CCN meeting | Rachel Rac-Lubashevsky (Brown University, USA)

When
27-01-2022 from 16:00 to 17:00
Where
https://ugent-be.zoom.us/j/94439892435?pwd=dnJRN0ViYWpSV0RnOGRUNFZwRUF3dz09
Language
English

Temporal dynamics of selective gating in working-memory: How do humans decide which information is relevant to attend to in memory, which cognitive operation to take, and when?

Optimal flexible behavior requires an agent to not only respond to incoming sensory events but also adaptively adjust action selection based on context. Moreover, while some events in memory need to be robustly maintained over time in the face of distracting interference, sometimes sensory events dictate that such memories should be disrupted and rapidly updated. These challenges highlight the need for a context-dependent control mechanism that selectively gates information into and out of working-memory (WM) to guide actions. This kind of adaptive control is thought to be managed by a hierarchical corticostriatal gating system. Despite converging evidence supporting the role of the gating system in managing higher order learning and cognitive flexibility, the core assumption that WM gating is an elaboration of the more established mechanism of motor gating has not been rigorously tested. Furthermore, very little is known on the order in which selective cognitive and action decisions are operating and how they unfold in time. In my talk, I will present our recent study showing that selective gating in WM (input, output) and response gating are managed by analogous computations and that they are distinguishable temporally. I will first describe the novel paradigm that we used, the “reference-back-2 task” and how we employed it to study the functional role of WM gating and response gating. I will show evidence from our EEG decoding approach, that switching between responses and switching between multiple items or task rules held in WM lead to a boost in the neural signature of the representation (WM or motor action) selected by the corresponding gate. Finally, I’ll show evidence from the hierarchical drift-diffusion model (HDDM) that the same control mechanism (elevated motor decision threshold) is engaged by response conflict during motor switching as recruited in response to cognitive conflict arising from switching between multiple representations held in WM. Taken together, these results support the notion that cognitive gating operations scaffold on top of mechanisms involved in motor gating. I will discuss the implications of these findings for our understanding of cognitive flexibility and WM prioritization.