As in the first experiment, subjects incidentally
studied colored objects, but in this experiment the
colors were randomly selected by the experimenter.
Thus, personal color preferences could not be a
reason for any implicit color effect. For reasons of
distraction, subjects had to name the objects during
study. We decided on a naming task in this experiment
instead of the color choice task used Experiment
2 in order to avoid ceiling-effects in explicit
color memory. In this experiment, it was important
to attain mid-level color memory performance. Of
course, color should be remembered above chance so
we could assume that color was actually encoded at
all, but still there should be enough cases where
color is not remembered explicitly and so the possibility
remained to select the “old” color solely implicitly
without having remembered it explicitly before.
If there is an implicit color effect going beyond
the cases where colors can be remembered explicitly,
then there should be a preference for the old color
in cases where no explicit recollection was possible.Hence, the interesting cases are those in which subjects
select a wrong color in the explicit part or state
that they cannot remember. For these cases, we
wanted to check whether there would nevertheless be
an implicit preference for the old color. It was not
apparent to subjects whether they had correctly remembered
the former color or not, since they were
not given any feedback about their responses.