Panitz Martin, Deserno Lorenz, Kaminski Elisabeth, Villringer Arno, Sehm Bernhard, Schlagenhauf Florian
Department of Neurology, Max-Planck-Institute for Human Cognitive and Brain Sciences, 04103 Leipzig, Germany.
Department of Psychiatry and Psychotherapy, Campus Charité Mitte, Charité Universitätsmedizin Berlin, 10117 Berlin, Germany.
Cereb Cortex Commun. 2022 Jan 27;3(1):tgac006. doi: 10.1093/texcom/tgac006. eCollection 2022.
The medial prefrontal cortex (mPFC) is thought to be central for flexible behavioral adaptation. However, the causal relationship between mPFC activity and this behavior is incompletely understood. We investigated whether transcranial direct current stimulation (tDCS) over the mPFC alters flexible behavioral adaptation during reward-based decision-making, targeting Montreal Neurological Institute (MNI) coordinates = -8, = 62, = 12, which has previously been associated with impaired behavioral adaptation in alcohol-dependent patients. Healthy human participants ( = 61) received either anodal ( = 30) or cathodal ( = 31) tDCS versus sham tDCS while performing a reversal learning task. To assess the mechanisms of reinforcement learning (RL) underlying our behavioral observations, we applied computational models that varied with respect to the updating of the unchosen choice option. We observed that anodal stimulation over the mPFC induced increased choice switching after punishments compared with sham stimulation, whereas cathodal stimulation showed no effect on participants' behavior compared with sham stimulation. RL revealed increased updating of the unchosen choice option under anodal as compared with sham stimulation, which accounted well for the increased tendency to switch after punishments. Our findings provide a potential model for tDCS interventions in conditions related to flexible behavioral adaptation, such as addiction.
内侧前额叶皮质（mPFC）被认为是灵活行为适应的核心。然而，mPFC活动与这种行为之间的因果关系尚未完全明确。我们研究了经颅直流电刺激（tDCS）作用于mPFC时，是否会在基于奖励的决策过程中改变灵活行为适应能力，刺激靶点为蒙特利尔神经病学研究所（MNI）坐标：x = -8，y = 62，z = 12，该坐标先前已被证实与酒精依赖患者的行为适应受损有关。61名健康人类参与者在进行逆向学习任务时，分别接受了阳极（n = 30）或阴极（n = 31）tDCS刺激以及伪刺激。为了评估我们行为观察结果背后的强化学习（RL）机制，我们应用了在未被选择的选项更新方面存在差异的计算模型。我们观察到，与伪刺激相比，mPFC上的阳极刺激在惩罚后会导致选择转换增加，而阴极刺激与伪刺激相比对参与者的行为没有影响。强化学习显示，与伪刺激相比，阳极刺激下未被选择的选项更新增加，这很好地解释了惩罚后选择转换增加的趋势。我们的研究结果为tDCS干预与灵活行为适应相关的疾病（如成瘾）提供了一个潜在模型。
