Is the motion system relatively spared in amblyopia? Evidence from cortical evoked responses

Visual evoked potentials (VEPs) produced by pattern reversal were compared with those elicited by onset of motion in 37 amblyopic children (20 with anisometropic amblyopia, seven with strabismic amblyopia and 10 with both anisometropia and strabismus), The amplitudes and peak latencies of the main P-1 peak in the pattern-reversal VEP and of the motion-specific N-2 peak in the motion-onset VEP through the amblyopic eye were compared with those through the normal fellow eye. Regardless of the type of amblyopia, the amplitude of the pattern-reversal VEP for full-field stimulation was significantly smaller and its latency significantly longer through the amblyopic eye (P < 0.001). In contrast, neither the amplitudes nor the latencies of the N-2 motion-onset VEPs differed significantly between amblyopic and non-amblyopic eyes. For pattern-reversal VEPs through the amblyopic eyes, the extent to which amplitude was reduced and latency prolonged correlated well with the reduction of visual acuity, whereas the amplitudes and latencies of motion-onset VEPs did not vary with visual acuity. Even for stimuli restricted to the central visual field (5 or 2 deg diameter) or to the peripheral field (excluding the central 5 deg), motion-onset responses were indistinguishable through the two eyes, while pattern-reversal responses always differed significantly in amplitude. These results suggest that the source of motion-onset VEPs (probably an extrastriate motion-sensitive area) is less affected in amblyopia than that of pattern-reversal VEPs (probably the striate cortex). The motion pathway, presumably deriving mainly from the magnocellular layers of the lateral geniculate nucleus, may be relatively spared in amblyopia.