Is the interrogating/interviewing of crime perpetrators no longer suspect? A personal commentary on 25 years of progress

Abstract

This article focuses on some of the main developments regarding the interviewing of suspects during the past 25 years. It contains a personal commentary on: the 25 years ago publication of the 1999 Milne and Bull book; the evolution of the KREATIV approach in Norway; when to disclose information to suspects; a major decision by Wicklander-Zulawski & Associates; the High-Value Detainee Interrogation Group; the Council of Europe's document; the views of jail inmates; the Méndez Principles; the use of cognitive empathy; the American Psychology-Law Society's consultation document.

At the beginning of the 1990s, I assisted the development by the police in England and Wales of the very innovative PEACE method of investigative interviewing (for a recent mention of this see Bull, 2023), which was produced largely in response to weaknesses found in the late 1980s in the interviewing of suspects. I was then asked by the government to analyse tape-recorded interviews of suspects in complex cases conducted by those first trained in this new method (experienced interviewers) (Bull and Cherryman, 1995). Around the same time Becky Milne (assisted by myself and Gunter Koehken as supervisors) commenced work on her PhD (on the cognitive interview) that was awarded in the mid-1990s. One major result of Becky and I working together was the idea for us to write a book overviewing this new concept of investigative interviewing that was published 25 years prior to the current Special Issue of the International Journal of Police Science & Management (IJPSM).

The 1999 publication of the Milne and Bull book

It is a coincidence that this Special Issue of the IJPSM marking its 25 years takes us back to 1999 – the year in which also appeared the book Investigative Interviewing: Psychology and Practice (Milne and Bull, 1999) that some have said was important [e.g. ‘Interesting writeup of where police interviewing usually goes wrong (and is trained wrong) and how it can go correctly…’; see https://www.goodreads.com/en/book/show/513860]. On its back cover this book noted that,

Well-informed and skilled interviewing is a key factor in dealing with suspects, victims and witnesses … there is now a substantial body of research and theory in psychology which … can guide both training and the development of investigative work. The purpose … is to provide a concise and clearly written guide to the psychological concepts and research-based knowledge that can support and guide investigative interviewing…This book will be of interest and value to a wide range of professionals involved in training and practice in the police or other agencies, as well as social workers, lawyers, psychologists and psychiatrists involved in forensic work.

In the years since the book's publication, there have been large increases in the number of relevant research publications and major changes in many countries regarding a move from coercive interrogations to ethical interviewing. The first country (after England and Wales in 1992) to make such a major move was Norway.

‘KREATIV’ in Norway (2001)

In 2019 (in our co-authored publication) a very experienced Norwegian police officer (Asbjorn Rachlew) stated that,

… it should be pointed out that in the early 1990s, new recruits to the police service in Norway were taught to believe that the job was to get suspects to confess. Shortly thereafter in 1997, officers in a high-profile murder investigation for the Norwegian Police obtained a confession that was later retracted. This became a seminal case that shocked the justice system because it was subsequently demonstrated that this admission was false – caused by manipulative, confession-oriented policing techniques. The case exposed the fact that Norway had little formal expertise and aptitude in forensic psychology and that this gap in knowledge was having the adverse effect of pursuing innocent people for crimes while allowing the true culprits to remain at large. (Bull and Rachlew, 2019: 182)

Partly as a result of this, in 2001 the Norwegian Police introduced their KREATIV method in Norway (an updated version of the PEACE method). Essentially, this required a change of mindset in investigators and this method operationalises the ‘presumption of innocence’. The primary role of an investigator is to gather as much accurate and reliable information as possible rather than merely to gain a confession. Rachlew also stated that ‘In Norway, this change in policy was a rather painful process for some, but the general feedback from investigators has been very positive’ (Bull and Rachlew, 2019: 184). In recent years the Norwegian Police and the Norwegian Centre for Human Rights have been disseminating their KREATIV method to many countries (e.g. Fiji, Indonesia, Thailand, Vietnam).

When to disclose information to suspects?

