Back

Dynamic prioritisation of sensory and motor contents in working memory

Echeverria-Altuna, I.; Boettcher, S. E. P.; van Ede, F.; Nobre, A. C.

2024-11-14 neuroscience
10.1101/2024.11.12.623203 bioRxiv
Show abstract

Internal selective attention prioritises both sensory and motor contents in working memory to guide prospective behaviour. Prior research has shown how attention modulation of sensory contents is flexible and temporally tuned depending on access requirements, but whether the prioritisation of motor contents follows similar flexible dynamics remains elusive. Also uncharted is the degree of co-dependence of sensory and motor modulation, which gets at the nature of both working-memory representations and internal attention functions. To address these questions, we independently tracked the prioritisation of sensory and motor working-memory contents as a function of dynamically evolving temporal expectations. The design orthogonally manipulated when an item location (left vs right side) and associated prospective action (left vs right hand) would be relevant. Contralateral modulation of posterior alpha (8-12 Hz) activity in electroencephalography (EEG) tracked prioritisation of the item location, while contralateral modulation of central mu/beta (8-30 Hz) activity tracked response prioritisation. Proactive and dynamic alpha and mu/beta modulation confirmed the flexible and temporally structured prioritisation of sensory and motor contents alike. Intriguingly, the prioritisation of sensory and motor contents was temporally uncoupled, showing dissociable patterns of modulation. The findings reveal multiple modulatory functions of internal attention operating in tandem to prepare relevant aspects of internal representations for adaptive behaviour.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.