cessor. However, such an argument is suspect because automated activities can
sometimes be disrupted if too much attention is paid to them. For example, it
can become more difficult to walk down a steep spiral staircase if attention is
paid to the leg movements involved. Moreover, Reason’s diarists produced an
average of only one action slip per day, which does not indicate that their usual
processing strategies were ineffective. Indeed, most people seem to alternate
between the automatic and attention-based modes of control very efficiently.
The optimal strategy involves very frequent shifts from one mode of control to
the other, and it is noteworthy that these shifts are performed with great suc-
cess for the most part.
Action slips are the consequences of a failure to shift from automatic to at-
tention-based control at the right time. Although they are theoretically impor-
tant, action slips usually have a minimally disruptive effect on everyday life.
However, there may be some exceptions, such as absent-minded professors who
focus on their own profound inner thoughts rather than on the world around
them!
Section Summary
Action slips (i.e. the performance of actions that were not intended) have been
investigated by means of diary studies in which subjects keep daily records
of any slips they make. Various categories of action slip have been identified,
but they all typically involve highly practised activities. Highly practised skills
mostly do not require detailed attentional monitoring except at critical decision
points. Failures of attention at such decision points cause many action slips.
Failure to remember what was done a few seconds previously is responsible for
many other action slips.
Evaluation of Theories of Attention
Attention: Unitary or Multiple Systems?
Most research has been based on the notion that there is a single, limited-
capacity, attentional system. So far as focused attention tasks are concerned, the
limitations of this system allegedly produce bottlenecks in processing. So far
as divided attention tasks are concerned, attentional limitations often prevent
successful performance of two tasks together, and lead to the development of
automatic processes that are not reliant on attentional capacity.
One of the reasons for the long-lasting popularity of the view that attention
is unitary (i.e. there is a single system) is that it fits well with introspective
evidence. It seems as if we have a single attentional system which can (in
the visual modality) be directed like a variable-beam spotlight to some part of
the environment. However, this view is wrong. As was discussed earlier in
the chapter, Posner and Petersen (1990) have identified three separate atten-
tional processes: disengagement of attention from a stimulus; shifting of atten-
tion from one stimulus to another; and engagement of attention on a new
stimulus.
The fact that attention is not unitary has grave implications for most theory
and research on attention. The notion that any given process either requires
Attention and Performance Limitations 393