Neuron-type-specific Contributions to Dynamic Coding during Flexible Sensorimotor Decisions in Frontoparietal Cortex

Neocortical circuits consist of multiple neuronal cell types, each likely playing distinct roles in flexible behavior. However, studies of decision-making have often overlooked these cell types, limiting our understanding of their specific contributions to local circuit functions. To address this, we simultaneously recorded neuronal activity from the frontal eye field (FEF), lateral PFC, and lateral intraparietal area (LIP) in a macaque monkey performing a visuomotor decision-making task. We used extracellular spike waveforms to reliably identify two functional classes of neurons: broad-spiking (BS) putative pyramidal cells and narrow-spiking (NS) putative interneurons. These cell types exhibited distinct response dynamics and choice-related information encoding across cortical regions. NS neurons in LIP and PFC showed higher choice-related activity and contributed to early encoding of decisions, whereas in FEF, NS neurons demonstrated dynamic encoding patterns, with BS neurons exhibiting significantly more stable encoding. Our findings reveal that choice information is represented differently across cell types and cortical regions, with NS neurons favoring early population coding in PFC and LIP and BS neurons exhibiting more static encoding in FEF. This heterogeneous coding strategy suggests that decision-related dynamics in the frontoparietal network are shaped by interactions between these distinct neuronal populations. The results provide new insights into cortical circuit dynamics and cell-type-specific contributions to decision-making.