Alcohol-intoxicated eyewitnesses in the field: Do high intoxication level, emotional distress and choice of investigative method affect the memory of a crime?
Eyewitness information is arguably one of the most powerful pieces of evidence during criminal investigations. Research on witness memory has been at the center of memory research for decades. The prevalence of alcohol-intoxicated witnesses and victims is very high. However, there are very few studies on alcohol and witness memory, especially at higher intoxication levels. The purpose of this project is, through quasi-experiments in the field, to examine how different intoxication levels interact with emotional distress at the crime scene, interview method and line-up procedure, and how these factors in turn affects witnesses’ event and face memory. Those factors are important to examine given that they are frequently encountered in real-life witness events, and there are no previous studies that have looked at those interactions. Participants will be recruited at real bars in Sweden and witness a mock crime. They will then be asked to identify the perpetrator in a line-up, and recall what they remember in an interview. Knowledge from this project will lead to a deepened theoretical understanding, but will also benefit the work of legal practitioners who will gain knowledge about which interview methods and line-ups that are most appropriate to use when intoxicated witnesses, who potentially have witnesses an emotionally distressing crime, are involved in the criminal investigation. This knowledge can lead to a higher resolution rate of crimes in the society.
Final report
THE PROJECT’S AIM AND DEVELOPMENT
Sub-project 1: How alcohol-intoxication affect witnesses’ memory of crimes
There is a strong link between alcohol consumption and crimes, and the prevalence of alcohol-intoxicated eyewitnesses, victims and perpetrators is high. However, there is a lack of knowledge about how alcohol affect memory in a legal context and how these intoxicated persons should be interviewed to gain as much correct information as possible, in an ethical manner, to aid the police investigation further (see Hagsand et al., 2021). Therefore, the aim of the project was to study how intoxication, from low to high breath alcohol concentration levels, affect witnesses’ memory. The project started the 1st of January 2019 and developed well. However, in March 2020, the covid-19 pandemic hit which meant that the data collection immediately had to be halted, since it was not safe to conduct the study anymore as it included physical contact with participants in the field. After that, an adjustment period followed where the whole project’s aim had to be revised and a new aim and orientation was granted by the funding agency.
Sub-project 2: How Swedish police interrogate intoxicated and sober suspects of crimes
The new sub-project’s aim was, through archival studies, to study Swedish police interrogations with alcohol-intoxicated and sober suspects of crimes. This aim is of high importance because there is very limited research on how Swedish police interrogation is conducted. The reason why the research focus on eyewitnesses was not maintained in this new sub-project was because the Swedish Police Authority rarely video/audio record interrogations with witnesses, why the focus was changed to study suspects of crimes. This sub-project developed well despite the pandemic, as it did not require any physical contact with participants.
METHOD
Sub-project 1: How alcohol-intoxication affect witnesses’ memory of crimes
Through quasi-experimental method in the field, participants were recruited in bars. After fulfilling the inclusion criteria and informed consent, participants breath alcohol concentration level was measured via a breathalyzer. Participants also answered some background information and filled out the Big Five personality test. Shortly thereafter, they witnessed an event on film and was then asked to recall what they remembered during an interview and then to identify the suspect in a lineup. Data was processed and then quantitatively analyzed in the statistical program SPSS.
Sub-project 2: How Swedish police interrogate intoxicated and sober suspects of crimes
This sub-project included several archival studies and police interrogations with suspects was obtained from the Swedish Police Authority. These interrogations were coded based on a taxonomy about interrogation methods and degree of suspect cooperation. The taxonomy has been developed by Professor Christopher Kelly in the U.S., who became an external expert in this new sub-project as the covid-19 pandemic gave the project a new research aim. Data was processed and quantitatively analyzed in the statistical programs SPSS, Stata and R.
THE THREE MOST IMPORTANT RESULTS OF THE PROJECT
Main result 1: The field study (Hagsand & Schreiber Compo, 2024a) recruited 65 participants in bars, with a variety of blood alcohol concentration levels (0.0–0.14%). Alcohol did not have a negative effect on the number of recalled details during the interview. However, personality affected witness’ memory. Witnesses with high scores on the Big Five-dimension Openness remembered more details about the event compared to witnesses with low scores, regardless of intoxication level. There was also a significant interaction between high scores on the Big Five-dimension Conscientiousness and high breath alcohol concentration, where these witnesses were more careful and therefore reported fewer details about the event. Beyond this, the study also found that the accuracy of the witness testimony during the interview decreased as the intoxication level increased. The study concluded that alcohol and personality, to some extent, can affect witness memory of crime.
