a well known illusion is the Kanizsa triangle, a
brighter than white triangle is perceived in the
middle, even though the edges of the shape are
not physically there
illusions like this can be explained by Gestalt - when we see a figure which is
incomplete, our perception makes a whole shape using CLOSURE, we complete the
edges to make a regular or familiar shape, this is the figure of the figure-ground
relationship
with the Kanizsa triangle, the arrow heads, 'bitten'
cirrcles and the space around them become the ground
when the circles are replace with dots, we see 2 interlocking triangles, similarity organises the dots and
lines and continuity joins the lines to form a triangle
Which do we recognise first?
Peterson (1994) conducted an experiment to see which we do first, divide the stimulus into figure-ground, or
recognise the object
participants briefly saw images and judges whether a black or white area was
the figure, they generally said it was the black area, however, when the
images were turned upside down, it took them longer which shows they were
recognising the object before they organised it into figure-ground
Explaining Distortions
Gestalt theory can explain the Muller-Lyer and still works when the
fins are replaced with circles, this suggests that in perceiving the
figure as a whole, we tend to add the fins or circles to the central
lines, in the 'shorter' lines, it pulls the whole image together making it
look smaller, in the 'longer' lines, the ends extend beyond the central
lines, stretching the figure out and making it look longer
Explaining Ambiguous Figures
we separate the figure from the ground and normally
an image cannot be both, when something can be
figure or ground, it is an ambiguous figure
it is an illusion because we cannot tell
whether the black or the white area is the
figure, both are meaningful objects
Evaluating
Strengths: it provides a good explanation for ambiguous figures when they are
black and white, explains fictions well (see weaknesses)
Weaknesses: cannot explain any distortions other then the Muller-Lyer, it suggests we should see
a 6 pointed star in the Kanisza triangle, but we don't - we see 2 triangles, it seems to use different
explanations at different times