SFN 2013 Poster: Components of the neural dynamics during visual perceptual learning
Perceptual learning, an improvement with training in perceptual ability, is an important form of adaption in sensory systems and beyond. Learning-induced changes have been observed in the primary visual cortex (V1) of monkeys, but the temporal dynamics of a network of neurons during the training is poorly understood.
Using chronically implanted microelectrode arrays, we recorded a population of V1 neurons in monkeys over the course of perceptual training. The task was to find an elongated visual contour made of collinear bars embedded in a complex background.
High dimensional neural population responses to the stimulus pattern can be described by the dynamics of two main components: a facilitatory component involving cells having receptive fields (RFs) over the contour to be detected, and a suppressive component involving cell having RFs on the background. The components’ strengths depend on the saliency of the embedded contour, and that the suppressive activity systematically lags the facilatory drive. Furthermore, their weights progressively increase over the course of learning and correlate well with the animal’s behavioral performance.
To better understand the mechanism of the learning-induced network changes we constructed a simple rate-based model that can replicate the general dynamics of the two components, and thereby suggesting a neural mechanism for how perceptual learning induces changes to the network activity.
Components of the neural dynamics during visual perceptual learning – Malte J. Rasch,Yin Yan, Chen Minggui, Si Wu & Wu Li (SFN 2013 poster PDF)