Consciousness

interest took an average of seventeen alternations,
with some participants taking up to eight alternations
to notice a change that was obvious once seen. This
suggests that for unattended parts of the image, peo-
ple have to do a slow serial search to find the change.

But even highly salient features can be subject to
change blindness. For example, very gradual changes
in facial expression can go unnoticed while still affect-
ing people’s subsequent behaviour (Laloyaux et al.,
2008). The possibility of implicit change detection, or
‘mindsight’ (Simons and Ambinder, 2005), adds to the
idea of illusion by suggesting that what people do is
not always a consequence of what they consciously
see, or report seeing.

Change blindness could have serious consequences
in ordinary life, such as failing to detect changes while
driving. Experiments have used natural traffic scenes
with changes made during blinks, blanks, and sac-
cades. Relevant changes are detected more quickly
than irrelevant ones, but even so can take 180 ms lon-
ger to detect than when seen without any kind of dis-
ruption (Domhoefer, Unema, and Velichkovsky, 2002).
Not only do we blink and move our eyes while driv-
ing, but mud splashes on the windscreen can disrupt
change detection, too. O’Regan, Rensink, and Clark
(1999) showed that small shapes briefly spattered over
a picture could prevent even large changes elsewhere
being noticed. Comparable events happen all the time on the road and in the air,
suggesting that dangerous mistakes might be made by drivers or pilots if a crucial
event occurs just as some mud or a large insect splats onto the windscreen. Later
experiments found, unexpectedly, that driving expertise made no difference to
change blindness for driving-relevant changes, and also that relevant changes
near the periphery of a driving scene were detected faster than those near the
central vanishing point of the road ahead, where we would expect drivers to be
focused (Galpin, Underwood, and Crundall, 2009).

These findings raise questions about how closely static images viewed on a
screen can really replicate the conditions of real-life driving. One general crit-
icism of these experiments is that the changes  – like the height of a railing
moving up a foot or two behind a couple’s heads – are implausible, and there
is no reason why a well-adapted visual system should be able to detect them.
Some of the research on inattentional blindness does better at bringing the
laboratory into the world.

INATTENTIONAL BLINDNESS

Could it be that if you don’t pay attention to something, you simply do not see it?
Just how paying attention to something relates to being conscious of it is hard
to pin down (Chapter  7), but one way of thinking about the connection comes
from studies of the odd phenomenon of inattentional blindness, pioneered by

Noncritical trial

Critical trial

Fixation

1500 msec

Stimulus

200 msec

Mask

500 msec

Time

Fixation

1500 msec

Stimulus

200 msec

Mask

500 msec

Time

Critical

stimulus

FIGURE 3.9 • Displays for the critical and non-

critical trials in Mack and Rock’s

experiments. In this experiment

the critical stimulus is in the

parafovea. In other experiments

the cross was in the parafovea

and the critical stimulus was at the

fixation point (Mack and Rock,

1998).