Back

Distinct beta frequencies reflect categorical decisions

Rassi, E.; Zhang, Y.; Mendoza, G.; Mendez, J. C.; Merchant, H.; Haegens, S.

2022-10-21 neuroscience
10.1101/2022.10.17.512497 bioRxiv
Show abstract

Beta oscillations are involved in a variety of cognitive functions beyond their traditional sensorimotor role. Based on prior findings of content-specific beta synchronization during working memory and decision making, we hypothesized that beta activity supports the activation and reactivation of cortical representations by mediating neural ensemble formation within and between brain regions. We here found that beta activity in monkey dorsolateral prefrontal cortex (dlPFC) and in pre-supplementary motor area (preSMA) reflects the content of a stimulus in relation to the task context, regardless of its objective properties. In multiple versions of a categorization task, we changed the boundary between categories from one block of trials to the next, such that a stimulus which belonged to one of two categories during one block could belong to the other category during the next one. During a delay in which monkeys had to hold their categorical decision in mind, we found that two distinct beta-band frequencies were consistently associated with the same two relative categories, and that activity in these respective bands predicted the animals responses. We characterized beta at these frequencies as transient bursts with distinct temporal profiles. We further showed that dlPFC and preSMA are connected via these distinct frequency channels, with dlPFC driving the frequency separation, a result supported by Granger causality and spike-field coherence analyses. In sum, these results provide support for the role of beta in forming neural ensembles, and further show that such ensembles synchronize at different beta frequencies.

Matching journals

The top 2 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.