Volume 20, Issue 4, May 1997, Pages 147–152

Perspectives

To see but not to read; the magnocellular theory of dyslexia

• John Stein¹, 
• Vincent Walsh²

• ¹ Dept of Physiology, Parks Rd, University of Oxford, Oxford, UK OX1 3PT
• ² Dept of Psychology, South Parks Rd, University of Oxford, Oxford, UK OX1 3UD

• http://dx.doi.org/10.1016/S0166-2236(96)01005-3, How to Cite or Link Using DOI

• Permissions & Reprints

View full text

Purchase $31.50

---

Developmental dyslexics often complain that small letters appear to blur and move around when they are trying to read. Anatomical, electrophysiological, psychophysical and brain-imaging studies have all contributed to elucidating the functional organization of these and other visual confusions. They emerge not from damage to a single visual relay but from abnormalities of the magnocellular component of the visual system, which is specialized for processing fast temporal information. The m-stream culminates in the posterior parietal cortex, which plays an important role in guiding visual attention. The evidence is consistent with an increasingly sophisticated account of dyslexia that does not single out either phonological, or visual or motor deficits. Rather, temporal processing in all three systems seems to be impaired. Dyslexics may be unable to process fast incoming sensory information adequately in any domain.

---

Figures and tables from this article:

Fig. 1. Words can be hard to read for several different reasons. Visual confusions can cause letter reversals (‘worbs’), distortion and blurring (‘can be hard to read’) and superimposition (‘for several different’).

Figure options

Fig. 2. The magnocellular laminae of the lateral geniculate nucleus (mLGN) and visual deficits. The lateral geniculate nucleus (LGN) is a six-layered structure that receives input from the retinal ganglion cells and the primary visual cortex. There are important differences between the relative preferences of the four dorsal parvocellular (P) laminae and the two ventral magnocellular (M) laminae. Cells in the P, but not the M laminae, receive spatially segregated conederived inputs from small (P) retinal ganglion cells¹¹. M cells have larger receptive fields than P cells and a higher achromatic contrast sensitivity¹². From the point of view of dyslexia research, the most important differences between the two groups of cells are the faster conduction velocities and the high temporal (transient) sensitivity of the M cells compared with the sustained responses of the P neurones¹³. The laminar organization of the LGN can be exploited to demonstrate the functional significance of the differences between P and M cells. The figure shows the effect of LGN lesions on flicker (left) and motion (right) detection judgements in monkeys. When the parvocellular LGN was lesioned (A), the monkeys were unimpaired in the contralateral visual field (filled squares) on either task, compared with their performance in the spared ipsilateral (open circles) visual field. However, when a lesion was made in the magnocellular laminae (B), the monkeys were greatly impaired in flicker and motion judgements in the contralateral ‘magno-blind’ visual field. The opposite pattern of results was obtained when the monkeys were required to detect or discriminate colour, form, size, texture or disparity¹⁴. The specificity of these findings is mirrored in the specificity of visual deficits reported in dyslexia. Data replotted from Ref. 14.

Figure options

Fig. 3. Functions of the posterior parietal cortex (PPC) and their relations to dyslexia. The different shading patterns show the separate subareas of the PPC. Damage to each region can produce specific deficits: spatial mislocalization may be associated with damage to areas 5 and 7; spatial disorientation with damage to areas 5, 7 and 39; neglect with damage to areas 39 and 40; ‘cocktail party’ problems with damage to area 40; visuomotor co-ordination with damage to areas 5, 7 and 39; and visuoverbal association with damage to area 40.

Figure options

Fig. 4. Left neglect in a dyslexic child. The child's age was 7 years 11 months, with a reading age that was retarded by 20 months. The child's I.Q. was 92.

Figure options