Chapter
Seven
Attention
functional roles of attention; others try to connect attention with other faculties
like memory or action. We will start with a brief round-up of the main theories,
and then return to the central question of this chapter: how are consciousness
and attention related?
THEORIES OF ATTENTION
In the nineteenth century, Helmholtz, Hering, and Wundt were among the physi-
ologists and psychologists who experimented with attention. In the 1950s, many
ingenious experiments were performed with a method called dichotic listening,
in which two different streams of sound are played to each ear. Normally only
one stream can be tracked at once, but certain kinds of stimuli can break through
from the non-attended ear, and others can have effects on behaviour without
being consciously heard. If the message being listened to moves from one ear
to the other, people usually follow the meaning and don’t even notice they have
swapped ear. This raised the question of whether selection operated early on or
after much processing had already taken place, leading to the early versus late
selection debate which has never really been resolved – although more recently,
as we will see in a moment, it has been sidestepped by the concept of perceptual
load.
For a long time most theories treated attention as a bottleneck, with preconscious
sensory filters needed to decide what should be let through to the deeper stages
of processing (Broadbent, 1958). This makes sense because clearly the brain has a
limited capacity for detailed processing, and is a massively parallel system which
produces serial outputs, such as speech and sequential actions. So somehow
many parallel processes have to be brought together, or selected, to ensure that
a sensible serial output occurs.
The main problem with such theories was that to cope with the evidence, the
proposed filters became more and more complicated, until the pre-attentive
processing began to look as complex as the deeper processing to which it was
supposed to give access. These models then gave way to those based on subtler
ways of allocating processing resources. The spotlight of attention was then seen
as less like a narrow beam or single bottleneck and more like the outcome of
many mechanisms by which the nervous system organises its resources, giving
more to some items (or features or senses) than others. But for some people, the
whole topic was becoming so unwieldy that perhaps the very concept of atten-
tion was at fault (Allport, 1993; Pashler, 1998). At some point in its history, the
science of attention arguably began to study – or create – something that bears
little relation to the intuitive idea of attention as a sharpening of focus (Watzl,
2011). Scientists have redefined attention as a perceptual filter, a feature-binding
mechanism, a broadcaster to working memory, or a competitive bias process. Is
the gap between the intuitions and the science a problem?
Perceptual load theory was proposed by psychologist Nilli Lavie as an attempt
to return to the intuitive idea of a bottleneck of attention, and to rethink it in a
simpler way. In this theory, perceptual processing has limited capacity, and when
a task involves dealing with a large amount of information (high perceptual load),
that capacity is fully exhausted by the processing of the attended-to informa-
tion: this results in early, top-down selection effects from our current goals and