Chapter
Nine
Agency and free will
In experiments on voluntary action and conscious aware-
ness, Patrick Haggard and colleagues at University College
London used Libet’s clock method for participants to time
the onset of four single events: a voluntary key press, a
muscle twitch produced by stimulating their motor cortex
with transcranial magnetic stimulation (tms), a click made
to sound like tms, and a tone. next, in the voluntary con-
dition, they pressed a key and a tone sounded 250 msec
later. In the tms condition, their finger twitched involun-
tarily and the tone followed, and in a control condition just
the click was used. In each case, they reported the time of
the first event and when they heard the tone.
In this second stage, large perceptual shifts were found as
compared with the single-event case. the voluntary key
press and the time of the tone were reported as being
closer together, whereas the involuntary twitches (caused
by tms) and the tone seemed further apart. there was
no effect for sham tms and the effect was greatest for
shorter time intervals. the effect is known as ‘intentional
binding’ and its strength can be affected by predictability,
feedback, and beliefs (moore and obhi, 2012).
What does this imply for consciousness? the experiment-
ers themselves claimed thatthe perceived time of intentional actions and of their
sensory consequences [.. .] were attracted together
in conscious awareness, so that subjects perceived
voluntary movements as occurring later and their
sensory consequences as occurring earlier than they
actually did.
(Haggard, Clark, and Kalogeras, 2002, p. 382)this interpretation is a form of Cartesian materialism,
implying that events are perceived and manipulated ‘in
conscious awareness’. A more sceptical interpretation is
that the important processes of timing and discriminating
between self-caused and external events happen without
anything being ‘in’ or ‘out’ of consciousness.
Haggard asks whether the conscious experience of owning
an action depends on predicting the coming action or infer-
ring agency afterwards. From these and further experi-
ments on timing, he concludes that ‘the phenomenologysimultaneity took to become conscious because
they could report this spot position at their leisure.
The whole method of timing W was also criticised, as
was the adequacy of using a skin stimulus as a con-
trol to test the accuracy of the timing, and the failure
to allow for delays involved in each, or in switching
attention between the spot and W (Breitmeyer, Roll-
man, Underwood, Niemi). There have also been pro-
posals that instead of reflecting preconscious motor
preparation, the readiness potential might instead
be the result of an averaging of random noise that
exceeds a certain threshold – that is, it might not
tell us anything about readiness for a specific action
(Schurger, Sitt, and Dehaene, 2012).
Some of these criticisms are undermined by subse-
quent replications. For example, British psychologist
Patrick Haggard and his colleagues not only repli-
cated the basic findings, but showed that awareness
of one’s own actions is associated with a premotor
event (lateralised RP) after the initial intention and
preparation, but before the motor command is sent
out (Haggard, Newman, and Magno, 1999). Compar-
ing trials with early and late awareness, they found
that the time of awareness covaried not with the RP
but with the lateralised RP, concluding that ‘the pro-
cesses underlying the LRP may cause our awareness
of movement initiation’ (Haggard and Eimer, 1999,
p. 128). Haggard and Libet (2001) then debated the
implications of these results.
A 2008 study by Chun Siong Soon and colleagues
in Leipzig updated Libet’s experiment using fMRI,
tweaking the design to try to circumvent some of
the criticisms, particularly with regard to the timing
of W. Instead of using a clock face for timing, partici-
pants were presented with consonants in the middle
of a screen, one at a time for 500 ms each, and asked
to passively observe the stream of letters. This made
the sequence unpredictable, unlike the hand mov-
ing round the clock, to avoid them anticipating their
decision or choosing in advance a time at which to
move. They were told to relax (not to be too eager to
press a button when the letters first appeared, nor
to maintain a constant state of readiness to move),
and to press either the left or the right button with
the index finger of the corresponding hand as soon
as they felt the urge to do so. They were asked to
remember the letter that was on the screen when
they decided which button to press, not when they