Neural basis of deduced ordinal judgments in the primate prefrontal cortex

根据先前获得的信息推断两个对象之间的顺序关系是演绎推理的一种形式，称为传递性推理（TI）。例如，如果A>B且B>C，则可以推断A>C。尽管序数信息和TI对智能行为的重要性，其神经机制仍然在很大程度上未被探索。在这个项目中，前额叶机制的规则为基础的推理能力将研究领域的序数数值能力。恒河猴将被训练来推断和记忆作为五个类别的范例的图片的顺序位置，并基于规则提示来决定是否根据较低或较高的等级对类别进行排序。这项任务需要工作记忆，灵活分配新的图片到各自的视觉类别，以及掌握有序性，以允许决策。根据核心假设1，我们怀疑行为猴子的前额叶神经元集合在任务的不同时期变得动态协调，以灵活地编码新类别成员的推断和相对排名。将通过细胞外记录评估推定的兴奋性锥体神经元和抑制性中间神经元的贡献，以了解引起TI过程的微电路。通过评估局部场电位与单单位记录相结合，神经元的同步性在分配神经元到不同的合奏作为一个假定的机制，灵活性在ordinality判断的作用将被探讨。为了探索核心假设2和3，同时记录从潜在的皮层输入结构到PFC，如后顶叶皮层，和潜在的输出区域，如运动前皮层，将有助于解决输入和输出定义的合奏在灵活的行为。这些数据将有助于破译非符号逻辑推理过程中认知灵活性的前额叶机制和回路。

Inferring the ordinal relationship between two objects based on earlier acquired information is a form of deductive reasoning called transitive inference (TI). For instance, if A>B and B>C, it can be inferred that A>C. Despite the importance of ordinal information and TI for intelligent behaviors, its neuronal mechanisms remain largely unexplored. In this project, the prefrontal mechanisms of a rule-based deduction capability will be studied in the realm of ordinal numerical competence. Rhesus monkeys will be trained to deduce and memorize the ordinal position of pictures that are exemplars of five categories, and decide based on a rule cue whether to order the categories according to lower or higher rank. This task requires working-memory, flexible assignment of novel pictures to the respective visual categories, as well as a grasp of ordinality to allow decision-making. In line with core hypothesis 1, we suspect that prefrontal neuron ensembles of the behaving monkeys become dynamically coordinated during different epochs of the task to flexibly encode the inferred and relative rank of novel category members. The contributions of putative excitatory pyramidal neurons and inhibitory interneurons will be assessed via extra-cellular recordings to learn about the micro-circuitry giving rise to TI processes. By assessing local field potentials in combination with single-unit recordings, the role of neuronal synchrony in assigning neurons to different ensembles as a putative mechanism for flexibility during ordinality judgments will be explored. To explore core hypotheses 2 and 3, simultaneous recordings from potential cortical input structures to the PFC, such as the posterior parietal cortex, and potential output areas, such as the premotor cortex, will help to resolve input- and output-defined ensembles during flexible behavior. The data will be instrumental in deciphering the prefrontal mechanisms and circuits for cognitive flexibility during nonsymbolic logical reasoning.