Brain-machine interfaces (BMIs) have demonstrated how they can be used for reaching tasks with both invasive and non-invasive signal recording methods. Despite the constant improvements in this field, there still exist diverse factors to overcome before achieving a natural control. In particular, the high variability of the brain signals often leads to the incorrect decoding of the subject intentions, producing unreliable behaviours in the controlled device. A possible solution to this problem would be that of correcting this erroneous decoding using a feedback signal from the user. In this work, we evaluate the possibility of decoding neural signals associated to performance monitoring (EEG-recorded error-related potentials) during a reaching task. Compared to previous works where these error potentials were recorded under scenarios with discrete movements performed by the cursor, under real conditions the cursor is moving continuously and thus the system is required to asynchronously detect any possible error. To this end, we simulated two different erroneous events during the monitoring of a reaching task: errors at the beginning of the movement, and errors happening in the middle of the trajectory being executed. Through the analysis of the recorded EEG of three subjects, we demonstrate the existence of neural correlates for the two types of elicited error potentials, and we are able to asynchronously detect them with high accuracies.
Looking for publications? You might want to consider searching on the EPFL Infoscience site which provides advanced publication search capabilities.
Objective. Decoding forelimb movements from the firing activity of cortical neurons has been interfaced with robotic and prosthetic systems to replace lost upper limb functions in humans. Despite the potential of this approach to improve locomotion and facilitate gait rehabilitation, decoding lower limb movement from the motor cortex has received comparatively little attention. Here, we performed experiments to identify the type and amount of information that can be decoded from neuronal ensemble activity in the hindlimb area of the rat motor cortex during bipedal locomotor tasks. Approach. Rats were trained to stand, step on a treadmill, walk overground and climb staircases in a bipedal posture. To impose this gait, the rats were secured in a robotic interface that provided support against the direction of gravity and in the mediolateral direction, but behaved transparently in the forward direction. After completion of training, rats were chronically implanted with a micro-wire array spanning the left hindlimb motor cortex to record single and multi-unit activity, and bipolar electrodes into 10 muscles of the right hindlimb to monitor electromyographic signals. Whole-body kinematics, muscle activity, and neural signals were simultaneously recorded during execution of the trained tasks over multiple days of testing. Hindlimb kinematics, muscle activity, gait phases, and locomotor tasks were decoded using offline classification algorithms. Main results. We found that the stance and swing phases of gait and the locomotor tasks were detected with accuracies as robust as 90% in all rats. Decoded hindlimb kinematics and muscle activity exhibited a larger variability across rats and tasks. Significance. Our study shows that the rodent motor cortex contains useful information for lower limb neuroprosthetic development. However, brain-machine interfaces estimating gait phases or locomotor behaviors, instead of continuous variables such as limb joint positions or speeds, are likely to provide more robust control strategies for the design of such neuroprostheses.