The Department of Psychology, University of Oslo, Oslo, Norway.
Norwegian Center for Addiction Research, Department of Clinical Medicine, University of Oslo, Oslo, Norway.
Neuropsychopharmacology. 2017 Aug;42(9):1833-1840. doi: 10.1038/npp.2017.58. Epub 2017 Mar 15.
Modifying behavior to maximize reward is integral to adaptive decision-making. In rodents, the μ-opioid receptor (MOR) system encodes motivation and preference for high-value rewards. Yet it remains unclear whether and how human MORs contribute to value-based decision-making. We reasoned that if the human MOR system modulates value-based choice, this would be reflected by opposite effects of agonist and antagonist drugs. In a double-blind pharmacological cross-over study, 30 healthy men received morphine (10 mg), placebo, and the opioid antagonist naltrexone (50 mg). They completed a two-alternative decision-making task known to induce a considerable bias towards the most frequently rewarded response option. To quantify MOR involvement in this bias, we fitted accuracy and reaction time data with the drift-diffusion model (DDM) of decision-making. The DDM analysis revealed the expected bidirectional drug effects for two decision subprocesses. MOR stimulation with morphine increased the preference for the stimulus with high-reward probability (shift in starting point). Compared to placebo, morphine also increased, and naltrexone reduced, the efficiency of evidence accumulation. Since neither drug affected motor-coordination, speed-accuracy trade-off, or subjective state (indeed participants were still blinded after the third session), we interpret the MOR effects on evidence accumulation efficiency as a consequence of changes in effort exerted in the task. Together, these findings support a role for the human MOR system in value-based choice by tuning decision-making towards high-value rewards across stimulus domains.
改变行为以最大化奖励是适应性决策的关键。在啮齿动物中,μ-阿片受体(MOR)系统编码对高价值奖励的动机和偏好。然而,目前尚不清楚人类 MOR 是否以及如何参与基于价值的决策。我们推断,如果人类 MOR 系统调节基于价值的选择,那么激动剂和拮抗剂药物的相反作用将反映出来。在一项双盲药物交叉研究中,30 名健康男性接受了吗啡(10 毫克)、安慰剂和阿片拮抗剂纳曲酮(50 毫克)。他们完成了一个二择一决策任务,该任务已知会导致对最常奖励的响应选项产生相当大的偏差。为了量化 MOR 在这种偏差中的作用,我们使用决策的漂移扩散模型(DDM)拟合了准确性和反应时间数据。DDM 分析揭示了两个决策子过程的预期双向药物效应。吗啡刺激 MOR 增加了对高奖励概率刺激的偏好(起点转移)。与安慰剂相比,吗啡还增加了,纳曲酮减少了,证据积累的效率。由于两种药物都不影响运动协调、速度-准确性权衡或主观状态(实际上,参与者在第三次会议后仍然处于蒙蔽状态),我们将 MOR 对证据积累效率的影响解释为任务中努力变化的结果。总之,这些发现支持人类 MOR 系统在基于价值的选择中的作用,通过调整决策以在刺激域中偏向高价值奖励。