|
Department of Psychology Cognitive Science Program Ecological Psychology Resources International Society for Ecological Psychology Ecological Psychology (Journal) James Gibson's Purple Perils Perception/Action Labs Action Research Lab - Reading, UK Center for Complex Systems and Brain Science - Florida Center for Ecological Psychology - Portsmouth, UK Center for the Ecological Study of Perception and Action - University of Connecticut Gustav Fechner Perception Lab - University of Western Kentucky Haskins Laboratories - New Haven, CT Human Factors Research Laboratory - University of Minnesota Infant Motor Development Laboratory - Indiana University, Bloomington Perception-Action Lab - Aberdeen, UK Perceptual-Motor Dynamics Laboratory - University of Cincinnati Postural Stability Laboratory - University of Cincinnati VENLab - Brown University Robotics MIT Leg Laboratory MIT COG Lab BabyBot BioRobotics Lab - Case Western Reserve University Passive Dynamic Walking - Cornell Wireframe Walker Biomch-L Newsgroup Optic Flow (Java Demo) So What Is a Polar Planimeter, Anyway? Illusory Sculpture Other Resources Visual Psychophysics Toolbox Some of our experiments are done on Macintosh computers using a Visual Psychophysics Toolbox for Matlab by David Brainard and Denis Pelli. The Toolbox provides excellent control of visual displays for both the Mac and Windows. ExpLib Our phase perception experiments run on a version of this free C++/Direct X based software, created by Andrew Cohen and Michael Sautner. Circular Statistics Our recent research in phase perception produces relative phase data. This data is circular, i.e. it is drawn from a distribution in which the two ends are in fact the same value - 360° is the same as 0°. Notions such as a mean become complicated to compute - a sensible mean of the directions 1° and 359° is 0°, while taking an arithmetic mean yields 180°. There are some resources available to deal with this problem. Circular statistics is a set of trigonometric solutions to the problem of computing mean and variability measures. Batschelet, Edward (1981). Circular statistics in biology. Academic Press: New York. Jammalamadaka, S. Rao (2001). Topics in circular statistics. World Scientific: River Edge, NJ. Andrew's Matlab code for computing confidence intervals for the circular mean and concentration (inverse variance) parameters. Implements the resampling, bootstrapping technique required by this type of data. Bjorn Holmquist's MATLAB Orientation Toolbox - Implements algorithms from circular statistics in Matlab . Daniel Rizzuto's PHASEPack - Also implements circular statistical algorithms in Matlab. Resampling Statistics We are beginning to explore using resampling and bootstrapping techniques in statistics. Resampling is a non-parametric method that compares obtained data to custom built 'population' distributions obtained via resampling. In principle, this approach avoids almost all of the limitations to the GLM approach. Lunneborg, Clifford E. (2000). Data analysis by resampling: Concepts and Applications. Duxbury Press:Pacific Grove, CA. Resampling Statistics in Matlab Resampling Statistics in Excel Numerical Recipes Sculptures |