Abstract
The involuntary attention effect is the finding of faster RT to identify targets presented in a cued than uncued location even when the cue location is random with respect to the target location. We proposed that there are at least two separate mechanisms responsible for involuntary attention: (1) a serial search mechanism that accounts for performance when the target is difficult to locate, and (2) a decision mechanism that accounts for performance when the target is easy to locate. The serial model predicts that as the number of possible target locations increases, the cueing effect increases. The decision model makes the opposite prediction.
We previously demonstrated that when the target is easy to locate, in the absence of distractors in the display, the decision model accounts for the involuntary attention effect. However, when the target is made difficult to locate by presenting distractors with the target, the serial model accounts for the involuntary attention effect (Prinzmetal & Ha, 2008).
Here, we investigated two findings associated with involuntary attention: (1) IOR (inhibition of return) is the finding that when the cue-target interval is relatively long, observers are slower to identify targets in the cued than in the uncued location. (2) Contingent capture is the finding that the involuntary attention effect is larger when the cue and target are similar than when they are not similar.
We obtained IOR only when there were no distractors in the display. IOR was larger with 2 than 6 possible target locations. Thus IOR is better described by the decision mechanisms. We obtained contingent capture only when there were distractors, and the effect was larger with 6 than 2 locations. Contingent capture is due to a serial mechanism. Hence there are at least two separate mechanisms responsible for involuntary attention.