PT  - JOURNAL ARTICLE
AU  - Nicholas J. Sexton
AU  - Bradley C. Love
TI  - Directly interfacing brain and deep networks exposes non-hierarchical visual processing
AID  - 10.1101/2021.06.28.450213
DP  - 2021 Jan 01
TA  - bioRxiv
PG  - 2021.06.28.450213
4099  - http://biorxiv.org/content/early/2021/07/03/2021.06.28.450213.short
4100  - http://biorxiv.org/content/early/2021/07/03/2021.06.28.450213.full
AB  - One reason the mammalian visual system is viewed as hierarchical, such that successive stages of processing contain ever higher-level information, is because of functional correspondences with deep convolutional neural networks (DCNNs). However, these correspondences between brain and model activity involve shared, not task-relevant, variance. We propose a stricter test of correspondence: If a DCNN layer corresponds to a brain region, then replacing model activity with brain activity should successfully drive the DCNN’s object recognition decision. Using this approach on three datasets, we found all regions along the ventral visual stream best corresponded with later model layers, indicating all stages of processing contained higher-level information about object category. Time course analyses suggest long-range recurrent connections transmit object class information from late to early visual areas.Competing Interest StatementThe authors have declared no competing interest.