51

Procedure Inclusion criteria were scoring above 70 in the Guide for Understanding and Expression of Basic Spanish.

Individuals with mental illness and/or craniocephalic trauma were discarded. The subjects had to be functionally independent, without history of neurological and psychiatric conditions such as brain injury, cerebrovascular disease, epilepsy, Parkinson’s disease, psychiatric hospitalizations, etc.

The participation of the subjects was voluntary; what the assessment involved was explained, and they gave their verbal consent. The administration of the instruments was done individually in a place chosen by the subjects, where they felt at ease (i.e., in their house, in the shade of a tree; noisy places were avoided).

Statistical analysis Descriptive statistics were obtained for each of the neuropsychological variables per group. An analysis with a T-test for related groups and a T-test for independent groups was used to compare the effects of education and culture independently. The groups were compared in the statistical analysis as follows: (1) indigenous subjects with no education against control subjects with no education; (2) indigenous subjects with 1–4 years of education against control subjects with the same education; (3) the T-test was used to analyse the effect of education in independent groups of indigenous subjects with no education against indigenous subjects with 1–4 years of education; and finally (4) illiterate control subjects were compared to control subjects with 1–4 years of education. The significance level was established at p<.05 for all the statistical analyses.

RESULTS The effects of the culture, Maya vs. control, mantaining the level of education, can be observed in Tables 2 and 3. Table 2 shows the mean, standard deviation, and significance level obtained in the NEUROPSI subtests for illiterate groups (control and indigenous). Significant differences were found only in 2 of the 21 subtests. The differences were found in copy of the semi-complex figure and delayed verbal memory. The indigenous subjects scored higher TABLE 2 Mean, standard deviation and significance level obtained in the NEUROPSI subtests for illiterate groups (control and indigenous)a Subtest

Indigenous illiterates N=7 (±SD)

Control illiterates N=7 (±SD)

t

P

Orientation time Orientation space Orientation person Digit backwards Visual detection 20 minus 3 Immediate verbal memory Copy of semicomplex figure

2.86±0.38 2.00±0.00 1.00±0.00 1.43±1.40 6.43±3.15 1.29±2.21 2.86±1.35

2.43±0.79 1.86±0.38 1.00±0.00 1.57±1.13 9.17±4.75 1.57±2.07 4.29±0.76

1.441 1.000 – −0.225 −1.656 −0.203 −2.085

.200 .356 – .829 .149 .846 .082

8.35±0.90

6.21±1.41

3.198

.019

52

EFFECTS OF CULTURE AND EDUCATION

Subtest

Indigenous illiterates N=7 (±SD)

Control illiterates N=7 (±SD)

t

P

Naming 7.71±0.49 7.57±0.79 3.540 .736 Repetition 3.43±0.53 3.86±0.38 −2.121 .078 Comprehension 4.14±0.69 4.14±1.57 0.000 1.000 Semantic fluency 11.00±2.58 12.00±2.45 −0.536 .611 Similarities 3.57±1.90 1.86±1.68 1.816 .119 Hand position 1.71±0.49 1.00±1.00 1.698 .140 Alternating 0.71±0.76 0.57±0.53 0.420 .689 movements Opposite reactions 1.43±0.79 1.29±0.76 0.548 .604 Delayed visuospatial 6.78±1.34 4.14±2.15 2.334 .058 memory Delayed verbal 0.1 4±0.38 2.5±1.90 −3.232 .018 memory Cue recall 1.43±1.40 2.29±1.80 −1.072 .325 Recognition recall 5.57±0.53 5.86±0.38 −1.549 .172 NEUROPSI total 64.71±3.21 66.07±4.77 −0.894 .406 aSignificant differences were found only in 2 of the 21 subtests: copy of the semi-complex figure and delayed verbal memory. The indigenous subjects scored higher in copying; however, in delayed verbal memory they scored lower.

TABLE 3 Mean, standard deviation, and significance level obtained in the NEUROPSI subtests for subjects with 1–4 years of education (control and indigenous)a Subtest

Indigenous 1–4 N=15 (±SD)

Controls 1–4 N=15 (±SD)

t

P

Orientation time Orientation space Orientation person Digit backwards Visual detection 20 minus 3 Immediate verbal memory Copy of semi-complex figure Naming Repetition Comprehension Semantic fluency Phonological fluency Similarities Hand position

2.93±0.26 2.00±0.00 0.87±0.35 2.67±0.90 11.50±4.31 3.36±2.34 4.00±0.85 9.66±1.69

2.87±0.35 1.87±0.35 1.00±0.00 3.00±0.65 11.50±4.75 4.00±1.57 4.53±0.99 10.10±1.54

0.564 1.468 −1.468 −1.160 0.000 −1.662 −2.256 −0.958

.582 .164 .164 .265 1.000 .120 .041 .354

7.80±0.41 3.80±0.41 4.80±0.86 12.73±3.3 3 4.88±5.11 3.64±1.78 2.40±1.40

7.67±0.62 3.93±0.26 4.60±0.99 18.13±4.69 6.38±3.46 3.79±2.12 2.53±0.99

0.695 −1.000 0.899 −3.965 −0.846 −0.219 −0.299

.499 .334 .384 .001 .425 .830 .769

OSTROSKY-SOLÍS ET AL.

Subtest

Indigenous 1–4 N=15 (±SD)

Controls 1–4 N=15 (±SD)

t

53

P

Alternating movements 0.80±0.86 0.80±0.68 0.000 1.000 Opposite reactions 1.67±0.62 1.60±0.51 0.367 .719 Delayed visuospatial 8.53±2.44 8.46±1.96 0.095 .926 memory Delayed verbal memory 2.67±1.99 2.33±2.41 0.523 .609 Cue recall 3.33±2.06 3.27±1.91 0.113 .912 Recognition recall 5.40±1.12 5.00±1.69 1.146 .271 NEUROPSI total 84.10±12.65 88.26±15.61 −3.450 .004 aThe indigenous subjects obtained significantly lower scores on immediate verbal memory, semantic fluency, and in the NEUROPSI total score.

TABLE 4 Effects of education on the cognitive profile of the Mayan indigenous subjects: Mean, standard deviation, and significance level obtained in the NEUROPSI subtests for illiterate indigenous subjects and indigenous subjects with 1– 4 years of educationa Subtest

Indigenous illiterates n=7 (±SD)

Indigenous 1–4 n=15 t (±SD)

P

Orientation time Orientation space Orientation person Digit backwards Visual detection 20 minus 3 Immediate verbal memory Copy of semicomplex figure Naming Repetition Comprehension Semantic fluency Similarities Hand position Alternating movements Opposite reactions Delayed visuospatial memory Delayed verbal memory Cue recall Recognition recall

2.86±0.38 2.00±0.00 1.00±0.00 1.43±1.40 6.43±3.15 1.29±2.21 2.86±1.35

2.93±0.26 2.00±0.00 0.87±0.35 2.67±0.90 11.50±4.31 3.36±2.34 4.00±0.85

−0.483 – 1.468 −2.146 –3.094 –2.127 −2.066

.641 – .164 .063 .007 .055 .071

8.35±0.90

9.66±1.69

−2.361

.029

7.71±0.49 3.43±0.53 4.14±0.69 11.00±2.58 3.57±1.90 1.71±0.49 0.71±0.76

7.80±0.41 3.80±0.41 4.80±0.86 12.73±3.33 3.64±1.78 2.40±1.40 0.80±0.86

−0.402 −1.625 −1.917 −1.333 −0.083 −1.686 −0.237

.696 .137 .075 .202 .935 .108 .816

1.43±0.79 6.78±1.34

1.67±0.62 8.53±2.44

−0.706 −2.111

.497 .049

0.14±0.38

2.67±1.99

−4.737

.000

1.43±1.40 5.57±0.53

3.33± 2.06 5.40±1.12

−2.542 0.486

.021 .633

54

EFFECTS OF CULTURE AND EDUCATION

Subtest

Indigenous illiterates n=7 (±SD)

Indigenous 1–4 n=15 t (±SD)

P

NEUROPSI total 64.71±3.21 84.10±12.66 −5.561 .000 aThe group with 1–4 years of education obtained higher scores on visual detection, copy and delayed recall of the semicomplex figure, delayed verbal memory in spontaneous and cued recall, and on the total NEUROPSI score.

TABLE 5 Effects of education on the cognitive profile of the control subjects: Mean, standard deviation, and significance level obtained in the NEUROPSI subtests for illiterate control subjects and control subjects with 1–4 years of educationa Subtest

Control illiterates N=7 (+SD)

Controls 1–4 n=15 (+SD)

t

P

Orientation time 2.43±0.79 2.87±0.35 −1.409 .201 Orientation space 1.86±0.38 1.87±0.35 −0.506 .956 Orientation person 1.00 ± 0.00 1.00±0.00 – – Digit backwards 1.57±1.13 3.00±0.65 −3.101 .015 Visual detection 9.17±4.75 11.50±4.75 −0.812 .433 20 minus 3 7.57±2.07 4.00±1.57 −2.736 .022 Immediate verbal 4.29±0.76 4.53±0.99 −0.646 .528 memory Copy of semi6.21±1.41 10.10±1.54 −5.831 .000 complex figure Naming 7.57±0.79 7.67±0.62 −0.282 .784 Repetition 3.86±0.38 3.93±0.26 −0.483 .641 Comprehension 4.14±1.57 4.60±0.99 −0.707 .499 Semantic fluency 12.00±2.45 18.13±4.69 −4.025 .001 Similarities 1.86±1.68 3.79±2.12 −2.357 .033 Hand position 1.00±1.00 2.53±0.99 −3.360 .006 Alternating 0.57±0.53 0.80±0.68 −0.856 .406 movements Opposite reactions 1.29±0.76 1.60±0.51 −1.000 .345 Delayed visuospatial 4.14±2.15 8.46±1.96 −4.423 .001 memory Delayed verbal 2.57±1.90 2.33±2.41 0.250 .806 memory Cue recall 2.29±1.80 3.27±1.91 −1.168 .265 Recognition recall 5.86±0.38 5.00±1.69 1.867 .080 NEUROPSI total 66.07±4.77 88.26±15.61 −5.024 .000 aSignificant differences in favour of subjects with 1–4 years of education were present in digit backwards, consecutive subtraction (20–3), copy and delay recall of semi-complex figure, semantic fluency, similarities, hand position, and in the total NEUROPSI score.

in copying; however, they scored lower in delayed verbal memory.

OSTROSKY-SOLÍS ET AL.

55

Table 3 shows the neuropsychological subtests that turned out to be significantly different in indigenous and in controls with 1–4 years of education. The indigenous subjects obtained significant lower scores in immediate verbal memory, semantic fluency, and in the total NEUROPSI score.

Tables 4 and 5 shows the effect of education on the cognitive profile of the illiterate subjects once they acquired reading and writing skills. Table 4 shows the neuropsychological profile of the indigenous population with no education against the indigenous population with 1–4 years of education. The group with 1–4 years of education obtained higher scores on visual detection, copy and delayed recall of a semicomplex figure, delayed verbal memory in spontaneous and cue recall, and in the total NEUROPSI score.

Table 5 shows comparative data of the illiterate control subjects vs. the control subjects with 1–4 years of education. Significant differences in favour of subjects with 1–4 years of education were present in the eight subtests: digit backwards, consecutive subtraction (20–3), copy and delayed recall of the semicomplex figure, semantic fluency, similarities, hand position, and in the total NEUROPSI score.

Figure 1 shows the effect of culture, Maya vs. controls, in the illiterates and in those with 1–4 years of education. Figure 2 shows the effect of education on the cognitive profile of the illiterate subjects (in both the Maya and the control group) once they acquired reading and writing skills.

DISCUSSION Examining the influence of the cultural factor, the results obtained in this research indicate that illiterate indigenous subjects showed better execution in visuoperceptual tasks (copy of semi-complex figure), but obtained lower scores on subtests related to delayed verbal memory. These results suggest that the cultural environment in which the indigenous people live has a significant influence on their cognitive organization and, therefore, on the expression of their skills. Their culture demands the use of visuospatial skills, since they are people devoted to farming and basketry manufacturing for economic survival. Nevertheless, delayed verbal memory skills are probably not used constantly or demanded in their environment. On the contrary, control subjects who live in the city probably require more verbal memory skills more than visuospatial skills, and therefore showed significantly higher scores.

These results concur with Ardila and Moreno (2001) who, in their study on Arauco indigenous people devoted to fishing and hunting, found good execution of ideomotor skills; however, the opposite result was found when copying figures. Although they reported poor performance, in this study we found that the performance of indigenous subjects was above that of the control subjects. Probably these differences are due to the dissimilar demands of the environment.

Although they had the same level of education (1–4years), when we compared indigenous subjects with controls we found significant difficulties in delayed verbal memory, semantic fluency, and total NEUROPSI. This leads us to believe that although both groups have acquired reading and writing, culture still exerts influence on the use of different skills, such as delayed verbal memory and semantic fluency.

These differences could also be due to the fact that the indigenous subjects were bilinguals (Maya-Spanish) and, although fluent in Spanish, the use of both languages might interfere with their semantic fluency performance. Differences also might be due to the fact that in their day-to-day functioning, illiterate subjects do no have to use delayed verbal memory skills as much as the control group. Not only does culture intervene in the development and use of cognitive processes, but education also influences the appearance of certain cognitive skills. It has been said that education is not limited to the acquisition of reading, writing, and calculus; it also requires the knowledge of the practical use and adaptation of such skills to the context and situation where they are required (Manly et al., 1999; Morais, Kolinsky, Alegria, & Scliar-Cabral, 1998). Learning the use of skills acquired during the literacy process is a challenge for any

56

EFFECTS OF CULTURE AND EDUCATION

Figure 1. Effect of the culture, Maya vs. control, maintaining the level of education. 1(a) shows the neuropsychological profile of the illiterate groups. Significant differences were found in copy of the semi-complex figure and delayed verbal memory. The indigenous subjects scored higher in copying; however, in delayed verbal memory they scored lower. 1(b) shows the neuropsychological profile of subjects with 1â&#x20AC;&#x201C;4 years of education. The indigenous subjects obtained significant lower scores in immediate verbal memory, semantic fluency, and in the NEUROPSI total score.

individual, which also makes possible the necessary modifications and adjustments in order to perform adequately the tasks of the neuropsychological assessment.

OSTROSKY-SOLÍS ET AL.

57

Figure 2. Effect of education on the cognitive profile of the illiterate subjects (in both the Maya and the control group) once they acquired reading and writing skills. 2(a) shows the neuropsychological profile of the indigenous population with no education against the indigenous population with 1–4 years of education. The group with 1–4 years of education obtained higher scores on visual detection, copy and delayed recall of the semi-complex figure, delayed verbal memory in spontaneous and cued recall, and in the total NEUROPSI score. 2(b) shows comparative data of the illiterate control subjects vs. the control subjects with 1–4 years of education. Significant differences in favour of subjects with 1–4 years of education were present in the eight subtests: digit backwards, consecutive subtraction (20–3), copy and delayed recall of the semi-complex figure, semantic fluency, similarities, hand position, and in the total NEUROPSI score.