In many countries, interviewers are required to disclose to suspects some of the information that justifies them being suspects. However, especially here in the United Kingdom (UK), not every piece of information (that should have been) amassed during the proper ‘planning and preparation’ phase (e.g. of the PEACE method) has to be disclosed at the outset. Indeed, one of the seven fundamental principles of the PEACE method (Milne and Bull, 1999) is that ‘Information obtained from the person who is being interviewed should always be tested against what the investigator already knows or what can reasonably be established’. In 2003, Stavroula Soukara (during her PhD with me) drew my attention to her newly collected data arising from a special sample of (recorded) police interviews with suspects in which the suspects had actually ‘shifted’ from saying nothing (or ‘no comment’) to admitting to the crime. She had noticed that in the minutes leading up to the ‘shifting’ the most frequently demonstrated interviewing skills were: (a) the continued asking of open questions (that is part of the training of the PEACE method); and (b) the still disclosing of relevant information (that is part of the ‘advanced’ training of the PEACE method) (Bull and Soukara, 2010). Much of this ‘shifting’ only took place after many minutes/some hours into the interviews. She and I immediately realised that if the interviewers were disclosing (new) information long after the interview had commenced, this meant that they had not disclosed it all at the beginning (which is what untrained interviewers usually do) (Bull, 2014). This led me to try to obtain funding to examine this phenomenon in greater depth. After failing for some years to obtain funding, I finally managed to do this from my government's greater focus on supporting research relating to terrorism. In three related studies involving experienced police interviewers, we found that such ‘gradual’ disclosure was more effective in aiding decisions about ‘truth-teller versus liar’ than the (typical) early disclosure or very late disclosure (Dando and Bull, 2011; Dando et al., 2015; Sandham et al., 2022).

In their pioneering study, Hartwig et al. (2005) found that early disclosure created a significantly lower truths/lies detection rate than did disclosure later in the interview. Since then, Professor Par Anders Granhag has led an inspiring team of quality researchers who have extensively studied the many effects of different timings/types of evidence/information disclosure by interviewers (Granhag et al., 2020).

In Samantha Leahy-Harland's 2017 analyses of tape-recorded interviews with 56 suspects regarding very serious crimes (e.g. murder, rape), who were interviewed (as was usual) by well-trained (‘level 3’) police in England, we examined the strategies used by interviewers and the responses made by the suspects. By far the most frequently demonstrated interviewers’ skills were the gradual disclosing of information and the contrasting of some of this with what the suspect had so far said (such skills were part of ‘level 3’, advanced training in the UK). We examined whether any skills were associated with suspects responding relevantly and found positive associations for interviewers’ use of ‘presentation of evidence’, ‘rapport/empathy’ and ‘requesting attention’ (Leahy-Harland and Bull, 2017). However, we found negative associations for ‘explicitly asking to tell the truth’, ‘emphasising the seriousness of the offence’ and ‘situational futility’ (tactics that used to be/still are trained in those countries that employ what was known as the Reid method).

‘Wicklander-Zulawski & Associates’

The organisation Wicklander-Zulawski & Associates (WZ, that used to train the Reid method) was set up in 1982 by its co-founders Doug Wicklander and Dave Zulawski (see their 1993 book). On its website (at the time of writing this article) it was stated that ‘Get the Skills That Get the Truth. Learn techniques that get results in the workplace and equip your team for success. From Government to Private Sector, WZ is the standard in professional training', and that ‘WZ facilitates over 450 programs annually for clients in human resources, loss prevention, executive management, compliance, law enforcement and government agencies’.

In 2017, WZ issued a press release announcing that it would discontinue teaching the Reid method of interrogation after having taught it for many years. This press release stated that ‘WZ has been licensed by John E. Reid and Associates, Inc., originator and developer of the Reid Method, and had included this direct, positive interrogation method in their standard public sector curriculum with WZ's non-confrontational techniques for the last 33 years’. and that ‘It's human nature to deny and defend oneself. Confrontation is not an effective way of getting truthful information’, said WZ President and CEO Shane Sturman, CFI. ‘Rather than primarily seeking a confession, it's an important goal for investigators to find the truth ethically through a respectful, non-confrontational approach’. This press release also stated that.

