Hand Position Fields of Neurons in the Premotor Cortex of Macaques during Natural Reaching

Abstract

Planning and execution for goal-directed movements requires the integration of the current position of body or body parts with various kinematic parameters of the movement itself. In the hippocampus, the field-based representation of spatial information plays an essential role in this process during navigation. However, if a similar, field-based encoding framework is also utilized by the motor areas during hand reaching remains unclear. In this study, we investigated the hand position tuning in the dorsal premotor cortex (PMd) neurons (n = 601) when four monkeys performing a naturalistic reach-and-grasp task. We show that 132/601 (22%) of PMd neurons increased their firing rates when the monkeys hand occupied specific positions in the space, forming the "position fields" in their spatial firing maps that can be well described by 2D Gaussian functions/kernels. We further analyzed how the field-tuning of hand position is co-represented with other task-related parameters including the hand moving direction, speed, and reward location in the same population of PMd neurons, revealing a mixed-selective framework also similar to that discovered in the hippocampus. These position-tuned cells demonstrated high efficiency in encoding hand position, with ~10% of overall recorded neurons contributing >80% of accuracy in decoding instantaneous hand moving trajectories. These results suggest that a field-based encoding framework of position may be a common component to representing spatial information and integrating it with kinematic parameters for guiding goal-directed movements of the body or body parts.