Perceptual learning of fine contrast discrimination changes neuronal tuning and population coding in macaque V4. Issue 1 (December 2018)

• Record Type: Journal Article
• Authors: Sanayei, Mehdi; Chen, Xing; Chicharro, Daniel; Distler, Claudia; Panzeri, Stefano; Thiele, Alexander
• Is Part Of: Nature communications. Volume 9:Issue 1(2018)
• Page Start: 1
• Page End: 15
• Publication Date: 2018-12
• Subjects: Biology -- Periodicals; Physical sciences -- Periodicals; 505
• DOI: 10.1038/s41467-018-06698-w
• Languages: English
• ISSNs: 2041-1723

Abstract

Abstract Perceptual learning, the improvement in perceptual abilities with training, is thought to be mediated by an alteration of neuronal tuning. It remains poorly understood how tuning properties change as training progresses, whether improved stimulus tuning directly links to increased behavioural readout of sensory information, or how population coding mechanisms change with training. Here, we recorded continuously from multiple neuronal clusters in area V4 while macaque monkeys learned a fine contrast categorization task. Training increased neuronal coding abilities by shifting the steepest point of contrast response functions towards the categorization boundary. Population coding accuracy of difficult discriminations resulted largely from an increased information coding of individual channels, particularly for those channels that in early learning had larger ability for easy discriminations, but comparatively small encoding abilities for difficult discriminations. Population coding was also enhanced by specific changes in correlations. Neuronal activity became more indicative of upcoming choices with training. Perceptual learning, the improvement in perceptual abilities with training, is thought to involve changes in neuronal 'tuning'. Here, the authors show that perceptual learning works by making neurons increasingly sensitive to task-relevant differences in stimuli, and by improving population coding mechanisms.