Because of the possible abuses inherent in the confrontational Reid style, we believe it is time to move away from the practices of the 1970s when it was developed… The Reid Method has remained relatively unchanged since the 1970s, and it conflicts with the progressive nature of how people communicate today. The Reid Method does not reflect updates in our legal system and does not acknowledge the availability of scholarly work on the subject.

One of the ways in which the WZ organisation made itself aware of relevant scholarly work commenced in 2012 at the annual conference of the International Investigative Research Group held in Toronto at which Dave Zulawski approached me and asked if we could chat. During that chat, at Dave Zulawski's request, I quickly overviewed some of the recent scholarly work. In response, I was invited in the next couple of years to attend one of WZ's two-day workshops in the United States (US) and to write a comprehensive overview of it and of their 1993 book entitled Practical Aspects of Interview and Interrogation that focused (as requested) on what within the workshop and book: (a) is in line with scholarly work/published research, (b) is not in line with scholarly work/published research, and (c) is not (as yet) covered by scholarly work/published research. I sent this very extensive overview in 2015 to Dave Zulawski (confirming that to remain independent I wanted no fee), which may well have helped this major 2017 development in the US.

The High-Value Detainee Interrogation Group

Also in the US, the ‘High-Value Detainee Interrogation Group’ (HIG) was established in 2009. The HIG is a three-agency entity involving the Federal Bureau of Investigation, the Central Intelligence Agency and the Department of Defense. The HIG's research programme studies the effectiveness of various approaches and techniques. HIG researchers identify and test existing techniques and develop lawful, new techniques that may be more effective. The HIG identifies research gaps and commissions research to fill those gaps. To carry out the research, the HIG contracts world-renowned, scientists known for their expertise (see https://www.fbi.gov/investigate/terrorism/high-value-detainee-interrogation-group).

Soon after it was set up, the HIG invited me to the US on two occasions to: (a) describe the history behind the 1992 PEACE method, (b) outline the contents of the PEACE method, (c) describe its training, and (d) talk about its effectiveness. After that, I was asked by the HIG to write a comprehensive overview of the published literature on what is believed and what is actually known about the characteristics that may contribute to people being good/effective investigative interviewers (Bull, 2013). In my review, I noted that in a survey of interviewers (from a variety of countries) working on criminal investigations or on ‘intelligence’ issues, it was reported that being patient, showing kindness, respect and concern, and appealing to cooperation were highly rated by these investigators – as were confronting with evidence of guilt and identifying contradictions in the account. Since that time, the HIG has funded more than 100 research projects (for information on these see Brandon and Meissner, 2023; Mindthoff and Meissner, 2023). This large body of research has broadened and deepened an investigative interviewing (non-coercive) approach to dealing with suspects that is having positive effects within and beyond the US including Europe.

The Council of Europe document

The Council of Europe is the continent's major human rights organisation. It comprises 47 Member States, 28 of which are Member States of the European Union. All Council of Europe Member States have signed up to the European Convention on Human Rights, a treaty designed to protect human rights, democracy and the rule of law. In 2018, the Council published a 34-page document entitled ‘A brief introduction to investigative interviewing: A practitioner's guide’ (Boyle and Vullierme, 2018). This guide aimed to give police officers an effective tool, plus the knowledge and skills needed to apply that tool in the course of their investigations. It was informed by scientific research as to what works and by the real-life experiences of expert practitioners. The primary focus of the guide was the questioning of suspects. As did the 1992 PEACE method, it stated seven ‘Principles of investigative interviewing’:

The aim is to obtain accurate, reliable and comprehensive information.

The purpose of the interview is to obtain a full, truthful and accurate account of what has transpired and other pertinent information of which the interviewee has knowledge.

Maintain an open but investigative mind that accepts the innocence of a person until proven guilty.

The interviewer must not allow personal or preconceived views to have influence but must instead maintain an open mind. This will enable a more investigative approach that allows for the questioning of accounts to uncover a complete and reliable version. In so doing the interviewer can test the account given against information already in their possession.

Behave impartially and with respect for the dignity of all.

It is essential that the interviewer behave at all times in a professional manner that is compliant with the human rights of the suspect and is responsive to individuals who may have special needs.

A silent suspect can still be questioned, as you are obliged to investigate.

