Auto-deleting brain machine interface: Error detection using spiking neural activity in the motor cortex

Abstract

Brain machine interfaces (BMIs) aim to assist people with paralysis by increasing their independence and ability to communicate, e.g., by using a cursor-based virtual keyboard. Current BMI clinical trials are hampered by modest performance that causes selection of wrong characters (errors) and thus reduces achieved typing rate. If it were possible to detect these errors without explicit knowledge of the task goal, this could be used to automatically "undo" wrong selections or even prevent upcoming wrong selections. We decoded imminent or recent errors during closed-loop BMI control from intracortical spiking neural activity. In our experiment, a non-human primate controlled a neurally-driven BMI cursor to acquire targets on a grid, which simulates a virtual keyboard. In offline analyses of this closed-loop BMI control data, we identified motor cortical neural signals indicative of task error occurrence. We were able to detect task outcomes (97% accuracy) and even predict upcoming task outcomes (86% accuracy) using neural activity alone. This novel strategy may help increase the performance and clinical viability of BMIs.

Motor Cortical Visuomotor Feedback Activity Is Initially Isolated from Downstream Targets in Output-Null Neural State Space Dimensions

“…If a parallel BMI ‘execution error decoder’ could infer perturbation specifics or the desired correction quickly and with high accuracy, the system could automatically correct the error faster than the BMI user would otherwise be able to. This proposed motor execution error auto-correction has parallels to our ongoing work to detect and correct for task outcome errors (Even-Chen et al, 2015). …”

Augmenting intracortical brain-machine interface with neurally driven error detectors

“…The decoder output a two-dimensional BMI cursor velocity control signal. Our experiment was designed to resemble a ‘typing task’: the monkeys had to acquire a specific target cued in green amongst a keyboard-like grid of selectable yellow targets using the BMI-controlled cursor [43,75,76] (figure 1a, Methods). We delayed reward and auditory feedback for 600 ms following target selection (figure 1a–iv) to temporally separate neural activity reflecting the monkey’s (presumed) recognition of the task’s outcome from neural activity related to explicitly receiving the liquid reward on successful trials.…”

“…They were free to move their arm even during BMI control [40,74,75]. A keyboard-like task was modeled after the task described in [75,76]. The goal of the task and the experiment timeline is depicted in the Behavioral Task section of the Results and figure 1.…”

“…When comparing BMI error detection online using the FIT-KF decoder, we modified the monkeys’ typing task in two ways to make it more analogous to human typing [43,81]. First, to simulate how users must press the ‘delete’ key after an incorrect selection and then correctly select the missed key, we cued a predefined delete key after every incorrect selection.…”

Virtual typing by people with tetraplegia using a self-calibrating intracortical brain-computer interface

“…However, more sophisticated methods could be incorporated that improve the estimate of the person's true aiming direction by, for example, iteratively recomputing the aiming direction and tuning models until they converge (28) or estimating and taking into account the person's internal model of the cursor's expected behavior under neural control (36,37). Also, the selection of particular segments of data to be included in filter calibration was based on a few simple heuristics in our study, but could conceivably be refined by taking into account information that can be inferred from the neural signals about the person's attentional state, intention to move, or error signals in local field potentials (38)(39)(40)(41)(42)(43). Finally, the time constants and other parameters determining the behavior of each of our methods have been hard-coded to values that were anecdotally found to work well across many sessions and several participants.…”