This experiment implements reverse correlation following wavelet transformation of stimulus and response data. It reproduces the model figures from a paper in press in Journal of Neuroscience Methods. The method is simple but effective for certain kinds of problems. In particular, nonstationary systems can be analyzed by extensions of this method. For neuronal data that include limited samples with noise, this method can be quite useful.

This update includes code for generating a set of figures from a manuscript that I've submitted. More features are included, but I haven't spent enough time reorganizing things and writing efficient code in this version.

This module performs correlations between stimulus and response data, producing estimates of the linear transformation ("kernel") that produces the responses from the stimuli. Two methods are included: 1) the conventional correlation, performed in the frequency domain; and 2) a novel method that uses wavelet transforms.

This version fixes some bugs, and adds a few drivers that produce figures for a manuscript.

This is the initial release to IgorExchange. It is subject to frequent changes.