Choice blindness
We'll defend a choice we never actually made.
What it means
Choice blindness is the finding that people often fail to detect when the outcome of a choice they just made has been covertly swapped for a different one — and will then proceed to explain and justify the choice they did not actually make. The mechanism reveals that our access to our own reasons is largely reconstructive: rather than retrieving the true cause of a preference, people confabulate a plausible-sounding rationale after the fact, generating confident explanations for an option they had in fact rejected. In the original demonstrations, participants shown two faces and asked why they 'chose' one frequently did not notice when handed the rejected face, and they elaborated reasons for it as readily as for genuine choices; the effect has since been extended to consumer tastes, moral and political attitudes. A nuance is that detection rates rise when the mismatch is large or the original preference was strong, so blindness is not total. It matters for the validity of stated preferences and self-reports, for understanding introspection's limits, and for fields from market research to political polling that take people's account of their own choices at face value.
The method behind it
The paradigm is a piece of applied stage magic. In the face task the experimenter uses a hidden duplicate card and a black-backed sleight to return the very photograph the person rejected. In the supermarket, jam jars and teapots are built with two compartments so the opposite sample is served on the re-taste, and the self-transforming survey hides a doctored patch that flips an answer as the page turns. The machinery holds everything constant except the outcome, isolating whether people notice that their action and its result have come apart. Detection is scored twice: concurrently, when they justify the choice, and retrospectively, in a debrief that asks whether anything seemed wrong.
How reliably it replicates
The numbers are consistent across settings. In the founding study most face swaps went unnoticed, and people supplied reasons for the photograph they had turned down. In the supermarket only a small fraction of jam and tea swaps were caught in the moment, even for tastes as far apart as cinnamon-apple and bitter grapefruit. On a moral-opinion survey, sixty-nine percent of participants missed at least one of two reversed answers; during a Swedish election, only about a fifth of altered political replies were detected. The effect also travels beyond its founding lab: forensic researchers find eyewitnesses will accept and defend a lineup face swapped for the one they rejected. It is one of the sturdier results in the confabulation literature.
What the confabulations look like
The explanations people give for choices they never made are neither hesitant nor thin. When coders compared them with reasons for genuine choices, the manipulated ones were about as long, specific, confident and emotionally engaged, and blind raters could not reliably tell the two apart. People sometimes reach for features unique to the option they rejected, praising the smoky aroma of a tea they did not pick or the sweetness of the jam they passed over. This is not lying; the speaker believes it. Introspection here is generative rather than a readout of stored reasons, the mechanism behind Nisbett and Wilson's old observation that we often tell more than we can know.
It can change real attitudes
A missed reversal does not merely pass unnoticed; it can move the person who missed it. When participants built arguments for a political stance that had been secretly flipped, many drifted toward it and still held the shifted view a week later, the largest change coming from those who had confabulated support. Self-transforming surveys nudged voter intentions across a left-right divide more readily than conventional polling implied was possible, and the same tool has been turned to cooling partisan hostility by feeding both sides falsely moderate feedback. Reasons voiced aloud, even for a position one rejected a minute earlier, tend to be adopted as one's own.
Boundaries and what it does not prove
Blindness is never complete, and its size tracks conditions. Detection rises when the two options are far apart perceptually, when the original preference was strong or freshly formed, when attention is high, and when people are warned to watch for a switch. Memory failure on its own does not produce the effect. And the paradigm shows that introspective access is fallible and partly reconstructive, not that it is absent: people catch a real share of the manipulations. Reported rates are drawn from trials where the sleight succeeded, so a headline figure describes accepted swaps rather than any general measure of how gullible the mind is.
Examples
Shown the face they 'picked' (secretly switched), most people don't notice and explain why they chose it.
In a supermarket taste test, shoppers sample two jams, are handed back the jar they rejected, and are asked why they liked it. Most don't notice, and explain the choice confidently.
Survey respondents have their own answers covertly reversed and are then asked to justify them. Many argue fluently for the political position they had opposed a minute earlier.
A witness picks suspect A from a photo lineup; shown 'their pick' minutes later, it is suspect B, and they detail the jaw and eyes that made B the culprit.
Reviewing two resumes, a manager flags one as the stronger hire; handed back the candidate they passed over, they list the very skills that make this one the obvious choice.
First described in Johansson, Hall, Sikström & Olsson (2005).
Key references
- Strandberg, T., Sivén, D., Hall, L., Johansson, P., & Pärnamets, P. (2018). False beliefs and confabulation can lead to lasting changes in political attitudes. Journal of Experimental Psychology: General, 147(9), 1382-1399. doi.org/10.1037/xge0000489
- Sagana, A., Sauerland, M., & Merckelbach, H. (2013). Witnesses' blindness for their own facial recognition decisions: A field study. Behavioral Sciences & the Law, 31(5), 624-636. doi.org/10.1002/bsl.2082
- Hall, L., Johansson, P., & Strandberg, T. (2012). Lifting the veil of morality: Choice blindness and attitude reversals on a self-transforming survey. PLoS ONE, 7(9), e45457. doi.org/10.1371/journal.pone.0045457
- Hall, L., Johansson, P., Tärning, B., Sikström, S., & Deutgen, T. (2010). Magic at the marketplace: Choice blindness for the taste of jam and the smell of tea. Cognition, 117(1), 54-61. doi.org/10.1016/j.cognition.2010.06.010
- Johansson, P., Hall, L., Sikström, S., & Olsson, A. (2005). Failure to detect mismatches between intention and outcome in a simple decision task. Science, 310(5745), 116-119. doi.org/10.1126/science.1111709