Driving beyond consciousness:
A test of Multiple Drafts theory

Toward a Science of Consciousness – Tucson 2006

Mary Osborn and Susan Blackmore

“there are no fixed facts about the stream of consciousness independent of particular probes.”
(Dennett, 1991, p 138)

In the well-known ‘unconscious driving phenomenon’ experienced drivers can arrive at their destination to realise that they have no recollection of having driven there. They may say that they were conscious of listening to music or having a conversation, but not of driving. In a questionnaire study we collected accounts of the phenomenon, and found that it happens mostly to experienced drivers on familiar routes.

Unconscious driving is often described as a case of automatised or skilled action without consciousness, but is this the right interpretation? We contrast two  ways of conceptualising this experience, based on Dennett’s distinction between Cartesian materialism and his own multiple drafts theory. The common, Cartesian materialist, view implies that either the experience of driving or the experience of listening must have been “in consciousness”, while the other was unconscious. Multiple drafts theory, by contrast, claims that there is no fact of the matter about which was in consciousness because there is no theatre of the mind in which experiences are displayed for an inner observer; instead, a decision is made, after the fact, about which was conscious. This decision depends on how the system is probed at the end of the journey.

In an experiment to test this theory, participants played an X-box driving game  while listening to a CD and were probed in two different ways to report what they were conscious of. Multiple drafts theory predicts that the answers given should depend on the nature of the probe used.

17 participants (11 female, 6 male, average age 28) played “Gotham Racing” using an X-box console, driving around Princes Street Edinburgh in an SRR Chevrolet, while listening to a CD of “The Hitch Hiker’s Guide to the Galaxy” through headphones. After sufficient practise, they spent 20 minutes driving during which they were probed 16 times either by a light appearing on the driving screen or by a buzzer in the headphones (8 of each, in random order). When they either saw the light or heard the buzzer they were asked to say what they were aware of (what they were concentrating on or conscious of thinking about) immediately prior to the prompt. Their responses were recorded and, after the experiment was completed, independent judges read all the responses and categorised each as having been either concerned with driving or not.

If the type of probe affected the response given we would expect more answers concerned with driving (D) after the light probe (L), and more not concerned with driving (nD), after the buzzer (B). This prediction was not confirmed (results for 255 responses: LD 74, LnD 54, BD 70, BnD 57; chi squared, 1df, = 0.188).

The results did not provide evidence for Dennett’s MD theory. However, we think this technique could be developed to provide better tests of whether there really is a fact of the matter about what is in a person’s consciousness at any time.