In order to determine the influence of education on the cognitive profile of both groups of subjects (indigenous and controls) once they acquired reading and writing skills, we compare the neuropsychological profile of illiterate vs. those educated for 1–4 years. We found significant differences, in

58

EFFECTS OF CULTURE AND EDUCATION

favour of the subjects with 1 to 4 years of schooling, in attention and visuoperceptual processing (visual detection, copy of a figure), visual and verbal memory (delayed recall of complex figure, verbal memory), and the NEU-ROPSI total score; these data show that even if it is true that culture influences the application and development of certain (visuospatial) skills, education influences it too. Results lead us to suppose that education drives the acquisition of specific skills such as attention and memory abilities.

The former data agree with several investigators who have found lower performance in illiterates in memory tasks, visuospatial skills, and digit retention (Ardila et al, 1989, 2000; Castro-Caldas & Reis, 2000; Ostrosky-Solís et al., 1999, 1998; Rosselliet al., 1990). As Morais and Kolinsky (2000) pointed out, the written language and its inherent characteristics have deep consequences for the ability to process linguistic and non-linguistic information. For example, the linguistic domain affects phonological and lexical knowledge; semantics influences the ability to categorize and conceptual representation as well as the strategies used for codification and recall during memory; and executive functions are expressed in selective attention and in the inhibition of inappropriate responses. Likewise, education provides training and improves the ability to process information from concrete stimulus to a model of abstract representation of the real world. Those skills acquired at school are essential to perform the operations required for the execution of neuropsychological tests (Grossi et al., 1993). Thus, once reading and writing are acquired, we observe a significant change in the way stimuli are memorized and conceptualized.

Our data show that culture can influence different skills; although both groups were illiterates, the Maya group performed better on visuospatial tasks whereas the control group scored higher on delayed verbal memory. Education trains working memory as well as strategies to improve both visual and verbal delayed memory. No significant differences were found in other cognitive processes (orientation, comprehension, and some executive functions). Our data show that culture dictates what it is important for our survival, and that education could be considered as a subculture that emphasizes the development of certain skills instead of others; however, the influences of both variables on cognitive skills are different, so both should be considered when assessing subjects from different cultures. The interpretation of neuropsychological tests, and thus accurate assessment of cognitive dysfunction, is dependent on both education and cultural skills.

REFERENCES Ardila, A. (1996). Toward a cross-cultural neuropsychology. Journal of Social and Evolution Systems, 19, 237–249.

Ardila, A., & Moreno, S. (2001). Neuropsychological test performance in Arauco Indians: An exploratory study.

Journal of the International Neuropsychological Society, 7, 510–515.

Ardila, A., Ostrosky-Solís, F., Rosselli, M., & Goacutemez, C. (2000). Age related cognitive decline during normal aging: The complex effect of education. Archives of Clinical Neuropsychology, 15, 495−514.

Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in illiterates: Visuospatial and memory abilities. Brain and Cognition, 11,147−166.

Castro-Caldas, A., Petersson, K.M., Stone-Elander, S., & Ingvar, M. (1998). The illiterate brain. Learning to read and write during childhood influences the functional organization of the adult brain. Brain, 121, 1053–1063.

Castro-Caldas, A., & Reis, A. (2000). Neurobiological substrates of illiteracy. Neuroscientist, 6, 475–482.

Chinoy, E. (1992). La sociedad. Una introducción a la sociología. Mexico: Fondo de Cultura Económica.

Grossi, D., Correra, G., Calise, C., Ruscitto, M.A., Vecchione, V., Vigliardi, M.V., & Nolfe, G. (1993). Evaluation of the influence of illiteracy on neuropsychological performances by elderly persons. Perceptual and Motor Skills, 77, 859–866.

INEGI (Instituto Nacional de Estadística, Geografía e Informática). (2000). Conteo de Población y Vivienda. Mexico: Author.

OSTROSKY-SOLÍS ET AL.

59

INI (Instituto Nacional Indigenista). (2002). Subdirección de Investigación, IBAI, Base de Localidades y Comunidades Indigenas. Mexico: Author.

Luria, A.R. (1976). Cognitive development, its cultural and social foundations. Cambridge, MA: Harvard University Press.

Manly, J.J., Jacobs, D.M., Sano, M., Bell, K., Merchant, C.A., Small, S.C., & Stern, Y. (1999). Effect of literacy on neuropsychological test performance in nondemented, education-matched elders. Journal of the International Neuropsychological Society, 5, 191–202.

Matute, E., Leal, L., Zarabozo, D., Robles, A., & Cedillo, C. (2000). Does literacy have an effect on stick constructions tasks? Journal of the International Neuropsychological Society, 6, 668–672.

Mejia, S., Gutierrez, L., & Ostrosky-Solis, F. (in press). Validity of diagnostic tests for dementia and mild cognitive impairment in Spanish-speaking elderly population. Journal of Psychogeriatrics.

Morais, J., & Kolinsky, R. (2000). Biology and culture in the literate mind. Brain and Cognition, 42, 47–49.

Morais, J., Kolinsky, R., Alegria, J., & Scliar-Cabral, L. (1998). Alphabetic literacy and psychological structures. Letras de Hoje, 33, 61–79.

Ostrosky, F., Canseco, E., Quintanar, L., Navarro, E., & Ardila, A. (1985). Sociocultural effects in neuropsychological assessment. International Journal of Neuroscience, 27, 53–66.

Ostrosky, F., Quintanar, L., Canseco, E., Meneses, S., Navarro, E., & Ardila, A. (1986). Habilidades cognoscitivas y nível sociocultural (Cognitive abilities and sociocultural level). Revista de Investigación Clínica (Mexico) , 38, 37–42.

Ostrosky-Solís, F. (2002). Education effects on cognitive function: Cognitive reserve, compensation or testing bias?

Journal of the International Neuropsychological Association, 8, 290–291.

Ostrosky-Solís, F., Ardila, A., & Rosselli, M. (1997). NEUROPSI: Una batería neuropsicológica breve. Mexico: Laboratorios Bayer.

Ostrosky-Solís, F., Ardila, A., & Rosselli, M. (1999). NEUROPSI: A brief neuropsychological test battery in Spanish with norms by age and educational level. Journal of the International Neuropsychological Society, 5, 413−433.

Ostrosky-Solís, F., Ardila, A., Roselli, M., López-Arango, G., & Uriel Mendoza, V. (1998). Neuropsychological test performance in illiterates. Archives of Clinical Neuropsychology, 13, 645–660.

Rosselli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in illiterates II: Language and praxic abilities. Brain and Cognition, 12, 281–296.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 47–60

Literacy and cognitive change among ethnically diverse elders Jennifer J.Manly, Desiree Byrd, Pegah Touradji, Danurys Sanchez, and Yaakov Stern Columbia University College of Physicians and Surgeons, New York, USA

Research concerned with relations between literacy level and assessment of cognition among ethnically diverse elders is presented. The evidence suggests that literacy has a profound effect on neuropsychological measures across verbal and nonverbal domains, and that this effect is independent of other demographic and experiential factors such as age, years of education, sex, ethnicity, and language use. It appears that reading level is a more sensitive predictor of baseline test performance, and also that literacy skills are protective against memory decline.

Adjustment for reading level, which in part reflects quality of education, overcomes the limitations of years of education as an index of educational experience among multicultural elders and thus can improve the specificity of certain neuropsychological measures. Differences in organization of visuospatial information, lack of previous exposure to stimuli, and difficulties with interpretation of the logical functions of language are possible factors that affect test performance of elders with low levels of literacy.

Une recherche portant sur les relations entre le niveau d’alphabétisation et l’évaluation de la cognition chez les aînés de diverses ethnies est présentée. Les études antérieures suggèrent que I’alphabétisation a un effet marqué sur les mesures neuropsychologiques a travers les domaines verbaux et non verbaux et que cet effet est indépendant d’autres facteurs démographiques et expérimentaux tels que l’âge, le nombre d’années de scolarité, le sexe, le groupe ethnique et la langue utilisée. Il apparaît que le niveau de lecture est un prédicteur plus sensible de la performance de base a un test et aussi que les habiletés d’alphabétisation protègent contre le déclin de la mémoire. L’adaptation au niveau de lecture, lequel reflète en partie la qualité de l’éducation, surpasse les limites associées au nombre d’années de scolarité en tant qu’indice d’expérience educative chez des personnes âgées de différentes cultures et, ainsi, peut améliorer la spécificité de certaines mesures neuropsychologiques. Les differences dans l’organisation d’informations visuospatiales, le manque d’exposition antérieure aux stimuli et les difficultés avec l’interprétation des fonctions logiques du langage sont des facteurs pouvant potentiellement avoir un impact sur la performance au test chez les aînés ayant de faibles niveaux d’alphabétisation.

Se presenta una investigación acerca de la relación entre el nivel educativo y la evaluación de habilidades cognitivas en adultos de diferentes grupos culturales (etnias). Los hallazgos sugieren que la educación tiene un profundo efecto en medidas neuropsicológicas tanto en areas verbales como no verbales, y que este efecto es independiente de otros factores demográficos y de experiencia como la edad, años de educación, sexo, etnicidad y uso del lenguaje. Al parecer,

LITERACY AND COGNITIVE CHANGE

61

el nivel de lectura es un predictor más sensible del desempeño inicial en las pruebas, y las habilidades de lectura y escritura (literacy skills) protegen del deterioro de la memoria. Ajustar puntajes de acuerdo al nivel de lectura, lo cual refleja la calidad de la educación, supera las limitaciones de tomar los años de educación como un índice de experiencia educacional entre adultos de diferentes grupos culturales mejorando la especificidad de ciertas medidas neuropsicológicas. Los factores que afectan el desempeño de adultos con bajo nivel educativo incluye diferencias en la organización de información visoespacial, falta de exposición previa a los estímulos y a dificultades en la interpretación de las funciones lógicas del lenguaje.

During the infancy of the field of neuropsychology, Vygotsky (1962, 1978) suggested that the development and organization of basic psychological processes such as abstraction, inference, and memory depended on the type of symbols (e.g., writing systems) used by individuals in their environment. Luria (1976) followed this work with a study of illiterate, unschooled individuals; he found that they solved cognitive problems in a context-bound manner and were more influenced by the perceptual and functional attributes of a stimulus than were schooled literates, who were more responsive to abstract concepts and logical relationships among stimuli. As a result of their studies of the Vai people in Liberia, Scribner and Cole (1981) concluded that although literacy is not necessary for the development of logic, abstraction, memory, and communication skills, the nature of writing systems and the way in which they are used affect the organization and expression of these cognitive abilities.

These early studies led a number of investigators to describe the influence of literacy level on specific neuropsychological measures as a reflection of underlying brain function. Several authors have reported poor performance of illiterate individuals on language tasks such as repetition of psuedowords, recall of phonologically related word associates, word list, sentence, and story recall, phonemic or letter fluency tasks, naming, and auditory comprehension (Lecours et al, 1987; Reis & Castro-Caldas, 1997; Reis, Guerreiro, & Castro-Caldas, 1994). The authors concluded that the illiterates’ lack of grapheme-phoneme correspondence explained their performance on these language-based measures. However, discrepancies in the cognitive test performance of literates and illiterates are not restricted to tasks involving phonemic skills; effects of reading and writing ability have been reported on measures of figure memory, visuospatial ability (copy of simple and complex figures, time-telling, recognition of super-imposed figures, stick constructions), digit span, naming, calculation, praxis, alternating movements, and cancellation tasks (Ardila, Rosselli, & Rosas, 1989; Rosselli, Ardila, & Rosas, 1990).

The purpose of this article is to summarize our research on literacy and neuropsychological test performance, and place this work in the context of prior studies on the effect of reading and writing skill on

Correspondence should be sent to Jennifer Manly, PhD, G. H. Sergievsky Center, 630 West 168th Street P& S Box 16, New York, NY 10032, USA (E-mail: jjm71@columbia.edu).

This research was supported by federal grants AG16206 (J.Manly), AG07232 (R.Mayeux), the Alzheimer’s Association, and the New York City Speakers Fund for Biomedical Research—Toward the Science of Patient Care. The authors thank Rosann Costa for her help with data management, Maria Gonzalez-Diaz, Cherita McDowell, and Judes Fleurimont for assistance with scheduling and interviewing participants. © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000286

62

MANLY ET AL.

cognitive ability. Our work has been concerned with improving the specificity of neuropsychological measures among ethnically diverse elders; therefore, the discussion will focus on the interrelationships between age, race, culture, formal schooling, and quality of education.

The participants for our research described below were selected from the Washington Heights Inwood Columbia Aging Project (WHICAP), a community-based, epidemiological study of dementia in the ethnically diverse neighbourhoods of Northern Manhattan, NY. The WHICAP study follows a random sample of elderly Medicare recipients above age 65 residing in selected census tracts of Washington Heights and Inwood. The population from which participants were drawn is comprised of individuals from several different countries of origin, representing three broadly defined ethnic categories (i.e., Hispanic, African American, and white). Approximately 6.5% of the overall cohort report that they are illiterate.

Interviews and neuropsychological testing were conducted in English or Spanish, according to the subject’s wishes.

The neuropsychological measures used in these studies were selected to assess cognitive functions that are typically affected in dementia and have been shown to effectively distinguish between normal ageing and dementia in this community (Stern et al., 1992). The evaluation included measures of learning and memory, orientation, abstract reasoning, language, and visuospatial ability. Specific ability areas and tests administered include: verbal list learning and memory (Selective Reminding Test [SRT], Buschke & Fuld, 1974); nonverbal memory (multiple choice version of the Benton Visual Retention Test [BVRT], Benton, 1955); orientation (items from the Mini Mental State Examination [MMSE], Folstein, Folstein, & McHugh, 1975); verbal reasoning (Similarities subtest of the Wechsler Adult Intelligence Scale-Revised [WAIS-R], Wechsler, 1981); nonverbal reasoning (Identities and Oddities subtest of the Mattis Dementia Rating Scale, Mattis, 1976); naming (15-item version of the Boston Naming Test, Kaplan, Goodglass, & Weintraub, 1983); letter fluency (Controlled Word Association, Benton & Hamsher, 1976; Jacobs, Sano, Albert, Schofield, Dooneief, & Stern, 1997); category fluency (animals, food, and clothing, using procedures from the Boston Diagnostic Aphasia Examination [BDAE], Goodglass & Kaplan, 1983); repetition (high-frequency phrases of the BDAE, Goodglass & Kaplan, 1983); auditory comprehension (first six items of the Complex Ideational Material subtest of the BDAE, Goodglass & Kaplan, 1983); visuoconstruction (Rosen Drawing Test, Rosen, 1981); and visuoperceptual skills (multiple-choice matching of figures from the BVRT, Benton, 1955).

All interview questions, test instructions, and stimuli were translated into Spanish by a committee of Spanish speakers from Cuba, Puerto Rico, Spain, and the Dominican Republic, and then back-translated to ensure accuracy. Test items were translated literally. Where necessary, scoring criteria were modified so as to give credit for responses reflecting regional idioms. The Spanish version of the battery is described in detail elsewhere (Jacobs et al., 1997). Evaluations were conducted in either English or Spanish, based on the subject’s opinion of which language would yield the best performance. Examiners were balanced bilinguals who spoke both English and Spanish daily with friends, family, and colleagues.

