Fortune Telling: How to Test a Prediction, Not Believe It

About this tool

Fortune telling is the cognitive distortion of forecasting a bad outcome and then treating the forecast as established fact. David Burns called it the fortune teller error in Feeling Good in 1980 and paired it with mind reading under the heading jumping to conclusions. The sentence usually arrives without any hedging: this interview will go badly, the test results will be bad, they will say no. Aaron Beck and Gary Emery described the same process in 1985 as the hallmark of anxious thinking, in which the mind runs ahead to danger and the body responds to the forecast as if it were the event.

Predicting is not the problem. Forecasting is one of the most useful things a mind does, and a person who could not anticipate consequences would be in trouble by lunchtime. The distortion lies in two places: the certainty attached to the forecast, and the absence of any record. Nobody writes their predictions down, so the forecasting system never receives feedback and never improves. A weather service that kept no record of its own accuracy would be trusted far less than most people trust their own worst-case forecasts.

In practice the forecast changes behavior, and the changed behavior protects the forecast. If a party is predicted to be humiliating, the invitation gets declined, and the prediction is never tested. If a conversation is predicted to end badly, it gets postponed, and the problem grows. Avoidance feels like a reasonable response to a certainty, and it removes the only evidence that could have corrected it. Worse, nothing bad then happens, which often gets credited to the worrying rather than to the fact that the forecast was wrong. What Is Worry? sets out that cycle in more detail.

There is a direct test of how often worries come true, and it is not flattering to fortune telling. Lucas LaFreniere and Michelle Newman asked people in treatment for generalized anxiety disorder to record their worries as they occurred and then check the outcomes, and reported in 2020 that roughly nine in ten of the recorded worries did not come to pass. When something did happen, participants most often rated the outcome as about as bad as they had expected, and rated it better than expected slightly more often than worse. The sample was small and specific to people seeking treatment, and the records were self-reported, so the exact proportion should not be treated as a general law. The direction of the finding is the useful part.

Why forecasts skew is partly a matter of how prediction works. Daniel Kahneman and Amos Tversky showed in 1973 that people predict from how well a scenario fits a vivid story rather than from how often such outcomes actually occur, which leaves base rates out of the calculation. A detailed disaster scene feels more probable than a dull likely outcome because it is easier to picture. Anxiety sharpens this further, since threatening material is easier to bring to mind, and ease of imagining is routinely mistaken for likelihood.

Fortune telling contains a second prediction that is almost never examined: the forecast of how you would cope. Timothy Wilson and Daniel Gilbert's work on affective forecasting, summarized in 2005, found that people tend to overestimate how intense and how long-lasting their emotional reaction to a future event will be. Both good and bad events are absorbed faster than expected. That matters here because the dread in a prediction is not really about the event; it is about an imagined version of yourself who cannot handle it. The record generally shows that the person in the forecast copes worse than the person who actually turns up.

Fortune telling, worry and catastrophizing are relatives rather than synonyms. Fortune telling is the prediction itself: this will go badly. Catastrophizing escalates the consequence and shrinks your capacity to cope with it: it will go badly, I will lose the contract, and I will never work again. Worry is the verbal chain that runs between them, usually in what-if form, and tends to continue long after it has stopped producing anything useful. Decatastrophizing Worksheet works on the escalation, and Catastrophizing Test offers a structured look at how often the escalation happens.

The remedy is boring and effective: a prediction log. Before an event, write what you predict in a form that could be checked by someone else, attach odds as a percentage, and name the single observable outcome that would count. Afterwards, write what actually happened and how bad it really was out of ten. Ten or twenty rows later the pattern is visible, and it is the pattern that moves belief, not any single disconfirmation. Most people find their predicted severity runs well above the recorded severity, and that gap is the thing they can use next time dread arrives.

Two questions help in the moment, before anything is written. The odds question: out of a hundred times this situation has occurred, how many ended the way I am predicting, and what is the most likely outcome rather than the worst one? The record question: what happened the last five times I predicted this? Most people can list the five occasions and most of them do not match the forecast. A third question is useful when the stakes are real: if the outcome did happen, what would I actually do about it? A prediction with a plan attached loses most of its grip.

Some predictions can only be settled by running them. If the forecast is about your own performance or about how others will react, no amount of reasoning will resolve it, which is why cognitive therapy uses behavioral experiments: make the prediction specific, do the thing without the usual protective measures, and record the result. The protective measures matter. Over-preparing, arriving late, saying almost nothing, bringing someone along: any of these lets a person conclude that only the precaution prevented disaster, which leaves the forecast intact for next time.

This is not an argument for optimism, and it is not advice to ignore risk. Thinking ahead about what could go wrong is how a project gets finished and how a household stays solvent. The useful version states odds rather than certainties, names a plan, and then stops. The costly version states certainties, produces dread, and replaces action with avoidance. When a forecast is genuinely accurate, as in an unsafe relationship or a job that really is ending, it is information to act on, not a distortion to argue with. If predictions of disaster are running most days, or if avoidance has started shrinking your life, a licensed therapist can work through them systematically rather than one forecast at a time.

Reviewed by Michael Callans, MSW
Checked the fortune teller error against Burns's original list and kept the worry outcome finding stated with the limits of the study it comes from.

References

1. Burns DD. Feeling Good: The New Mood Therapy. New York: William Morrow; 1980.
2. Beck AT, Emery G, Greenberg RL. Anxiety Disorders and Phobias: A Cognitive Perspective. New York: Basic Books; 1985.
3. Kahneman D, Tversky A. On the psychology of prediction. Psychological Review. 1973;80(4):237-251.
4. LaFreniere LS, Newman MG. Exposing worry's deceit: percentage of untrue worries in generalized anxiety disorder treatment. Behavior Therapy. 2020;51(3):413-423.
5. Wilson TD, Gilbert DT. Affective forecasting: knowing what to want. Current Directions in Psychological Science. 2005;14(3):131-134.
6. Beck JS. Cognitive Behavior Therapy: Basics and Beyond. 2nd ed. New York: Guilford Press; 2011.

Fortune Telling (Cognitive Distortion) FAQ