Elizabeth Loftus
Elizabeth Loftus is a cognitive psychologist at the University of California, Irvine, and the researcher who established that memory is reconstructed rather than retrieved. Her best-known experiment showed that asking how fast cars were going when they "smashed" into each other produced higher speed estimates than "hit", and that a week later the participants given "smashed" were more likely to report broken glass that was never in the film. Her work changed how courts treat eyewitness testimony. It is also routinely overstated, and the scope is worth getting right.
Who they are
Loftus is Distinguished Professor at UC Irvine, with appointments across psychology and social behaviour, criminology, law and society, and the School of Law. That spread is not incidental. The finding she is known for is a laboratory result about memory with immediate consequences in a courtroom, and she has spent much of her career at the point where the two meet.
The 1974 study with John Palmer is the canonical demonstration. Participants watched films of car accidents and were then asked about them, with one word varied in the question. The verb changed the estimate. More striking, it changed what participants later reported having seen, producing a detail that had not been there. The mechanism the result points to is that recollection is an act of assembly performed at the moment of asking, drawing on fragments plus whatever has arrived since, including the wording of the question itself.
The programme that followed pushed on how far that assembly can be moved, up to and including the implantation of memories for events that did not happen. The findings replicate, which puts them in a different category from a good deal of what social psychology produced in the same decades.
What the work establishes needs stating precisely, because it gets stretched. It shows that memory is open to influence at retrieval, and that the influence can be very light, a single leading verb in a single question. It does not show that old memories are generally wrong, and it licenses no verdict on any particular recollection. Someone's memory of a specific afternoon in 1994 may be perfectly accurate; Loftus's work is about the vulnerability of the process, not about the falsity of its outputs.
On this site the relevance is to nostalgia. Nostalgic memories sit in exactly the conditions the reconstructive account says matter most: they are old, so there is more assembly to do, and they carry emotional weight, so the assembly has a direction. The direction itself was measured elsewhere, by Terence Mitchell and colleagues, in the rosy-view studies.
There is a practical lesson too, and it is not about courtrooms. There is no neutral way to ask a question. Anyone writing a customer survey or a feedback form is choosing between "hit" and "smashed" whether they know it or not.
Key papers
Loftus, E. F., & Palmer, J. C. (1974), "Reconstruction of automobile destruction: An example of the interaction between language and memory," Journal of Verbal Learning and Verbal Behavior, 13(5), 585-589. The study that established the misinformation effect: the verb in the question changed both the speed estimate and, a week later, what participants reported having seen.
Loftus has published extensively beyond this paper, including on the implantation of false memories and on the treatment of eyewitness evidence in court. Only work cited on this site is listed here.
Key ideas
Memory is reconstructive. Recollection is assembled at the moment of asking rather than played back.
The misinformation effect: information supplied after an event, including the wording of a question about it, can alter what is subsequently remembered.
The intervention required is very small. A single verb is enough.
The scope of the claim: retrieval is open to influence. This is not the claim that old memories are generally wrong, and it settles nothing about any particular memory.
There is no neutral question. Survey and testimonial design carry the finding whether or not the designer intends it.
Concepts on this site
Declared vs Revealed Preferences
Find them elsewhere
The X account states faculty at University of California, Irvine and its content is plainly hers. Her UCI pages and her 2025 CV link no LinkedIn or Scholar profile, so neither was recorded
Checked 2026-09-10. Handles and affiliations go out of date, so treat this block as a snapshot rather than a standing fact.
Sources
- Loftus, E. F., & Palmer, J. C. (1974), "Reconstruction of automobile destruction: An example of the interaction between language and memory," Journal of Verbal Learning and Verbal Behavior, 13(5), 585-589.
- University of California, Irvine, faculty listings for Psychology and Social Behavior, Criminology, Law and Society, and the School of Law. (Affiliation and title.)