Perceptual Learning of Motion Direction Discrimination Induced by True and False Feedback

Abstract

Top-down factors such as feedback plays an important role in learning. However, in the literature of perceptual learning, only a few studies have focused on this topic. In the present study, we investigated how different types of feedback affect perceptual learning of motion direction discrimination. We trained participants for six days to discriminate whether the direction of moving dots was the same as the orientation of a fixed reference bar that preceded. The direction was either the same or deflected 1° to one side of the reference orientation during training. The participants were assigned to three groups based on the types of feedback they received during training: true feedback, false feedback, and random feedback. For the true feedback and false feedback groups, we analyzed the data from the learners who correctly reported the deflection and improved discrimination sensitivity, in accordance with their received feedback, after the training. A three-way repeated measure ANOVA for learners in both true and false feedback conditions revealed significant triple interactions among training, direction, and deflection (trueFB:F(1,11)= 13.512, p=0.004; falseFB: F(1,11)= 17.179, p=0.002). Additional analysis revealed that d' significantly changed only for trained direction and deflection after training, but not for other directions or deflections. These findings suggest that feedback can exert significant effect on the learning of motion direction discrimination. Particularly, the false feedback to the subthreshold stimuli during training led to wrong perception and such effect was limited to the trained condition. Meeting abstract presented at VSS 2016