Foveal feedback in perceptual processing: Contamination of neural representations and task difficulty effects

Visual object recognition was traditionally believed to rely on a hierarchical feedforward process. However, recent evidence challenges this notion by demonstrating the crucial role of foveal retinotopic cortex and feedback signals from higher-level visual areas in processing peripheral visual infor...

Full description

Saved in:
Bibliographic Details
Published in:PloS one 2023-10, Vol.18 (10), p.e0291275-e0291275
Main Authors: Contemori, Giulio, Oletto, Carolina Maria, Battaglini, Luca, Motterle, Elena, Bertamini, Marco
Format: Article
Language:English
Subjects:
Biology and Life Sciences
Chi-square test
Contamination
Cortex
Decision making
Evaluation
Feedback
Hypotheses
Image processing
Information processing
Investigations
Object recognition
Pattern recognition
Physical Sciences
Research and Analysis Methods
Retina
Social Sciences
Visual effects
Visual perception
Visual signals
Visual tasks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Visual object recognition was traditionally believed to rely on a hierarchical feedforward process. However, recent evidence challenges this notion by demonstrating the crucial role of foveal retinotopic cortex and feedback signals from higher-level visual areas in processing peripheral visual information. The nature of the information conveyed through foveal feedback remains a topic of debate. To address this, we conducted a study employing a foveal mask paradigm with varying stimulus-mask onset asynchronies in a peripheral same/different task, where peripheral objects exhibited different degrees of similarity. Our hypothesis posited that simultaneous arrival of feedback and mask information in the foveal cortex would lead to neural contamination, biasing perception. Notably, when the two peripheral objects were identical, we observed a significant increase in the number of "different" responses, peaking at approximately 100 ms. Similar effect was found when the objects were dissimilar, but with an overall later timing (around 150 ms). No significant difference was found when comparing easy (dissimilar objects) and difficult trials (similar objects). The findings challenge the hypothesis that foveation planning alone accounts for the observed effects. Instead, these and previous observations support the notion that the foveal cortex serves as a visual sketchpad for maintaining and manipulating task-relevant information.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0291275