Planning in Nonhuman Primates Emerges from Structure Knowledge and is Distinct from Attention, Working Memory, Effort Control, and Learning
Wen, X.; Neumann, A.; Dhungana, S.; Tiesinga, P.; Womelsdorf, T.
Show abstract
Forward planning involves mentally simulating future choices. Such mental simulation may constitute an independent cognitive operation, or it may be intricately linked to canonical cognitive operations supporting working memory of choice options, covert attentional information sampling, forming memory models of problem structure and effort control for pursuing future goals. Here, we test the relative independence of planning in non-human primates with an object-sequence planning task assessed together with an assay of simpler cognitive tasks. The sequence task unexpectedly masked objects at future serial positions probing NHPs to plan ahead serial future choices. We found that five of six subjects planned ahead 2-3 ordinal positions. Planning depth was predicted by knowledge of the serial structure but constituted a unique cognitive factor that was distinct from sequence memory, object memory, and learning efficacy. Similarly, planning was segregated from working memory capacity, attention, inhibitory control and effort control. These findings suggest that planning ahead constitutes a separate construct not reducible to other domain-general cognitive operations, and emerges in NHPs that successfully form an internal model of the problem structure.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Location- and feature-based selection histories make independent, qualitatively distinct contributions to urgent visuomotor performance 96%
- Neural mechanisms of credit assignment for delayed outcomes during contingent learning 96%
- Phasic arousal suppresses biases in mice and humans across domains of decision-making 96%
"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.