Emotional design
Norman splits emotional response into gut reaction, use over time, and the story afterwards.
Norman describes three levels of processing. Visceral is the immediate, pre-conscious reaction to appearance, sound and feel. Behavioural is the felt quality of use: control, fit, competence. Reflective is the slower judgement about what owning or using the thing says, which is where meaning and self-image sit.
Three levels, three different failures
How it shows up in software
The three levels are a checklist for where an experience is failing. A product can look right and feel wrong in use, or work well and mean nothing, and the fix is different in each case. Most teams over-invest in visceral and under-invest in reflective, which is where long-term attachment actually lives.
Using it well
- Diagnose complaints by level. Looks cheap is visceral, fights me is behavioural, and I do not want to be the kind of person who uses this is reflective.
- Build the behavioural layer around competence: responsiveness, clear system state, and recoverable mistakes.
- Give the reflective layer something honest to hold, such as a record of what the person has actually made or learned.
- Check the levels do not contradict each other. A serious tool wrapped in playful illustration confuses what the product is for.
Where it turns manipulative
- Using reflective framing to sell identity the product does not earn, such as calling routine usage a movement.
- Papering over behavioural failure with visceral polish, which raises expectation and deepens the eventual disappointment.
- Attaching self-worth to usage, so a missed day reads as a personal failing rather than a skipped session.
Where you have seen it
Things
Motion and typography carry the visceral level, keyboard-first interaction carries the behavioural, and the review ritual gives the reflective level something to be about.
Strava
The activity feed handles visceral and behavioural, while yearly and segment records give users a durable story about the kind of athlete they are.
What the research says
- Norman, 2004Contested
Introduces the three-level model and argues that attractive things work better because positive affect broadens problem solving, while anxiety narrows it.
A framework, not a tested theory. The levels are a useful vocabulary; they were not derived from an experiment and there is no measurement instrument that separates them cleanly.
- Desmet and Hekkert, 2007Mixed evidence
Propose a framework of product experience with aesthetic, meaning and emotional components, developed independently and broadly compatible with Norman's levels.
Also a framework. Convergence between two frameworks is not the same as empirical support.
- Fredrickson, 2001Mixed evidence
The broaden-and-build theory, which Norman leans on for the claim that positive affect widens thought and action repertoires.
Parts of this literature have had replication trouble, particularly the positivity ratio work associated with it. The broad claim is plausible; specific numbers from it are not usable.
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.