The affect heuristic
People read a quick good or bad feeling off a thing and reason backwards from it.
When judging something, people consult a fast affective tag before they consult the facts. That tag then shapes the facts they accept: liked things are judged safer and more beneficial, disliked things riskier and less useful. The effect grows under time pressure and when the person lacks a way to evaluate the thing directly.
How it shows up in software
A person forms an impression of a product in the first seconds and then interprets everything after through it. A pricing page that feels honest gets read charitably, and an identical page on a site that feels cheap gets read as a trap. The same permission prompt reads as reasonable or invasive depending on how the app felt beforehand.
Using it well
- Put the honest version of the hard information in front of people early, while their affective read is still forming, rather than at the end of a flow.
- Match the emotional register of a screen to its stakes. A delete confirmation that uses the same cheerful tone as onboarding makes people misjudge the risk.
- Test first impressions separately from task success. Show a screen for five seconds, ask what the product does and whether they would trust it with a payment.
- When you change something that carries risk, such as data sharing, rewrite the surrounding copy too. Facts alone will not move an affective read.
Where it turns manipulative
- Wrapping a genuinely risky action in warm illustration and soft language so people underweight what they are agreeing to.
- Using the affective glow of a free tier to get consent that people would refuse if the same request arrived cold.
- Manufacturing dread around a competitor or around inaction, so the feeling does the arguing that the evidence cannot.
Where you have seen it
Apple
Privacy-related prompts on iOS use plain black text on a neutral sheet with no brand warmth, which separates the affective read of the app from the read of the permission.
Monzo
The app shows a transaction's merchant logo and location on a map before any of the numeric detail, so the first read of a charge is recognition rather than arithmetic.
What the research says
- Finucane, Alhakami, Slovic and Johnson, 2000Well evidenced
Participants judged risk and benefit of technologies. The two moved inversely, and the inverse link got stronger when people were rushed, which is hard to explain if they were weighing evidence.
- Slovic, Finucane, Peters and MacGregor, 2007Mixed evidence
A review pulling the affect heuristic together across risk perception, insurance and financial judgement. Frames affect as a primary input to judgement rather than a by-product.
This is a synthesis, not a single test. The general pattern replicates well. Claims about the size of the effect in any specific decision do not transfer from these studies.
Grades are a judgement about the evidence, not about how useful the idea is. Plenty of contested effects are still worth knowing, as long as you do not cite them as settled.