Main result 2: An archival study (Hagsand et al., 2022) included a minor pilot study and a main study with 97 interrogations with alcohol- and drug intoxicated as well as sober suspects concerning low-stakes crimes such as driving under the influence and illegal sale of drugs. The result showed that the police used more confrontational interrogation techniques with alcohol- and drug intoxicated suspects, compared to sober ones. Also, suspects displaying signs of substance use disorder had a higher degree of cooperation with the police during the interrogations, compared to suspects without this disorder. The study concludes that some suspects seem to be more vulnerable during interrogations.
Main result 3: Swedish police interrogation methods with sober suspects in high-stakes crimes, such as murder and sexual assault cases, were analyzed in two studies.
The first, and more minor, archival study (Hagsand et al., 2023a) with 19 interrogations can be viewed as a pilot study. The study found that when the interrogators focused on rapport and relationship building to create a respectful working relationship with the suspects, the suspects in turn cooperated more. When the police instead used more confrontational interrogation methods and presented evidence, the suspects’ degree of cooperation decreased. The study concluded that the choice of interrogation method is important to be able to collect as much information as possible, in an ethical manner, during interrogations to aid the police investigation.
The second larger archival study (Hagsand et al., 2024b) included an analysis of 72 interrogations with the aim to examine if the interrogators gender and crime type affected the interrogators’ choice of interrogation methods and the suspects’ degree of cooperation. The result showed that the police presented more evidence at the end of the interrogations for murder suspects, compared to sexual assault suspects. The study also found that murder suspects cooperated to a lesser degree with the police, compared to sexual assault suspects. The study concluded that murder suspects probably are less cooperative than sexual assault suspects because of differences in the length of penalty.
NEW RESEARCH QUESTIONS
Sub-project 1 has led to the insight that witnesses’ personality is a key factor to study to understand better how they remember crimes. It is also important to expand the witness research to also include examinations of how intoxicated victims remembers, something that is mentioned as a suggestion for future research in an editorial article by Hagsand et al. (2023b).
Sub-project 2 have given rise to several new research questions about which factors that affect the interaction between police and suspects. For example, a new study has been conducted to examine who is responding to whom during interrogations. In our newest study (Hagsand et al., 2024c), we found partial support for the hypothesis that the police affect the suspects during the interrogations, but we found most support for the hypothesis that it is what the suspects does that affect the police to a greater extent. The later finding can be viewed as a paradigm shift within the research field.
DISTRIBUTION OF RESEARCH FINDINGS
The research results from the present project have been distributed via publications in scientific international peer-reviewed journals with open-access, as well as through international conference presentations, and by collaborations with the Swedish Police Authority. For example, project leader Hagsand evaluated the new national policy guidelines on interrogations with alcohol-intoxicated persons. The research from this project has also been highlighted in media, for example in Swedish Television. Finally, some further work remains in the project with two manuscripts that are anticipated to be submitted to peer-review journals during 2024. In total, it is anticipated that the present project will produce 6 scientific articles.
REFERENCES
Hagsand, A.V., & Schreiber Compo, N. (2024a). The big five traits openness and conscientiousness can affect alcohol-intoxicated eyewitnesses’ memory. Revise and resubmit decision (in June 2024) to Journal of Memory and Learning.
Hagsand, A.V., Halili, A., Nilsson, S., Kelly, C., Evans, J., Schreiber Compo, N., Zajac, H., & Lidell, L. (2024b). How crime type and police gender affect interrogator-suspect interactions in high-stake cases in Sweden: Analyze of recorded police interrogations with suspects. To be submitted to a journal during 2024.
Hagsand, A.V., Jansson, L., Kelly, C., Kapetanovic, S., Evans, J., & Schreiber Compo, N. (2024c). Temporal ordering effects in suspect interrogations, or, who is responding to whom? Examination of two relationships using cross-lagged path model analysis. To be submitted to a journal during 2024.
