It is evident, then, that 
investigators in perception 
have not solved the problem of 
the geometrical illusions, 
although progress has been 
made. Probably no one theory 
can do justice to all of the 
known illusions, and it may 
well be the case that even a 
single illusion is based on 
several simultaneously 
occurring factors. Of the 
illusions considered here, it 
seems warranted to suggest that 
the Ponzo illusion (and the 
corridor illusion, which is 
similar to it) is jointly 
determined by depth processing 
and by contrast and 
assimilation, that the 
Poggendorff illusion is 
explicable in terms of 
unconscious depth processing, 
and that the Müller-Lyer 
illusion is primarily the result 
of an inability to isolate the 
test lines or, otherwise stated, 
of an incorrect comparison. But 
these conclusions are put forth 
only tentatively.

    To come back to a point made 
at the outset, it does seem to be 
true that we are learning 
something relevant to the 
problem of veridical perception 
in daily life by the exploration 
of these illusions and, 
conversely, that what we have 
learned from the study of 
nonillusory phenomena has 
helped us in trying to explain 
the illusions. But even if we 
learned nothing of value in 
investigating illusions, they 
would still constitute puzzles 
that curiosity would drive us to 
unravel.

    We have been concerned here 
with what might be called 
static illusions—–that is, 
illusions of extent or line 
direction in stationary 
patterns. But many of the most 
striking illusions are those 
based on motion. While the 
emphasis in the next chapter is 
not on illusion but on the 
search for general principles of 
motion perception, it will be 
seen that almost every 
phenomenon considered there 
is, in fact, an illusion of 
motion.