Abstract
We showed previously that the perception of motion-defined form is immature in school-age children and deficient in both eyes of children with amblyopia. The extent to which these developmental changes reflect the functioning of the M/dorsal or the P/ventral pathway is not known. We used dot speed to bias the relative contribution of the M/dorsal (fast motion) and P/ventral (slow motion) pathways to the processing of motion-defined form. Our first attempt at this, using a sparse dot pattern (VSS 2009), produced inconclusive results. Here we report the effect of speed, age and amblyopia, using a higher-density pattern.
Motion-defined rectangles (vertical or horizontal) were created by moving signal dots inside the shape in one direction (up or down) and signal dots outside the shape in the opposite direction. Noise dots inside and outside the shape moved in random directions. The proportion of signal dots specified the coherence of the pattern. All dots moved at the same speed throughout a run. Monocular motion coherence thresholds, for discriminating rectangle orientation, were determined at slow (0.1 deg/s), medium (0.9 deg/s) and fast speeds (5.0 deg/s) using a staircase procedure.
First we examined typical development in young children (4-6 years), older children (7-10 years), adolescents and adults. Coherence thresholds were: (1) highest at the fast and lowest at the medium speed for all four groups, and (2) higher in the young children than in the three older groups.
Next we examined the fellow eyes of patients with anisometropic and/or strabismic amblyopia. Coherence thresholds were elevated, relative to age-matched controls, in most of the patients at the slow and medium speeds only.
These results suggest that the P/ventral components of the task are deficient in patients with amblyopia. We find no clear maturational differences between the M/dorsal and P/ventral pathways in children with typical visual development.
Natural Science and Engineering Research Council of Canada.