Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.10955 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916617369157632 |
|---|---|
| author | Morgan, Jonathan Albanna, Badr Herman, James P. |
| author_facet | Morgan, Jonathan Albanna, Badr Herman, James P. |
| contents | Attention is fundamental to both biological and artificial intelligence, yet research on animal attention and AI self attention remains largely disconnected. We propose a Recurrent Vision Transformer (Recurrent ViT) that integrates self-attention with recurrent memory, allowing both current inputs and stored information to guide attention allocation. Trained solely via sparse reward feedback on a spatially cued orientation change detection task, a paradigm used in primate studies, our model exhibits primate like signatures of attention, including improved accuracy and faster responses for cued stimuli that scale with cue validity. Analysis of self-attention maps reveals dynamic spatial prioritization with reactivation prior to expected changes, and targeted perturbations produce performance shifts similar to those observed in primate frontal eye fields and superior colliculus. These findings demonstrate that incorporating recurrent feedback into self attention can capture key aspects of primate visual attention. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_10955 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A recurrent vision transformer shows signatures of primate visual attention Morgan, Jonathan Albanna, Badr Herman, James P. Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition Attention is fundamental to both biological and artificial intelligence, yet research on animal attention and AI self attention remains largely disconnected. We propose a Recurrent Vision Transformer (Recurrent ViT) that integrates self-attention with recurrent memory, allowing both current inputs and stored information to guide attention allocation. Trained solely via sparse reward feedback on a spatially cued orientation change detection task, a paradigm used in primate studies, our model exhibits primate like signatures of attention, including improved accuracy and faster responses for cued stimuli that scale with cue validity. Analysis of self-attention maps reveals dynamic spatial prioritization with reactivation prior to expected changes, and targeted perturbations produce performance shifts similar to those observed in primate frontal eye fields and superior colliculus. These findings demonstrate that incorporating recurrent feedback into self attention can capture key aspects of primate visual attention. |
| title | A recurrent vision transformer shows signatures of primate visual attention |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition |
| url | https://arxiv.org/abs/2502.10955 |