[Comp-neuro] Toolbox available for linear and nonlinear modeling of sensory neuron spike trains

Daniel A. Butts dab at umd.edu
Mon Sep 23 01:22:54 CEST 2013

Dear Colleagues,

To understand the elements of visual and auditory stimuli that drive sensory neuronal responses, receptive field estimation using spike-triggered averaging has been a traditional approach. However, a range of maximum-likelihood based approaches have been adapted for neuroscience which are able to surmount many of the limitations of spike triggered averaging. For example, the Generalized Linear Model (GLM) allows for receptive field estimation in the context of natural (or highly correlated stimuli) including the incorporation of spike history effects [1,2], quadratic models can capture nonlinear stimulus processing through a generalization of spike-triggered covariance analysis [3], and our Nonlinear Input Model (NIM) allows for describing nonlinear processing in terms of the integration of physiologically interpretable excitatory and inhibitory inputs [4,5]. The latter is described in our recent paper [5]:


We have created Matlab toolbox, which implements these three approaches, which we have made available on our lab website:


In addition to this toolbox, we also provide several tutorials illustrating applications of these techniques to different types of sensory neurons. We encourage experimental labs to apply these models to their own data, and computational labs to explore and further develop these models -- and will be glad to incorporate feedback and post new applications from the community.

Happy modeling!

Dan Butts (dab at umd.edu)
Assistant Professor, Dept of Biology
University of Maryland, College Park

References cited:
[1] Paninski L (2004) Maximum likelihood estimation of cascade point-process neural encoding models. Network 15: 243-262.
[2] Truccolo W, Eden UT, Fellows MR, Donoghue JP, Brown EN (2005) A point process framework for relating neural spiking activity to spiking history, neural ensemble, and extrinsic covariate effects. J Neurophysiol 93: 1074-1089.
[3] Park IM, Pillow JW (2011) Bayesian spike-triggered covariance analysis. Adv Neural Inf Process Syst (NIPS) 24: 1692-1700.
[4] Butts DA, Weng C, Jin JZ, Alonso JM, Paninski L (2011) Temporal precision in the visual pathway through the interplay of excitation and stimulus-driven suppression. J Neurosci 31: 11313-27.
[5] McFarland JM, Cui Y, Butts DA (2013) Inferring nonlinear neuronal computation based on physiologically plausible inputs. PLoS Computational Biology 9: e1003142.

More information about the Comp-neuro mailing list