HM Hospitales - HM CINAC, 28938 Móstoles, and CEU-San Pablo University, 28003 Madrid, Spain.
Clinical & Experimental Neuroscience (NiCE-IMIB-IUIE), Department of Human Anatomy & Psychobiology, School of Medicine, Campus Espinardo, University of Murcia, 30071 Murcia, Spain.
Neuroscience. 2021 Mar 15;458:166-179. doi: 10.1016/j.neuroscience.2020.12.035. Epub 2021 Jan 19.
In everyday life, risky decision-making relies on multiple cognitive processes including sensitivity to reinforcers, exploration, learning, and forgetting. Neuroimaging evidence suggests that the dorsolateral prefrontal cortex (DLPFC) is involved in exploration and risky decision-making, but the nature of its computations and its causal role remain uncertain. We provide evidence for the role of the DLPFC in value-independent, directed exploration on the Iowa Gambling Task (IGT) and we describe a new computational model to account for the competition of directed exploration and exploitation in guiding decisions. Forty-two healthy human participants were included in a right DLPFC, left DLPFC or sham stimulation groups using continuous theta-burst stimulation (cTBS). Immediately after cTBS, the IGT was completed. Computational modelling was used to account for exploration and exploitation with different combinations with value-based and sensitivity to reinforcers for each group. Applying cTBS to the left and right DLPFC selectively decreased directed exploration on the IGT compared to sham stimulation. Model-based analyses further indicated that the right (but not the left) DLPFC stimulation increased sensitivity to reinforcers, leading to avoidance of risky choices and promoting advantageous choices during the task. Although these findings are based on small sample sizes per group, they nevertheless elucidate the causal role of the right DLPFC in governing the exploration-exploitation tradeoff during decision-making in uncertain and ambiguous contexts.
在日常生活中,风险决策依赖于多种认知过程,包括对强化物的敏感性、探索、学习和遗忘。神经影像学证据表明,背外侧前额叶皮层(DLPFC)参与探索和风险决策,但它的计算性质及其因果作用仍不确定。我们提供了证据表明 DLPFC 在独立于价值的、有导向的探索中的作用,即在爱荷华赌博任务(IGT)上,并且我们描述了一个新的计算模型来解释在指导决策时,有导向的探索和利用之间的竞争。使用连续 theta 爆发刺激(cTBS),42 名健康人类参与者被纳入右 DLPFC、左 DLPFC 或假刺激组。在 cTBS 之后,立即完成 IGT。使用计算模型来解释探索和利用,每个组都有不同的基于价值和对强化物的敏感性的组合。将 cTBS 应用于左右 DLPFC 会选择性地降低 IGT 上的有导向探索,与假刺激相比。基于模型的分析进一步表明,右(但不是左)DLPFC 刺激增加了对强化物的敏感性,导致在任务中避免风险选择并促进有利选择。尽管这些发现基于每个组的小样本量,但它们阐明了右 DLPFC 在不确定和模糊情境下决策中管理探索-利用权衡的因果作用。