Variability in the Simon effect

The automatic-controlled distinction provides one way to explore movement disorders such as apraxia. For such a population it would be useful to have a simple task and measure to distinguish automatic from controlled movements. The Simon task, which compares performance on congruent versus noncongruent tasks, is thought to have both automatic and controlled components. As a measure, Segalowitz and colleagues (Gilbert & Shallice, 2002; Norman, 1968, 1981; Norman & Shallice, 1986; Shallice, 1994) suggested the coefficient of variation (CV); responses resulting from controlled processing are thought to have more variability than those resulting from automatic processes. Two experiments evaluated the use of the CV as an indicator of automatic versus controlled processing in right-handed WSU kinesiology students, using a one-handed Simon task. It was predicted that noncongruent Simon trials would require more controlled processing than congruent trials, and thus would be more variable as measured by CVs. Experiment 1 (N=42) revealed the expected Simon effect of faster reaction time for congruent trials as opposed to noncongruent trials occurred. However, congruent trials were more variable than noncongruent trials when evaluated with distributional analysis, and were more positively skewed, indicating that the congruent trials require more processes than noncongruent trials. Experiment 2 (N = 58) used a dual task procedure to validate conclusions based on CVs in Experiment 1. When a resource demanding counting task was added to the Simon task, congruent trials slowed more than noncongruent trials, although congruent trials remained faster than noncongruent trials. This result confirmed the CV results and indicated that congruent trials require more resources and thus more controlled processing than noncongruent trials