Contaminated record · genuine or implanted?
Misinformation Effect
The distortion of a person's memory of an event after they are exposed to misleading information about that event, causing them to incorporate the false information into their recollection.
Exhibit · the account as remembered
Elizabeth Loftus established the misinformation effect in a landmark 1974 study with John Palmer, showing that post-event leading questions altered eyewitness memory for a filmed car accident. Loftus, Miller & Burns (1978) extended this by showing that verbal misinformation about a stop sign could be integrated into visual memory, causing witnesses to incorrectly identify a yield sign. The effect has been replicated hundreds of times and is a major factor in wrongful convictions based on eyewitness testimony, as documented by the Innocence Project. ← reconstructed, not recorded
How the contamination enters
Two main mechanisms are proposed. The memory impairment hypothesis (Loftus) suggests that misinformation overwrites or alters the original memory trace, so the original is genuinely lost. The source-monitoring account (Lindsay & Johnson; Johnson et al., 1993) proposes that both the original and post-event information are retained, but the source of each is confused at retrieval, a failure of source monitoring. Neuroimaging evidence supports both accounts: post-event misinformation can suppress hippocampal reactivation of the original memory, but source-monitoring failures are also observed when both memories are accessible.
Competing account — The response-bias account suggests misinformation only affects what people report, not their underlying memory. The source-monitoring account holds that original and post-event information coexist but sources are confused. The memory impairment account proposes original memories are genuinely overwritten. Current evidence supports a dual-mechanism view where both source confusion and memory alteration operate depending on conditions.
Seams the examination found
In Experiment 1, mean speed estimates varied: 'smashed' = 40.8 mph, 'collided' = 39.3 mph, 'bumped' = 38.1 mph, 'hit' = 34.0 mph, 'contacted' = 31.8 mph. In Experiment 2, 32% of 'smashed' participants falsely recalled broken glass versus 14% of 'hit' participants and 12% of the control group who were not asked about speed.
Elizabeth F. Loftus & John C. Palmer · 1974 · n = 45 (Experiment 1), n = 150 (Experiment 2)
75% of participants in the consistent condition correctly identified the stop sign. Only 41% of participants in the misinformation condition chose the stop sign; the remaining 59% chose the yield sign, having integrated the false verbal information into their visual memory. This demonstrated the semantic integration of verbal information into a visual memory trace.
Elizabeth F. Loftus, David G. Miller & Helen J. Burns · 1978 · n = 75 (Experiment 1)
Participants often misidentified narrative details as having come from the slides, demonstrating a failure of source monitoring. Forewarning reduced but did not eliminate the misinformation effect, suggesting it operates partly through genuine memory alteration and partly through source confusion.
D. Stephen Lindsay & Marcia K. Johnson · 1989 · not reported in available sources
Where provenance breaks
Be alert when eyewitnesses describe events after discussing them with others, viewing media coverage, or receiving feedback from authorities. Notice when details in someone's account change after exposure to new information. In legal contexts, question whether identifications were made under suggestive conditions with confirming feedback.
Authenticity — robust. The misinformation effect has been replicated hundreds of times across varied paradigms and populations. A 1994 meta-analysis by Steblay found the effect to be reliable and robust. However, the precise mechanism remains debated: some replications support the memory-overwriting account, others the source-confusion account. Zwaan (2025) has proposed a large-scale replication of the original Loftus & Palmer study. The effect's magnitude varies with delay between event and misinformation, with the strength of original memory, and with individual differences in suggestibility.
Records later found altered
law · 1984-1995
Ronald Cotton wrongful conviction
Jennifer Thompson identified Ronald Cotton as her rapist in 1984 with complete certainty. Cotton was convicted and served over 10 years in prison. DNA testing in 1995 exonerated Cotton and identified Bobby Poole as the actual perpetrator. Eyewitness misidentification and post-event contamination (repeated lineup exposure, feedback from detectives) contributed to Thompson's false memory. The case is one of the most famous examples of eyewitness error leading to wrongful conviction.
law · 1989-present
75% of DNA exonerations involved eyewitness misidentification
The Innocence Project has documented that eyewitness misidentification played a role in approximately 75% of the 375+ DNA exonerations in the United States. Post-event information (suggestive lineup procedures, confirming feedback, media exposure) systematically contaminates eyewitness memory through the misinformation effect, making originally uncertain identifications seem highly confident at trial.
To authenticate the memory
Minimize exposure to post-event information before giving or recording testimony. Document witness accounts immediately after events before contamination can occur. Use double-blind lineup procedures where the administrator does not know the suspect's identity, and avoid giving confirming feedback after an identification.
The National Research Council (2014) recommends these procedures based on decades of research. Double-blind sequential lineups and video recording of identification procedures reduce misinformation contamination. Wells et al. (1998) system variables framework provides evidence-based recommendations for preserving eyewitness memory integrity.
Easily confused with
- Source Monitoring Errormechanistically-linkedSource monitoring failures are a key mechanism underlying the misinformation effect: witnesses confuse information from the actual event with information learned later.
- SuggestibilitysiblingBoth are 'sins of commission' in Schacter's framework. Suggestibility refers to memory distortion from suggestion or social pressure, while the misinformation effect focuses specifically on post-event information.
- False MemorychildThe misinformation effect is one major pathway through which false memories are created; it specifically involves distortion from post-event information rather than spontaneous confabulation.