Investigative Ophthalmology & Visual Science, 39:631. 1998. [ARVO 1998; Ft. Lauderdale, FL.]

Purpose. To determine whether an observer can have an accurate feeling about the state of a visual stimulus (sensing) without an accompanying visual experience (seeing).

Methods. An original and a modified image (each of duration 240 ms) were continually alternated, with a uniform gray field (duration 80 ms) between each display. Images were all of real-world scenes. Changes were large and easily seen once noticed. Observers viewed the display and pressed a key twice, first when they sensed a change was occurring, and then when they saw it.

Results. Consistent with previous results (Rensink et al., 1997), all observers found it difficult to see change under these conditions. For approximately two-thirds of observers, there was little difference in reaction times between the two key presses. For the other third, however, reaction time differences exceeded 1 s in a significant proportion of trials, and often exceeded 5 s. In these trials, observers had a strong sense that something was changing, even though they could not see the change itself. Performance on catch trials indicated that the accuracy of their sensing was good.

Conclusions. At least one-third of normal observers are able to sense change in a visual stimulus without seeing it. Although similar to blindsight in that there is no accompanying visual experience, this phenomenon is accompanied by a strong mental experience of relatively abstract nature. Given that attention is required to see change, this "mindsight" may correspond to an alerting signal from other, nonattentional streams that have already detected the change.