rightmost stimulus calling for response of the rightmost finger) or crossed (e.g.
leftmost stimulus calling for response of the rightmost finger). Performance was
rather poor when the relationship between stimulus and response was corre-
sponding for one stimulus but crossed for the other. In these circumstances, the
subjects were sometimes confused, as indicated by the fact that the errors were
largely those expected if the inappropriate stimulus–response relationship had
been selected. Thus, the uncertainty caused by mixing two different stimulus–
response relationships added a complexity to performance that did not exist
when only one of the tasks was performed.
Theoretical Accounts of Dual-Task Performance
Several theories of dual-task performance have been proposed over the years,
and some of the main theoretical approaches are discussed here. As we will
see, there have been theoretical disagreements about the relative importance of
general and specific processes in this area. However, we will first of all consider
the work of Welford (1952), who provided one of the first systematic attempts
to account for dual-task performance.
Bottleneck Theories Welford (1952) argued that there is a bottleneck in the pro-
cessing system which makes it difficult (or impossible) for two decisions about
the appropriate responses for two different stimuli to be made at the same time.
Much of the supporting evidence for this theory came from studies of thepsy-
chological refractory period. In the standard task, there are two stimuli (e.g. two
lights) and two responses (e.g. button presses), and the subject’s task is to re-
spond to each stimulus as rapidly as possible. When the second stimulus is pre-
sented very shortly after the first stimulus, there is generally a marked slowing
of the response to the second stimulus: this is known as the psychological re-
fractory period effect (see Welford, 1952).
Although the existence of this psychological refractory period effect is con-
sistent with the notion of a bottleneck in processing, it could be argued that it
occurs because people are not used to having to respond to two immediately
successive stimuli. However, Pashler (1993) discussed one of his experiments in
which the effect was still observable after more than 10,000 trials of practice.
Another objection to the notion that the delay in responding to the second
stimulus reflects a bottleneck in processing is that the effect may instead be due
to similarity of stimuli and/or similarity of responses.
Pashler (1990) carried out a study to decide between the bottleneck and sim-
ilarity-based accounts of the psychological refractory period effect. According
to the bottleneck theory, the effect should be present even when the two stimuli
and the two responses differ considerably. In contrast, the effect should disap-
pear if similarity is crucial to its existence. In one of Pashler’s (1990) experi-
ments, the stimuli were a tone requiring a vocal response and a visual letter
requiring a button-push response. Some of the subjects were told the order in
which the stimuli would be presented, whereas others were not. The findings
are shown in figure 15.7. In spite of a lack of either stimulus or response simi-
larity, there was a psychological refractory period effect, and it was somewhat
greater when the order of the stimuli was known than when it was not. Thus,
the findings provided strong support for the bottleneck position.
Attention and Performance Limitations 381