magnitude of the effect of irrelevant information is a function of the temporal overlap of the activation produced by it with that produced by the relevant stimulus information ( Hommel, 1993b).
Most people think of compatibility effects as reflecting relatively fixed perceptual-motor relations. However, compatibility effects vary as a function of the goal of an action. Guiard ( 1983) had subjects move a cursor to the left or right, in response to a high or low pitch tone, by turning a steering wheel. This produces a standard Simon effect, responses are faster when the tone occurs to the side to which the cursor must be moved. Of most interest is a task in which subjects were required to hold the wheel at the bottom. In this case, the hands move left to shift the cursor to the right. Yet, a standard spatial correspondence effect was obtained such that responses were faster when the pitch assigned to the rightward cursor movement occurred in the right ear rather than in the left ear, and vice versa.
Hommel ( 1993a) conducted a standard Simon task with auditory stimuli (high and low pitch tones) in which the keypress lit a light on the opposite side. When subjects were told to ignore the light, responses were faster when the stimulus and response locations corresponded. However, when subjects were told to trigger the assigned light as quickly as possible in response to the stimulus, responses were faster when the stimulus location and response light location corresponded than when they did not. The important point from Guiard's ( 1983) and Hommel's studies for system designers is that very different spatial compatibility effects can be obtained as a function of the task goals of the performers.
One of the adages in Human Factors is that training cannot overcome the consequences of a bad design. That is clearly illustrated in studies of S-R compatibility. Even in two-choice reaction tasks, subjects who practice for up to 2,400 trials with an incompatible mapping never catch up with subjects who practice the same amount with a compatible mapping ( Dutta & Proctor, 1992). This is true both for the standard left- right task and for the task in which the stimuli are up and down locations and the responses are left and right locations. It also holds for tasks in which stimulus location is irrelevant. Contrary to what one might think, even the simplest compatibility effects are not transient phenomena.
In this paper we have only touched on several theoretical issues and findings of contemporary research on S-R compatibility. We have provided some illustrations of implications for Human Factors, and there are likely many more that could be derived. The compatibility literature contains a wealth of information about which anyone who is interested in the design of displays and controls should be aware.
Dutta A., & Proctor R. W. ( 1992). Persistence of stimulus-response compatibility effects with extended practice. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18, 801-809.
Fitts P. M., & Deininger R. L. ( 1954). S-R compatibility: Correspondence among paired elements within stimulus and response codes. Journal of Experimental Psycholoy, 48, 483-491.
Hommel B. ( 1993a). Inverting the Simon effect by intention: Determinants of direction and extent of effectsof irrelevant spatial information