Introspection -- the process of thinking about yourself, and figuring out why you do the things you do -- is an immediately obvious and seemingly important part of human consciousness. Our natural curiosity provokes us to try and figure ourselves out, to try and understand the physics of our personality as we understand those of our world. Unfortunately, there is evidence that what we have come to believe (through introspection) to be the reasons for our actions may be just as incorrect as an outside observer trying to explain the same actions. That is, while we believe we have privileged access to our reasoning, we may not have as much as it seems.

The first evidence of this is from the Maier string experiment, performed in 1931. This experiment asked participants to tie together two strings which were hanging from the ceiling at such a distance that one string couldn't be reached while holding the other. There were three possible solutions to the problem using the props in the room: some weights, a chair, and another piece of string. Each time the participant solved the problem, he was asked to do it again a different way, and the solutions were valued from easy to hard. Hardest was the solution of adding weights to the strings and setting them in motion like pendulums, so by standing in the middle one could eventually grab both strings. Maier found that most people (ie, the control group) were unable to come up with this solution on their own, but if the experimenter set one string swinging by bumping into it, the participant would often find the solution immediately. When asked how they figured it out, the participants replied with things like "It just came to me," and "Using the weights seemed like a good idea," -- they uniformly didn't recognize that the experimenter's movement had triggered their intuition.

Nisbett & Wilson's 1977 moon-ocean experiment showed the same lack of understanding of one's own reasoning. It was ostensibly a test of medium-term memory where participants were asked to memorize a list of word pairs. In the experimental group, the list contained the pair moon-ocean, whereas the control group's list did not. After the results were taken -- and nearly all of the experimental group could recall the ocean-moon pair -- both groups were asked to name whatever kind of laundry detergent came to mind. Overwhelmingly, the experimental group named Tide, while the control group had a more even distribution. When asked why they thought of Tide, the experimental group mentioned its packaging, its name-brand recognizability, an ad they saw on TV last night, etc. As in Maier's experiment, they didn't (with two or three exceptions) mention the experiment as a cause of their action.

Another experiment by Nisbett & Wilson asked participants -- passersby in a mall who volunteered -- to select the best stocking from four identical stockings attached to a board. The subjects chose the rightmost one much more than the others (a four to one ratio), an expected result known as the position effect. As usual, when the participants were asked to explain their motivation, they came up with all kinds of reasons that weren't the position, and were in fact hostile to the suggestion that position had anything to do with their choice.

Nisbett & Wilson's work precipitated two rules about introspection which are held to be fact by the psychological community. In their words:

  1. People often cannot report accurately on the effects of particular stimuli on higher order, inference-based responses. The accuracy of subjective reports is so poor as to suggest that any introspective access that may exist is not sufficient to produce generally correct or reliable reports.

    and

  2. When reporting on the effects of stimuli, people may not interrogate a memory of the cognitive processes that operated on the stimuli; instead, they may base their reports on implicit a priori theories about the causal connection between stimulus and response.