NEUROIMAGING STUDIES OF LITERATE AND ILLITERATE SUBJECTS

between the literate and illiterate group were generally independent of whether performance was included in the analysis or not (Petersson et al., 2000). Comparing the PET data between groups (Figure 5.7) suggested that there was a more prominent left-sided inferior parietal (BA 40) activation in the words vs. pseudowords comparison in the literate group. In the reverse comparison (pseudowords vs. words), the literate group displayed a significant activation in the anterior insular cortex (BA 14/15) bilaterally and in the right inferior frontal/frontal opercular cortices (BA 44/45/47/49), left perigenual anterior cingulate cortex (BA 24/32), left basal ganglia, midline anterior thalamus and midline cerebellum. In the illiterate group, significant activation was only observed in the middle frontal/frontopolar region (BA 10). These results represented the first indication that the functional architecture of auditory-spoken language processing is influenced by literacy.

This suggests that there exists a relation between the acquisition of alphabetic orthographic knowledge and aspects of phonological processing in terms the functional brain organization, consistent with behavioral findings outlined above.

5.3.1 A NETWORK ANALYSIS OF IMMEDIATE VERBAL REPETITION IN LITERATE AND ILLITERATE SUBJECTS

Complementary to the approach and results outlined in the previous section is to take a network perspective on cognitive brain function. In general, information is thought to be represented as distributed activity in the brain while information processing, subserving cognitive brain functions, is thought to emerge from the interactions between different functionally specialized regions or neural groups. When trying to understand cognitive processing as instantiated in the brain it is therefore natural to analyze functional interactions from a network perspective (Ingvar & Petersson, 2000).

Structural equation modeling (SEM, cf., chapter 3) provides one approach to test for differences in network interactions explicitly. Petersson et al. (2000) attempted to characterize the functional organization of immediate verbal repetition in literate and illiterate subjects in terms of effective connections between regions in a given functional-anatomical model (cf., section 6.6). In terms of network interactions, the results showed no significant difference in the literate group when comparing the word and pseudoword condition. Neither was there any significant difference between the literate and illiterate

group in the word repetition condition. In contrast, there were significant differences between word and pseudoword repetition in the illiterate group and between the illiterate and literate group in the pseudoword condition. The differences between groups were mainly related to the phonological loop, in particular, the interaction between Broca’s region and the inferior parietal region.

[Figure 5.8] Differences between literacy groups in the local thickness (circle) of the corpus callosum indicate that this is thinner in illiterate compare to the literate subjects (P <

0.01).

The absence of significant difference between word and pseudoword repetition in the literate group relates to the fact that the network interactions were similar during both word and pseudoword repetition. This indicates that the literate subjects automatically recruit the same processing network during immediate verbal repetition for words and pseudowords. In contrast, this was not the case for the illiterate group. This is consistent

with the suggestion that phonological processing is differently organized in illiterate individuals due to a different developmental background related to the acquisition of alphabetic reading and writing skills. Based on this and in conjunction with the behavioral results outlined in sections 5.2.2-4, we suggest that these differences in phonological loop interactions might represent a primary difference between the two literacy groups related to differences in sub-lexical phonological processing. This is in line with the suggestion that the parallel interactive processing characteristics of the language system differ between literate and illiterate subjects (Petersson et al. 2000).

5.3.2 NEUROANATOMICAL FINDINGS RELATED TO THE CORPUS CALLOSUM

One may wonder whether there are neuroanatomic correlates corresponding to the literacy status. It is well-known that the corpus callosum, the large fiber bundle that interconnects the two brain hemispheres, develops during childhood and into young adulthood. In particular, there is an active myelination process of the neuronal axons running through this structure in order to establish efficient communication between the brain’s two hemispheres (Giedd et al., 1996). Recent results suggest that the posterior mid-body part of the corpus callosum undergoes extensive myelination during the years of reading acquisition in children, that is, during 6-10 years of age (Thompson et al., 2000) and the fibers that cross over in this region of the corpus callosum interconnect the parieto-temporal regions (for a general review see e.g., Zaidel & Iacoboni, 2003). A recent study of the morphology of the corpus callosum suggested that the posterior mid-body region of the corpus callosum (Figure 5.8), that is, the part that interconnect the left and right parieto-temporal cortices, is thinner in illiterate compare to the literate subjects (Castro-Caldas et al., 1999). Petersson et al. (1998) hypothesized that this may be related to a difference in the inter-hemispheric interactions between literacy groups with respect to the parieto-temporal cortices.

