Although Treisman and 
Schmidt’s initial finding on 
illusory conjunctions has been 
confirmed and is now widely 
accepted, the possibility 
remains that the effect is more 
a matter of subjects guessing in 
the face of uncertainty about 
what they had seen than it is a 
genuine perception. Treisman 
made several attempts to rule 
out this possibility. 
Nonetheless not all reports of 
what one believes one has 
perceived in the brief exposure 
period of several shapes of 
differing colors may reflect a 
genuine perceptual experience. 
So the interpretation of this 
discovery should be placed on 
"hold" for the time being.

    Arien Mack of the New School 
for Social Research and I have 
taken a different tack in 
exploring the role of attention 
in perception. Not satisfied 
with any existing method of 
eliminating attention to a 
display, we sought a more 
stringent way of doing so in 
order to investigate the 
question of what, if anything, 
is perceived when attention 
can be prevented from being 
deployed to some specific 
stimulus item. In most studies, 
the only way of getting at this 
question has been to give a 
subject two tasks 
simultaneously, usually 
emphasizing to the subject that 
one task is important and the 
other of only secondary 
interest. This is a "divided 
attention" paradigm. A task 
Treisman and Schmidt used, in 
which subjects are to identify 
two digits, one on each side of 
colored letters, can be thought 
of as the primary task. 
Reporting what colored letters 
were seen between the 
numbers can be thought of as 
the secondary task.