The goal of our project is to develop general methods for estimating the stimulus-response transfer function (the spatio-temporal or spectro-temporal receptive field, STRF) of sensory neurons. Our estimation techniques are based on the well-known reverse correlation methods, but include two significant extensions. First, our estimation algorithms can be used in conjunction with arbitrary, complex stimuli, including natural stimuli. Second, we are developing a battery of complementary non-linear estimation techniques. These data analysis tools will be implemented in a user-friendly software package with extensive help and examples. We will also make available the complete source code in the form of a Matlab toolbox. These software tools should enable a wide range of sensory physiologists to perform STRF estimation, regardless of their level of expertise in computational neuroscientists.
<p>
We have derived and successfully implemented routines for estimating linear stimulus-response transfer functions for neural responses to arbitrary stimulus ensembles. The underlying theories and derivation are described in: "Estimating spatio-temporal receptive fields of auditory and visual neurons from their responses to natural stimuli" (2001), Theunissen et al., Network: Comp. Neural Syst., 12, 289-316. In addition, we are currently researching several non-linear estimation methods and are developing validation methods that can be used to quantify the goodness of fit of estimated STRFs.
|