Assessing the Relative Impact of Grasp and Object on Inferior Frontal Gyrus Activity during a Grasping Task

The lateral grasp network, responsible for translating visual properties of an object to execution of a motor act, is comprised of the anterior intraparietal area (AIP), area F5, and the primary motor cortex (M1). Non-human primate studies of F5 have shown that it encodes a wide range of hand positions and object properties. Human studies in F5’s human homologue, the inferior frontal gyrus (IFG), have leveraged this area’s ability to encode grasp-object pairs for the purposes of Brain Machine Interface (BMI) control. However, whether modulation is driven by grasp, object, or the interaction between grasp and object is unclear. In the present study, sixty-four features were recorded from IFG during a motor visualization task where grasp and object were varied. Grasp was found to be the predominant factor driving modulation of IFG signals. Object was found to only be weakly represented in neural data. However, object contribution peaked earlier than grasp contribution, indicating early integration of object information. Grasp-object interactions were also found to have a significant impact. Cortical separation between grasping conditions varied based on the object presented. In addition, subspace analysis showed that the underlying neural population structure associated with each object type was significantly different from one another. Despite the impact of object type, the present study suggests that due to the significantly larger impact of grasp, BMI decoders can be used to decode grasp with above chance accuracy across a variety of grasp-object pairs.