Other evidence is less direct. 
If fatigue is cumulative rather 
than quickly dissipated, a 
subject shown an ambiguous 
figure A—–B that can be 
perceived as either A or B 
should, after a number of 
previous reversals, switch 
between A and B with 
increasing rapidity. Such 
acceleration of the rate of 
reversal has been reported by 
many investigators.

    There are several reasons to 
be skeptical of the satiation 
theory, however. Among them 
is that reversal often fails to 
occur if the subject does not 
realize that the figure is 
ambiguous. Ordinarily, in 
experiments on reversal, the 
subject is first shown the figure 
and told (or shown) the two 
different ways in which it can 
be seen. Then the ambiguous 
figure is presented 
continuously, and the subject 
is instructed to respond every 
time it is perceived to reverse. 
Since the investigators want to 
study perceptual reversal, it is 
natural enough that they first 
make sure the subject knows 
what the alternatives are. But 
giving subjects such 
information not only 
establishes the alternatives in 
the memory of the subject. It 
may also create what 
psychologists call a "demand 
character" in the experiment—–
the subject assumes that 
reversal is what is expected.