English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

It will be possible to decode cortical motor signals to enable a robotic arm to accurately and quickly assist spinal cord injured patients. Understanding how quickly the brain plans reaches is important to quantify the performance limits of such a system. METHODS We trained a rhesus monkey to touch a center point on a video screen and to hold for 200-1000 ms while one of several targets arranged around a circle on a touch screen was shown. These included layouts with 2, 4, or 8 directions at a 10 cm radial distance and layouts with 4 or 8 directions at 6 and 12 cm radial distances. After a cue, he then performed reaches out to the specified target. Given a target location, the distribution of spike rates for each trial was modeled as a multivariate Gaussian. Models were computed on neural data from a microsurgically implanted a 96-channel electrode array into the premotor cortex. We then computationally predicted the upcoming reach from 100-275 ms of neural activity starting 150 ms after the target was shown. Given a successful prediction, the subject was rewarded without executing a real reach. RESULTS Using just 200 ms of neural activity the correct target location could be predicted using a maximum likelihood estimator on 80% of trials for the 1 of 8 target layout. When wrong, the predicted target was typically adjacent to the actual target. Prediction computation required ~20 ms. Up to 10 targets in a row were predicted from neural planning activity alone with no real reaches performed. Over 3.5 reach plans could be decoded in under one second for the 1 of 2 target layout. CONCLUSIONS Our prosthetic system is capable of fast and accurate positioning of prosthetic icons, faster than real reach capabilities.

785 The Speed at Which Reach Movement Plans Can Be Decoded from the Cortex and Its Implications for High-performance Neural Prosthetic Arm Systems