| |
Abstract:
The purpose of the experiments was to investigate more
closely the effects of different kinds of training on a listener's
sensitivity to auditory stimuli. Auditory stimuli consisted of
bandpass filtered white noise, differing along the dimension of the
center frequency of the pass band. Subjects' sensitivities to these
auditory stimuli were measured before and after training using a
"same/different" discrimination task to provide d' measures. In the
first experiment, subjects were trained to recognize stimuli that
fell within a small region of frequency space as members of a
category. Comparison of the pre- and post-training discrimination
tests showed that subjects became worse at discriminating between
stimuli that fell in this category range as a result of the
training process. No decrease in discriminability was found for
stimuli in a control region of frequency space that was not present
during training. This form of deformation has been referred to as
"acquired similarity" or "acquired equivalence", although to our
knowledge this is the first clear demonstration of the effect as
the result of learning in the laboratory. A second experiment
showed that the opposite effect, "acquired distinctiveness", is
achieved if subjects are trained on a discrimination task rather
than a categorization task. Additional experiments were run to
refine our understanding of the training conditions required to
produce acquired similarity.
|
