Assimilation and Contrast
Two opposed predictions about what an expectation does to the experience that follows it. Assimilation pulls the judged experience towards the expectation, so a product you were told was good feels good; contrast pushes it away, so a product that falls short of a raised expectation is punished for the gap rather than judged on its merits. They predict opposite signs from the same setup, which is how a satisfaction literature can run for decades on both at once. The only quantitative adjudication available found the assimilation sign and no evidence supporting contrast effects.
What it is
The practitioner version of contrast, that raising expectations is dangerous because people will notice the shortfall, is widely repeated as though it were established. Tom Schiebler, Nick Lee and Felix Brodbeck set up exactly that question in a 2025 meta-analysis in the Journal of the Academy of Marketing Science, noting in their own words that it is possible for expectations to have opposing effects on consumer satisfaction depending on the underlying psychological processes presupposed, and that a comprehensive assessment of the empirical evidence for those positions was missing. Their meta-analytical path analysis covers one hundred and fifty records and just under fifty-nine thousand people. Across that aggregate, raising expectations raised satisfaction, at a correlation they describe as moderate, and the predicted backlash did not appear.
In effect
Two caveats travel with the result and both go on the page rather than in a footnote. The positive relationship was significantly stronger in cross-sectional studies than in longitudinal and experimental ones, which means the effect is largest in the weakest designs. The authors also report an unusual downward publication bias, running the opposite way to the familiar one, which on their reading suggests the true correlation between disconfirmation and satisfaction is higher still than their already high estimate. The moderators worth carrying are that the correlation was stronger for predictive than for normative expectations, and stronger for services than for goods.
What it does not say
It does not establish who originated either prediction. The paradigm long predates the meta-analysis and this vault has not audited its lineage, so no originator is recorded here rather than crediting whoever is nearest to hand; Schiebler and colleagues adjudicate the two predictions and did not invent either. The finding also does not say contrast never happens, only that it does not show up in the aggregate, and the design-strength moderator means the aggregate is weighted towards the studies least able to rule out alternatives. One further limit: the ambiguity moderator from the advertising literature, which decides whether an expectation gets to act at all, has never been meta-analysed, so this synthesis is about expectation and satisfaction rather than about that condition.
Sources
- Schiebler, T., Lee, N. J., & Brodbeck, F. C. (2025). "Expectancy-disconfirmation and consumer satisfaction: A meta-analysis." Journal of the Academy of Marketing Science, 54(1), 91-112. doi:10.1007/s11747-024-01078-x. Open access, full text read. One hundred and fifty records, 58,597 participants, meta-analytical path analysis. Author order confirmed, Schiebler first, Ludwig-Maximilians-Universität München.
- Hoch, S. J., & Ha, Y.-W. (1986). "Consumer Learning: Advertising and the Ambiguity of Product Experience." Journal of Consumer Research, 13(2), 221-233. doi:10.1086/209062. Abstract read in full. The source of the ambiguity moderator this synthesis does not cover.