The interviewer should put questions to a suspect even when that suspect refuses to answer. The suspect may later claim that they were not asked certain questions. This is an opportunity for them to give an account.

You can ask whatever questions you wish provided they are pertinent to the investigation.

You are required to investigate the crime in question and to that end are free to ask any question that is relevant to the resolution of the case.

You do not need to accept the answer given and can persist in questioning provided such questioning does not become oppressive.

The interviewer can probe and challenge accounts given in order to clarify contradictions and inconsistencies in their search for a truthful and accurate account.

Recognise that a truthful confession of guilt has benefits both for the justice system and the victim.

While you must be cautious as to the possibility of a false confession, an admission of guilt can limit anxiety and stress for the victim and bring justice in a more efficient and timely manner than a contested court case.

Admissions of guilt have been a focus of the few available studies involving the views of jail inmates.

The views of jail inmates

Many people believe that crime perpetrators (especially of serious crimes) would not during an interview/interrogation voluntarily provide highly incriminating information and/or admit to their crime (Cleary and Bull, 2019). However, research has found this belief is not well-founded. Obtaining the views of crime perpetrators has rarely been achieved (partly because of access difficulties), although this was pioneeringly achieved by Ulf Holmberg during the time that he was a police officer in Sweden. He analysed the questionnaire responses of men in prison who had been convicted of murder or serious sexual offences. In the questionnaire, 22 items focused on how the interviewers had behaved when interviewing these men, and a further 16 items focused on the men's reactions. Another question asked about the extent to which the prisoners had admitted to (or denied) the crime during the police investigation (around half had admitted). Analysis of what the prisoners indicated regarding how interviewers had interviewed them revealed that some of the interviewers were reported to have behaved largely in a dominating way and others in a humane way. Analysis of the interviewees’ reports of their own reactions revealed that some of them had been obliging, friendly and felt respected, whereas others had been frightened, stressed and felt insulted. Whether denial or admittance occurred was then related to the reported style of interviewing. A relationship was found between the interviewees’ reactions and denial/admittance, in that those who reported being frightened, stressed or insulted were less likely to have admitted (Holmberg and Christianson, 2002).

Similarly, in 2005 O’Connor and Carson (both also highly experienced professional interviewers) found in the US that the predominant reason the confessors (to child molestation) gave for why they confessed was the respect shown to them by the interviewers. In Australia Kebbell et al. (2006) (also see Kebbell et al., 2008) found that only half of the convicted sex offenders with whom they held a research interview said that they had entered the police interview having already decided whether to deny or confess – less than 20% had planned to deny and around 30% had planned to confess; the other 50% entered the police interview not yet having decided whether to deny or confess. In Canada, St-Yves and Deslauriers-Varin (2009) also found that only a minority say that they entered the police interview/interrogation with their mind set on denial and a survey of prison inmates found that evidence-using and humanitarian interviewing techniques were associated with confessions and with suspects’ self-reported cooperation during interviews (Snook et al., 2015).

Most of these prior studies are to an extent limited by small sample sizes and/or specialised participants (e.g. sex offenders). Interviewers may approach these suspects differently or the offenders themselves may differ in their reactions to police interviewing, raising questions about generalizability. Such studies have mostly occurred outside the US, where legal procedures, interviewing strategies and cultural norms may well differ. In a country such as the US (where in society the presence of guns and the use of coercion is more prevalent than in, say, Australia or Canada), we wondered whether the views/intentions of jail inmates would differ from those found in prior research (Cleary and Bull, 2021). We found that 39% had planned to deny, 36% had entered the interview not yet having decided whether to deny or confess, and thus 25% had entered the interview having already decided to confess. Such research has clear implications for how suspects should be interviewed/interrogated that fit with the (1992) PEACE method of investigative interviewing and the more recent 2021 Méndez Principles of effective interviewing (now to be mentioned).

The ‘Méndez Principles’

In light of psychological research, a growing number of countries/organisations have decided to adopt a model/approach of ‘investigative interviewing’ of suspects that does not rely on a coercive or oppressive approach (Bull, 2019, 2014; Bull and Rachlew, 2019). Indeed, in 2016, the United Nations (UN) Special Rapporteur on torture and other cruel, inhumane or degrading treatments (law professor Juan Méndez) submitted his report that was then transmitted by the UN Secretary-General to the UN General Assembly (available at http://antitorture.org/wp-content/uploads/2016/09/Report_A-71-298_English.pdf).

