Main content



Loading wiki pages...

Wiki Version:
**motion visibility project data** Motion visibility is used as a basic tool in cognitive neuroscience. This repository contains measurements of human cortical response to motion visibility and functions to generate predictions for future projects. **experiment design** In this experiment eleven observers (8 female, 3 male; mean age, 26; age range 19-36) were shown a constant baseline of incoherent motion at 25% contrast. Every few seconds the contrast and/or coherence were incremented, resulting in a change in the blood oxygentation level dependent (BOLD) signal in retinotopically defined visual cortex. We measured these changes and used them to build a quantitative framework which you can use to predict how cortex will respond in future experiments. **data** In the file data.mat you will find the event-related response which was measured for each of the forty contrast, coherence, and duration conditions in this experiment for each of the eleven observers. **prediction** In the function cohcon_predict.m you will find a function which can be used to predict the response in retinotopically defined cortical areas to different changes in contrast or motion coherence, based on the data we measured. This function relies on the quantitative framework which we fit to the measurements. The framework assumes that the response due to contrast is shaped as a sigmoidal function and the response due to coherence is linear (or a saturating exponential in MT). We assumed that there is some nonlinearity due to duration, such that increasing duration evokes responses that are less than that expected by a linear model. Finally, we assume that you are starting from a baseline contrast similar to ours. If your stimulus starts from a different baseline (for example, a gray screen would have a baseline contrast of 0) we have built the function to try to interpolate what would occur. cite Please cite this work when you use it. (This work is currently in preparation, a citation will be added upon submission to a preprint archive).