From an examination of medical, psychiatric, and neurological functioning, as well as from a measurement of functional status (Blessed, Tomlinson, & Roth, 1968; Boller, Mizutani, Roessmann, & Gambetti, 1980), a physician independently determined whether each participant met criteria for dementia using the Diagnostic and Statistical Manual of Mental Disorders-Revised, 3rd ed. (American Psychiatric Association, 1987) criteria. Participants were excluded from the research described below if they had a history of Parkinson’s disease, stroke, alcohol abuse, or major psychiatric illness.

LITERACY AND COGNITIVE CHANGE

63

COGNITIVE TEST PERFORMANCE AMONG ILLITERATES AND LITERATES WITH NO FORMAL EDUCATION Results of prior research suggest that literacy level has a significant influence on the nature of performance on traditional neuropsychological measures of verbal and nonverbal skills; however, many studies were unable to distinguish between the effects of literacy and the effects of little or no exposure to formal education. In our group’s earliest work in this area (Manly et al., 1999) we intended to address this limitation by comparing the test performance of unschooled nondemented literate and illiterate elders on a dementia battery. That is, we were able to distinguish the effects of literacy alone versus those associated with formal education, since in addition to those who never learned to read and write there were many elders in our sample who were literate but who had received little or no formal education.

At the time of the neuropsychological evaluation, we asked participants “Did you ever learn to read and write?” as a part of a survey questionnaire. Generally, illiterate elders in the study were born and raised in rural communities in the Caribbean or the southern United States, where formal schooling was unavailable or where they were required to work at an early age. When these individuals came to New York City, they usually obtained jobs that did not require reading or writing skills, and many of the women worked as housewives. Literates with no formal education usually learned to read and write from siblings at home.

Twenty-six literate and 47 illiterate nondemented elders with no formal education, who were equivalent on age and functional status, participated in the study. There were equal proportions of women in each group, but there was a higher proportion of Spanish-speakers in the illiterate group (92%) as compared to the literate group (65%). Significantly more illiterates (66%) could not perform, or refused to perform, the letter fluency task as compared to literates (19%). Multivariate analyses revealed significant effects of both language and literacy on overall test performance, with no significant interaction effect. Univariate testing showed that, independent of language, illiterates obtained significantly lower scores on BVRT recognition memory, WAIS-R Similarities, BDAE Repetition, and BVRT matching than literates (Table 1). There were no significant effects of language on individual test scores, and the results did not change when we limited the analyses to Spanish-speakers.

We concluded that overall, illiterates obtained significantly lower neuropsychological test scores than education-matched literates. The overall effect of literacy status remained significant when the analysis was limited to those with no formal schooling and when the potential effect of language of test administration was controlled. Because our sample included individuals who had learned to read and write, yet had received no formal schooling, we were able to control for the major environmental influence of formal education and test for the effects of literacy status alone. Our finding that literacy status had no effect on delayed recall or semantic fluency is promising for assessing illiterate elders for dementia, since memory dysfunction and semantic fluency impairment are hallmark signs of dementia and rapid forgetting is particularity important for the diagnosis of Alzheimer’s disease (e.g., Bondi, Salmon, & Butters, 1994; Troster et al., 1993; Welsh, Butters, Hughes, Mohs, & Heyman, 1992).

TABLE 1 Cognitive test scores of literate (n=26) and illiterate (n=47) elders with no formal education Effect of literacya Test

Literate

Illiterate

F

P

Learning/memory SRT total recall SRT delayed recall

32.08 (10.90) 4.31 (2.69)

28.11 (7.29) 3.57 (1.92)

5.06 3.26

.028 .076

64

MANLY ET AL.

Effect of literacya Test

Literate

Illiterate

F

P

BVRT recognition 4.96 (2.40) 3.47 (1.72) 8.06 .006 memory Orientation MMSE orientation 9.32 (0.90) 8.44 (1.52) 4.83 .031 Abstract reasoning WAIS-R similarities 7.04 (3.25) 5.33 (1.57) 10.27 .002 DRS identities & 12.22 (3.16) 12.77 (2.04) 0.24 .627 oddities Language Boston Naming 12.74 (1.54) 11.24 (2.50) 4.89 .031 Category fluency 11.34 (3.48) 12.15 (3.07) 0.28 .596 BDAE repetition 7.58 (0.83) 7.11 (1.15) 7.56 .008 BDAE 4.54 (1.38) 3.43 (1.39) 3.47 .067 comprehension Visuospatial ability Rosen drawing 1.48 (1.04) 0.88 (0.88) 4.96 .030 BVRT matching 6.52 (2.23) 4.79 (1.93) 16.98 .000 SRT=Selective Reminding Test; BVRT=Benton Visual Retention Test; MMSE=Mini-Mental State Examination; WAIS-R=Wechsler Adult Intelligence Scale—Revised; DRS=Dementia Rating Scale; BDAE=Boston Diagnostic Aphasia Examination. aThe effect of literacy (literate vs. illiterate) on neuropsychological test score after the effect of language (English vs.

Spanish) is accounted for using 2×2 ANOVA.

SPANISH READING LEVEL, NEUROPSYCHOLOGICAL TEST PERFORMANCE Interpretation of the results of our first study was limited because literacy was determined by self-report rather than measured directly. Recent work, reported here in preliminary form, addresses this limitation by assessing literacy using a reading measure. In this work, we evaluated the relationship of reading level to cognitive test performance among nondemented Spanish-speakers from the same community-based, epidemiological study. Potential participants self-identified as Hispanic and performed the neuropsychological battery in Spanish.

Reading level was measured using the 30-item Word Accentuation Test (WAT; Del Ser, GonzalezMontalvo, Martinez-Espinosa, Delgado-Villapalos, & Bermejo, 1997). This measure was designed to be equivalent to English language measures of reading recognition such as the NART (Nelson, 1982; Nelson & O’Connell, 1978), which consists of words with an irregular pronunciation whose proper reading would depend on previous knowledge or exposure to the words. This technique is not possible in the Spanish language, since the orthography of Spanish correlates with pronunciation in a regular, rule-based way. The authors of the WAT, which was developed in Spain, developed a measure in which the reader is confronted with an ambiguous graphic clue: infrequent, irregularly stressed words written in capital letters without their accent marks. This format allows for some correspondence with the NART, as correct pronunciation depends on previous knowledge of the words. The authors of the WAT found a Cronbach’s α coefficient of

LITERACY AND COGNITIVE CHANGE

65

internal consistency of .91 and found the measure useful in estimating premorbid intellectual functioning among demented patients. A measure similar to the WAT, but developed for use in Buenos Aires, was reported to have high internal consistency and good concurrent validity with the WAIS Vocabulary subtest and number of years of formal education (Burin, Jorge, Arizaga, & Paulsen, 2000).

We attempted to administer the WAT to a total of 780 Spanish-speaking elders. However, we did not obtain WAT scores for 30 individuals (3.9%). Most of these elders (n=28) had visual deficits and were unable to read the words. The remaining two elders refused to complete the test. Of the 750 elders with WAT scores, 92 were independently diagnosed as demented by a physician. Demographic data for the nondemented and demented participants is presented in Table 2. The majority of these elders identified as Dominican (59.3%), TABLE 2 Demographics for demented and nondemented Hispanic elders N % female % Dominican Mean age (SD) Mean years of education (SD) Mean WAT reading score (SD) WAT=Word Accentuation Test.

Nondemented

Demented

658 67.3 59.1 74.8 (5.7) 6.4 (4.2) 13.6 (8.6)

92 77.2 60.5 77.7 (6.9) 4.8 (3.8) 9.1 (8.5)

t or Ď&#x2021;2

p

3.821

.05

4.3 3.4 4.7

<.001 .001 <.001

while there were also Puerto Ricans (13.3%), Cubans (19.0%), and other Hispanics (8.4%) in the sample.

The majority of the sample (99%) reported that Spanish was their first language.

Our first goal was to explore the relationship of WAT scores to overall years of education among the 658 nondemented elders. The raw correlation between the two variables was significant (r=.53; p<.001).

Figure 1 depicts WAT values for each grade level; the box represents the interquartile range that contains 50% of values, the lines that extend from the box indicate the highest and lowest values (excluding outliers), and the dark line across the box indicates the median. This figure shows that despite the strong, linear relationship between the two variables, there is a great deal of variability in actual reading ability within each grade level, suggesting that the WAT is tapping into an aspect of educational experience that is not captured by years of schooling alone.

The next question was whether Spanish reading level was a significant independent predictor of neuropsychological test score over and above years of education, age, and sex among nondemented elders.

Multiple regression was used to predict each of the 13 neuropsychological test scores. In each regression equation, demographic variables (age, education, and gender) were entered as a first set of predictors, and then WAT reading level was entered in a final step. The proportion of test variance explained by the two sets of predictors is presented in Table 3. As is shown, WAT score accounted for a small but significant amount of unique variance measures across cognitive domains, ranging from 1 to 10% of the overall variance.

These results are similar to those obtained in the Manly et al. (1999) study using self-reported literacy level, and support the conclusion that Spanish literacy has a significant, independent effect on cognitive test performance beyond that predicted by years of education, age, or sex. This effect appears to be strongest on language-based measures such as naming, verbal abstraction, and comprehension, but is also significant in

66

MANLY ET AL.

Figure 1. Word Accentuation Test median, interquartile range, and high/low values by grade level.

nonverbal domains including drawing and figure matching. It is likely that literates have better developed skills in organization and analysis of certain types of visuospatial information than do individuals who have not learned how to read and write. Alternatively, literates may be successfully using linguistic skills to mediate nonverbal tasks, skills that illiterates cannot access. Line drawings may have been more ambiguous or less recognizable for illiterates, and thus more difficult to name even with a stimulus cue. The relationship of literacy with naming is not a new finding; Reis and her colleagues (Reis et al., 1994) discussed the possibility that learning conventional representations of familiar objects is comparable to learning representations of letters or words. Additionally, illiterates may have less exposure to, or familiarity with, the objects themselves (e.g., camel, cactus, accordion, harp) as a result of their reading limitations. The construction of written language provides literate individuals with practice in the interpretation of complex sentences in which TABLE 3 Effect of demographics and WAT reading on neuropsychological test score among nondemented Spanish-speaking elders Test Learning/Memory SRT total recall SRT delayed recall BVRT recognition memory Orientation MMSE Orientation Abstract reasoning WAIS-R Similarities DRS Identities & Oddities Language

Effect of age, years of education, sex R2

Independent effect of WAT scorea Î&#x201D;R2

.10 .06 .12

.03 .02 .05

.06

.01

.19 .04

.08 .04

LITERACY AND COGNITIVE CHANGE

Test

Effect of age, years of education, sex R2

67

Independent effect of WAT scorea ΔR2

Boston Naming .08 .10 Letter fluency .07 .03 Category fluency .08 .02 BDAE repetition .03 .04 BDAE comprehension .11 .08 Visuospatial skills Rosen drawing .12 .06 BVRT matching .19 .07 SRT=Selective Reminding Test; BVRT=Benton Visual Retention Test; MMSE=Mini-Mental State Examination; WAIS-R=Wechsler Adult Intelligence Scale-Revised; DRS=Dementia Rating Scale; BDAE=Boston Diagnostic Aphasia Examination. aExcept for the effect of WAT on Orientation, all R2 are significant below the .001 level.

subject-object order is varied, and in decoding logical relationships from language. We suspect that the relationship of literacy to WAIS-R Similarities subtest performance is driven by factors first described by Luria (1976) and Scribner and Cole (1981): Illiterates may have focused on more practical, concrete aspects of the items or lack the vocabulary to obtain higher scores on the measure.

SPANISH READING LEVEL AND DEMENTIA Our next goal was to determine the effect of literacy on decline in cognitive function over time. Low education has been established as a significant risk factor for AD and other dementias (Kawas & Katzman, 1999). A higher prevalence of Alzheimer’s disease and dementia among elders with low levels of education has been found in Brazil (Caramelli et al., 1997), China (Hill et al., 1993; Zhang et al., 1990), Finland (Sulkava et al., 1985), France (Dartigues et al., 1991), Italy (Bonaiuto et al., 1990; Prencipe, Casini, Ferretti, Lattanzio, Fiorelli, & Culasso, 1996), Israel (Bowirrat, Treves, Friedland, & Korczyn, 2001; Korczyn, Kahana, & Galper, 1991), the Netherlands (Ott et al, 1995), Sweden (Fratiglioni et al., 1991; Gatz, Svedberg, Pederson, Mortimer, Berg, & Johansson, 2001), and the United States (Callahan, Hall, Hui, Musick, Unverzagt, & Hendrie, 1996; Gurland et al., 1995; Mortel, Meyer, Herod, & Thornby, 1995).

Higher incidence of dementia has been demonstrated in several studies (Evans et al., 1993; Letenneur, Commenges, Dartigues & Barberger-Gateau, 1994; Stern, Gurland, Tatemichi, Tang, Wilder, & Mayeux, 1994; White et al, 1994). Cognitive decline appears to be faster (Stern, Albert, Tang, & Tsai, 1999; Teri, McCurry, Edland, Kukull, & Larson, 1995; Unverzagt, Hui, Farlow, Hall, & Hendrie, 1998) and associated with increased risk of mortality (Stern, Tang, Denaro, & Mayeux, 1995) among highly educated minorities with Alzheimer’s disease, which suggests that the level of brain pathology is greater by the time welleducated individuals show the signs of dementia.

Cognitive reserve has been suggested as the mechanism for the link between low education and higher risk of dementia observed in these studies (Mortimer, 1988; Satz et al., 1993; Stern, 2002). Reserve, or the brain’s ability to tolerate the effects of dementia pathology, may result from native ability or from the effects of lifetime experience. Years of education may serve as a proxy for reserve, whether it results from ability or experience. In passive models of reserve (Stern, 2002), education would be a proxy for the brain’s capacity (synaptic density or complexity) to tolerate either gradual or sudden insult. In active models, years

68

MANLY ET AL.

of education would be an indicator of the brain’s ability to compensate for pathology through more efficient use of existing cognitive networks or recruitment of alternate networks.

However, there are cases in which the relationship between education and risk for cognitive impairment or dementia is weakened or absent. Two large international studies of incident dementia found that illiteracy or low levels of education did not increase the risk of Alzheimer’s disease among elders in India (Chandra et al., 2001) and West Africa (Hall et al., 1998; Hendrie, 2001). In fact, these studies had the lowest prevalence and incidence rates of dementia observed to date despite the fact that a large proportion of the populations lacked formal schooling or literacy training. This paradox serves to illuminate the difficulty in comparing cultural groups with disparate backgrounds. Reserve is measured by proxy variables (such as years of education, occupational level, or IQ measures), but there are a number of ways in which cultural, racial, and economic factors may affect the predictive power of these proxies. First, it is possible that raceand income-based limits on educational opportunity weaken the relationship between years of education and native ability, leading to underestimates of the relationship between education and cognitive decline.

