← The Casebook

Contaminated record · genuine or implanted?

Suggestibility

The tendency for memory to be altered by leading questions, social cues, or post-event information; people can absorb suggestions and later remember them as if they were part of the original event.

Exhibit · the account as remembered

Elizabeth Loftus and colleagues demonstrated in the 1970s that eyewitness memory is highly malleable to post-event suggestion. In Loftus & Palmer (1974), participants who were asked about cars that 'smashed' into each other gave higher speed estimates and were more likely to falsely recall broken glass than those asked about cars that 'hit' each other. The misinformation effect has been replicated in hundreds of studies across species and populations, and Gudjonsson's work on interrogative suggestibility links it directly to false confessions in criminal cases. ← reconstructed, not recorded

How the contamination enters

Suggestibility operates through multiple mechanisms. Source misattribution occurs when post-event information is mistakenly remembered as having come from the original event. Semantic integration theory proposes that misleading details are blended with original memories into a unified representation. More recent accounts emphasize that misinformation can impair access to original traces (retrieval impairment) without necessarily destroying them. Gudjonsson's model adds that interrogative suggestibility involves a 'Yield' component (accepting leading questions) and a 'Shift' component (changing responses under negative feedback), driven by cognitive and social factors.

Competing account — McCloskey & Zaragoza (1985) argued that misinformation effects need not imply memory impairment—they could reflect demand characteristics, response bias, or lower rates of correct guessing on misled items. Subsequent work (e.g., Lindsay & Johnson 1989; Belli 1989) provided evidence for genuine memory impairment under at least some conditions, leading to a reconciliation: both memory impairment and response bias contribute.

Seams the examination found

seam 01Loftus & Palmer speed estimationnot reported

Mean speed estimates ranged from 40.8 mph ('smashed') to 31.8 mph ('contacted'). One week later, 32% of the 'smashed' group falsely recalled broken glass versus 14% of the 'hit' group, though no glass was present.

Loftus & Palmer · 1974 · n = 45 (Experiment 1) and n = 150 (Experiment 2) undergraduates

seam 02Loftus, Miller & Burns semantic integrationnot reported

Misled participants performed significantly worse on recognition of the original event detail; they often chose the suggested item over the originally seen item, supporting the memory impairment hypothesis.

Loftus, Miller & Burns · 1978 · n = 180 undergraduates across multiple experiments

seam 03McCloskey & Zaragoza modified recognition testnot reported

In six experiments, no difference was found between misled and control subjects when the suggested item was not an option, challenging the memory impairment hypothesis and arguing that effects could be due to response bias or source confusion.

McCloskey & Zaragoza · 1985 · multiple experiments with adult subjects

Where provenance breaks

When a witness's account changes after they hear others describe the event; when leading questions ('Was the car going fast when it smashed?') produce details (broken glass) that were not present; when therapy clients 'recover' memories after suggestive prompting.

alteration confirmed

Authenticity — robust. The misinformation effect is one of the most robust findings in memory research, replicated in hundreds of studies across humans and non-human species, using diverse materials (slides, videos, real events) and populations. McCloskey & Zaragoza's (1985) challenge prompted refinement rather than rejection; subsequent research established that genuine memory impairment occurs under certain conditions (e.g., delayed testing, low source discriminability) while response bias also contributes. Meta-analyses confirm a reliable effect.

Records later found altered

  • law · 1989-1991

    Guildford Four and Birmingham Six wrongful convictions

    Innocent suspects confessed to IRA bombings in the UK after coercive interrogation and leading questioning. Their convictions were overturned in 1989-1991; Gudjonsson's expert testimony on suggestibility and psychological vulnerability was pivotal in the appeals.

  • medicine · 1990s

    Recovered memory therapy controversy (1990s)

    In the 1990s, thousands of patients in therapy 'recovered' memories of childhood sexual abuse and Satanic ritual abuse under suggestive questioning and hypnosis. Many of these memories were later shown to be false, leading to wrongful accusations, retracted confessions, and multi-million-dollar lawsuits against therapists.

  • law · ongoing

    Eyewitness misidentification in criminal trials

    Eyewitness misidentification is the single leading cause of wrongful conviction in the United States, contributing to approximately 70% of convictions overturned by DNA evidence; suggestive police lineups and leading questions are primary contributors.

To authenticate the memory

Warn witnesses about the possibility of misinformation before exposure, use open-ended rather than leading questions, and record initial statements as soon as possible before post-event information can contaminate memory.

Warnings given before misinformation exposure can reduce the effect; the NICHD protocol for child interviewing, which emphasizes open-ended questions and avoids suggestive prompts, has been shown to protect children from misinformation contamination.

Easily confused with

File your own case

Open the same case on your own draft.

Paste a memo, a research draft, or a strategy argument. It is scored against all 175 cards, and the strongest two or three risks come back with the evidence quoted and one practical next check.

Open a case on your draft →