 |
 |
 |
|
The final and most intriguing results concerned the effect that prior
knowledge or memory of an object has upon the categorization or perception
of its colour. The hues were now presented within the outlines of
fruit and vegtables. The results suggested that the expectation of
the fruit's
colour heavily biased the categorisation in the direction the colour
expected. For example, many more test hues were categorised as yellow
when presented
within the outline of a banana than if presented in a circular patch.
This
finding was further investigated in a colour adjustment paradigm.
The subject was presented with a target patch and asked to adjust
the colour to mean grey. The target patch was either in the shape of a
fruit
or one of several
control conditions. The subject was also asked to adjust the colour
to match the typical colour of the presented fruit or vegetable outline.
Typical
findings are shown in figure
3. Remarkably the colours that are judged
to be grey by the subjects are in fact shifted well away from the
true achromatic point, in the opposite direction the fruit's typical
colour in colour contrast space. The memory, and therefore expectation
of
the
typical colour of the fruit has strongly biased the subjective categorisation
of grey.
Figure
4 shows how the results hold for all the fruit and vegetables
tested, with the possible exception grapes where the typical colour
is, of course, ambiguous. From these results Professor Gegenfurtner
concluded that photographs of fruits and vegetables are perceived
in the characteristic
color of the object if the average color of the pixels in each photo
is
grey. For the objects to appear grey, subjects adjusted the average
color 10% - 30% in the colour direction opposite to the remembered
colour.
|
Constancy |
|
|
^ Back to top
|