Minorities with strong intellectual abilities may not achieve high levels of academic or occupational status because their opportunities are limited by societal forces (e.g., racism, poverty) unrelated to their native intellect or drive to succeed. Although such individuals may be powerful or influential in their community, their abilities may not be reflected in years of schooling or traditional indicators of occupational status.

Alternatively, rather than a reflection of innate ability, years of education could be an indicator of lifetime experiences that change the brain during childhood or adult life and thus create a reserve against disease pathology. However, use of years of education to represent a direct effect of experience on the brain or cognition is also problematic when employed among ethnic minorities and immigrants due to the increased discordance between years of education and quality of education among these groups. For example, there is abundant evidence to suggest (a) that years of education is not a commensurate measure between African Americans and whites and is thus an inadequate estimate of educational experience (Margo, 1985, 1990; Smith, 1984; Smith & Welch, 1977; Welch, 1966, 1973), and (b) that African Americans have reading skills significantly below their self-reported education level (Albert & Teresi, 1999; Baker, Johnson, Velli, & Wiley, 1996; Parasuraman & Haxby, 1993).

We propose that literacy could be a more sensitive proxy for reserve than years of education because it more accurately reflects the quality of the educational experience provided to ethnic minority elders. In addition, literacy could be a more accurate reflection of native ability because it does not assume that all individuals learn the same amount from a certain grade level, or that some excel more than others; also, the seeking of learning outside of school will be reflected in measurements of literacy. We tested this hypothesis among Hispanic elders using the Spanish literacy measure.

We first wanted to determine whether there was any change in WAT score over time. For this analysis, we used a subset of 126 elders who were administered the WAT twice. The test-retest interval ranged from 1 to 4 years, with a mean of 2.5 years (SD=6 months). The majority of these elders were not demented at baseline (88%); however, eight of the nondemented elders converted to dementia at follow-up. The overall test-retest reliability of the WAT scores was good (Pearson’s r=.78). Of the eight elders who converted to dementia at their follow-up visit, the test-retest reliability was high (Pearson’s r=.91).

Given that the WAT did not show significant change over time among demented or nondemented elders, we used baseline WAT scores in the larger sample (n=750, described above) to determine if Spanish reading level was a significant independent predictor of dementia status. A stepwise binary logistic regression was used to determine the best predictors of dementia status (nondemented vs. demented) among the following independent variables: age, sex, years of education, and Spanish reading level. Criteria for entry was set at p<.01 in order to strike a balance between the likelihood of committing Type I and Type II

LITERACY AND COGNITIVE CHANGE

69

errors. This analysis revealed that WAT score was the strongest independent predictor of dementia status, such that Spanish-speaking elders with lower literacy levels were more likely to be demented (β=−.06, p<.

001). Of the possible predictors, age was the only additional variable that entered the equation (β=−.07, p<.

001), indicating that older elders were more likely to be diagnosed with dementia. Our study thus supports a link between literacy level and dementia, and indicates that among Spanish-speaking elders, literacy level is a more sensitive predictor of presence of dementia than years of education. If, as our preliminary data suggests, WAT scores remain stable in the early stages of dementia, literacy may be a more accurate reflection of cognitive reserve than years of schooling.

LITERACY AND MEMORY DECLINE The studies reviewed in the previous section focus on the relationship of schooling and literacy to dementia or Alzheimer’s disease. But there is also evidence for a role of education in age-related cognitive decline.

Several studies of normal ageing have reported more rapid cognitive and functional decline among individuals with lower educational attainment (Albert et al., 1995; Butler, Ashford, & Snowdon, 1996; Christensen et al., 1997; Chodosh, Reuben, Albert, & Seeman, 2002; Farmer, Kittner, Rae, Bartko, & Regier, 1995; Finley, Ardila, & Roselli, 1991; Snowdon, Ostwald, & Kane, 1989). These studies suggest that the same education-related factors that delay the onset of dementia also allow individuals to cope more effectively with changes encountered in normal ageing. We designed a study to explore the relationship of literacy level to change in memory ability over time among an ethnically diverse sample of Englishspeaking nondemented elders (Manly, Touradji, Tang, & Stern, in press). Specifically, we wanted to determine if literacy was a stronger predictor of memory decline (and thus a more sensitive indicator of reserve) than years of education or racial/ethnic classification, although each of these variables were expected to influence baseline scores. We focused our analyses on immediate and delayed recall measures from a verbal word list learning task, since these measures are sensitive to age-related memory decline and the earliest signs of Alzheimer’s disease.

A total of 136 participants with four completed evaluations were stratified into two literacy groups based on the median performance of the group on the WRAT-3 reading test (median=47). Table 4 presents the demographics of the two groups. As expected, the low literacy group had fewer years of education and low literacy participants were more likely to be ethnic minorities than the high literacy group. The groups did not differ from each other on age or gender composition. Mean follow-up duration was 5.1 years (SD=1.1) and did not differ within the two literacy groups.

Generalized estimating equations (GEE) analyses were performed to determine differences in change in Selective Reminding Test (SRT) total recall performance between the two literacy groups. There was no effect of gender, ethnicity, or interaction of any of these covariates with time on SRT total recall. The final model indicated that age and education had main effects on SRT total recall score in the expected directions, and that there was a significant literacy group effect and a significant time effect. There was also a significant positive Literacy×Time effect, indicating that the low literacy group had a steeper decline in total recall scores compared with the high literacy group. These results are presented graphically in Figure 2.

Using SRT delayed recall as the dependent variable in a GEE analysis yielded similar results; only years of education was significant among the added covariates, but there was a significant

70

MANLY ET AL.

Figure 2. Change in Selective Reminding Test total recall score over time.

TABLE 4 Characteristics of the two literacy groups used for analysis Low literacy group (n=67)

High literacy group (n=69)

M

SD

Range

M

SD

Range

79.8 5.8 10.6 3.4 40.3 7.8 76.1 76.1 WRAT-3=Wide Range Achievement Test-Version 3. at(134)=4.34; p<.001. bt(134)=13.22; p<.001. cχ2(1)=19.56; p<.001.

67–94 0–16 0–47

80.3 13.0 51.2 63.8 39.1

5.2 3.2 2.2

68–94 1–18 48–56

Age Years of educationa WRAT-3 Reading scoreb Sex, % female Ethnicity, % minorityc

literacy group effect and a significant time effect. There was also a significant positive Literacy× Time interaction, indicating that the low literacy group had a steeper decline in delayed recall scores compared with the high literacy group. The profile of SRT delayed recall scores over time for high and low literacy groups is shown in Figure 3.

These results indicate that elders with both high and low levels of literacy declined in immediate and delayed memory over time; however, the decline was more rapid among low literacy elders. This suggests that high literacy skills do not provide complete preservation of memory skills but rather a slowing of agerelated decline. All participants had normal overall cognition and were functioning normally in daily activities; thus the decline in memory scores was not associated with the onset of a dementia disorder. There were no interactions between time and either years of education or ethnicity, suggesting that in this diverse population of normal elders, literacy was the most sensitive predictor of memory decline. Unlike many prior studies that examined the relationship of education to dementia or normal ageing, we did not find that low education (less than 12 years) was a risk factor for cognitive decline.

LITERACY AND COGNITIVE CHANGE

71

Figure 3. Change in Selective Reminding Test delayed recall score over time.

RACE, READING LEVEL, AND NEUROPSYCHOLOGICAL TEST PERFORMANCE Another related line of research has focused on determining the factors that underlie racial group differences in cognitive test performance at a single point in time. It is now well known that African Americans obtain lower scores on neuropsychological tests as compared to non-Hispanic whites (see Manly & Jacobs, 2001, for review). Investigators frequently use covariance or matching procedures in order to equate racial groups on years of education before comparing neuropsychological test performance, since education often differs between groups of African Americans and whites. Therefore, support of racial group differences depends on whether we can successfully “adjust” for years of education. However, along with several other authors (Kaufman, Cooper, & McGee, 1997; Loewenstein, Arguelles, Arguelles, & LinnFuentes, 1994; Whitfield & Baker-Thomas, 1999), we argue that due to the disparities in quality of education reviewed above, matching on quantity of formal education does not necessarily mean that the quality of education received by each racial group is comparable. The variable “years of education” systematically differs between African Americans and whites and is also related to cognitive test performance. If this variable is not commensurate between racial groups, residual confounding will occur and spurious racial differences will be interpreted despite matching groups on years of education. Therefore, the purpose of our next study (Manly, Jacobs, Touradji, Small, & Stern, 2002) was to determine if discrepancies in quality of education could explain differences in cognitive test score between African American and white elders matched on years of education. We assessed English reading level using the Reading Recognition subtest from the Wide Range Achievement Test-Version 3 (Wilkinson, 1993). Participants were asked to name letters and pronounce words out of context. The words are listed in order of decreasing familiarity and increasing phonological complexity. WRAT-3 grade equivalent scores were derived from the normative values for people aged 65–74 from the manual.

Groups of 192 African American and 192 white TABLE 5 Demographics and reading level for African American and white elders Variable

African American mean (SD)

White mean (SD)

t or χ2

p

N Female

192 68.2%

192 68.2%

–

–

72

MANLY ET AL.

Variable

African American mean (SD)

White mean (SD)

t or Ď&#x2021;2

p

Age Years of education WRAT-3 reading score Reading level=reported grade Reading level>reported grade Reading level<reported grade

73.9 (5.8) 12.8 (2.8) 44.2 (7.2) 47% 20% 33%

74.6 (5.9) 13.0 (3.0) 49.3 (4.1) 56% 38% 7%

1.0 0.6 8.4

.30 .55 < .001

48.98

<.001

elders matched on educational attainment and sex distribution were formed through a stratified random sampling procedure. Table 5 compares the demographic characteristics of the two groups. When raw WRAT-3 scores were converted into grade-equivalent scores and compared with reported years of education, self-reported years of education gave an overestimate of actual reading level for a higher proportion of African Americans (33%) than whites (7%).

A MANCOVA revealed that African Americans scored significantly lower on the neuropsychological test battery overall as compared to whites. Follow-up univariate tests revealed significant differences between African Americans and whites on measures of word list learning and memory (SRT total immediate and delayed recall), figure memory (BVRT recognition), abstract reasoning (WAIS-R Similarities and DRS identities and oddities), letter fluency, category fluency, and visuospatial skill (Rosen drawings and BVRT matching). Table 6 shows racial group means and standard deviations on each test within the battery.

WRAT-3 scores accounted for 34% of the variance of test scores overall. When WRAT-3 scores were entered into the multivariate model, the overall effect of race remained significant, F(13, 368)=2.7, p=.001, but the effect size was significantly reduced. Follow-up univariate comparisons (Table 6) showed that after accounting for WRAT-3 score, racial differences on all measures except category fluency and the Rosen drawing test were no longer significant.

TABLE 6 Effect of race and WRAT-3 reading score on neuropsychological test score Effect of race Test

African American

Learning/Memory SRT total recall 39.8 (10.1) SRT delayed 5.8 (2.7) recall BVRT 7.4 (1.8) recognition memory Orientation MMSE 9.7 (0.7) orientation Abstract reasoning WAIS-R 12.0 (6.9) Similarities

After covarying for WRAT reading score

White

F

P

F

P

43.5 (10.1) 6.7 (2.9)

12.9 8.9

.000 .003

1.7 0.7

.194 .407

8.1 (1.5)

20.4

.000

2.5

.115

9.8 (0.5)

3.5

.062

1.0

.310

16.0 (6.3)

34.8

.000

3.7

.055

LITERACY AND COGNITIVE CHANGE

Effect of race Test

African American

White

F

73

After covarying for WRAT reading score P

F

P

DRS identities 14.6 (1.7) 15.1 (1.3) 10.8 .001 1.8 .186 & oddities Language Boston naming 14.0 (1.4) 14.1 (1.5) 0.5 .480 2.8 .093 Letter fluency 9.9 (3.8) 12.2 (4.1) 31.8 .000 3.0 .083 Category 14.6 (3.8) 16.8 (3.8) 31.4 .000 10.0 .002 fluency BDAE 7.8 (0.6) 7.8 (0.5) 0.0 .848 1.3 .250 repetition BDAE 5.5 (0.9) 5.8 (0.7) 13.4 .000 3.2 .073 comprehension Visuospatial ability Rosen drawing 2.6 (0.9) 3.0 (0.8) 30.0 .000 10.9 .001 BVRT 8.9 (1.4) 9.4 (1.2) 13.1 .000 3.4 .065 matching SRT = Selective Reminding Test; BVRT = Benton Visual Retention Test; MMSE = Mini-Mental State Examination; WAIS-R = Wechsler Adult Intelligence Scale-Revised; DRS = Dementia Rating Scale; BDAE = Boston Diagnostic Aphasia Examination.

These findings suggest that the full extent of discrepancies in educational experience between African Americans and whites is not captured by a simple “highest grade attained” variable, and thus residual confounding may explain findings of “persistent” race effects after matching groups on years of education.

CONCLUSION The research reviewed in this article demonstrates that literacy level is a crucial predictor of cognitive test performance among ethnically diverse elders. Not only does literacy level influence the specificity of neuropsychological measures. but it is also a powerful predictor of cognitive decline. Perhaps the most important finding from the studies described above is the large discrepancy between years of education and actual literacy level among ethnic minorities and immigrants. This suggests that race- or ethnicity-specific norms that correct for years of education may be less accurate than norms that correct for quality of education and are not specific to one racial or ethnic classification. Just as age and sex are expected to adjust expectations of an individual’s performance, years of education have traditionally been used to adjust for changes in baseline knowledge, strategy, and skill that are accompanied by formal schooling. However, based on the results of this work, we propose that regardless of race/ethnicity, literacy measures educational experience more accurately than years of education, and thus is a superior assessment of the knowledge, strategy, and skills needed to perform well on traditional neuropsychological tasks. Test scores adjusted for reading level can be used to predict performance more accurately than if only years of education and racial classification were used.

74

MANLY ET AL.

Of course, both native ability and educational experience contribute to an individual’s literacy level.

However, the factors that determine the strength of each contribution tend to differ among ethnic or racial groups. When comparing between ethnic groups, differences in the availability of educational opportunity are likely to predominate. Racism, poverty, and other societal forces may have prevented some individuals with high native ability from gaining literacy skills, so literacy scores are a better proxy for quality of education in between-group comparisons. Within the same ethnic or racial group, disparities in educational opportunity may not be as strong, making literacy more likely to reflect ability to achieve academic success.

In this case, literacy may be more appropriately used as a proxy for an individual’s innate intellectual ability.

Literacy involves not only the ability to read and write script, but also the knowledge of how and in what context to apply literacy skills for specific purposes. All reading and writing tasks involve specific skills such as encoding language into graphic symbols, the visual and motor abilities involved in forming and decoding characters, words, or sentences, and retrieving word representations from memory (Scribner & Cole, 1981). Since each of these skills could potentially have an effect on cognitive test performance, future investigations of literacy could measure these specific abilities in order to clarify their relationships to performance on neuropsycholo gical measures. Assessment of the context in which literacy skills are used (or not used) may improve our knowledge of how literacy influences performance on measures of abstraction, comprehension, and problem solving. If cultural and individual factors can mediate the relationship between literacy and cognitive ability, the effect on neuropsychological test performance may not be straightforward or universal.

