Abstract
Purpose: Motion repulsion (MR) refers to the exaggeration of perceived angular separation between two simultaneously presented stimuli that move in different directions. This phenomenon has been interpreted as an evidence for the mutual inhibition between direction-tuned neurons in visual cortex. Previous studies on perceptual learning (PL) have suggested that a consequence of extensive perceptual training is the modification of interneuron interaction. However, experimental evidence that supports this proposal is limited. Our study aimed to investigate the effect of PL of motion direction discrimination on the mutual inhibition between direction-tuned neurons by measuring the change in MR size before and after the training.
Methods: The experiment consisted of a pre-training test phase (session 1-2), a training phase (session 3-8) and a post-training test phase (session 9-10). In the training phase, 32 participants were randomly assigned to one of four training groups (2 tasks × 2 base direction; tasks: luminance or direction discrimination; base direction: 240 deg or 300 deg). Classical 3-up-1-down staircase procedure was used during training. Constant stimuli method was used to measure participants’ discrimination threshold and size of MR in the pre- and post- training tests. The stimuli were dynamic random-dot displays (coherence level: 100%, speed: 4 deg/s, stimulus diameter: 10°) generated with Movshon /Newsome algorithm (Pilly & Seitz, 2009).
Results: (1) Training significantly decreased motion direction discrimination threshold along the trained direction. This learning effect transferred little to the untrained direction. (2) Motion direction discrimination training reduced bilateral MR size. (3) Luminance discrimination training did not affect the measured MR size, confirming that the aforementioned reduction of bilateral MR size cannot be attributed to the training induced attentional shift or adaptation.
Conclusion: These results demonstrate that the lateral inhibition between motion direction-tuned neurons can be modulated by PL.
Meeting abstract presented at VSS 2013