Chapter
Seven
Attention
(without eye movements) and one for overt shifts of attention (with eye move-
ments) (Thompson, Biscoe, and Sato, 2005). That is, counter to what the premotor
theory predicts, the FEF neurons driving saccades are separate from those driving
attentional selection. Instead, spatially selective activity in the FEF may serve as
a visual salience map, identifying potential targets for eye movements without
being an explicit saccade plan. The same experiments have also found that when
attention shifts covertly to a target in a pop-out visual search task, activity in FEF
movement neurons is actively suppressed, with no spatial selectivity. This has led
researchers to conclude that activity in the visual, not the motor, FEF neurons is
what ‘corresponds to the mental spotlight of attention’ (2005, p. 9479).
Findings like these suggest that not all areas involved in motor preparation are
involved in covert attention, and not all regions involved in covert attention have
motor functions. Regions may be involved in both, but in the weaker sense of
creating a ‘priority map which signals the location of behaviourally relevant stim-
uli’ (Smith and Schenk, 2012, p. 1106). Hence perhaps why dissociations between
eye-movement preparation and attention allocation have been found for both
overt and covert attention (Hunt and Kingstone, 2003). So the proposal that
attention and motor control use the same neural circuits, as well as the stronger
claim that motor activation is both necessary and sufficient for spatial attention,
may be going too far.
The ‘biased competition’ (or ‘integrated competition’) theory, which originated in
the 1990s, is one possible alternative. The basic idea is that attention is a neural
competition mechanism biased by feedback from a person’s goals, expecta-
tions, emotional states, and so on (Ruff, 2011). What does this theory say about
action control? Here, action preparation increases the probability of the goal of
the action being selected for attention and processing, but does not guarantee
it, any more than the absence of motor preparation prevents a location being
attended to. In this theory, ‘attention is the consequence of competition within
and across different sensory-motor systems’ (Smith and Schenk, 2012, p. 1112).
The inputs compete for neural representation, which is allocated on the basis of
physical salience, current goals, and working-memory contents, and the winner
of the competition is attended to, ‘in the sense that it becomes available to higher
cognitive processes such as awareness and response systems’ (p. 1112). The idea
of winning a competition to be broadcast is reminiscent of global workspace’s
‘fame in the brain’ (Chapter 5).
One more theory in the broad cognitive-neuroscience category, and adding an
evolutionary spin, is Michael Graziano’s ‘attention schema theory’ of conscious-
ness, which builds on the biased-competition model and on IIT and global work-
space theories, and connects consciousness directly with attention. Indeed, for
Graziano, ‘awareness is the internal model of attention’ (Webb and Graziano, 2015,
p. 1), and it evolved as a way of modelling and controlling attention: top-down
control is improved when the brain can use a simplified model of attention itself.
This is the attention schema, and the theory explains ‘how the human machine
claims to have consciousness and assigns a high degree of certainty to that con-
clusion’ (Graziano, 2016, p. 98). This theory is broadly illusionist, but Graziano
prefers to say that consciousness is ‘a useful caricature of something real and
mechanistic’ (p. 112): ‘Subjective awareness – consciousness – is the caricature of
attention depicted by that internal model’ (p. 98).
‘attention is the
consequence of
competition within and
across different sensory-
motor systems’(Smith and Schenk, 2012,
p. 1112)‘awareness is the
internal model of
attention’(Webb and Graziano, 2015, p. 1)‘consciousness is not
an illusion but a useful
caricature of something
real and mechanistic’(Graziano, 2016, p. 112)