48 Should I Stay or Should I Go? Neural Circuits Underlying Decisions to Explore or Exploit

Objective: Adaptive decision-making is necessary to sustain functional independence. Maladaptive decisions are among the most prevalent features of psychological and neurological disorders. One crucial aspect of decision-making involves arbitrating between exploring new avenues with risky but potentially lucrative outcomes or exploiting prior knowledge and endorsing predictable outcomes. Balancing this dichotomy creates a behavioral tension that shapes all decisions and is termed the exploration-exploitation trade-off. This trade-off has been linked to reward and affective drives and associated neural circuitry as well as neuropsychological dysfunction. However, the neural mechanisms underlying the exploration-exploitation trade-off are still uncertain, due to the scarcity of literature and the heterogeneity of paradigms. This study aimed to systematically quantify and disambiguate neuroanatomical correlates of the exploration-exploitation tradeoff in a normative adult sample. These findings provide a necessary starting point for future investigations of this fundamental aspect of decision-making across clinical populations, with potential implications for assessment and intervention. Participants and Methods: We used the effect-location method of meta-analysis to analyze data from 10 functional neuroimaging studies investigating the exploration-exploitation tradeoff in non-clinical samples. We analyzed the location and frequency of significant neural activations across studies for both explorative and exploitative decisions and characterized them as core and non-core regions. Core activations were defined as those reported in over 50% of studies. Secondary and tertiary activations were defined as those reported in 40% and 30% of studies, respectively. The present review was conducted in accordance with the guidelines of the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Results: The results revealed that explorative and exploitative choice behaviours differed markedly with respect to associated patterns of task-related brain activity. Exploration was associated with activity in brain regions implicated in externally directed, goal-based attentional processing and reward-related uncertainty, mainly tapping bilateral parietal and frontal circuitry, with relatively high consistency across studies. A core explorative network was revealed, consisting of activity in the frontal polar cortex, the dorsal anterior cingulate cortex, the bilateral medial frontal gyrus, the bilateral precuneus, and the bilateral intraparietal sulcus. Secondary and tertiary regions were also detected, including the bilateral anterior insula, the left precentral gyrus, the bilateral superior frontal gyrus, the right inferior frontal gyrus, the left supplementary motor area, the bilateral superior parietal lobule, and the bilateral thalamus. Exploitation was associated with brain regions implicated in internally directed processes including reward valuation, motivation, and memory. Core exploitative activations included the ventromedial prefrontal cortex, the bilateral anterior cingulate cortex, and the bilateral orbitofrontal cortex. Secondary and tertiary activations included the bilateral hippocampus, the left middle temporal gyrus, the bilateral angular gyrus, the left posterior cingulate cortex, the left superior frontal gyrus, and the bilateral superior temporal gyrus. Conclusions: The exploration-exploitation trade-off provides a novel paradigmatic approach to study adaptive and maladaptive decision-making behaviour in humans. Our findings support the neural dichotomization of exploration and exploitation and illuminate potential neural networks underlying this fundamental feature of decision-making. Understanding these mechanistic networks opens a new avenue of inquiry into decision-making deficits in clinical populations, including neurodegenerative, neurodevelopmental, and neuropsychiatric syndromes.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.