People are very good at statistics-when they do not actually think they are doing statistics
Our January speakers are Max Berniker and Ian Stevenson, who specialize in physical medicine and rehabilitation at Northwestern University. Their talk is entitled: People are very good at statistics-when they do not think they are actually doing statistics.
Abstract:
The world is complex and variable, and our perception of it is noisy. In a recent study we have formalized motor adaptation as the process of optimally inferring changes in the world and our bodies given our observed motor errors. In the first part of our talk, we will present results that demonstrate our approach. This model makes predictions that are consistent with a wide range of experimental data from numerous research groups. What's more, this approach offers a principled explanation for motor adaptation and generalization as the result of an inference strategy for the nervous system.
In the second half of our talk, we will briefly discuss how Bayesian statistics can be used to understand how the brain "works". Neurons communicate with each other using spikes of electrochemical activity (essentially point processes). Recently, we have been using Generalized Linear Models with regularization to understand what causes a neuron to spike and how neurons interact with one another. Inferential statistics is becoming an increasingly important tool in understanding these complex, high-dimensional systems.
Lunch is $30 for CCASA members, $35 for nonmembers. Nonmembers, join the chapter for a year for only $15 and get the discount plus all the other benefits of membership! As usual, the Lucille Derrick Fund will purchase a limited number of tickets for students who wish to attend. If you are a student and would like to take advantage of this offer, please register online below, and contact Lou Fogg, expressing your interest.
Please register for the luncheon by Friday January 16, 2009.