DEEP BACKGROUND 04 OF 13 . THE ROOM Can you trust what customers tell you about their own reasons? Not about reasons, no. People sincerely report explanations they never used. In one study shoppers picked the rightmost of four identical items by almost four to one, and when asked whether position had influenced them, virtually all denied it. WHY THIS IS WORTH MONEY Most of what gets bought as audience understanding is people being asked why. Why did you switch. Why do you like it. Why would you buy it. If the answer to that question is unreliable in a specific and well documented way, then a large part of what organisations spend on understanding their audience is producing confident, sincere, well articulated fiction, and then briefing off it. That is not a small saving. It is the difference between a method that works and one that feels like it does. WHAT THE RECORD SHOWS The stockings, and the worried glance Nisbett and Wilson (Psychological Review, 1977) ran two studies in commercial premises under the guise of a consumer survey. Passersby were asked to say which article of clothing was best quality. Four nightgowns in one study, four identical pairs of nylon stockings in the other. In their words: "There was a pronounced left-to-right position effect, such that the rightmost object in the array was heavily overchosen", with "the right-most stockings being preferred over the left-most by a factor of almost four to one." Then the part that matters. "No subject ever mentioned spontaneously the position of the article in the array." And when asked directly whether position might have mattered, "virtually all subjects denied it, usually with a worried glance at the interviewer suggesting that they felt either that they had misunderstood the question or were dealing with a madman." The general claim is broader than a shopping quirk. People often cannot report what actually influenced them, and what they offer instead is a plausible theory about why somebody might do such a thing. It is offered sincerely. It is not introspection. And you cannot get round it by imagining The usual objection to fieldwork is that it is expensive, and that a workshop could do the same job. Get the team in a room, run a perspective-taking exercise, imagine our way into the customer. Eyal, Steffel and Epley (Journal of Personality and Social Psychology, 2018) ran twenty five experiments on exactly that. Predicting emotions from faces and postures, telling real smiles from fake, telling lies from truth, predicting a spouse's preferences and attitudes. Their finding: "we failed to find any consistent evidence that it actually did so. If anything, perspective taking decreased accuracy overall while occasionally increasing confidence in judgment." Across the first fifteen experiments the effect was negative, d = -0.26. And the resolution, from the same paper: "getting another person's perspective directly, through conversation, increased accuracy but that perspective taking did not." The title carries it. Accuracy comes from getting perspective, meaning going and obtaining the information, not from taking it, meaning imagining yourself into it. Note what they also found. People were confident either way. The team who spent an afternoon in somebody's kitchen and the team who spent an afternoon imagining it come back equally certain. Only one of them comes back correct. THE STRONGEST OBJECTION, AND THE ANSWER TO IT Nisbett and Wilson was attacked hard and quickly, and the attacks are worth naming. Smith and Miller (Psychological Review, 1978) argued the paper used inappropriate statistical tests, defined causality incorrectly, and stated its argument in a form that could not be falsified. That is a serious methodological objection and it has never fully gone away. Renero (Philosophy, 2025) went at the stocking case directly, arguing that researchers have aimed introspection at the wrong target, treating the psychological causes of a mental process as the thing being introspected rather than the process itself. Ericsson and Simon (Psychological Review, 1980) is usually cited as the rehabilitation of self report, and it is the objection most people expect. Read properly it does not rescue asking why. Their claim is that a verbal report is valid data when it reports information currently heeded and held in short term memory, verbalised while the task is happening. Reports of motives and explanations, produced afterwards about information not in attention, are the category they themselves excluded. So the strongest available objections are about how far the 1977 finding generalises and how it was analysed, not about whether asking a customer why is reliable. On that specific question the literature has not moved. Russo, Johnson and Stephens (Memory and Cognition, 1989) tested it directly across four tasks and concluded that "Retrospective protocols yielded substantial forgetting or fabrication in all tasks, supporting the consensus on the nonveridicality of these methods." WHERE THE EVIDENCE STOPS The stockings study is from 1977, in a shop, about a trivial choice. It is a demonstration, not a law of everything, and the sample sizes reported for the two studies are counterintuitive enough that we quote the ratio rather than the numbers. Eyal and colleagues tested instructed perspective taking. It does not follow that empathy is worthless, or that thinking about somebody else is a waste of time. It follows that imagining is not a substitute for asking. The final experiment in that paper, the one showing conversation improved accuracy, is reported here at the level of its conclusion. The figures were not obtainable and none are quoted. None of this tests a meet and greet. It tests the reliability of asking people why, which is the thing a meet and greet is built to avoid. WHERE THE CONE SITS A meet and greet is built around observing a life rather than interrogating a decision, and it happens where people actually live rather than behind glass. Three moves in order. Observe, and be interested in the life. Notice, bringing the conversation gently towards the domain without leading. Then notice what is important and true, which is where the insight is. Not what they need, not what they want. What is true, because what is true is the only thing that generates ideas nobody else could have had. And it rules out sending somebody else. The insight arrives in the room, in the moment a person says the thing they had not planned to say. It does not survive being written down by a third party and presented back three weeks later. WHAT TO DO ON MONDAY Book two visits. Not a focus group and not a facility. Their kitchen or their cafe. Ban the word why for the first half hour and ask them to show you instead. Then write down only what you saw, before you write down what you think it means. Next is the top of the spectrum, and why the people who care most give you the answers you can least use. SOURCES Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84(3), 231-259. doi:10.1037/0033-295X.84.3.231 Eyal, T., Steffel, M., & Epley, N. (2018). Perspective mistaking: Accurately understanding the mind of another requires getting perspective, not taking perspective. Journal of Personality and Social Psychology, 114(4), 547-571. doi:10.1037/pspa0000115 Smith, E. R., & Miller, F. D. (1978). Limits on perception of cognitive processes: A reply to Nisbett and Wilson. Psychological Review, 85(4), 355-362. doi:10.1037/0033-295X.85.4.355 Ericsson, K. A., & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87(3), 215-251. doi:10.1037/0033-295X.87.3.215 Russo, J. E., Johnson, E. J., & Stephens, D. L. (1989). The validity of verbal protocols. Memory and Cognition, 17(6), 759-769. doi:10.3758/BF03202637 Renero, A. (2025). Telling just what we know: Revisiting Nisbett and Wilson's stocking case. Philosophy, 100(4), 727-753. doi:10.1017/S0031819125100995 Superfandom . Deep Background . https://superfandom.ai/science/the-room/ Matt Hart . published 2026-09-08 . last changed 2026-09-09