Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911660210388992 |
|---|---|
| author | Ventura, Linda Ariel Bosch, Victoria Kietzmann, Tim C Thorat, Sushrut |
| author_facet | Ventura, Linda Ariel Bosch, Victoria Kietzmann, Tim C Thorat, Sushrut |
| contents | Adaptive cognition requires structured internal models of objects and their relations. Predictive neural networks are often proposed to learn such world models, but how these are instantiated and how they support prediction remain unclear. We investigate this in a minimal in-silico setting. A recurrent neural network samples tokens sequentially from 2D continuous token scenes and is trained to predict the upcoming token from the current input and a saccade-like displacement. On novel scenes, prediction accuracy improves across the sequence, indicating in-context learning. Decoding analyses reveal path integration and dynamic binding of token identity to position. Interventional analyses show that new bindings can be learned late in sequence and that out-of-distribution bindings can be learned as well. Together, these findings show how structured representations relying on flexible binding emerge to support prediction, offering a mechanistic account of sequential world modeling relevant to cognitive science. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03490 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network Ventura, Linda Ariel Bosch, Victoria Kietzmann, Tim C Thorat, Sushrut Machine Learning Neurons and Cognition Adaptive cognition requires structured internal models of objects and their relations. Predictive neural networks are often proposed to learn such world models, but how these are instantiated and how they support prediction remain unclear. We investigate this in a minimal in-silico setting. A recurrent neural network samples tokens sequentially from 2D continuous token scenes and is trained to predict the upcoming token from the current input and a saccade-like displacement. On novel scenes, prediction accuracy improves across the sequence, indicating in-context learning. Decoding analyses reveal path integration and dynamic binding of token identity to position. Interventional analyses show that new bindings can be learned late in sequence and that out-of-distribution bindings can be learned as well. Together, these findings show how structured representations relying on flexible binding emerge to support prediction, offering a mechanistic account of sequential world modeling relevant to cognitive science. |
| title | Path Integration and Object-Location Binding Emerge in an Action-Conditioned Predictive Sequence Network |
| topic | Machine Learning Neurons and Cognition |
| url | https://arxiv.org/abs/2602.03490 |