Skip to Main Content
Multi-unit, intra-cortical recordings from the primary motor cortex have been shown to provide information about functional movement of the body, and thus, have been used as command signals for control of an external robotic arm in rat and monkey. However, study of the M1 responses has shown that movement encoding may be dependent on both the functional and behavioral context of the intended motion. The main objective of the present work was to determine if self-paced, repetitive forelimb movements are effectively encoded in multiple-unit recordings from the primary motor cortex (M1) in freely moving, non-constrained rats. Four rats were chronically implanted with 7-channel, 50 μm tungsten micro-wire arrays. Standard psychophysical techniques were first used to train the rats to depress a response paddle in return for a food reward. We computed peri-event time histograms and found both statistically significant excitatory (24/49) and inhibitory (9/49) pre-paddle activity up to 200 ms before a paddle hit. On average, responses from 161±37 individual paddle hits were necessary in order to detect statistically significant (> 95%), excitatory pre-paddle action. Thus, while it is possible to detect self-paced, forelimb movements in multi-unit recordings of M1, the high number of repetitions required would limit the efficacy of a real-time cortical neuroprosthesis.