home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Illusion - Is Seeing Really Believing?
/
Illusion - Is Seeing Really Believing (1998)(Marshall Media)[Mac-PC].iso
/
pc
/
illusion
/
rock_fie.cxt
/
00446_Field_446.txt
< prev
next >
Wrap
Text File
|
1996-12-31
|
2KB
|
66 lines
Because these world-mode
perceptions seem to be
preferred, the organization at
this stage must be based upon
some underlying principle that
governs preference, such as
simplicity, stabilization
resulting from recognition, or
the selection of those percepts
that explain certain
regularities or simultaneously
occurring events in the
stimulus. By this I mean that
those perceptions are preferred
that account for co-occurences
or regularity in the stimulus
that would remain purely
coincidental were the
alternative perception to occur.
For example, in the kinetic
depth effect, the retinal image
of a line rotating in the third
dimension simultaneously
changes its length and
orientation. Were one to
perceive the line literally as
doing just that, the two co-
occurring changes would be
coincidental--that is, they
would have no particular
relation to one another. But
when one perceives the line
rotating in depth, these two
transformations are both the
result of a common cause, the
perspective representation of
such a rotating line. Or, to give
another example, perceiving
literally several regions of
differing lightness in either of
the transparency figures would
not explain the fact that the
lightness values in the smaller
regions change along the same
straight line as in the larger
regions. But this regularity in
the pattern is fully explained
by the transparency percept. On
the other hand, the literal
percept would seem to be
preferred in the illusory figure
based on crosses at the corners
discussed earlier, perhaps
because the familiar
symmetrical shape and parallel
contours of each corner
fragment would be unexplained
coincidences if an illusory
contour figure were perceived.
Therefore, I suggest that a
general principle of perception
is the preference for
"solutions" that account for
such stimulus regularities or
co-occurrences.