摘要
奖励价值编码是价值决策过程的核心组成部分,大脑通过评估不同选项的奖励价值来指导行为选择。前额叶皮层是大脑高级认知功能的核心区域,先前众多研究结果表明,前额叶皮层中的多个子区域都参与了对奖励信息的编码。然而,这些研究往往只关注特定实验中单个脑区或两个脑区的活动,在不同的实验设计中同一脑区的功能可能有不同的解释。为了同时探究前额叶皮层的多个子区域是如何编码奖励价值信息的,本文基于一个自由选择(free-choice)的猕猴奖励预测实验,使用四片脑皮层电图(electrocorticography,ECoG)电极阵列(每个阵列有64通道)同时记录了前扣带回皮层(anterior cingulate cortex,ACC)、外侧前额叶皮层(lateral prefrontal cortex,LPFC)、内侧前额叶皮层(medial prefrontal cortex,MPFC)和眶额叶皮层(orbitofrontal cortex,OFC)的电信号。该实验设计了五种刺激-奖励组合,使猕猴能将刺激与对应果汁奖励关联起来,根据猕猴对每种果汁奖励的选择频率判断喜爱程度及所代表的价值。比较四个脑区记录的ECoG在不同频带和不同奖励条件下的平均功率强度发现:OFC的alpha频带和LPFC的theta、alpha、beta频带的平均功率不仅能区分五种奖励,而且其强度排序与猕猴对不同果汁奖励的喜好程度一致;MPFC和ACC的alpha、beta频带则受到刺激的调制。特别地,与奖励价值编码相关的平均功率强度发放大部分集中在刺激消失后、奖励发放前的延迟期阶段。以上结果表明,在同一个奖励任务下,尽管前额叶皮层的四个脑区都与奖励编码相关,但侧重的功能有所不同,这有助于后续探究前额叶皮层各个子区域在奖励价值编码中的相互作用。
Calculating reward value is a core component in the value-based decision-making process,where the brain guides behavioral choices by evaluating the reward values of different options.The prefrontal cortex(PFC),a central hub for higher-order cognitive functions,has been involved in reward encoding across multiple subregions,as evidenced by numerous experimental studies.However,these studies often focused on the activity of a single or two brain regions within specific experimental paradigms,leading to potential variability in the functional interpretation of the same region across different experimental designs.To simultaneously investigate how multiple subregions of the PFC encode reward values,our study was based on a free-choice reward prediction task in macaques,recording neural signals from the anterior cingulate cortex(ACC),lateral prefrontal cortex(LPFC),medial prefrontal cortex(MPFC),and orbitofrontal cortex(OFC)by a 64-channel electrocorticography(ECoG)electrode array in each area.The task designed five stimulus-reward pairings,so that macaques could associate stimuli with corresponding juice rewards.The preferences and values of each reward were determined by the macaques'choice frequencies.We compared the mean power intensity of ECoG in different frequency bands and reward conditions in the four subregions.On the alpha band of OFC and the theta,alpha,beta bands of LPFC,we found that the mean power not only distinguished the five reward conditions,but also exhibited the consistency between the rankings of mean power intensity and the macaques'preferences of juice rewards.Meanwhile,the mean power on the alpha and beta bands of MPFC and ACC were modulated by the stimuli.Notably,the mean power related with reward value encoding was predominantly concentrated during the delay period,which is between the stimulus period and the reward period.These results suggest that,within the same reward task,the four PFC subregions are differentially involved in reward encoding,highlighting their distinct functional contributions.This finding provides a foundation for further exploration of the interactions among PFC subregions in reward value encoding.
作者
林申华
王毅泓
徐旭颖
王如彬
潘晓川
Masamichi Sakagami
LIN Shen-Hua;WANG Yi-Hong;XU Xu-Ying;WANG Ru-Bin;PAN Xiao-Chuan;SAKAGAMI Masamichi(Institute of Cognitive Neurodynamics,School of Mathematics,East China University of Science and Technology,Shanghai 200237,China;Brain Science Institute,Tamagawa University,Tokyo 194-8610,Japan)
出处
《生理学报》
北大核心
2025年第6期1171-1186,共16页
Acta Physiologica Sinica
基金
supported by the National Natural Science Foundation of China(No.12172132,12272136,12472054).
