Abstract
In the debate about the relation between attention and awareness a curious observation is that the two phenomena have opposite effects on afterimage formation: unperceived inducing images leave a weaker afterimage whereas unattended inducing images leave a stronger afterimage. This qualitative difference stands out among other findings that generally show attention and awareness to be similar, albeit not identical. Our starting point is the observation that, beside an afterimage, inducing images also cause contrast threshold elevation. Indeed, this threshold elevation impairs visibility of the afterimages themselves. This renders inconclusive existing reports of opposite effects of attention and awareness on subsequent afterimage perception: a fainter afterimage could either indicate lessened afterimage formation or augmented threshold elevation. We present a new psychophysical method to tease apart these two factors and thereby clarify the effects of attention and awareness on afterimage formation. Using this method, which centers on nulling the afterimage using a physical image, we demonstrate that attention and awareness have similar effects on afterimage formation, and also on threshold elevation. Both are augmented when the inducing image is attended, as well as when the inducing image is perceived. The impression of opposite effects of attention and awareness only arises when afterimage strength and threshold elevation are not distinguished, such as in measures of afterimage visibility. In addition, we show that inter-observer differences in the effects of attention and awareness on threshold elevation are correlated. Inter-observer differences in the effects on afterimage formation, however, are not. Our results indicate that attention and awareness are qualitatively similar in their effects on afterimage formation and threshold elevation, and that this similarity is particularly pronounced at the level of threshold elevation.
Rubicon Grants from the Netherlands Organisation for Scientific Research (JB and JvB), NIH EY 13358 (RB).