Effects of Hemisphere-Specific and Hemisphere-Alluding Stimulation on the Reading Performance of Dyslexic Subtypes

Abstract

Bakker's (1979) 'balance model' of dyslexia postulates that P (perceptual)-type and L (linguistic)-type dyslexias are related to functionally overdeveloped right (RH) and left (LH) cerebral hemispheres, respectively. P-type dyslexia is characterised by an accurate but slow and fragmented reading style, while L-type dyslexia is associated with a fast but inaccurate reading style. Twenty-one children with P-type dyslexia (mean age= 9.9 years) and l9 children with L-type dyslexia (mean age=9.7 years) were treated with hemisphere-specific stimulation (HSS) and hemisphere-alluding stimulation (HAS). Visual HSS was produced by presenting words to either the left or right visual half-fields, while tactile HSS was produced by presenting words to either the left or right hand of the subject. HAS training was achieved through the presentation of either semantically/phonetically demanding (alluding to the LH) or perceptually demanding (alluding to RH) text. Subjects were randomly assigned to two training conditions (P-type training or L-type training). Subjects in the PP (n= 11) and LL (n=9) groups received an intervention programme that was consistent with their subtype (i.e., PP received stimulation of the LH whereas, LL received stimulation of the RH), while subjects in the PL (n= 10) and LP (n= 10) groups were given a programme aimed at stimulating the already 'overdeveloped' hemisphere (i.e., PL received stimulation of RH and LP received stimulation of LH). After treatment, the subject's reading performance was evaluated in terms of the number of time-consuming and substantive errors on a passage-reading test. Reading performance was further assessed on standardized tests of reading-related skills. All four groups showed improvement over time. However, contrary to predictions from Bakker's theory, the PP and LL groups did not improve significantly more than the PL and LP groups, respectively. In addition, P-type and L-type dyslexics were compared on a range of laterality measures (i.e., dichotic listening task, visual half-field task, peg-moving task and lateral preference observations). The two dyslexia subtypes did not differ on any of the laterality measures. The inferred over-involvement of the RH in mediating reading in P-type dyslexics was not reflected in their performance on these measures. No subtype differences were observed on the WISC-R (Wechsler, 1974), except that, contrary to expectations, P-type dyslexics were found to be better on the Digit Span subtest than L-type dyslexics. A detailed evaluation of Bakker's L/P classification system revealed problems with internal validity. It was argued that the reliability and accuracy of this classification system needs to be improved before further external validation studies are conducted.