Roy Baumeister and E.J. Masicampo (2010) have been persuaded by evidence of this kind, and have proposed that the function of consciousness is not the immediate initiation of actions. Instead, it is to create simulations of ourselves that inter-relate memory, current understandings of the social world, and evaluation of future actions. Conscious thought enables us to explain ourselves verbally to ourselves and others, and also to compose our minds. Here's what Baumeister and Masicampo say:
The influence of conscious thought on behavior can be vitally helpful but is mostly indirect. Conscious simulation processes are useful for understanding the perspectives of social interaction partners, for exploring options in complex decisions, for replaying past events (both literally and counterfactually) so as to learn, and for facilitating participation in culture in other ways (p. 945).
For a long time fiction has been seen as augmenting the functions of consciousness. Baumeister and Masicampo give us a good way of thinking about how this might occur. Fiction consists of crafted and externalized simulations: twins of our internal simulations. Plays, novels, short stories, and films, can be internalized by us readers and audience members to add to the construction of our own inner simulations.
Baumeister and Masicampo gives an example of a use of consciousness in planning:
... when one has a plane to catch tomorrow, one typically engages in a simulation that calculates backward from the plane’s takeoff time, allowing for airport procedures, the trip to the airport, and perhaps the hotel checkout before that, so one knows at what time to commence the sequence of acts. All the information used for this simulation is already in the mind, so conducting the simulation does not bring in new information from the environment ... These simulations work remarkably well in enabling people to be on time for their flights without having to spend many extra hours at the airport (p. 955).
When tracking the actions of a character in fiction, a reader or audience member uses this planning ability within his or her mental simulation: an author gives the reader information about the setting and the protagonist’s goals, then starts readers off on a plan. Then the author says what is accomplished and what events occur, including those in the minds of the protagonist and other characters. As we put aside our own goals and plans, and take on those of a protagonist, we enter that character’s patterns of conscious thought and come to know outcomes of actions. But the emotions that result are our own.
Baumeister and Masicampo's theory of the simulations of consciousness is very similar to the theory of the simulations of fiction with which we (at OnFiction) work. How encouraging, then, that these two kinds of simulation are so closely interchangeable.
Roy Baumeister & E.J. Masicampo (2010). Conscious thought is for facilitating social and cultural interactions: How mental simulations serve the animal-culture interface. Psychological Review, 117, 945-971.
Appendix: Galdi et al. (2008) studied people's conscious thinking and sub-consciously activated associations about a controversial proposal to enlarge a US military base in their city. Conscious beliefs were assessed by a 10-item questionnaire on environmental, social, and economic consequences of the proposal. Associations were measured by an implicit association test, in which people pressed buttons in response to pictures of the military base and evaluative words. People's decisions were measured by them saying whether they were (a) in favour of the base's enlargement, (b) undecided, or (c) against the enlargement. For the decided people, their conscious thoughts were highly predictive (p < 0.001) of both their decision and their automatic associations one week later (p < 0.01); but these people's automatic associations at the time of first assessment were not predictive of their decision a week afterwards. By contrast, among undecided participants, their conscious thoughts at first did not significantly predict their later decision, but their unconscious associations did significantly predict their decision (p < 0.05) and their conscious beliefs (p. < 0.01) a week later. This study may seem rather complicated. My view is that it indicates that our conscious thoughts are affected by processes of which we are not consciously aware—including the processes that produce our emotions—and that, rather than conscious thoughts being always able immediately to control what we decide they can, instead, work over a longer term to affect the structure of our minds, for instance as we think about an issue, become aware of our emotions about it, and discuss it with others.
Silvia Galdi, Luciano Arcuri & Bertram Gawronski (2008). Automatic mental associations predict future choices of undecided decision-makers. Science, 321, 1100-1102.
Image: From Raymond Mar: pooled fMRI results to show overlaps in the human brain between areas involved in thinking about people and areas involved in comprehending narrative.