🔬

Neuroprosthetics

1

Open Unknowns

0

Cross-Domain Bridges

2

Active Hypotheses

Open Unknowns (1)

Unknown What are the relative contributions of electrode impedance increase, neuronal loss from glial encapsulation, neural plasticity, and decoder parameter drift to the long-term performance degradation of intracortical neuroprosthetic decoders, and which failure mechanism should adaptive algorithms prioritise? u-neuroprosthetic-decoder-long-term-stability-mechanisms

Active Hypotheses

Hypothesis Fitts' law applies universally to brain-computer interface (BCI) cursor control: the information throughput of BCI pointing systems is bounded by the cortical motor channel capacity (~4 bits/second for intracortical BCIs, ~1 bit/second for EEG BCIs) regardless of the decoding algorithm used, because the bottleneck is biological (neural variability) rather than computational. medium

Hypothesis Motor cortex population activity during reaching lies on a low-dimensional (d ~ 6-12) smooth manifold embedded in the full neural state space, and neuroprosthetic decoders trained to operate in this manifold subspace show >2x improvement in robustness to neuron loss and inter-session non-stationarity compared to full-space Kalman filter decoders. high

Know something about Neuroprosthetics? Contribute an unknown or hypothesis →

Generated 2026-05-10 · USDR Dashboard