REFERENCES Albert, M.S., Jones, K., Savage, C.R., Berkman, L., Seeman, T., Blazer, D., & Rowe, J.W. (1995). Predictors of cognitive change in older persons: MacArthur studies of successful aging. Psychology and Aging, 10, 578–589.

Albert, S.M., & Teresi, J.A. (1999). Reading ability, education, and cognitive status assessment among older adults in Harlem, New York City. American Journal of Public Health, 89, 95–97.

American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Press.

Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in illiterates: Visuospatial and memory abilities. Brain and Cognition, 11, 147–166.

Baker, F.M., Johnson, J.T., Velli, S.A., & Wiley, C. (1996). Congruence between education and reading levels of older persons. Psychiatric Services, 47, 194–196.

Benton, A.L. (1955). The Visual Retention Test. New York: The Psychological Corporation.

Benton, A.L., & Hamsher, K.D. (1976). Multilingual Aphasia Examination. Iowa City, IA: University of Iowa.

Blessed, G., Tomlinson, B.E., & Roth, M. (1968). The association between quantitative measures of senile change in the cerebral grey matter of elderly subjects. British Journal of Psychology, 114, 797–811.

Boller, F., Mizutani, T., Roessmann, U., & Gambetti, P. (1980). Parkinson’s disease, dementia, and Alzheimer’s disease: Clinicopathological correlations. Annals of Neurology, 1, 329–335.

Bonaiuto, S., Rocca, W.A., Lippi, A., Luciani, P., Turtu, F., Cavarzeran, F., & Amaducci, L. (1990). Impact of education and occupation on prevalence of Alzheimer’s disease (AD) and multi-infarct dementia (MID) in Appignano, Macerata Province, Italy. Neurology, 40, 346.

Bondi, M.W., Salmon, D.P., & Butters, N. (1994). Neuropsychological features of memory disorders in Alzheimer’s disease. In R.D.Terry, R.Katzman, & K.L.Bick (Eds.), Alzheimer’s disease (pp. 41–63). New York: Raven Press.

Bowirrat, A., Treves, T., Friedland, R.P., & Korczyn, A.D. (2001). Prevalence of Alzheimer’s type dementia in an elderly Arab population. European Journal of Epidemiology, 8, 119−123.

Burin, D.I., Jorge, R.E., Arizaga, R.A., & Paulsen, J.S. (2000). Estimation of premorbid intelligence: The word accentuation test-Buenos Aires version. Journal of Clinical and Experimental Neuropsychology, 22, 677–685.

LITERACY AND COGNITIVE CHANGE

75

Buschke, H., & Fuld, P.A. (1974). Evaluating storage, retention, and retrieval in disordered memory and learning.

Neurology, 24, 1019–1025.

Butler, S.M., Ashford, J.W., & Snowdon, D.A. (1996). Age, education, and changes in the Mini-Mental State Exam scores of older women: Findings from the Nun Study. Journal of the American Geriatrics Society, 44, 675–681.

Callahan, C.M., Hall, K.S., Hui, S.L., Musick, B.S., Unverzagt, F.W., & Hendrie, H.C. (1996). Relationship of age, education, and occupation with dementia among a community-based sample of African Americans. Archives of Neurology, 53, 134–140.

Caramelli, P., Poissant, A., Gauthier, S., Bellavance, A., Gauvreau, D., Lecours, A.R., & Joanette, Y. (1997).

Educational level and neuropsychological heterogeneity in dementia of the Alzheimer type. Alzheimer Disease and Associated Disorders, 11, 9−15.

Chandra, V., Pandav, R., Dodge, H.H., Johnston, J.M., Belle, S.H., DeKosky, S.T., & Ganguli, M. (2001). Incidence of Alzheimer’s disease in a rural community in India: The Indo-US study. Neurology, 57, 985–989.

Chodosh, J., Reuben, D.B., Albert, M.S., & Seeman, T.E. (2002). Predicting cognitive impairment in high-functioning community-dwelling older persons: MacArthur Studies of Successful Aging. Journal of the American Geriatrics Society, 50, 1051−1060.

Christensen, H., Korten, A.E., Jorm, A.F., Henderson, A.S., Jacomb, P.A., Rodgers, B., & Mackinnon, A.J. (1997).

Education and decline in cognitive performance: Compensatory but not protective. International Journal of Geriatric Psychiatry, 12, 323−330.

Dartigues, J.F., Gagnon, M., Michel, P., Letenneur, L., Commenges, D., Barberger-Gateau, P., Auriacombe S., Rigal, B., Bedry, R., Alperovitch, A., Orgogozo, J.M., Henry, P., Loiseau, P., Salamon, R., & Groupe d’Etude Paquid (1991). Le programme de recherche paquid sur 1’epidemiologie de la demence methodes et resultats initiaux.

Review of Neurology, 147, 225−230.

Del Ser, T., Gonzalez-Montalvo, J.-I., Martinez-Espinosa, S., Delgado-Villapalos, C., & Bermejo, F. (1997). Estimation of premorbid intelligence in Spanish people with the Word Accentuation Test and its application to the diagnosis of dementia. Brain and Cognition, 33, 343−356.

Evans, D.A., Beckett, L.A., Albert, M.S., Hebert, L.E., Scherr, P.A., Funkenstein, H.H., & Taylor, J.O. (1993). Level of education and change in cognitive function in a community population of older persons. Annals of Epidemiology, 3, 71−77.

Farmer, M.E., Kittner, S.J., Rae, D.S., Bartko, J.J., & Regier, D.A. (1995). Education and change in cognitive function: The epidemiologic catchment area study. Annals of Epidemiology, 5, 1–7.

Finley, G.E., Ardila, A., & Roselli, M. (1991). Cognitive aging in illiterate Colombian adults: A reversal of the classical aging pattern? Revista Interamericana de Psicologia, 25, 103–105.

Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). ‘Mini-mental State’: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198.

Fratiglioni, L., Grut, M., Forsell, Y., Viitanen, M., Grafstrom, M., Holmen, K., Ericsson, K., Backman, L., Ahlbom, A., & Winblad, B. (1991). Prevalence of Alzheimer’s disease and other dementias in an elderly urban population: Relationship with age, sex and education. Neurology, 41, 1886–1892.

Gatz, M., Svedberg, P., Pederson, N.L., Mortimer, J.A., Berg, S., & Johansson, B. (2001). Education and the risk of Alzheimer’s disease: Findings from the study of dementia in Swedish twins. Journals of Gerontology, 56B, 292−300.

Goodglass, H., & Kaplan, E. (1983). The assessment of aphasia and related disorders. Philadelphia: Lea & Febiger.

Gurland, B.J., Wilder, D., Cross, P., Lantigua, R., Teresi, J.A., Barret, V., Stern, Y., & Mayeux, R. (1995). Relative rates of dementia by multiple case definitions, over two prevalence periods, in three cultural groups. American Journal of Geriatric Psychiatry, 3, 6–20.

Hall, K.S., Gureje, O., Gao, S., Ogunniyi, A., Hui, S.L., Baiyewu, O., Unverzagt, F.W., Oluwole, S., & Hendrie, H.C. (1998). Risk factors and Alzheimer’s disease: A comparative study of two communities. Australian and New Zealand Journal of Psychiatry, 32, 698–706.

Hendrie, H.C. (2001). Exploration of environmental and genetic risk factors for Alzheimer’s disease: The value of cross-cultural studies. Current Directions in Psychol Science, 10, 98–101.

76

MANLY ET AL.

Hill, L.R., Klauber, M.R., Salmon, D.P., Yu, E.S. H., Liu, W.T., Zhang, M., & Katzman, R. (1993). Functional status, education, and the diagnosis of dementia in the Shanghai survey. Neurology, 43, 138−145.

Jacobs, D.M., Sano, M., Albert, S., Schofield, P., Dooneief, G., & Stern, Y. (1997). Cross-cultural neuropsychological assessment: A comparison of randomly selected, demographically matched cohorts of English- and Spanishspeaking older adults. Journal of Clinical and Experimental Neuropsychology, 19, 331–339.

Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston Naming Test. Philadelphia: Lea & Febiger.

Kaufman, J.S., Cooper, R.S., & McGee, D.L. (1997). Socioeconomic status and health in blacks and whites: The problem of residual confounding and the resilience of race. Epidemiology, 8, 621–628.

Kawas, C.H., & Katzman, R. (1999). Epidemiology of dementia and Alzheimer’s Disease. In R.D.Terry, R.Katzman, S.S.Sisodia, & K.L.Bick (Eds.), Alzheimer’s disease (pp. 95−116). Philadelphia: Lippincott Williams & Wilkins.

Korczyn, A.D., Kahana, E., & Galper, Y. (1991). Epidemiology of dementia in Ashkelon, Israel. Neuroepidemiology, 10, 100.

Lecours, A.R., Mehler, J., Parente, M.A., Caldeira, A., Cary, L., Castro, M.J., Dehaut, F., Delgado, R., Gurd, J., Karmann, D.F., Jakubovitz, R., Osorio, Z., Cabral, L.S., & Junquerira, A.M.S. (1987). Illiteracy and brain damage.

1. Aphasia testing in culturally contrasted populations (control subjects). Neuropsychologia, 25, 231–245.

Letenneur, L., Commenges, D., Dartigues, J.F., & Barberger-Gateau, P. (1994). Incidence of dementia and Alzheimer’s disease in elderly community residents of South-Western France. International Journal of Epidemiology, 23, 1256–1261.

Loewenstein, D.A., Arguelles, T., Arguelles, S., & Linn-Fuentes, P. (1994). Potential cultural bias in the neuropsychological assessment of the older adult. Journal of Clinical and Experimental Neuropsychology, 16, 623–629.

Luria, A.R. (1976). Cognitive development, its cultural and social foundations. Cambridge, MA: Harvard University Press.

Manly, J.J., & Jacobs, D.M. (2001). Future directions in neuropsychological assessment with African Americans. In F.R.Ferraro (Ed.), Minority and cross-cultural aspects of neuropsychological assessment (pp. 79–96). Leiden, The Netherlands: Swets & Zeitlinger.

Manly, J.J., Jacobs, D.M., Sano, M., Bell, K., Merchant, C.A., Small, S.A., & Stern, Y. (1999). Effect of literacy on neuropsychological test performance in nondemented, education-matched elders. Journal of the International Neuropsychological Society, 5, 191–202.

Manly, J.J., Jacobs, D.M., Touradji, P., Small, S.A., & Stern, Y. (2002). Reading level attenuates differences in neuropsychological test performance between African American and White elders. Journal of the International Neuropsychological Society, 8, 341–348.

Manly, J.J., Touradji, P., Tang, M.X., & Stern, Y. (in press). Literacy and memory decline among ethnically diverse elders. Journal of Clinical and Experimental Neuropsychology.

Margo, R.A. (1985). Disenfranchisement, school finance, and the economics of segregated schools in the United States South, 1980–1910. New York: Garland Publishing.

Margo, R.A. (1990). Race and schooling in the South, 1880–1950: An economic history. Chicago: University of Chicago Press.

Mattis, S. (1976). Mental status examination for organic mental syndrome in the elderly patient. In L.Bellak & T.B.Karasu (Eds.), Geriatric psychiatry (pp. 77–121). New York: Grune & Stratton.

Mortel, K.F., Meyer, J.S., Herod, B., & Thornby, J. (1995). Education and occupation as risk factors for dementia of the Alzheimer and ischemic vascular types. Dementia, 6, 55–62.

Mortimer, J.A. (1988). Do psychosocial risk factors contribute to Alzheimer’s disease. In A.S. Henderson & J.H.Henderson (Eds.), Etiology of Dementia of Alzheimer’s Type (pp. 39–52). Chichester, UK: John Wiley.

Nelson, H.E. (1982). The National Adult Reading Test (NART): Test Manual. Bury St Edmunds, UK: Thames Valley Test Company. Nelson, H.E., & O’Connell, A. (1978). Dementia: The estimation of premorbid intelligence levels using the National Adult Reading Test. Cortex, 14, 234–244.

LITERACY AND COGNITIVE CHANGE

77

Ott, A., Breteler, M.M., Van Harskamp, F., Claus, J.J., Van der Cammen, T.J., Grobbee, D.E., & Hofman, A. (1995).

Prevalence of Alzheimer’s disease and vascular dementia: Association with education. The Rotterdam study.

British Medical Journal, 310, 970–973.

Parasuraman, R., & Haxby, J. (1993). Attention and brain function in Alzheimer’s disease: A review. Neuropsychology, 7, 242–272.

Prencipe, M., Casini, A.R., Ferretti, C., Lattanzio, M.T., Fiorelli, M., & Culasso, F. (1996). Prevalence of dementia in an elderly rural population: Effects of age, sex, and education. Journal of Neurology, Neurosurgery and Psychiatry, 60, 628−633.

Reis, A., & Castro-Caldas, A. (1997). Illiteracy: A cause for biased cognitive development. Journal of the International Neuropsychological Society, 3, 444−450.

Reis, A., Guerreiro, M., & Castro-Caldas, A. (1994). Influence of educational level of non-brain-damaged subjects on visual naming capacities. Journal of Clinical and Experimental Neuropsychology, 16, 939–942.

Rosen, W. (1981). The Rosen Drawing Test. Bronx, NY: Veterans Administration Medical Center.

Rosselli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in illiterates. II. Language and praxic abilities. Brain and Cognition, 12, 281–296.

Satz, P., Morgenstern, H., Miller, E.N., Selnes, O.A., McArthur, J. C., Cohen, B.A., Wesch, J., Becker, J.T., Jacobson, L., D’Elia, L.F., Van Gorp, W., & Visscher, B. (1993). Low education as a possible risk factor for cognitive abormalities in HIV-1: Findings from the Multicenter AIDS Cohort Study (MACS). Journal of Acquired Immune Deficiency Syndromes, 6, 503–511.

Scribner, S., & Cole, M. (1981). The psychology of literacy. Cambridge, MA: Harvard University Press.

Smith, J.P. (1984). Race and human capital. American Economic Review, 4, 685–698.

Smith, J.P., & Welch, F. (1977). Black-White male wage ratios: 1960–1970. American Economic Review, 67, 323–328.

Snowdon, D.A., Ostwald, S.K., & Kane, R.L. (1989). Education, survival and independence in elderly Catholic sisters, 1936–1988. American Journal of Epidemiology, 130, 999–1012.

Stern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve concept. Journal of the International Neuropsychological Society, 8, 448–460.

Stern, Y., Albert, S., Tang, M.X., & Tsai, W.Y. (1999). Rate of memory decline in AD is related to education and occupation: Cognitive reserve? Neurology, 53, 1942–1947.

Stern, Y., Andrews, H., Pittman, J., Sano, M., Tatemichi, T., Lantigua, R., & Mayeux, R. (1992). Diagnosis of dementia in a heterogeneous population. Development of a neuropsychological paradigm-based diagnosis of dementia and quantified correction for the effects of education. Archives of Neurology, 49, 453−460.

Stern, Y., Gurland, B., Tatemichi, T.K., Tang, M.X., Wilder, D., & Mayeux, R. (1994). Influence of education and occupation on the incidence of Alzheimer’s disease. Journal of the American Medical Association, 271, 1004–1010.

