Task-switching DyVA framework (IMAGE)
Caption
(Main) Deep learning models developed with TACC supercomputers to mimic human behavior on the task-switching game Ebb and Flow (Lumosity). This basic research can be applied to help understand cognitive control, the mental processes we use in daily life to adaptively switch between different tasks, and which are disrupted in certain mental disorders. Schematic of task-DyVA framework shown, Upper, Ebb and Flow gameplay data from one participant. Lower left, the transformed stimuli supplied as inputs to the model. Lower right, model outputs (blue) and model output targets (grey). Red dots indicate the model’s responses.
Credit
DOI: 10.1038/s41562-022-01510-8 Jaffe, P.I., Poldrack, R.A., Schafer, R.J. et al.
Usage Restrictions
by permission
License
Original content