2021
DOI: 10.1523/jneurosci.1338-21.2021
|View full text |Cite
|
Sign up to set email alerts
|

A One-Shot Shift from Explore to Exploit in Monkey Prefrontal Cortex

Abstract: Much animal learning is slow, with cumulative changes in behavior driven by reward prediction errors. When the abstract structure of a problem is known, however, both animals and formal learning models can rapidly attach new items to their roles within this structure, sometimes in a single trial. Frontal cortex is likely to play a key role in this process. To examine information seeking and use in a known problem structure, we trained monkeys in an explore/exploit task, requiring the animal first to test objec… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

2
10
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
3
2

Relationship

0
5

Authors

Journals

citations
Cited by 7 publications
(12 citation statements)
references
References 61 publications
(80 reference statements)
2
10
0
Order By: Relevance
“…Using fMRI, we investigated the neural correlates of the assessment of the possibility to use the information collected during the choice in the future, manipulated through horizon length, as well as the assessment of the contingency between choice and information, manipulated through the availability of the counterfactual outcome. Modulation of activity associated with exploratory behavior in an uncertain environment has been recorded in humans and monkeys in both the ACC and the MCC [42,[52][53][54], but here, we found interesting anatomical distinctions. We found that the pgACC was more active when the information could be used in the future in the long horizon.…”
Section: Strategic Exploration Signals In Acc/mcc and Dlpfcsupporting
confidence: 51%
See 1 more Smart Citation
“…Using fMRI, we investigated the neural correlates of the assessment of the possibility to use the information collected during the choice in the future, manipulated through horizon length, as well as the assessment of the contingency between choice and information, manipulated through the availability of the counterfactual outcome. Modulation of activity associated with exploratory behavior in an uncertain environment has been recorded in humans and monkeys in both the ACC and the MCC [42,[52][53][54], but here, we found interesting anatomical distinctions. We found that the pgACC was more active when the information could be used in the future in the long horizon.…”
Section: Strategic Exploration Signals In Acc/mcc and Dlpfcsupporting
confidence: 51%
“…Specifically, a network consisting of the MCC, the dlPFC, and, potentially, the locus coeruleus could support the relaxation of the effect of expected value on choices based on the context. Altogether, these results illustrate how ACC/MCC and dlPFC might dynamically switch modes to pursue different goals depending on the task demands [ 42 , 52 , 54 ]. Future studies will aim at testing whether switching mode is dependent on noradrenergic inputs and which causal role both regions play in changing into and out of strategic exploration.…”
Section: Discussionmentioning
confidence: 96%
“…Using fMRI, we investigated the neural correlates of the assessment of the possibility to use the information collected during the choice in the future, manipulated through horizon length, as well as the assessment of the contingency between choice and information, manipulated through the availability of the counterfactual outcome. Modulation of activity associated with exploratory behavior in an uncertain environment has been recorded in humans and monkeys in both the ACC and the MCC ( 5, 24, 39, 5153 ) but here we found interesting anatomical distinctions. We found that the pgACC was more active when the information could be used in the future in the long horizon.…”
Section: Discussionsupporting
confidence: 44%
“…Specifically, a network consisting of the MCC, the dlPFC and potentially the locus coeruleus could support the relaxation of the effect of expected value on choices based on the context. Altogether, these results illustrate how ACC/MCC and dlPFC might dynamically switch modes to pursue different goals depending on the task demands (24,39,53). Future studies will aim at testing whether switching mode is dependent on noradrenergic inputs and which causal role both regions play in changing into and out of strategic exploration.…”
Section: Strategic Exploration Signals In Acc/mcc and Dlpfcmentioning
confidence: 77%
“…Optimal ‘one-shot’ transitions would require only a single error trial to successfully switch behaviour 23,24 , as seen in Fig. 1d.…”
Section: Mainmentioning
confidence: 99%