Fisiologia del sistema di lettura nella normalità e nei disturbi evolutivi

Reading is a multicomponent task that involves basic visuoperceptive, oculomotor and attentional skills (Aghababian and Nazir 2000; Hyona and Olson 1995; Levy, et al. 2010; Rayner1986; Shaywitz and Shaywitz 2008), together with access to symbolic orthographic and lexico-semantic knowledge and the activation of phonological representations (Coltheart, et al. 1997; Plaut, et al. 1996; Zorzi, et al. 1998). Numerous studies have been published that investigated the neural network involved in single word or pseudoword reading in subjects with normal reading abilities and in subjects with developmental dyslexia, using both behavioural and neuroimaging tecnique (Cornelissen et al., 1995; Fawcett & Nicolson, 1994; Finch et al., 2002; Laycock et al., 2008; Nicolson & Fawcett, 1993; Ramus et al., 2003; Snowling, 1981; Stein, 2003; Steinbrink et al., 2008). Given the complexity of reading system and the heterogeneity of the behavioural pattern observed in dyslexic subjects, a wide range of possible explanations have been suggested for dyslexia, such as (i) a specific linguistic problem due to a deficit in phonological processing (Frith, 1999; Ramus et al., 2003; Snowling, 2001), (ii) an impairment of the magnocellular pathway (Eden, VanMeter, Rumsey, & Zeffiro, 1996; A. M. Galaburda, 1993; Hari & Renvall, 2001; Stein, 2001) and (iii) a disfunction of the cerebellar system (D. V. Bishop, 2002; Nicolson et al., 2001; Rae et al., 2002). Each of these theories has found support in behavioural and/or imaging experiments, suggesting that all the systems have an important role in reading process and that they interact with reading process throught anatomofunctional convergence mechanisms. Thus, we used fMRI to explore the extent of the anatomical overlap between the auditory phonological, the visual magnocellular and the motor/cerebellar systems with reading, in order to isolate “interface” areas, that is, regions that are involved in different cognitive systems relevant for reading (Damasio, 1989). Twenty-eight normal subjects performed, after a neuropsychological screening, five tasks during fMRI scans: word and non-word reading, auditory rhyming for letter names, visual motion perception and a motor sequence learning task. We found a rostro-caudal functional gradient in the left occipitotemporal cortex (van der Mark et al., 2011): an anterior area that was activated by both reading and auditory rhyming tasks; a posterior area, commonly activated by both reading and the motion perception task and a medial area, including the so called Visual Word Form Area (Cohen et al., 2002). that was specifically activated by the reading task. On the contrary, the reading and the motor/cerebellar systems showed an overlap in left premotor ventral area and in the cerebellum, bilaterally. These data were in line with the hypothesis that the left occipito-temporal cortex, broadly considered, is an interface area (Devlin et al., 2006) between the reading system and the phonological and magnocellular systems, even if no single area emerged as a convergence area of all the systems. In the light of these evidence, we investigated the areas of reduced activations in a group of dyslexic subjects during a non-word reading task. 12 dyslexic, well compensated, adults were tested with the same previous tasks tackling non-word reading, phonological awareness, visual motion perception and motor learning skills at both behavioural and neurofunctional levels. The behavioural tests confirmed, at the group level, the presence of a phonological disorder, together with the reading deficit, while no between-group differences emerged in the magnocellular and motor/cerebellar tasks. However, the single-subject analyses showed that reduced performance in reading and in phonological tasks was occasionally associated with a reduced performance in tasks designed to test magnocellular visual or the motor “cerebellar” systems. At the anatomofunctional level, fMRI data confirmed that dyslexics had reduced activation of the left inferior-temporal and ventral occipito-temporal cortices for word and non-word reading (Paulesu et al., 2001; Richlan et al., 2009; S. E. Shaywitz et al., 1998). We also observed that these reduced activations associated with non-word reading in dyslexics involved the visual/orthographic and the orthographic/phonological interface area, together with the ventral area specifically activated by the reading task, observed in the previous experiment. On the contrary, the reduced hypoactivations observed in dyslexic subjects during the word reading task not showed an overlap with the more posterior region, such as the visuo-orthografic interface area was involved in sublexical processes and was less stressed by this type of stimulus, while an overlap emerged between this hypoactivated region and the orthographic-phonological area and the reading per sè area. Moreover, there were also hyperactivations in domain specific regions in the visuo/magnocellular and in the motor/cerebellar tasks. Given the normal performance of dyslexic in these tasks at the behavioural level, we hypothesize that these hyperactivations may represent a possible sign of successfull compensatory processes. However, the phonological and the reading speed deficit did stand above other difficulties being evident at both behavioural and physiological level, while the other systems may have undergone a more efficient compensation because of a more redundant neuronal architecture. Finally, I describe the data of a single-case study on a “rare” left hemispherectomized patient. We studied the linguistic behaviour and neurolinguistic organization of a 14 years-old adolescent who underwent a left hemispherectomy at age 2.5. The aphasia observed after surgery was a clear sign of an initial left language lateralization. This study allowed us to evaluate the hypothesis that the occipito-temporal competence could develop in the non dominant hemisphere, together with the behavioural linguistic recovery. EB's neuropsychological pattern was akin to that of a surface dyslexia patient (K. Patterson, Marshall, & Coltheart, 1985): EB made significantly more errors than controls in a visual lexical decision task, in discriminating written homophones and in reading irregular and loan words, even if he did not show a pathological performance in regular word and non-word reading. At anatomofunctional level, EB’s fMRI patterns were similar to those observed in the dominant hemisphere of controls: in both EB and controls the frontal regions, the middle and inferior temporal gyri and the calcarine cortex were activated during word and non-word reading, while reduced activations were observed, in EB than controls, in the angular gyrus and in the occipito-temporal cortex. In particular, the occipito-temporal region was not completely deprived of its functional role, as demonstrated by its “normal” activation in a more elementary shape-matching task; nonetheless, there was no plain commitment of that brain region for reading, as observed in controls. Notwithstanding, the findings from a single case-study deserve great interpretative caution and any attempt toward general conclusions must be tempered with prudence, the communality between the EB’s activations and the haemodynamic response observed in the dominant hemisphere of controls, together with the EB’s behavioural profile, suggested that the neurolinguistic development of EB was similar to the one resulting from the left hemisphere.