guest ::
| Home > Publications database > Modeling Temporally Precise Spike Artefacts to Study Their Impact on Spike Correlation Analyses |
| Home > Publications database > Modeling Temporally Precise Spike Artefacts to Study Their Impact on Spike Correlation Analyses |
| Poster (Other) | FZJ-2020-04261 |
; ; ; ;
2020
This record in other databases:
Please use a persistent id in citations: http://hdl.handle.net/2128/26349 doi:10.12751/NNCN.BC2020.0095
Abstract: Due to technical advances, the number of neurons recorded in parallel increases drastically. This development reveals new types of artefacts: Common noise and cross-talk are observed in the raw parallel recording signals [1-3], as well as hyper-synchronous spike events at sampling rate precision in sorted spike data (‚synchrofacts‘; 3-5). These likely originate from environmental electromagnetic signals that couple into the recording signals. Here we concentrate on synchrofacts and their effects on results of spike data analyses, such as cross-correlation and higher-order synchronous or spatio-temporal spike patterns.In experimental data sets, e.g. recorded in macaque M1/PM1 with 100-electrode Utah arrays and manually spike sorted, we noticed synchrofacts in population histograms at a bin size matching the sampling rate (30 kHz). The complexity distribution [6], i.e. the histogram of synchronous events of a certain size (number of spikes across neurons) and their counts, shows entries up to 60, far larger than predicted by independent data of the same rate. Not all channels participate equally, but this is not related to the spatial electrode distribution.To systematically study the effects of the synchrofacts on analysis results, we formulate a stochastic data generation model in which we have control over synchrofacts, ‚neuronal‘ correlations and firing rates. We model background activity as independent Poisson processes and inject ‚neuronal‘ correlations and synchrofacts each formulated by separate Compound Poisson Processes (CPPs, 7-8). A CPP generates synchronous events with event sizes given by its amplitude distribution and inserts these spikes randomly into the neuronal spike trains. To model the observed synchrofacts we adapt the spike assignment with a non-uniform assignment distribution.In the next step, we apply various analysis methods to the artificial data to determine how synchrofacts affect the analysis results. Questions we are going to address are: a) does the presence of synchrofacts decrease the detectability of neuronal correlation activity, b) which type of correlation activity (pairwise or higher-order) is more diluted, and c) which synchrofact parameters (rate, correlation order, distribution over neurons, distribution over time) are mostly affecting the results. In order to do this we compare the analysis results from data with and without synchrofacts. This allows us to propose suitable methods of dealing with them.ReferencesMusial P, Baker S, Gerstein G, King E, Keating J (2002) Signal-to-noise ratio improvement in multiple electrode recording, Journal of Neuroscience Methods, Volume 115, Issue 1, Pages 29-43, ISSN 0165-0270, 10.1016/S0165-0270(01)00516-7Dann B, Michaels JA, Schaffelhofer S, Scherberger H (2016) Uniting functional network topology and oscillations in the fronto-parietal single unit network of behaving primates, eLife 2016;5:e15719, 10.7554/eLife.15719Essink S, Kleinjohann A, Barthélemy F, Ito J, Riehle A, Brochier T, Grün S (2019) Detection and Removal of Artefacts in Multi-Channel Electrophysiology Recordings, Bernstein Conference 2019, 10.12751/nncn.bc2019.0068Sprenger J (2014) Spatial Dependence of the Spike-Related Component of the Local Field Potential in Motor Cortex (Master’s thesis, RWTH Aachen).Torre E, Quaglio P, Denker M, Brochier T, Riehle A, Grün S (2016) Synchronous Spike Patterns in Macaque Motor Cortex during an Instructed-Delay Reach-to-Grasp Task, J. Neurosci. 36(32):8329-8340, 10.1523/JNEUROSCI.4375-15.2016Grün S, Abeles M, Diesmann M (2008) Impact of Higher-Order Correlations on Coincidence Distributions of Massively Parallel Data, in: Marinaro M, Scarpetta S, Yamaguchi Y (eds) Dynamic Brain - from Neural Spikes to Behaviors, NN 2007, Lecture Notes in Computer Science, vol 5286. Springer, Berlin, 10.1007/978-3-540-88853-6_8Kuhn A, Aertsen A, Rotter S (2003) Higher-Order Statistics of Input Ensembles and the Response of Simple Model Neurons, Neural Computation 15:1, 67-101, 10.1162/089976603321043702Staude B, Grün S, Rotter S (2010) Higher-Order Correlations and Cumulants, in: Grün S, Rotter S (eds) Analysis of Parallel Spike Trains, Springer Series in Computational Neuroscience, vol 7, Springer, Boston, MA, 10.1007/978-1-4419-5675-0_12
|
The record appears in these collections: |
PDF
PDF (PDFA)