A large number of neuropsychological studies of acquired reading and writing impairment (alexia and agraphia) describe neuroanatomic lesions most prominently centered on the parieto-temporal region, including the inferior parietal cortex and the posterior portions of the superior temporal gyrus, thus suggesting this region is important for the mapping of orthographic representations onto phonologic representations

(Friedman, Ween, & Albert, 1993). Ernest Weber suggested in 1904 that the left hemispheric language dominance might depend on the acquisition of reading and writing skills and early attempts to address the issue in aphasic patients appeared to support this hypothesis (Cameron, Currier, & Haerer, 1971; Wechsler, 1976). Specific differences in language processing between literate and illiterate aphasic subjects has been reported, in particular with respect to pseudoword repetition and verbal memory tasks (Coppens et al., 1998). Lecours (1989) suggested that illiterate subjects are more likely to use processing networks that include right-hemisphere regions when performing language tasks (Coppens et al., 1998). Moreover, a reversal of ear advantage for phonetically similar words in illiterate subjects has been reported (Damásio, Damásio, Castro-Caldas, & Hamsher, 1979).

However, the mechanisms influencing hemispheric specialization and the consequent interhemispheric interaction are not well-understood and both genetic as well as environmental factors appear to be relevant (Sommer, Ramsey, Mandl, & Kahn, 2002;

Thompson et al., 2001). Functional hemispheric lateralization has been shown to depend on several factors, including stimulus material (Kelley et al., 1998), experimental task (Stephan et al., 2003), and a recent review concluded that hemispheric specialization for language is multi-factorial and may depend on both task as well as brain region (Josse &

Tzourio-Mazoyer, 2004), and it is well accepted that both hemispheres play a role in language processing (Friederici, 2002; Knecht et al., 2002). Furthermore, computational modeling has indicated that several possible mechanisms can support hemispheric lateralization (Reggia & Schulz, 2002).

5.3.3 HEMISPHERIC DIFFERENCES RELATED TO LITERACY

In a recent study (Petersson et al., submitted) we investigated two different samples of illiterate female subjects and their matched literate controls with respect to the hemispheric lateralization of the inferior parietal cortex in two simple auditory-verbal language tasks encompassing four different conditions (cf., section 6.7). In brief, while the literate group showed a positive left–right difference, the illiterate subjects showed a negative left–right difference in the inferior parietal region (Figure 6.4 and 6.5). Detailed inspection of the results suggested that the degree of functional lateralization may be task dependent.

However, what stayed constant independent of task was the relation of the left-right

differences between the two literacy groups (i.e., group x hemisphere interaction), indicating that this relation generalizes over auditory-verbal tasks. Thus, it appears that literacy influences the functional hemispheric balance in the inferior parietal region. It is interesting to note that recent experimental results have suggested a rostral to caudal myelination process of the corpus callosum during childhood and early adulthood (Thompson et al., 2000; Zaidel & Iacoboni, 2003), indicating an ongoing developmental process to establish efficient interactions between the two hemispheres. The fibers that cross over in the posterior mid-body region of corpus callosum interconnect the parieto-temporal regions and undergoes extensive myelination during the years of reading acquisition (Thompson et al., 2000) and this is the same region in which recent evidence indicate that literate subjects are thicker compared to illiterate subjects (cf., section 5.3.2).

Thus, one may speculate that acquiring reading and writing skills at the appropriate age shapes not only the local morphology of the corpus callosum but also the degree of functional specialization as well as the pattern of interaction between the interconnected inferior parietal regions.

In conclusion, literacy represents an essential aspect of contemporary culture.

Formal education and the educational system can be viewed as an institutionalized process of structured cultural transmission. The results outlined in this chapter indicate that formal education and its use influence important aspects of cognition and behavior as well as structural and functional properties of the brain. Taken together, the evidence provides strong support for the hypothesis that the brain is modulated by literacy and formal education.