Asymmetric transfer between the learning of the complex stimulus

IntroductionPerceptual learning of complex stimulus (such as faces or houses) are shown to be specific to the stimulus, indicating the plasticity of the human high-level visual cortex. However, limited understanding exists regarding the plasticity of the representation of complex stimuli in visual w...

Full description

Saved in:
Bibliographic Details
Main Authors: Yangyang Du, Hui Kou, Huijie Liu, Taiyong Bi
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-04-01
Series:Frontiers in Neuroscience
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fnins.2025.1578862/full
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:IntroductionPerceptual learning of complex stimulus (such as faces or houses) are shown to be specific to the stimulus, indicating the plasticity of the human high-level visual cortex. However, limited understanding exists regarding the plasticity of the representation of complex stimuli in visual working memory (VWM) and its specificity.MethodsTo address this question, we adopted a delayed match-to-sample task to train the working memory for faces and houses. Subjects were trained for 6 days with neutral faces, happy faces, sad faces, and houses in Experiments 1, 2, 3, and 4, respectively.ResultsThe results revealed that training significantly increased the sensitivity (d’) to discriminate the visual representations in VWM in all four experiments. Furthermore, the learning effects of neutral faces were transferable to emotional faces and vice versa. However, the learning effects of emotional faces exhibited limited transfer to untrained emotional faces. More importantly, the transfer of learning effects between faces and houses was asymmetrical, i.e., only the learning effects of faces could transfer to houses, whereas the reverse was not true.DiscussionThese results highlight distinct cognitive processes underlying the training effects for different stimulus categories and provide valuable insights into the mechanisms of VWM improvement.
ISSN:1662-453X