About Entrez
Text Version
Entrez PubMed
Overview
Help |
FAQ
Tutorial
New/Noteworthy
E-Utilities
PubMed Services
Journals Database
MeSH Database
Single Citation Matcher
Batch Citation Matcher
Clinical Queries
LinkOut
Cubby
Related Resources
Order Documents
NLM Catalog
NLM Gateway
TOXNET
Consumer Health
Clinical Alerts
ClinicalTrials.gov
PubMed Central
|
|
Elimination of response latency variability in neuronal spike trains.
Nawrot MP, Aertsen A, Rotter S.
Neurobiology and Biophysics, Institute of Biology III, Albert-Ludwigs-University, 79104 Freiburg, Germany.
Neuronal
activity in the mammalian cortex exhibits a considerable amount of
trial-by-trial variability. This may be reflected by the magnitude of
the activity as well as by the response latency with respect to an
external event, such as the onset of a sensory stimulus, or a
behavioral event. Here we present a novel nonparametric method for
estimating trial-by-trial differences in response latency from neuronal
spike trains. The method makes use of the dynamic rate profile for each
single trial and maximizes their total pairwise correlation by
appropriately shifting all trials in time. The result is a new
alignment of trials that largely eliminates the variability in response
latency and provides a new internal trigger that is independent of
experiment time. To calibrate the method, we simulated spike trains
based on stochastic point processes using a parametric model for phasic
response profiles. We illustrate the method by an application to
simultaneous recordings from a pair of neurons in the motor cortex of a
behaving monkey. It is demonstrated how the method can be used to study
the temporal relation of the neuronal response to the experiment, to
investigate whether neurons share the same dynamics, and to improve
spike correlation analysis. Differences between this and other
previously published methods are discussed.
PMID: 12750895 [PubMed - indexed for MEDLINE]
|