Stern, Y., Tang, M.X., Denaro, J., & Mayeux, R. (1995). Increased risk of mortality in Alzheimer’s disease patients with more advanced educational and occupational attainment. Annals of Neurology, 37, 590–595.

Sulkava, R., Wikstrom, J., Aromaa, A., Raitasalo, R., Lahtinen, V., Lahtela, K., & Palo, J. (1985). Prevalence of severe dementia in Finland. Neurology, 35, 1025–1029.

Teri, L., McCurry, S.M., Edland, S.D., Kukull, W.A., & Larson, E.B. (1995). Cognitive decline in Alzheimer’s disease: A longitudinal investigation of risk factors for accelerated decline. Journals of Gerontology: Biological Sciences and Medical Sciences, 50A, M49-M55.

Troster, A. I Butters, N., Salmon, D.P., Cullum, C.M., Jacobs, D., Brandt, J., & White, R.F. (1993). The diagnostic utility of savings scores: Differentiating Alzheimer’s and Huntington’s diseases with the Logical Memory and Visual Reproduction tests. Journal of Clinical and Experimental Neuropsychology, 15, 773–788.

Unverzagt, F.W., Hui, S.L., Farlow, M.R., Hall, K.S., & Hendrie, H.C. (1998). Cognitive decline and education in mild dementia. Neurology, 50, 181–185.

Vygotsky, L.S. (1962). Thought and language. Cambridge, MA: MIT Press.

Vygotsky, L.S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

78

MANLY ET AL.

Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised. New York: The Psychological Corporation.

Welch, F. (1966). Measurement of the quality of education. American Economic Review, 56, 379–392.

Welch, F. (1973). Black-White differences in returns to schooling. American Economic Review, 63, 893–907.

Welsh, K.A., Butters, N., Hughes, J.P., Mohs, R.C., & Heyman, A. (1992). Detection and staging of dementia in Alzheimer’s disease: Use of the neuropsychological measures developed for the Consortium to Establish a Registry for Alzheimer’s Disease. Archives of Neurology, 49, 448–452.

White, L., Katzman, R., Losonczy, K., Salive, M., Wallace, R., Berkman, L., Taylor, J., Fillenbaum, G., & Havlik, R. (1994). Association of education with incidence of cognitive impairment in three established populations for epidemiological studies of the elderly. Journal of Clinical Epidemiology, 47, 363–374.

Whitfield, K.E., & Baker-Thomas, T. (1999). Individual differences in aging minorities. International Journal of Aging and Human Development, 48, 73–79.

Wilkinson, G.S. (1993). Wide Range Achievement Test 3—Administration Manual. Wilimington, DE: Jastak Associates.

Zhang, M., Katzman, R., Salmon, D., Jin, H., Cai, G., Wang, Z., Qu, G., Grant, I Yu, E., Levy, P., Klauber, M.R., & Liu, W.T. (1990). The prevalence of dementia and Alzheimer’s disease in Shanghai, China: Impact of age, gender and education. Annals of Neurology, 27, 428–437.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 61–67

There is not any specific brain area for writing: From cavepaintings to computers Alfredo Ardila Florida International University, Miami, USA

In this paper it is pointed out that human brain adaptation was accomplished to survive in certain living conditions that existed long before classical civilizations did. It is argued that there is no brain area specialized for writing, but rather that writing relies on some basic abilities that existed long before writing was invented. Pre-writing was initially a visuoconstructive and ideomotor ability, and only later did it become the language-related ability of writing. It is also emphasized that most of the neuropsychological syndromes, including agraphia, were described during the late 19th and early 20th century, but living conditions have changed dramatically during the last 100 years. Writing no longer means only using pencil and paper, but using computer word processing programs. Writing using paper and pencil does not require the same cognitive, motor, and spatial tasks as those required when using a computer keyboard. Although the conceptual knowledge of written language can be the same, the motor activity and the spatial abilities that are used are rather different. It can be anticipated that new neuropsychological syndromes resulting from these new living conditions will be described in the future.

Cet article attire l’attention sur le fait que le cerveau humain s’est adapté afin de survivre dans des conditions de vie existant longtemps avant les civilisations classiques. Il est propose qu’il n’y a aucune zone cérébrale spécialisée pour l’écriture, mais plutôt que l’écriture repose sur certaines habiletés de base existant bien avant que l’écriture soit inventée. La pré-écriture était initialement une habileté visuoconstructive et idéomotrice et ce n’est que plus tard qu’elle est devenue une habileté reliée au langage (écriture). Par ailleurs, cet article met l’emphase sur le fait que la plupart des syndromes neuropsychologiques, incluant l’agraphie, furent décrits au cours de la fin du 19éme siècle et le début du 20ème siècle. Cependant, les conditions de vie ont énormément évoluées durant les 100 dernières années. Écrire ne représente plus uniquement le fait d’utiliser un papier et un crayon, mais également l’utilisation d’un programme de traitement de texte informatisé. L’écriture au moyen d’un papier et d’un crayon ne fait pas reference aux mêmes tâches cognitives, motrices et spatiales que l’écriture a partir d’un clavier d’ordinateur.

Quoique la connaissance conceptuelle du langage écrit puisse être la même, l’activité motrice et les habiletés spatiales utilisées sont différentes. Il peut être anticipé que, dans le futur, les syndromes neuropsychologiques resultant de ces nouvelles conditions de vie seront décrits.

En n este artículo se señala que la adaptación del cerebro humano fue lograda para sobrevivir bajo ciertas condiciones de vida mucho antes de las civilizaciones clásicas. Se discute que no existe un area especializada para la escritura, sino que, la escritura se basa en habilidades

80

ARDILA

básicas que han existido desde antes que la escritura fuera inventada. La pre-escritura fue inicialmente una habilidad visuoconstructiva e ideomotora, y después se convirtió en una habilidad relacionada con el lenguaje (escritura). Se enfatiza que la mayoría de los síndromes neuropsicológicos, incluyendo la agrafía, fueron descritas durante finales del siglo 19 y principios del siglo 20, sin embrago, las condiciones de vida han cambiado dramáticamente durante los últimos cien años. La escritura ya no sólo significa el uso de papel y lápiz sino también el uso de programas de procesamiento de texto. La escritura con lápiz y papel no requiere las mismas tareas cognitivas, espaciales y motoras que se necesitan cuando se utiliza un teclado de computadora. A pesar de que el concepto de lenguaje escrito pueda ser el mismo, la actividad motora y las habilidades espaciales que son usadas son totalmente diferentes. Se anticipa que en el futuro se puedan describir nuevos síndromes neuropsicológicos resultado de las nuevas condiciones de vida.

Anthropology has striven to adequately understand man’s living conditions 10,000, 20,000, or 100,000 years ago. The Stone Age (usually divided into the early Stone Age, or Palaeolithic, and the later Stone Age, or Neolithic) extends to approximately 6,000 to 7,000 years ago (Hours, 1982). Agriculture appeared some 10,000 years ago. The first cities appeared some 6000 years ago and the first civilizations appeared about 5000 years ago. Writing only has a 5000- or 6000-year history and arithmetical abilities have a history of about 6000 years (Childe, 1936; Sampson, 1985). However, it has been thought that contemporary man (Homo sapiens sapiens) has lived on earth for at least 50,000 years but perhaps, according to current evidence, it may be since at least 100,000 years ago. We can state with some certainty that, during this time, the structural changes of the brain in man have been minimal (Harris, 1983; Kochetkova, 1973; Tomasello, 2000). Human brain adaptation was for survival in Stone-Age life conditions (existing for about 98% of this life span) rather than in those life conditions existing nowadays. Only by departing from the analysis of these original conditions can we understand the specific characteristics and idiosyncrasies of the brain’s adaptation. It would seem reasonable for any neuroscientist to put the question: “What type of information did the human brain become adapted to process?” Consequently, “What are man’s basic cognitive abilities?” The search for universals has guided an important proportion of anthropological and linguistic activity during the last decades.

Anthropologists and linguists have attempted to find out basic and universal ways of social organization in different human groups (Van den Berghe, 1979) and fundamental language characteristics (Greenberg, 1978; Hagége, 1982). Attempts are made to infer the social organization of prehistoric man and languages existing before living languages. An excellent example of this last point is the reconstruction obtained for the Indo-European language (Anderson, 1973; Lehmann, 1974; Martinet, 1975), whose last speaker passed away several thousands of years ago. Efforts are currently being made to reconstruct even older protolanguages (Shevoroshkin, 1990).

Correspondence should be sent to Alfredo Ardila, PhD, 12230 NW 8 Street, Miami, Florida 33182, USA (E-mail: alfredoardila@cs.com). © 2004 International Union of Psychological Science http: //www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000295

THE BRAIN AND WRITING

81

Undoubtedly, neuropsychology has advanced tremendously in some specific areas. Significant advancements have been in the assessment of the sequelae of brain pathology and in the establishment of clinical/anatomical correlations. However, we do not yet sufficiently understand or know what can be considered as “basic cognitive abilities.” Our understanding of cultural differences is optimistic yet limited (Ardila, 1995; Fletcher-Janzen, Strickland, & Reynolds, 2000; Uzzell, Pontón, & Ardila, in press).

Matthews (1992), in his International Neuropsychological Society presidential address, accurately observed that” …a very limited kind of neuropsychology, appropriate to only a fraction of the world’s population, is presented to the rest of the world as if there could be no other kind of neuropsychology, and as if the education and cultural assumptions on which…neuropsychology is based were obviously universals that applied everywhere in the world” (p. 421).

Furthermore, most of the neuropsychological syndromes were described during the late 19th and early 20th century: aphasia (Broca, 1863; Wernicke, 1874), alexia (Déjèrine, 1891, 1892), agraphia (Exner, 1881), acalculia (Henschen, 1925), apraxia (Liepmann, 1900), spatial orientation disturbances (Jackson, 1874/1932), prosopagnosia (Bodamer, 1947), visuoconstructive disturbances (Henschen, 1925; Poppelreuter, 1917), and executive functioning defects (Harlow, 1868), among others. Nonetheless, living conditions have changed dramatically during the last 100 years. Writing no longer means only the use of a pencil and a paper, but also the use of a computer word processor program. Arithmetical abilities, too, have changed; instead of writing numbers down on paper and applying certain computational rules, we more often require the ability to use a pocket calculator. One major source of knowledge of other people’s faces is through television, and a major source of knowledge of other people’s voices is through the telephone.

Intensive exposure to these media has been observed only over the last few decades.

The need for a clear understanding of the origins of current cognitive abilities is evident (Ardila, 1993a, 1993b, 1993c, 1993d). The example of reading and writing may be illustrative of the need for a historical/ anthropological analysis of neuropsychological syndromes. Varney (2002) points out that reading is a cultural, not an evolutionary, development. He emphasizes that “our capacity of reading did not evolve biologically; it evolved through cultural developments that were only acquired as ‘typical’ human abilities within the last 200 years in Europe and America, and only after World War II in the rest of the World” (p.

3). The origins of reading can be found in certain abilities that existed long before reading was developed.

Reading and writing were far from “universal” even at the beginning of the 21st century. According to the United Nations, “a person who is literate can, with understanding, both read and write a short simple statement on his or her everyday life…. A person is functionally literate who can engage in all of those activities in which literacy is required for effective function of his or her group and community and also for enabling him or her to continue to use reading, writing, and calculation for his or her own and the community’s development” (UNESCO, 2003). Surveys throughout the world have been conducted to observe populations speaking various languages and their inability to read or write a simple message. In the first survey (1950), at least 44% of the world’s population were found to be illiterate. A 1978 study showed the rate to have dropped to 32.5%. In 1990 illiteracy worldwide dropped to about 27%, and by 1998 to 16%. However, a study by the United Nations Children’s Fund (UNICEF) published in 1998 predicted that the world illiteracy rate would increase in the 21st century because only a quarter of the world’s children were in school by the end of the 20th century. The highest illiteracy rates were found in the less developed nations of Africa, Asia, and South America. The lowest illiteracy rates were found in Australia, Japan, North Korea, and the more technologically advanced nations of Europe and North America. Currently, there are an estimated 862 million illiterate adults in the world, of whom about two thirds are women (UNESCO, 2003). The mean educational level of contemporary man is only about 3–4 years of school!

82

ARDILA

Varney (2002) analysed the origins of reading ability. He suggested that the ancient skills of gesture comprehension and animal tracking were the underpinnings of brain organization that permitted reading to occur. He demonstrated that alexia is significantly associated with impaired pantomime and animal footprint recognition. Thus, these abilities, existing since early human history, were prerequisites that led the way to the cultural development of reading. Gesture recognition may have existed for several millions of years, but reading developed just a few millennia ago.

HOW DID WRITING APPEAR? Wall paintings appeared during the Palaeolithic era, some 30–35,000 years ago (Childe, 1936). Across Europe, particularly in France and Spain, cave paintings dating from the Palaeolithic age have been found.

Mainly animals, but also people, instruments, and environmental conditions, are represented in these paintings. Further evolution in pre-writing is represented by paintings becoming standardized for representing specific elements (i.e., a standard bird means “bird”). Lecours, Peña-Casanova, and Ardila (1998) point out that writing begins with concrete pictograms that reflect realities accessible to the senses, particularly to vision. These pictograms further evolved and became abstract, progressively separating from the concrete representation. This situation was observed in Sumer (contemporary Iraq) about 53 centuries ago, and it is usually regarded as the beginning of writing in human history. Symbols (graphemes) referred to the meaning of the words, so these original writing systems are regarded as logographic. Graphemes representing sounds (syllables) appeared later, about 4000 years ago in Phoenicia (Sampson, 1985), and graphemes representing phonemes appeared even later in Greece.

The sequence of the evolution of writing in consequence was: Drawings→pictograms→logograms→syllabic graphemes→phonemic graphemes Writing systems can be divided in different ways; a major distinction between logographic (representing meanings) and sonographic (representing sounds) systems can be established (OMNIGLOT, 2003; Sampson, 1985).

The fundamental difference between logographic writing systems and other scripts is that each logographic symbol means something. As a result, logographic writing systems generally contain a large number of symbols: anything from several hundred to tens of thousands. In fact there is no theoretical upper limit to the number of symbols in some logographic scripts, such as Chinese. Logographic scripts may include the following types of symbols.

1. Logograms—symbols that represent parts of words or whole words. Some logograms resemble the things they represent and are sometimes known as pictograms or pictographs.

2. Ideograms—symbols that graphically represent abstract ideas.

3. Semantic-phonetic compounds—symbols that include a semantic element, which represents or hints at the meaning of the symbol, and a phonetic element, which denotes or hints at the pronunciation.

4. Sometimes symbols are used for their phonetic value alone, without regard for their meaning.

In sonographic writing systems, syllables (syllabic alphabets) or phonemes (phonemic alphabets) can be used. Alphabetic writing systems come in two varieties.

1. Abjads (consonant alphabets) represent consonants only, or consonants plus some vowels. Even though not common, full vowel indication (vocalization) can be added, usually by means of diacritics.

