Adaptive decision-making requires the ability to weigh the relative costs versus the benefits of our actions and to flexibly respond to changing environmental contingencies. While midbrain dopamine (DA) has been a major focus of study for its role in associative learning, motivation, and reward seeking, far less is known regarding the contribution of inhibitory dopamine inputs in promoting behavioral inhibition and avoidance learning. The rostromedial tegmental nucleus (RMTg) sends dense GABAergic projections to DA neurons in the ventral tegmental area (VTA) and mounting evidence suggests the RMTg-VTA circuit is required to suppress reward seeking under punishment, but it remains unclear whether stimulation of this pathway is sufficient to drive learning and promote shifts in cost-benefit decisions. To test this, we developed two separate tasks of passive and active avoidance in rats where lever pressing for food reward was associated with repeated contingent optogenetic stimulation of the RMTg-VTA pathway. In a one-lever fixed ratio 5 food-seeking task, we found that pairing RMTg-VTA stimulation with reward delivery caused a robust, yet transient suppression of reward seeking that quickly returned to baseline after contingent stimulation ceased. When given an alternative reward choice, however, RMTg-VTA stimulation caused a rapid and enduring shift in subjective choice leading to persistent and selective avoidance of the stimulation-paired reward. These findings support a multifaceted role for the RMTg-VTA pathway in learning and decision-making and provide key insights into the neural mechanisms underlying behavioral avoidance and maladaptive reward seeking.

适应性决策需要权衡我们行动的相对成本与收益，并能够灵活地应对环境变化。虽然中脑多巴胺（DA）在关联学习、动机和寻求奖励方面的研究已经很深入，但关于抑制性多巴胺输入在促进行为抑制和回避学习中的作用知之甚少。背缝核（RMTg）向腹侧被盖区（VTA）的多巴胺神经元发送密集的GABA能投射，越来越多的证据表明，RMTg-VTA回路在惩罚条件下抑制奖励寻求中是必需的，但尚不清楚刺激该通路是否足以驱动学习并促进成本效益决策的变化。为了测试这一点，我们在大鼠身上开发了两个不同的回避任务（被动和主动），其中杠杆按压以获取食物奖励与反复关联性的光遗传学刺激RMTg-VTA路径相关联。在一个一杆固定比例5的食物寻求任务中，我们发现将RMTg-VTA刺激与奖励递送配对会导致强烈的但短暂的抑制奖励寻求，在关联性刺激停止后很快恢复到基线水平。然而，当提供一个替代的奖励选择时，RMTg-VTA刺激导致主观选择迅速而持久地转变，从而持续且有选择地避免与刺激相关的奖励。这些发现支持了RMTg-VTA路径在学习和决策中的多方面作用，并为理解行为回避和适应不良的奖励寻求的神经机制提供了关键见解。