Skip to main content
Log in

No evidence that low levels of intoxication at both encoding and retrieval impact scores on the Gudjonsson Suggestibility Scale

  • Original Investigation
  • Published:
Psychopharmacology Aims and scope Submit manuscript

Abstract

Rationale

It is not uncommon for police to question alcohol-intoxicated witnesses and suspects; yet, the full extent to which intoxication impacts individuals’ suggestibility in the investigative interviewing context remains unclear.

Objective

The present study sought to measure the effect of alcohol-intoxication on interviewee suggestibility by implementing a standardized suggestibility test with participants whose intoxication-state was the same at both encoding and recall.

Methods

We randomly assigned participants (N = 165) to an intoxicated (mean breath alcohol level [BrAC] at encoding = 0.06%, and BrAC at retrieval = 0.07%), active placebo (participants believed they consumed alcohol but only consumed an insignificant amount to enhance believability), or control (participants knowingly remained sober) group. An experimenter then implemented the Gudjonsson Suggestibility Scale (GSS), which produced free recall outcomes (number of correct details and memory confabulations) and suggestibility outcomes (yielding to leading questions and changing answers in response to negative feedback from the experimenter).

Results

Intoxicated participants recalled fewer correct details than did placebo and control participants but did not make more confabulation errors. No effects of intoxication on suggestibility measures emerged.

Conclusions

Moderately intoxicated interviewees may not be more suggestible during investigative interviews than sober interviewees. However, before concrete evidence-based policy recommendations are made to law enforcement, further research is needed examining the effects of alcohol on suggestibility in conditions that are more reflective of the legal context.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. BrAC, which we report throughout this paper in %, denotes grams of ethanol/210 L of breath

  2. Evans et al. (2019) also included a sober-at-encoding/intoxicated-at-delay condition; however, the delay in Evans et al.’s study was 1 week, whereas the delay in Santtila et al. (1999) was 50 min

  3. One participant experienced a negative reaction to consuming alcohol and one participant disclosed having falsified responses during the eligibility interview

  4. Given slow recruitment during the start of data collection, we added, towards the end of data collection, an additional $20 Amazon gift card to the compensation package for participants opting for cash. This change was approved by the relevant Institutional Review Board

  5. These questionnaires included the Consideration of Future Consequences scale (Strathman et al. 1994) and the Brief Self-Control Scale (Tangney et al. 2004)

  6. We initially used the Intoxilyzer 5000 to collect BrAC measurements. Due to technical difficulties, we switched to using handheld breathalyzers. Both machines reliably measure BrAC and have been used in multiple past studies (e.g., Altman et al. 2019; Schreiber Compo et al. 2011)

  7. A confederate was in the bar for half of participants’ 30-min drinking session, for reasons pertaining to the unrelated study that used the same sober, placebo, and intoxicated participants. The confederate’s only role during this time was to engage in small talk with participants and the bartender. Regardless of participants’ intoxication-state condition, the confederate drank pure orange juice. Confederates were trained not to announce what was in their drink, but to respond to participants’ drink-related inquiries in a manner consistent with participants’ intoxication-state condition (e.g., confederates indicated they tasted the vodka in the placebo condition or only orange juice in the control condition)

  8. This procedure is used in confession research to experimentally manipulate participants’ culpability by having the confederate either seek help on a problem that should be solved individually, or not seek help, as determined by the randomly assigned culpability condition. Implementation of the cheating paradigm and the culpability manipulation was not related to the purpose of the present study. Note that the primary outcome measures presented here did not significantly differ across culpability conditions administered as part of the cheating paradigm. The interrogation phase of the cheating paradigm took place after all GSS procedures were completed

  9. Mixed-model ANOVAs, with recall timing as a within-subjects variable and intoxication-state as a between-subjects variable, did not produce significant interactions for correct details or confabulations, nor did a mixed-model ANOVA, with Yield 1/Yield 2 scores serving as the within-subjects variable. Thus, to ease interpretation of the results, we report a series of Bonferroni-corrected one-way ANOVAs

  10. The pattern of differences across the intoxication-state conditions for all free recall and suggestibility outcomes remained the same as those presented in the main text when we ran ANCOVA analyses with verbal IQ included as a covariate

  11. We examined the correlation between the number of correct details and the number of confabulations separately for each intoxication-state condition at both immediate and delayed recall. None of the groups demonstrated significant correlations between these variables at either immediate (ps ≥ .132) or delayed recall (ps ≥ .226)

References

Download references

Acknowledgments

We thank the many research assistants in the TRIIIAD Lab at the Department of Psychology, Florida International University, USA, for their help with data collection and coding.

Funding

This study was funded by Grant SES-1556762 from the National Science Foundation (USA) and by Grant 2014–6693 from the Swedish Research Council (Sweden).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Jacqueline R. Evans.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

ESM 1

(DOCX 13 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mindthoff, A., Evans, J.R., Compo, N.S. et al. No evidence that low levels of intoxication at both encoding and retrieval impact scores on the Gudjonsson Suggestibility Scale. Psychopharmacology 238, 1633–1644 (2021). https://doi.org/10.1007/s00213-021-05797-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00213-021-05797-9

Keywords

Navigation