Brain-Mind Institute for Future Leaders of Brain-Mind Research

Home | About | Why BMI? | Why Me? | Programs | BMM | Press | Committees | Webinars | Classes | Founding | Login

Summer 2013 | Summer 2014 | BMI 811 | BMI 821 | BMI 831 | BMI 871 | Registration | Sponsors

Visual Perceptual Learning and Its Brain Mechanisms
A New Perspective

Cong Yu

Department of Psychology and Peking-Tsinghua Center for Life Sciences
Peking University, Beijing
yucong@pku.edu.cn

Abstract

Visual perceptual learning is regarded as a powerful tool to understand brain plasticity at the behavioral level. Learning is known to be specific to the trained retinal location and orientation, which places important constraints on perceptual learning theories, many of which assume that perceptual learning occurs in the early visual areas that are retinotopic and orientation selective.

However, we created new experimental paradigms to demonstrate that location and orientation specificities can be eliminated from perceptual learning. In a “double training” paradigm, location specific learning can transfer completely to a new retinal location following additional training at the new location with an irrelevant task. Similarly, in a training-plus-exposure (TPE) paradigm, orientation/direction-specific learning can transfer completely to an untrained new orientation/direction if an observer is also passively exposed to the new orientation/direction through an irrelevant task.

These results suggest that perceptual learning is more likely a high-level process that occurs beyond the retinotopic and orientation/direction selective visual cortex. What is being actually learned in perceptual learning? I will present evidence that perceptual learning may be a form of concept learning, in that the brain may learn a highly abstract “concept” of orientation/direction. On the other hand, why high-level perceptual learning shows specificity in the first place? I will present evidence that learning specificity may result from high-level learning not being able to functionally connect to the untrained visual inputs that are under-activated due to insufficient stimulation or suppression, as well as unattended, during training.  It is the double training and TPE paradigms that bottom-up and top-down reactivate untrained inputs to establish functional connections and enable learning transfer.

Short Bio

Cong Yu received his Ph.D. in experimental psychology from University of Louisville in 1995. After postdoc trainings in basic and clinical vision sciences at University of Houston and UC Berkeley, he joined in the Inst. of Neuroscience, Chinese Academy of Sciences in 2003, the Inst. of Cognitive Neuroscience and Learning, Beijing Normal University in 2006, and the Department of Psychology, Peking University in 2012. He is currently a professor with the Department of Psychology and an Investigator with the Peking-Tsinghua Center for Life Sciences at Peking University.


BMI-line-right