English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Feedforward CNN models have proven themselves in recent years asstate-of-the-art models for predicting single-neuron responses to naturalimages in early visual cortical neurons. In this paper, we extend these modelswith recurrent convolutional layers, reflecting the well-known massiverecurrence in the cortex, and show robust increases in predictive performanceover feedforward models across thousands of hyperparameter combinations inthree datasets of macaque V1 and V2 single-neuron responses. We propose therecurrent circuit can be conceptualized as a form of ensemble computing, witheach iteration generating more effective feedforward paths of various pathlengths to allow a combination of solutions in the final approximation. Thestatistics of the paths in the ensemble provide insights to the differentialperformance increases among our recurrent models. We also assess whether therecurrent circuits learned for neural response prediction can be related tocortical circuits. We find that the hidden units in the recurrent circuits ofthe appropriate models, when trained on long-duration wide-field imagepresentations, exhibit similar temporal response dynamics and classicalcontextual modulations as observed in V1 neurons. This work provides insightsto the computational rationale of recurrent circuits and suggests that neuralresponse prediction could be useful for characterizing the recurrent neuralcircuits in the visual cortex.

Recurrent networks improve neural response prediction and provide insights into underlying cortical circuits