2. Alphabets (phonemic alphabets) represent consonants and vowels.

THE BRAIN AND WRITING

83

Thus, initial writing (or rather, pre-writing) was a visuoconstructive ability (i.e., representing external elements visually), and only later did it become an ideomotor praxis ability (i.e., making certain learned and fixed sequences of movements with the hand to create a pictogram—a standardized representation of external elements). Still later, after writing became an ideomotor praxis ability, it became a linguistic ability (i.e., associating the pictogram with a word, and further analysing the word in its constituting sounds). It is not surprising that three major disorders in writing can be observed as a result of brain pathology: visuoconstructive (spatial or visuospatial agraphia), ideomotor (apraxic agraphia), and linguistic (aphasic agraphia). In addition, of course, writing requires visual and motor integrity.

HOW MANY PEOPLE CAN WRITE? Even though writing began several millennia ago, as recently as the 1950s about half of the world’s population was illiterate. The percentage of illiteracy dramatically increases as we go back in time, and up to only a couple of centuries ago, the overwhelming majority were illiterate. Until the 15th century, when the printing press was invented, writing may well have been limited to a few intellectual people and monks.

Even though there are no statistics available, it may be conjectured that 99% or more of the population was illiterate. Furthermore, it has to be kept in mind that the mean level of education of contemporary man is about 3 years of school, which may not be enough to develop automatic reading and writing.

It is evident that writing represents an unusual ability in humans. The overwhelming majority of members of our species who have lived could not read or write. Reading and writing is obviously far from being a “primary” or “biologically based” cognitive ability. Clearly, writing represents a cognitive ability that depends on the human cultural evolution (Vygotsky, 1962).

AGRAPHIA AS A NEUROPSYCHOLOGICAL SYNDROME Agraphia can be defined as the partial or total loss of the ability to produce written language, and is associated with brain pathology. The ability to write can be impaired as a result of linguistic defects (aphasia), but other elements not related to language (e.g., motor and spatial) also participate in the writing ability. It supposes at least a knowledge of the language codes (phonemes, words), an ability to convert language sounds in graphemes, a knowledge of the graphemic system (alphabet), an ability to perform fine movements, and an appropriate use of the space for distributing, joining, and separating letters. It is evident that diverse types of writing disturbances can be found in clinical practice.

Different attempts to classify writing disturbances are found in the history of neuropsychology. Goldstein (1948) distinguished two major types of agraphia: apractoamnesic and aphasic-amnesic. Luria (1976, 1980) referred to five different types of agraphia, three of them associated with aphasia (sensory agraphia, afferent motor agraphia, and kinetic agraphia) and two associated with visuospatial defects. Hécaen and Albert (1978) distinguished four types of agraphia: pure, apraxic, spatial, and aphasic.

Regardless of the diversity of classifications of agraphia, a basic distinction can be established between (1) agraphias due to a language impairment (linguistic or aphasic agraphias), (2) agraphias due to other types of impairments (most often, motor or spatial) disturbing the normal ability to write (Benson & Ardila, 1996), or simply (3) central and peripheral agraphias (Ellis, 1988). In the first case, agraphia is just a secondary manifestation of the aphasic syndrome. In the second, it can be interpreted as a result of a broader visuoconstructive/visuospatial impairment (Ardila & Rosselli, 1993), or motor-apraxic disturbance (Hécaen & Albert, 1978). Consequently, writing can be interpreted as a particular type of cross-modal learning. Certain visuo-constructive and ideomotor abilities become associated with language.

84

ARDILA

IS ANY AREA IN THE BRAIN SPECIALIZED FOR WRITING? Writing is a “functional system” (Luria, 1976) that requires, and is based on, some more fundamental abilities: praxis abilities (i.e., learning sequences of movements required to write the letters), spatial abilities (distributing letters and words in the space, and understanding the value of space in writing), constructional abilities (reproducing a model using certain movements), and obviously, the knowledge of the language and the association between verbal auditory elements and visual symbols.

Varney (2002) refers to the anthropological concept of pre-adaptation, which states that the evolution of a structure for one purpose can enable that structure to perform another purpose. This must be true for writing (as for any other abilities depending on cultural evolution). Visuo-constructive and ideomotor abilities represent prerequisites for writing; they are probably related to the ability to make tools and weapons and generally to use the hands in a skilled way.

FROM “AGRAPHIA" TO “DYSTYPIA" Contemporary literate man is using handwriting less and less, and relying on computers more and more. In an informal survey to 40 people with a college-level education background, they reported using a computer about 90% of the time when writing and handwrote only 10% of the time. Obviously, this sample does not represent all of humankind, and computers are not accessible to a large percentage of the human population.

But this sample seems to illustrate the way in which writing is evolving: from handwriting to typing on a computer.

Handwriting and using computers represent significantly different cognitive and motor abilities. During handwriting, fingers are maintained in a relatively steady position while the hand moves. In typing, the opposite pattern is observed. When typing, the right hand does not move from one side to the other and back as in handwriting, but the hands remain relatively stationary and only the fingers are moved. Letters are not written but selected. Both hands have to be used in a similar way when typing. Because of using both hands, we have to assume that a major interhemispheric integration is required. It is obvious to assume that right-hemisphere lesions located in the frontal and parietal areas should significantly impair the typewriting ability of the left hand.

Similarly, the use of the space is different. The normal spatial distribution of the words on the page is automatic on the computer and, hence, writing in this way cannot be spatially disorganized, as may be the case in handwriting. By the same token, letters are neatly written and easily recognizable. When typing, we are not using a space that is directly manipulated with the hands (“constructional space”), but only a “visual space.” Furthermore, typing is not a constructional task (we do not have to construct the letters) but rather a motor-spatial task.

Many people type using a spatial memory for the position of the letters in the keyboard. This is a type of memory not required in handwriting, and it probably depends on right hippocampal and parietal activity (Moser, Hollup, & Moser, 2002). Other people have to look at the keys to select the letters when typing. In this case, literal reading is a prerequisite for writing. Letters have to be recognized visually before they are written. In handwriting, we use a mental representation of the visual form of the letters. Interestingly, few people—if any, regardless of how well they can type—are able to reproduce (i.e., describe verbally or by drawing) how the different letters are arranged on the keyboard. Memory for their location seems to be a purely spatial and motor memory of which we are poorly aware.

For typing some special symbols (e.g., interrogation marks) and letters (the Spanish N), some relatively sophisticated motor manoeuvres are required, sometimes requiring the use of special keys or sequences of movements. In handwriting, however, special symbols are written using the mental forms that we have

THE BRAIN AND WRITING

85

learned. When typing, if a letter needs to be lower or upper case, a key has to be pushed. No other change to the movement is made. We can also select different writing styles and letter sizes using some special commands and menus, all without changing the sequences of the hand movements.

In cases of brain damage, how is typewriting altered? To the best of my knowledge, only one case of agraphia for typewriting has been published (Otsuki, Soma, Arihiro, Watanabe, Moriwaki, & Naritomi, 2002). Nonetheless, it can be assumed that different types of brain pathology may affect the ability for typing on a computer word processor. The following can be conjectured.

1. An anterior callosal lesion would impair the ability to coordinate the movements between the hands.

Furthermore, the left hand would be isolated from the linguistic left hemisphere, and would be unable to write. Left-hand hemiagraphia in callosal lesions has been observed (Benson & Ardila, 1996).

2. By the same token, it has been observed that damage in the supplementary motor area results in disturbances in the coordinated movements between both hands (Middleton & Strick, 2001). We can anticipate supplementary motor area typing agraphia.

3. Spatial memory disturbances should result in difficulties in recalling the positions of the letters on the keyboard. Typing would be slow, and would require a continual search for the letters.

Otsuki et al. (2002) reported on a 60-year-old right-handed Japanese man who showed an isolated persistent typing impairment without aphasia, agraphia, apraxia, or any other neuropsychological deficit.

They proposed the term “dystypia” for this peculiar neuropsychological manifestation. The symptom was caused by an infarction in the left frontal lobe involving the foot of the second frontal convolution and the frontal operculum. The patient’s typing impairment was not attributable to a disturbance of the linguistic process, since he had no aphasia or agraphia. Nor was it attributable to an impairment of the motor execution process, since he had no apraxia. Thus, it was deduced that his typing impairment was based on a disturbance of the intermediate process where the linguistic phonological information is converted into the corresponding performance. The authors hypothesized that the foot of the left second frontal convolution and the operculum may play an important role in the manifestation of “dystypia.”

Using a computer is somehow “equivalent” to a new writing system. Obviously, there is no brain area related to typing on a computer, as there is no brain area related to reading and writing. These are cultural and technological elements recently developed through human evolution. Rather, there are basic cognitive abilities (pre-adaptative abilities) that are required for the use of these new cultural elements: e.g., certain visuoperceptual abilities and cross-modal associations for reading, phonological awareness and some fine movements for writing, etc. Using computers is notoriously more complex, yet we can assume a “functional system” participating in their use.

It can be conjectured that using computers requires at least the following abilities.

1. A conceptual ability (executive functioning) to understand the principles governing the functioning of a computer.

2. Some visuoperceptual abilities to recognize icons, windows, etc.

3. Some skilled movements to type on the keyboard and manoeuvre the mouse correctly.

4. Some spatial abilities to handle the working space (monitor screen).

5. Some memory abilities to learn programs, to use the spatial position of the keys, etc.

Obviously, the ability to use computers can potentially be disrupted as a consequence of a failure in any one of these abilities (“acumputuria syndrome”). In the future, apart from “dystypia,” more complex disturbances in the ability to use computers will probably be established.

86

ARDILA

CONCLUSION The origins of writing can be traced back to cave paintings. Writing (or pre-writing) was initially a visuoconstructive ability, later involving some stereotyped movements to represent pictograms, and finally involving spoken language. It makes sense, therefore, that the ability to write can be disturbed in three major forms: as a visuospatial/ visuoconstructive dexterity, as an ideomotor skill, and as a linguistic ability.

Writing has followed a long evolution since cave painting during the Palaeolithic times. Different strategies have been used to represent spoken language visually (ideograms, alphabets, etc). Writing, however, has continued to evolve since its initial invention. The use of punctuation marks and the distinction between upper and lower case in writing—to mention just two examples—are relatively recent in history (Sampson, 1985). Evolution has continued with the development of different technical instruments for writing: the feather, the pencil, the typewriter, and the computer. Brain representation of written language has necessarily changed in some way, too. Neuropsychological syndromes associated with brain pathology have evolved over time. We can assume that the consequences of brain pathology in a Palaeolithic man were not the same as for a 19th-century individual (when agraphia was first described), or for contemporary man or woman (some of whom frequently spending most of their working day in front of a computer screen). It can be anticipated that in the future new neuropsychological syndromes resulting from new living conditions will be described.

REFERENCES Anderson, J.M. (1973). Structural aspects of language change. London: Longman.

Ardila, A. (1993a). Introduction: Toward a historical/ anthropological approach in neuropsychology. Behavioural Neurology, 6, 71–74.

Ardila, A. (1993b). Historical evolution of spatial abilities. Behavioral Neurology, 6, 83–88.

Ardila, A. (1993c). People recognition: A historical/ anthropological perspective. Behavioural Neurology, 6, 99–106.

Ardila, A. (1993d). On the origins of calculation abilities. Behavioural Neurology, 6, 89–98.

Ardila, A. (1995). Directions of research in cross-cultural neuropsychology. Journal of Clinical and Experimental Neuropsychology, 17, 143–150.

Ardila, A., & Rosselli, M. (1993). Spatial agraphia. Brain and Cognition, 22, 75–95.

Benson, D.F., & Ardila, A. (1996). Aphasia: A clinical perspective. New York: Oxford University Press.

Bodamer, J. (1947). Die Prosopagnosie. Archiv für Psychiatrie und Nervenkrankheiten, 179, 6−54.

Broca, P. (1863). Localization des fonctions cérébrales: Siege du langage articulé. Bulletin de la Societé d’Anthropologie, 4, 200–203.

Childe, V.G. (1936). Man makes himself. London: Pitman Publishing.

Déjèrine, J. (1891). Sur un cas de cécité verbale avec agraphie suivi d’autopsie. Comptes rendus. Societe de Biologie, 3, 197−201.

Déjèrine, J. (1892). Contribution a 1’etude anatomo-pathologique et clinique des differérents varietes de cécité verbale.

Comptes Rendus, Societe de Biologie, 4, 61–90.

Ellis, A.W. (1988). Normal writing processes and peripheral acquired dysgraphias. Language and Cognitive Processes, 3, 99–127.

Exner, S. (1881). Unersuchungen uber die lokalisation der Functionen in der Grosshimirinde des Menschen. Wien: Braumuller.

Fletcher-Janzen, E., Strickland, T.L., & Reynolds, C.R. (Eds.). (2000). Handbook of cross-cultural neuropsychology.

New York: Kluwer Academic/ Plenum Publishers.

Goldstein, K. (1948). Language and language disturbances. New York: Grune & Stratton.

Greenberg, J.H. (1978). Universals of human language. Stanford, CA: Stanford University Press.

THE BRAIN AND WRITING

87

Hagége, C. (1982). La structure des langues. Paris: Presses Universitaires de France.

Harlow, J.M. (1868). Recovery from the passage of an iron bar through the head. Massachusetts Medical Society Publication, 2, 327–346.

Harris, M. (1983). Cultural anthropology. New York: Harper & Row. Hécaen, H., & Albert, M.L. (1978). Human neuropsychology. New York: Wiley.

Henschen, S.E. (1925). Clinical and anatomical contributions on brain pathology. Archives of Neurology and Psychiatry, 13, 226–249.

Hours, F. (1982). Les civilisations du Paléolithique. Paris: Presses Universitaires de la France.

Jackson, J.H. (1932). Selected writings. London: Hodder & Stoughton. (First published 1874).

Kochetkova, V.I. (1973). Paleoneurology [in Russian]. Moscow: Moscow State University Press.

Lecours, A.R., Peña-Casanova, J., & Ardila, A. (1998). Orígenes y evolución de la escritura [Origins and evolution of writing]. In A.R.Lecours, J.Peña-Casanova, & F.Diéguez-Vide (Eds.), Dislexia y disgrafia: Teoría, formas clínicas y exploración (pp. 1–8). Barcelona: Masson.

Lehmann, W.P. (1974). Proto-Indian-European syntax. Austin, TX: University of Texas Press.

Liepmann, H. (1900). Das Krankheitsbild der Apraxie (motorische Asymbolie) auf Grund aines Falles von einseitiger Apraxie. Monatschrie Psychiatrie Neuroligie, 10, 214–227.

Luria, A.R. (1976). Basic problems of neurolinguistics. The Hague: Mouton.

Luria, A.R. (1980). Higher cortical functions in man. New York: Basic Books.

Martinet, A. (1975). Evolution de langues et reconstruction. Paris: Presses Universitaires de France.

Matthews, C.G. (1992). Truth in labeling: Are we really an international society? Journal of Clinical and Experimental Neuropsychology, 14, 418–426.

Middleton, F.A., & Strick, P.L. (2001). A revised neuroanatomy of frontal-subcortical circuits. In D.G.Lichter & J.L.Cummings (Eds.), Frontal-subcortical circuits in psychiatry and neurological disorders (pp. 44–58). New York: Guilford Press.

