EVIDENCE LIBRARY
After Drinking Following a Crash, Can a Later Interview Yield More Remembered Details?
A 60-person film experiment found more recalled details at a later session in those who drank after viewing. There was no significant alcohol-placebo difference at either session, and the design did not establish that drinking improves testimony.
A later retrieval attempt produced more recalled details in participants who drank after viewing a simulated crash, without a significant fall in their accuracy percentage. But the alcohol and placebo groups did not differ significantly in free recall at either interview. The finding supports a carefully limited view of repeated recall, rather than a claim that post-crash drinking improves memory.
The event happened before alcohol
Butterworth and colleagues studied 60 adults who completed two sessions. Participants described themselves as heavy drinkers, but those meeting specified alcohol-use-disorder or PTSD exclusion criteria were not eligible. The average age was about 29.
Participants watched a short dramatized car-accident film while sober. They were then randomly allocated to an alcoholic drink or a placebo. The first recall session followed beverage consumption and absorption; a second used the same memory tasks one week later.
This sequence matters. The study addressed drinking after an event was encoded, not drinking before a collision or being intoxicated during the event. Its findings cannot determine whether a real driver drank before or after a crash.
More details later did not mean a significant advantage over placebo
During five-minute free recall, the alcohol group reported an average of 25.10 details at the first session and 28.90 one week later. Its accurately recalled details increased from 23.23 to 26.67. Both within-group changes were significant.
The placebo group changed less. The group-by-session interactions were significant for total details, p = .015, and accurate details, p = .024. However, direct alcohol-placebo comparisons were not significant at either session.
Accuracy percentages also did not show a significant group-by-session interaction. More information and a greater proportion of correct information are separate outcomes. The results do not establish that the alcohol group’s later account was superior to the placebo group’s.
Confidence and the experience of remembering were separate outcomes
A cued task asked 16 questions about the film. Participants distinguished consciously remembering a detail from simply finding an answer familiar, and could respond that they did not know.
Some pooled response-category analyses suggested greater uncertainty in the alcohol group. Yet the participant-level analysis of “do not know” responses did not show a significant condition effect. Likewise, the group-by-session interaction for correct familiarity-based answers was p = .081, which did not meet .05. A strong claim that alcohol selectively improves one memory system would overstate these results.
The alcohol group’s average confidence declined between sessions even as free recall became more complete. Confidence, amount recalled and accuracy therefore should not be treated as interchangeable measures.
What limits the application to an actual crash?
The film was not an experienced collision, and the groups’ stress responses were not established as equivalent. Every participant had an initial recall attempt, so the experiment could not separate the effects of time and recovery from the benefit of prior retrieval practice.
The study was single blind. Post-film alcohol exposure was checked with a breath instrument; the target blood-equivalent concentration was not a direct blood measurement. The selected sample and frequent history of blackouts further limit generalization.
A later account can contain additional accurate details without proving the truth of every new statement. This experiment does not validate an individual account, prove a blackout or reconstruct BAC. For a separate memory question during intoxication, the site’s study of remembering delayed intentions examines prospective memory rather than recollection of a witnessed event.
Primary source
Butterworth B, Hand CJ, Lorimer K, Gawrylowicz J. The impact of post-encoding alcohol consumption on episodic memory recall and remember-know responses in heavy drinkers. Frontiers in Psychology. 2023;14:1007477. https://doi.org/10.3389/fpsyg.2023.1007477.