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

Working memory (WM), a fundamental cognitive process facilitating thetemporary storage, integration, manipulation, and retrieval of information,plays a vital role in reasoning and decision-making tasks. Robust benchmarkdatasets that capture the multifaceted nature of WM are crucial for theeffective development and evaluation of AI WM models. Here, we introduce acomprehensive Working Memory (WorM) benchmark dataset for this purpose. WorMcomprises 10 tasks and a total of 1 million trials, assessing 4functionalities, 3 domains, and 11 behavioral and neural characteristics of WM.We jointly trained and tested state-of-the-art recurrent neural networks andtransformers on all these tasks. We also include human behavioral benchmarks asan upper bound for comparison. Our results suggest that AI models replicatesome characteristics of WM in the brain, most notably primacy and recencyeffects, and neural clusters and correlates specialized for different domainsand functionalities of WM. In the experiments, we also reveal some limitationsin existing models to approximate human behavior. This dataset serves as avaluable resource for communities in cognitive psychology, neuroscience, andAI, offering a standardized framework to compare and enhance WM models,investigate WM's neural underpinnings, and develop WM models with human-likecapabilities. Our source code and data are available athttps://github.com/ZhangLab-DeepNeuroCogLab/WorM.

Decoding the Enigma: Benchmarking Humans and AIs on the Many Facets of Working Memory