| |
Abstract:
understand how this variability arises, one must examine each
stage of information processing. According to the conventional
view, precise temporal information will be gradually lost as the
information is passed through a layered network of mean-rate
"units." To test, in humans, whether populations of neurons at
different stages of processing behave like mean-rate "units," we
applied Blind Source Separation to MEG signals obtained during a
sensory-motor integration task. We identified multiple visual
sources differing in response latency and duration, and estimated
response time latency and response time variability by detecting
single-trial stimulus-locked events for each source. We found a
greater response time variability in the early visual responses
than the later visual responses. This finding supports the
hypothesis that variability in populational responses increases at
successive processing stages, despite evidence in the literature
that single neuron spike jitters at successive synaptic stages can
be preserved
|
