SIPR Newsletter Sign Up
You will be added to our mailing list to keep you updated with future events and activities from the Scottish Institute for Policing Research
The SIPR Evidence and Investigation Network is pleased to announce their 4th networking event: Applied Psychology in Policing Settings (APPS), taking place at the University of Aberdeen Kings College Conference Centre on Wednesday 16th April 2025, 10am to 4pm.
This event will bring together researchers and practitioners interested in applying psychological research in policing practice. The event will feature:
During these sessions, attendees will have a unique opportunity to network and develop key connections, start new collaborations, or identify new avenues for funding. Each one-hour keynote session will be followed by an hour reserved for discussion. To facilitate discussions, we reserved separate rooms for each speaker, so attendees may approach any or all the speakers during the hour. During the hackathons, practitioners will pitch a current issue, and all attendees will have the opportunity to contribute their ideas aimed at developing a research solution.
For this event, we have identified two core themes:
Novel Markers of Concealed Face Recognition: Emerging Technologies for Future Policing
Dr Ailsa Millen
In policing, suspects might deny recognition of criminal associates during interviews. To address this problem my research develops novel Concealed Information Tests (CITs) for face identification. In this talk, I will present my research on objective markers of concealed face recognition. The talk will highlight the potential of emerging technology for identifying concealed face recognition, and the importance of understanding individual differences in this context. I am looking forward to discussing how these novel approaches could be integrated into policing and investigative practices, and how AI might be used to create and optimise Concealed Information Test line-ups. I will also be happy to discuss my related research on “public perceptions of technology in the criminal justice system” and “advancing human-avatar interactions in policing and defence,” both of which highlight the importance of understanding cognitive and cultural diversity in these contexts.
What makes us look human? The curious case of hyperrealistic AI
Dr Clare Sutherland
In this talk I will discuss the advent of generative artificial intelligence (AI) models of human faces, and what this new technology means for people's ability to tell apart real human and ‘deepfake’ AI face images. I’ll discuss our recent international work showing that people now cannot tell the difference between real and AI faces, and that in some cases, AI faces look more realistic than do real human faces - a phenomenon we term ‘AI hyperrealism'. I’ll also draw on our investigations into the visual cues that distinguish real and AI faces, discuss which aspects of human perception are accurate (and which are inaccurate), and outline individual differences in how accurately different people are able to identify AI faces. I welcome an open discussion about the societal impacts of this technology and how we can collectively act to ensure that AI technology is used for the benefit of the public, minimising harm.
Exploring the Practice of Interviewing Witnesses in Cases of Domestic Abuse
Dr Eva Rubinova
Tackling domestic abuse, gender-based violence, and violence against women and girls is one of the main priorities for the Scottish Government (2018) and the Scottish Police Authority (2024). Within this multi-faceted issue, the present project takes the perspective of investigative interviewing to better understand how evidence in these cases is collected.
Domestic abuse is frequently reported only after multiple occurrences of abuse, sometimes after long delays. Research has shown that individuals (children and adults) remember recurring/repeated events differently than single/unique experiences. For example, when reporting on instances of repeated events, individuals may focus on reporting “what usually” happens without providing details of instances, and if they do report on instances, they frequently misattribute what happened when.
Emerging research indicates that some interviewing strategies may help individuals provide more detailed and accurate reports about repeated events, though this research typically lacks connection to real-world practices. The present project aims to collect information about practices currently used by Police Scotland officers when collecting statements from witnesses of domestic abuse. Knowledge of these practices will inform our future research focused on developing interviewing interventions that may facilitate reporting in cases of domestic abuse.
Enhancing Support for Vulnerable Individuals in Investigative Interviews: The Impact of Alcohol and Video-Recorded Testimonies on Justice Outcomes
Dr Julie Gawrylowicz
This presentation addresses key issues related to supporting vulnerable individuals in the criminal justice system, with a focus on investigative interviews. It will highlight research on the effects of alcohol on memory recall, examining the implications for interviewing intoxicated witnesses and victims. Additionally, the presentation will introduce a new line of research involving Video-Recorded Interviews (VRIs) with victims, exploring how the quality of these recordings influences jurors' perceptions of credibility. As technological advancements make VRIs increasingly common across Scotland, understanding their impact on jury decision-making is crucial to ensuring fair trials and achieving the best outcomes for victims.
Using virtual reality to study and improve eyewitness testimony
Dr Rui Paulo
Witnesses play a critical role in criminal investigations, but cognitive processes such as memory often influence their ability to recall important details. To address this, psychologists developed science-based interview techniques that improve witness memory and statements. Our team also created new retrieval strategies (e.g., Category Clustering Recall) that the police is already using to interview witnesses, victims, and informers in cases of kidnap, arson, extortion, fraud, and trafficking.
However, much of the research supporting the use of these techniques employed a mock-witness paradigm, where participants observed a crime on a 2D screen. This does not accurately reflect real-life eyewitness experiences, raising concerns about the effectiveness of these methods in real investigations.
With the rise of immersive virtual reality (VR), this project evaluated the effectiveness of different interviewing strategies recommended as best practice in a more realistic setting using a 360-degree mock-crime presented on a VR headset. Further, we tested for the first time whether virtual interview rooms that the police could access using existing technology and incorporate contextual information from the crime scene can be used to enhance eyewitness testimony further.
Improving eyewitness identifications with confidence
Dr Travis Seale-Carlisle
Outline: Many believe eyewitness memory is unreliable, and for good reason. Research shows memory is highly malleable, to the point where people can confidently recall traumatic events that never happened. Moreover, mistaken eyewitness identifications—often made with high confidence in court—have contributed to over 70% of the 358 wrongful convictions overturned by DNA evidence since 1989.
However, recent studies show that eyewitness confidence is a strong indicator of accuracy when measured during an initial, uncontaminated lineup. When these conditions are met confidence and accuracy are strongly related such that low confidence indicates low accuracy while high confidence indicates high accuracy. Crucially, in every DNA exoneration case where initial confidence was documented, the witness initially expressed low confidence, highlighting that their identification was error-prone and unreliable.
There is now widespread agreement among researchers on the strong link between initial confidence and identification accuracy. Many countries have updated their policies to require documentation of eyewitness confidence from the initial identification. Despite its rigorous lineup procedures, the UK does not mandate collection of confidence statements. Implementing this simple reform would improve the reliability of eyewitness identification evidence and help prevent wrongful convictions. The time has come for the UK to follow suit.
You will be added to our mailing list to keep you updated with future events and activities from the Scottish Institute for Policing Research