Recreation of the visual search paradigm experiment by Treisman and Gelade (1980) [1] [2].
Instructions
The .m MATLAB file can be run in the MATLAB environment, or the .app can be unzipped and run directly.
Background
The measurement of reaction times can function as a proximal measurement of underlying cognitive processes, specifically their number and complexity. This is based on assumptions that both the number of mental operations and their complexity are positively correlated with the time it takes to execute them. The landmark study by Treisman and Gelade (1980) provided experimental evidence in support of there being a feature primitive (core categories by which we group environmental stimuli) and serial search (one-by-one searching of stimuli) cognitive operations in the human brain.
Test Methodology
The test begins with prompting the user to begin the experiment, so as not to bias the initial measurements. The first stage of the experiment are 4 escalating levels (4 elements, 8 elements, 12 elements and finally 16 elements) of the Pop-Out test, where the target (a black X) is the only black element on screen. The user is asked to identify whether or not the target element is present, and the reaction time measured. The second stage, the Conjunction test, follows the same approach but there are also non-target black elements on screen, thus requiring a higher cognitive workload from the user.
Statistics
At the end of the test, a correlation test is carried out and inferences are displayed. A graph of reaction times against element size for each type of test is drawn. Replicating the study should show that the linear fit for the conjunction data has a positive slope, and the linear fit for the pop-out data is more flat.
References
[1] A. M. Treisman and G. Gelade, "A feature-integration theory of attention," Cognitive Psychology, vol. 12, (1), pp. 97-136, 1980.
[2] P. Wallisch et al, MATLAB for Neuroscientists, 2nd Edition. (2nd ed.) 2014.