Moser, E. I .,Hollup, S.A., & Moser, M.B. (2002). Representation of spatial information in dynamic neuronal circuits in the hippocampus. In L.R. Squire & D.L. Schacter (Eds.), Neuropsychology of memory (pp. 361–376). New York: Guilford Press.

OMNIGLOT. (2003). Retrieved May 1, 2003, from www.omniglot.com.

Otsuki, M., Soma, Y., Arihiro, S., Watanabe, Y., Moriwaki, H., & Naritomi, H. (2002). Dystypia: Isolated typing impairment without aphasia, apraxia or visuospatial impairment. European Neurology, 47, 136–140.

Poppelreuter, W. (1917). Die psychischen Schadigungen durch Kopfschuss im Kriege 1914–1916: Die Storungen der neideren und hoheren Schleitungen durch Verletzungen des Okzipitalhirns. Leipzig, Germany: Voss.

Sampson, G. (1985). Writing systems. Stanford, CA: Stanford University Press.

Shevoroshkin, V. (1990). The mother tongue: How linguistics have reconstructed the ancestor of all living languages.

The Sciences, May/June, 20–27.

Tomasello, M. (2000). The cultural origins of human cognition. Cambridge, MA: Harvard University Press.

UNESCO. (2003). Retrieved May 1, 2003, from www. portal.unesco.org.

Uzzell, B., Pontón, M., & Ardila, A. (Eds.) (in press). International handbook of cross-cultural neuropsychology.

Mahwah, NJ: Lawrence Erlbaum Associates Inc. Van den Berghe, P.L. (1979). Human family systems. New York: Elsevier.

Varney N.R. (2002). How reading works: Considerations from prehistory to present. Applied Neuropsychology, 9, 3–12.

Vygotsky, L.S. (1962). Thought and language. Cambridge, MA: MI T Press.

Wernicke, C. (1874). Der aphasiche Symptomencomplex. Breslau, Germany: Cohn & Weigert.

INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 68–70

INTERNATIONAL PLATFORM FOR PSYCHOLOGISTS

Congresses and scientific meetings

February 28–March 3, 2004 Compositionality, Concepts and Cognition: An Interdisciplinary Conference in Cognitive ScienceLocation: Düsseldorf, GermanyInformation: http:// www.phil.uni-duesseldorf.de/thphil/ compositionalityMarch 11–14, 2004Society for Research on Adolescence, 10th BiennialMeetingLocation: Baltimore, Maryland, USAContact: Thelma Tucker, Program Operations Manager, SRA, 3131 S. State St, Suite 302, Ann Arbor, MI 48101– 1623, USA;Fax: +1–734–998–6569;Email: tetucker@umich.eduURL: www.s-r-a.orgMarch 18–19, 2004Behavior Analysis Association of Michigan AnnualMeetingsLocation: Ypsilanti, Michigan, USAContact: http://www.baam.emich.edu/ baammainpages/confer1.HTMMarch 24–27, 20042004 Society for Behavioral Medicine AnnualMeeting and Scientific SessionsLocation: Baltimore, Maryland, USAURL:www.sbm.org/annualmeeting/ index.htmlMarch 24–27, 2004XIV. IFTA World Family Therapy CongressLocation: Istanbul, TURKEYInformation: www.ifta2004.orgApril 1–3, 200414th Biennial Meeting of the Society for Research in Human Development (formerly the Southwestern Society for Research in Human Development SWSRHD)Location: Park City, Utah, USACall for Papers: http://www.fcs.iastate.edu/ Swsrhd/2004/ default.htmURL: http://www.fcs.iastate.edu/swsrhd/Submission Deadline November 15, 2003April 1–4, 20047th European Conference on PsychologicalAssessment Location: Malaga, SPAINContact: Antonio Godoy, Facultad de Psicologia, Universidad de Malaga, 29071

INTERNATIONAL PLATFORM FOR PSYCHOLOGISTS

89

Malaga. (SPAIN)Tel. (34) 952 13 25 32; Fax (34) 95213 11 00E-mail: godoy@uma.es URL: http:// www.uma.es/petra/eapa2004/ index2.htmlApril 2–4, 2004Annual Conference Society for Industrial/Organizational Psychology (SIOP)Location: Chicago, Illinois, USAContact: lhakel@sipp.bgsu.eduURL: http://www.siop.orgApril 3, 2004The Twenty-First Annual Claremont Symposium on Applied SocialPsychology: Stigma and Group Inequality: Social Psychological ApproachesLocation: Claremont, California, USAInformation: Program Chairs: Shana Levin shana.levin@claremontmckenna.edu & Colette van Laar cvlaar@fsw.leidenuniv.nlApril 15–17, 2004British Psychological Society (BPS) Annual ConferenceLocation: London, UKContact: www.bps.org.ukDeadline for submissions: 20 October 2004April 15–18, 2004Society of Australasian Social Psychologists 33rd Annual MeetingLocation: Auckland, NEW ZEALANDContact: SASP@auckland.ac.nz URL: http:// www.psych.auckland.ac.nz/psych/SASP Deadline for submissions: 1 February 2004April 19–22, 2004WFPHA 10th International Congress on Public Health: Sustaining Public Health in a Changing World: Vision to ActionLocation: Brighton, UKContact: Allen K.Jones, PhD; Secretary General World Federation of Public Health Associations,Email: stacey.succop@apha.orgApril 23–25, 200418th Biennial Conference on Human Development (CHD 2004)Location: Washington, DCContacts: Susanne Denham (sdenham@gmu.edu); Elyse Lehman (elehman@gmu.edu); Adam Winsler (awinsler@gmu.edu)URL: http://adp.gmu.edu/CHD2004Submission Deadline: October 1, 2003 May 5–8, 2004XIVth Biennial International Conference on Infant StudiesLocation: Chicago, Illinois, USAInformation: http://sheu-hp.psy.depaul.edu /~gmichel/isis/ home.htmlDeadline for submission Nov 3, 2003May 6−7, 2004International Meeting for Autism Research (IMFAR)Location: Sacramento, California, USAContact: http://www.cevs.ucdavis.edu/ Cofred/Public/ Aca/WebSec.cfm ?confid=177&webid=829May 15–19, 200427th National AACBT Conference (Australian Association for Cognitive and Behavior Therapy)Location: Perth, Western Australia, AUSTRALIAContact: http://www.aacbt.orgMay 27–30, 200416th Annual Convention of the American Psychological SocietyLocation: Chicago, Illinois, USAContact: mweiner@aps.washington.dc.usURL: convention@aps.washington.dc.usJune 1–5, 2004American Association on Mental RetardationLocation: Chicago, Illinois, USAContact: AAMR: Tel: +1–202–387–1968; 1–800–424–3688; Fax: 202/387–2193URL: http://www.aamr.org/Events/ annual_meeting.shtmlJune 3–5, 2004Meetings of the Jean Piaget SocietyLocation: Toronto, Ontario, CANADAURL: www.piaget.orgJune 9–12 2004Small Group Meeting of the European Association of Experimental SocialPsychology: Understanding the Academic Underachievement of Low Status Group MembersLocation: Paris, FRANCEResearchers who are interested in participating in the meeting are invited to submit by the end of January 2004 a 1000-word summary of their proposed presentation to Jean-Claude Croizet (Laboratoire de Psychologie Sociale de la Cognition, Université Blaise Pascal, 34 avenue

90

INTERNATIONAL PLATFORM FOR PSYCHOLOGISTS

Carnot, 63000 Clermont-Ferrand, France) preferably byEmail: croizet@srvpsy.univ-bpclermont.frJune 16– 19, 20043rd Annual Hawaii International Conference on Social SciencesLocation: Honolulu Hawaii, USAContact: social@hicsocial.orgURL: http:// www.hicsocial.orgDeadline: Submissions—January 27, 2004June 22–26, 2004Sixth International Conference of the Learning Sciences (ICLS 2004): Embracing Diversity in the Learning SciencesLocation: Santa Monica, California, USAInformation: http://www.gseis.ucla.edu/ ~icls/ Paper and Poster Submission Deadline: November 15, 2003June 25–27, 2004Society for the Psychological Study of Social Issues (SPSSI) ConventionLocation: Washington, DC, USAContact: SPSSI web pageJune 28—July 1, 2004Head Start’s Seventh National Research ConferenceLocation: Washington, DC, USAContact: Dr. Faith Lamb-Parker, Scientific Director,Tel 212–305–4154;Email: flp1@columbia.eduURL: www.headstartresearchconf.netJuly 2–4, 200411th International Congress on Twin Studies (ICTS) 2004Location: Odense, DENMARKContact: http://www.icts2004.sdu.dkJuly 11–15, 200418th Biennial Meeting of the International Society for the Study of Behavioural Development (ISSBD)Location: Ghent, BELGIUMContact: Leni Verhofstadt-Deneve,Email: issbd@rug.ac.beURL: http://allserv.rug.ac.be/ISSBD2004July 27–31, 200427th Annual Colloquium of the International School Psychology Association (ISPA)Location: Exeter, UKContact: URL www.ispaweb.orgJuly 28—August 1, 2004112th Annual Convention of the American Psychological Association (APA) Location:Honolulu, Hawaii, USAContact: Convention Office, APA, 750 First Street NE,Washington DC 20002–4242 USATel: +1–202–336–5500URL: www.apa.org/ conventionJuly 29–August 1, 2004Sixth International Conference of Cognitive Modeling- ICCM-2004Location: Pittsburgh, Pennsylvania, USAContact: Email: iccm@pitt.edu.URL: http://simon.lrdc.pitt.edu/~iccmDeadline: Submissions 1 April 2004August 2–6, 200417th Congress of the International Association of Cross-Cultural Psychology (IACCP)Location: Xi’an, CHINAContact: Zheng Gang, Institute of Psychology, Chinese Academy of Sciences, 100101 Beijing, ChinaEmail: iaccp2004@psych.ac.cnURL: www.iaccp2004.orgAugust 3–6, 200462nd Annual Conference of the International Council of PsychologistsLocation: University of Jinan, Jinan, CHINAContact: Dr. Natividad Dayan, Scientific Chair, 99 General Ave, GSIS Village, Project 8 Quezon City, Metro Manila 01108 PhilippinesTel: 632–724–5358;Email: bereps@pacific.net.phAugust 8–13, 2004XXVIII International Congress of PsychologyLocation: Beijing, CHINAContact: XiaoLan Fu, Deputy Director, Committee forInternational Cooperation, Chinese Psychological Society, Institute of Psychology, Chinese Academy of Sciences, P.O. Box 1603, Beijing 100101, ChinaTel: ±86−10−6202−2071; Fax: +86–10–6202–2070URL: www.icp2004.psych.ac.cnAugust 18–20, 200422nd Nordic Congress of Psychology: Psychology in a World of Change and Diversity—Challenges for our ProfessionLocation: Copenhagen, DENMARKContact: Roal Ulrichsen, Chair, NPK2004 Organizing Committee, Danish Psychological Association, Stokholmsgade 27, DK-2100 Copenhagen Ø Denmark;Email pl@dp.dkURL: www.npk2004.orgNovember 18–21, 200445th Psychonomic Society Annual MeetingLocation: Minneapolis, Minnesota,

INTERNATIONAL PLATFORM FOR PSYCHOLOGISTS

91

USAContact: http://www.psychonomic.org/meet.htmNovember 18–21, 200438th Annual Conference, Association for Advancement of Behavior Therapy (AABT)Location: New Orleans, Louisiana, USAInformation: www.aabt.org2005 and beyondApril 7–10, 2005Biennial Meeting of the Society for Research in Child DevelopmentLocation: Atlanta, Georgia, USAContact: srcd@umich.eduApril 13–16, 20052005 Society for Behavioral Medicine Annual Meeting and Scientific SessionsLocation: Boston, Massachusetts, USAURL: www.sbm.org/ annualmeeting/index.htmlApril 15–17, 2005Annual Conference Society for Industrial/ Organizational Psychology (SIOP)Location: Los Angeles, California, USAContact: lhakel@siop.bgsu.eduURL: http://www.siop.orgMay 26–29, 200517th Annual Convention of the American Psychological SocietyLocation: Los Angeles, California, USAContact: mweiner@aps.washington.dc.usURL: convention@aps.washington.dc.us.washington.dc.usJuly 3–8, 20059th European Congress of PsychologyLocation: Granada, SPAINURL: www.efpa.orgAugust 18– 21, 2005113th Annual Convention of the American Psychological Association (APA)Location: Washington, DC, USAContact: Convention Office, APA, 750 First Street NE, Washington DC 20002–4242 USATel: +1–202–336–5500URL: www.apa.org/conventionSeptember 20– 24, 20051st Congress of the International Society for Cultural and Activity Research (ISCAR)Location: Seville, SPAINContact: iscar2005@iscar.orgNovember 10–13, 200546th Psychonomic Society Annual MeetingLocation: Toronto, Ontario, CANADAContact: http://www.psychonomic.org/meet.htmNovember 17–20, 200539th Annual Conference, Association for Advancement of Behavior Therapy (AABT)Location: Washington, DC, USAInformation: www.aabt.orgMarch 22–25, 2006Society for Behavioral Medicine Annual Meeting and Scientific Sessions Location: San Francisco, California, USAURL: www.sbm.org/annualmeeting/index.htmlJuly, 200626th International Congress of Applied PsychologyLocation: Athens, GREECEContact: Prof. James Georgas, Department of Psychology, School of Philosophy, University of Athens, Panepistemiopolis, Athens 15784 Greece;Tel. 30 1 7277524, FAX 30 1 7277534Email: dgeorgas@dp.uoa.grURL: http://www.erasmus.gr/dynamic/ conventions.asp?conv_id=21August 10–13, 2006114th Annual Convention of the American Psychological Association (APA)Location: New Orleans, Louisiana, USAContact: Convention Office, APA, 750 First Street NE, Washington DC 20002–4242 USATel: +1–202–336–5500URL: www.apa.org/conventionMarch 29— April 1, 2007Biennial Meeting of the Society for Research in Child DevelopmentLocation: Boston, Massachusetts, USAContact: srcd@umich.eduAugust 16–19, 2007112th Annual Convention of the American Psychological Association (APA)Location: San Francisco, California USAContact: Convention Office, APA, 750 First Street NE, Washington DC 20002–4242 USATel: ±1–202–336–5500URL: www.apa.org/conventionJuly 20–25, 2008XXIX International Congress of PsychologyLocation: Berlin, GERMANYContact: Ralf Schwarzer, Organizing Committee Chair, Freie Universität Berlin, Abteilung für

92

INTERNATIONAL PLATFORM FOR PSYCHOLOGISTS

Gesundheitspsychologie, Habelschwerdter Allee 45, 14195 Berlin, GermanyFax: +49 30 838 55634Email: health@zedat.fu-berlin.deURL: http:// www.icp2008.de

* Please send details of forthcoming events as far in advance as is possible to Dr Merry Bullock, Deputy SecretaryGeneral, International Union of Psychological Science and Associate Editor of the International Journal of Psychology, Science Directorate, APA, 750 First Street NE, Washington DC 20002, USA; Email: mbullock@apa.org; bullock@aca.ee; URL: http://www.iupsys.org