Hagsand, A. V., Kelly, C., Mindthoff, A., Evans, J. R., Schreiber Compo, N., Karhu, J., & Huntley, R. (2023a). The interrogator-suspect dynamic in custodial interrogations in Sweden: An application of the interrogation taxonomy framework. Scandinavian Journal of Psychology, 64, 352-367. https://doi.org/10.1111/sjop.12889
Hagsand, A. V., Flowe, H., Takarangi, M., & Gawrylowicz, J. (2023b). Editorial: The impact of alcohol and drugs on suspects', victims' and witnesses' cognition and memory. Frontiers in Psychology, 14: https://doi.org/10.3389/fpsyg.2023.1209406
Hagsand, A.V., Zajac, H., Lidell, L., Kelly, C.E., Schreiber Compo, N., & Evans, J. R. (2022). Police- suspect interactions and confession rates are affected by suspects’ alcohol and drug use status in low-stakes crime interrogations. Frontiers in Psychology, 13: 983362. https://doi.org/10.3389/fpsyg.2022.983362
Hagsand, A.V. (2021). Eyewitness testimony: How much alcohol is too much? Research Outreach, 126. https://doi.org/10.32907/RO-126-1847382832
Sub-project 1: How alcohol-intoxication affect witnesses’ memory of crimes
There is a strong link between alcohol consumption and crimes, and the prevalence of alcohol-intoxicated eyewitnesses, victims and perpetrators is high. However, there is a lack of knowledge about how alcohol affect memory in a legal context and how these intoxicated persons should be interviewed to gain as much correct information as possible, in an ethical manner, to aid the police investigation further (see Hagsand et al., 2021). Therefore, the aim of the project was to study how intoxication, from low to high breath alcohol concentration levels, affect witnesses’ memory. The project started the 1st of January 2019 and developed well. However, in March 2020, the covid-19 pandemic hit which meant that the data collection immediately had to be halted, since it was not safe to conduct the study anymore as it included physical contact with participants in the field. After that, an adjustment period followed where the whole project’s aim had to be revised and a new aim and orientation was granted by the funding agency.
Sub-project 2: How Swedish police interrogate intoxicated and sober suspects of crimes
The new sub-project’s aim was, through archival studies, to study Swedish police interrogations with alcohol-intoxicated and sober suspects of crimes. This aim is of high importance because there is very limited research on how Swedish police interrogation is conducted. The reason why the research focus on eyewitnesses was not maintained in this new sub-project was because the Swedish Police Authority rarely video/audio record interrogations with witnesses, why the focus was changed to study suspects of crimes. This sub-project developed well despite the pandemic, as it did not require any physical contact with participants.
METHOD
Sub-project 1: How alcohol-intoxication affect witnesses’ memory of crimes
Through quasi-experimental method in the field, participants were recruited in bars. After fulfilling the inclusion criteria and informed consent, participants breath alcohol concentration level was measured via a breathalyzer. Participants also answered some background information and filled out the Big Five personality test. Shortly thereafter, they witnessed an event on film and was then asked to recall what they remembered during an interview and then to identify the suspect in a lineup. Data was processed and then quantitatively analyzed in the statistical program SPSS.
Sub-project 2: How Swedish police interrogate intoxicated and sober suspects of crimes
This sub-project included several archival studies and police interrogations with suspects was obtained from the Swedish Police Authority. These interrogations were coded based on a taxonomy about interrogation methods and degree of suspect cooperation. The taxonomy has been developed by Professor Christopher Kelly in the U.S., who became an external expert in this new sub-project as the covid-19 pandemic gave the project a new research aim. Data was processed and quantitatively analyzed in the statistical programs SPSS, Stata and R.
THE THREE MOST IMPORTANT RESULTS OF THE PROJECT
Main result 1: The field study (Hagsand & Schreiber Compo, 2024a) recruited 65 participants in bars, with a variety of blood alcohol concentration levels (0.0–0.14%). Alcohol did not have a negative effect on the number of recalled details during the interview. However, personality affected witness’ memory. Witnesses with high scores on the Big Five-dimension Openness remembered more details about the event compared to witnesses with low scores, regardless of intoxication level. There was also a significant interaction between high scores on the Big Five-dimension Conscientiousness and high breath alcohol concentration, where these witnesses were more careful and therefore reported fewer details about the event. Beyond this, the study also found that the accuracy of the witness testimony during the interview decreased as the intoxication level increased. The study concluded that alcohol and personality, to some extent, can affect witness memory of crime.
Main result 2: An archival study (Hagsand et al., 2022) included a minor pilot study and a main study with 97 interrogations with alcohol- and drug intoxicated as well as sober suspects concerning low-stakes crimes such as driving under the influence and illegal sale of drugs. The result showed that the police used more confrontational interrogation techniques with alcohol- and drug intoxicated suspects, compared to sober ones. Also, suspects displaying signs of substance use disorder had a higher degree of cooperation with the police during the interrogations, compared to suspects without this disorder. The study concludes that some suspects seem to be more vulnerable during interrogations.
