December 5, 2025Journal of Neuroscience

Orthogonal-rotational Dynamics Supports Efficient Encoding and Updating for Streaming Information in Working Memory

View Full Paper

Ask AI

Bookmark

The orthogonal-rotational dynamics improve working memory efficiency, allowing for better organization of information.
Findings reveal the encoding mechanism where information is rapidly updated and stored in both RNN models and human brains.
Research involved comparing the dynamics of neural activities in recurrent neural networks and human subjects during an N-back task.
These insights may inform the design of artificial neural networks to optimize information processing and storage.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Analysis across RNN and human subjects shows orthogonal-rotational dynamics improve information encoding in working memory, suggesting implications for neural networks.

Cite This Study

Yang et al. (2025) studied this question.

synapsesocial.com/papers/693231368e51979591dceaff https://doi.org/10.1523/jneurosci.2220-24.2025

View Full Paper

Ask AI

Bookmark

Implication

Analysis across RNN and human subjects shows orthogonal-rotational dynamics improve information encoding in working memory, suggesting implications for neural networks.

Orthogonal-rotational Dynamics Supports Efficient Encoding and Updating for Streaming Information in Working Memory

Key Points

Discussion

Implication

Cite This Study

Orthogonal-rotational Dynamics Supports Efficient Encoding and Updating for Streaming Information in Working Memory

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider