[First published on Nathen’s Miraculous Escape, December 24, 2008.]

I do not believe in free will. I am what you might call a free-will agnostic. Hundreds of years of thought and debate have brought us no closer to agreement about whether an individual’s consciousness is capable of true agency. I can understand, though, how the idea that one might not control even one’s own actions is distressing; a perception of control makes stressful situations easier to cope with, while a perceived lack of control, and especially perceived loss of control, make stressful situations more distressing (Sapolsky, 2004). Note that it is our perception, however, not necessarily the reality of control, which is important. As it turns out, our perceptions of control are not reliable. The only situation in which we can be sure about how much control we have is when we demonstrably have no control at all, and in those situations people often still believe they have control.

Psychologists call this phenomenon the illusion of control: The tendency to overestimate how much control one has over an outcome. Participants in an experiment by Tennen and Sharp (1983), for example, believed that they controlled a blinking light about one third of the time, when in fact there was no connection between their button and the light. Even more surprising, experimental participants came to believe that they were good at guessing the outcome of the notoriously random coin toss (Langer & Roth, 1979). All it took were a few successes early on. Ellen Langer, in an early series of experiments (1975), found that the key to producing the illusion of control is the presence of skill cues, like practice, familiarity, choice, competition, and personal involvement. That is, if a task reminds you of a time when you used skill to influence what happened, you are likely to believe you have more control over the outcome of the task than you do. This tendency creates an inaccurate perception of reality which can have serious negative consequences but also may have beneficial effects.

The desire to be accurate in our perceptions and judgments is a basic human motivation, among psychological motivations second only to our need for self-esteem (Aronson, Wilkert, & Akert, 2007). This desire serves us well, especially in cases when a discovery of error on our part tempts us to distort reality to save our self-esteem; it would be much more difficult to learn from our mistakes if we were never compelled to admit them. In the case of the illusion of control, however, the mistaken belief in control will often not even come to our awareness. Usually there is no experimenter at the end of our task, gently debriefing us on how our button is not connected to the light. There are cases when the mistaken belief in control is benign enough—the experimental participants thinking they controlled a light were not hurting themselves or others, nor does a basketball player, swaying to the left, trying to move his shot toward the basket—but there are situations where having an inaccurate sense of control can be unfortunate.

The illusion of control can lead to overconfidence, a bad quality for those in dangerous or critical situations. A surgeon, for example, who believed she had more control over bleeding than she did, might kill her patient with a bad decision. Pilots overconfident in their control of an aircraft’s elevation or attitude are risking their own and their passengers’ lives. I expect that those who have lives explicitly in their hands are highly trained to know exactly where their control ends, so that they will take proper precautions, but it may be that the illusion of control causes injuries and fatalities in these and similar fields. And in less explicit situations, such as driving on the highway, the training may not be as thorough. A truck driver, for example, who fell asleep behind the wheel, likely did so believing that he had control over his staying awake. And even if life is not at stake, there can be loss; dice gamblers, for example, often believe that their strategy of playing increases their odds, when it demonstrably does not (Henslin, 1967, as cited by Langer, 1975). There may be interpersonal costs, too. If I believe that I am in control of a relationship, I might not give my partner the attention she needs, and eventually lose her. The illusion of control may play a part in addiction, too: The well-known first hurdle to recovery is admitting a lack of control.

I suspect that the best antidote to the illusion of control is to use knowledge of the phenomenon and how it operates to maintain skepticism about your level of control. Awareness alone should help. If the question of how much control I have comes up, I will be more skeptical, knowing that people tend to overestimate their level of control. The knowledge that the illusion tends to be stronger when skill cues are present should help, too: If I am feeling more confident about an outcome because I’ve had practice at a task or even because I am allowed to make a choice in it, this might trigger my memory of the illusion of control and lead me to carefully consider base rates. The trick is maintaining an empirical mindset, where questions like “How much control do I have here?” present themselves to the consciousness as questions, and knowledge of the ways we fools ourselves should help with that.

Another antidote to the illusion of control is more troubling: depression. It turns out that depressed people are realistic in their assessments of control where non-depressed people are not (Alloy & Abramson, 1979, as cited by Seligman, 2002). In another version of the blinking light experiment, non-depressed participants believed that they had control of the light about a third of the time, but depressed participants could tell that they had no control. This came to be called depressive realism. Happy people, on the other hand, tend to have a flexible locus of control. That is, they tend to believe that they are in control only when it serves their well-being to believe so. When things are going well, happy people will think they are responsible, but when things are not going well, they think that it is because of passing circumstances (Seligman, 2002). (Understand that this is not a conscious strategy, but a non-conscious tendency.)

It may be that depression causes depressive realism, but it also may be the other way around: A tendency to believe we have control keeps us from getting depressed, while a tendency to be realistic about control causes us to become depressed. Everyone is a victim of circumstances to some degree—social psychology is a mountain of evidence for that—and lack of control is stressful, especially if a situation is unpleasant. That stress may be ameliorated to some degree by an incorrect belief in control, as long as that belief has no serious negative consequences. In that case, the inability to fool ourselves may be the problem for depressed people. And if in fact we do not have free will, and any sense of control we have is illusory, that illusion may be what is keeping us happy.

References

Alloy, L. B., & Abramson, L. Y. (1979). Sadder but wiser.

Journal of Experimental Psychology, 108,

 441-485.

Aronson, E., Wilson, T. D., & Akert, R. M. (2007).

Social Psychology. 

Upper Saddle River, NJ: Pearson Prentice Hall.

Henslin, J. M. (1967). Craps and magic.

American Journal of Sociology, 73

, 316-330.

Langer, E. (1975). The illusion of control.

Journal of Personality and Social Psychology, 32,

 311-328.

Langer, E. & Roth, J. (1979). Heads I win, tails it’s chance: The illusion of control as a function of the sequence of outcomes in a purely chance task.

Journal of Personality and Social Psychology, 32,

 951-955.

Sapolsky, R. M. (2004). Why is psychological stress stressful? In

Why zebras don’t get ulcers: The acclaimed guide to stress, stress-related diseases, and coping 

(pp. 252-270). New York: Henry Holt.

Seligman, M. E. (2002). Happy but dumb? In

Authentic Happiness, 

(pp. 37-39). New York: Free Press.

Tennen, H. & Sharp, J. P. (1983). Control Orientation and the Illusion of Control.

Journal of Personality Assessment, 47,

 369-374.