Main result 3: Swedish police interrogation methods with sober suspects in high-stakes crimes, such as murder and sexual assault cases, were analyzed in two studies.
The first, and more minor, archival study (Hagsand et al., 2023a) with 19 interrogations can be viewed as a pilot study. The study found that when the interrogators focused on rapport and relationship building to create a respectful working relationship with the suspects, the suspects in turn cooperated more. When the police instead used more confrontational interrogation methods and presented evidence, the suspects’ degree of cooperation decreased. The study concluded that the choice of interrogation method is important to be able to collect as much information as possible, in an ethical manner, during interrogations to aid the police investigation.
The second larger archival study (Hagsand et al., 2024b) included an analysis of 72 interrogations with the aim to examine if the interrogators gender and crime type affected the interrogators’ choice of interrogation methods and the suspects’ degree of cooperation. The result showed that the police presented more evidence at the end of the interrogations for murder suspects, compared to sexual assault suspects. The study also found that murder suspects cooperated to a lesser degree with the police, compared to sexual assault suspects. The study concluded that murder suspects probably are less cooperative than sexual assault suspects because of differences in the length of penalty.
NEW RESEARCH QUESTIONS
Sub-project 1 has led to the insight that witnesses’ personality is a key factor to study to understand better how they remember crimes. It is also important to expand the witness research to also include examinations of how intoxicated victims remembers, something that is mentioned as a suggestion for future research in an editorial article by Hagsand et al. (2023b).
Sub-project 2 have given rise to several new research questions about which factors that affect the interaction between police and suspects. For example, a new study has been conducted to examine who is responding to whom during interrogations. In our newest study (Hagsand et al., 2024c), we found partial support for the hypothesis that the police affect the suspects during the interrogations, but we found most support for the hypothesis that it is what the suspects does that affect the police to a greater extent. The later finding can be viewed as a paradigm shift within the research field.
DISTRIBUTION OF RESEARCH FINDINGS
The research results from the present project have been distributed via publications in scientific international peer-reviewed journals with open-access, as well as through international conference presentations, and by collaborations with the Swedish Police Authority. For example, project leader Hagsand evaluated the new national policy guidelines on interrogations with alcohol-intoxicated persons. The research from this project has also been highlighted in media, for example in Swedish Television. Finally, some further work remains in the project with two manuscripts that are anticipated to be submitted to peer-review journals during 2024. In total, it is anticipated that the present project will produce 6 scientific articles.
REFERENCES
Hagsand, A.V., & Schreiber Compo, N. (2024a). The big five traits openness and conscientiousness can affect alcohol-intoxicated eyewitnesses’ memory. Revise and resubmit decision (in June 2024) to Journal of Memory and Learning.
Hagsand, A.V., Halili, A., Nilsson, S., Kelly, C., Evans, J., Schreiber Compo, N., Zajac, H., & Lidell, L. (2024b). How crime type and police gender affect interrogator-suspect interactions in high-stake cases in Sweden: Analyze of recorded police interrogations with suspects. To be submitted to a journal during 2024.
Hagsand, A.V., Jansson, L., Kelly, C., Kapetanovic, S., Evans, J., & Schreiber Compo, N. (2024c). Temporal ordering effects in suspect interrogations, or, who is responding to whom? Examination of two relationships using cross-lagged path model analysis. To be submitted to a journal during 2024.
Hagsand, A. V., Kelly, C., Mindthoff, A., Evans, J. R., Schreiber Compo, N., Karhu, J., & Huntley, R. (2023a). The interrogator-suspect dynamic in custodial interrogations in Sweden: An application of the interrogation taxonomy framework. Scandinavian Journal of Psychology, 64, 352-367. https://doi.org/10.1111/sjop.12889
Hagsand, A. V., Flowe, H., Takarangi, M., & Gawrylowicz, J. (2023b). Editorial: The impact of alcohol and drugs on suspects', victims' and witnesses' cognition and memory. Frontiers in Psychology, 14: https://doi.org/10.3389/fpsyg.2023.1209406
Hagsand, A.V., Zajac, H., Lidell, L., Kelly, C.E., Schreiber Compo, N., & Evans, J. R. (2022). Police- suspect interactions and confession rates are affected by suspects’ alcohol and drug use status in low-stakes crime interrogations. Frontiers in Psychology, 13: 983362. https://doi.org/10.3389/fpsyg.2022.983362
Hagsand, A.V. (2021). Eyewitness testimony: How much alcohol is too much? Research Outreach, 126. https://doi.org/10.32907/RO-126-1847382832