Abstract
Many investigators have assumed that there are at least two kinds of attention: voluntary and involuntary attention. Both forms of attention affect reaction time (RT), but it is not clear whether they do so with the same mechanism. Recent evidence suggests that voluntary attention affects the perceptual representation but that involuntary attention does not. For example, Prinzmetal et al. (2005) reported that while voluntary and involuntary attention both affected RT in spatial-cueing experiments, only voluntary attention affected accuracy when displays were data limited and there was no speed pressure. Voluntary attention enhances a stimulus specific ventral-stream area (FFA), while involuntary attention does not (Esterman, et al., 2005). Using high frequency gamma band activity, Laudau et al. (2006) found different patterns of activity for voluntary and involuntary attention. These and other results have lead to the conclusion that involuntary attention does not affect perception.
If involuntary attention does not affect the perceptual representation, what are the mechanisms by which it affects RT? In this talk, I present a model of attention that illustrates how involuntary attention affects RT without influencing perceptual processing. The theory, based on the “Leaky Accumulator” model (Usher, & McClelland, 2001), accounts for involuntary attention by postulating that involuntary attention primes responses to stimuli presented at the cued location, irrespective if their identity. Voluntary attention, on the other hand, affects the rate of accumulation of perceptual information. This theory was tested in several experiments in which subjects were put under speed pressure to respond before a deadline. Results were consistent with predictions of the model, which postulates different mechanisms for voluntary and involuntary attention. Furthermore, the model accounts for a variety of findings in the spatial-cueing paradigm for both voluntary and involuntary attention.