In this report, the summary stated that,

The Special Rapporteur … advocates the development of a universal protocol identifying a set of standards for non-coercive interviewing methods and procedural safeguards that ought, as a matter of law and policy, to be applied at a minimum to all interviews by law enforcement officials, military and intelligence personnel and other bodies with investigative mandates. (p. 2)

When mentioning this universal protocol in 2016 the UN Special Rapporteur noted that,

Encouragingly, some States have moved away from accusatorial, manipulative and confession-driven interviewing models with a view to increasing accurate and reliable information and minimizing the risks of unreliable information and miscarriages of justice. (p. 16)

and that

The essence of an alternative information-gathering model was first captured by the PEACE model of interviewing adopted in 1992 in England and Wales … investigative interviewing can provide positive guidance for the protocol…. (p. 16)

Around the time of the publication of this 2016 report, I was able to accept an invitation to go to a UN meeting in New York to give a presentation on the PEACE method.

Following on from the UN's acceptance of Professor Méndez's proposal that a universal protocol be written, a small international, multidisciplinary steering committee was set up in 2017 to draft such a document. I had the honour of being on this steering committee, which spent three years combining into one document information not only on interviewing/interrogating, but also on laws/safeguards. We were always aware that the finalised document should be shorter than 40 pages, but at one stage during our three years of deliberations, its draft length was more than 100 pages. The agreed, finalised version (of 35 pages) was published in June 2021 (and is available at https://www.wcl.american.edu/impact/initiatives-programs/center/publications/documents/Méndez-principles/).

In its section on effective practice, the protocol states that

Robust research supports the efficacy of an information-gathering approach to interviewing. Rapport-based, non-coercive methods offer effective techniques that can be successfully applied by trained professionals to gather criminal and intelligence information from interviewees. Establishing and maintaining rapport is an adaptive skill that helps create a working relationship between persons and enables better communication. (p. 7)

In its section on ineffective practice, it is stated that

Recent research conducted by professionals and practitioners from a wide range of disciplines provides significant evidence that the application of coercion can both initially enhance resistance on the part of the interviewee and, if continually applied, lead to the provision of false information or a false confession. The application of coercion can backfire by obstructing the collection of factual information. Moreover, coercive techniques interfere with memory retrieval. Research has proven that unreliable information and false confessions arising from abusive practices are a frequent and foreseeable consequence of poor interviewing techniques. They have led to wrongful convictions and faulty intelligence around the world, thus undermining the objectives and effectiveness of law enforcement and intelligence gathering. (p. 5)

On its page 13 it notes that,

An effective interview process will typically involve the following:

a. Thorough preparation and planning

b. Ensuring relevant safeguards are applied throughout

c. Keeping an open mind

d. Creating a non-coercive environment

e. Establishing and maintaining rapport

f. Using lawful and scientifically supported questioning techniques

g. Active listening and enabling the interviewee to speak freely and completely

h. Assessment and analysis of the information gathered and of the interviewing.

Cognitive empathy

In addition, interviewers’ use of ‘cognitive empathy’ (demonstrating a genuine understanding of the circumstances that the suspect is currently in and of the factors that may have caused the suspect to be in the current circumstances, often derived from the planning/preparation phase) and not ‘affective empathy’ (taking on the emotions of the suspect) is recommended in paragraph 30 of the Méndez Principles. Cognitive empathy is, in part, a component of building and maintaining rapport (the latter being especially difficult for interviewers to achieve if suspects have decided to talk about some terrible things they have done). Recently, Bianca Baker-Eck and I gained access to real-life recorded interviews with suspects in serious sex crime investigations. In both of our studies (Baker-Eck and Bull, 2022; Baker-Eck et al., 2021) we found that the amount of ‘investigation relevant information’ provided by the suspects was related to the amount of (cognitive) empathy demonstrated by the interviewers.

The American Psychology-Law Society

At the time of writing this article, the American Psychology-Law Society was disseminating a consultative document entitled ‘Police-Induced Confessions, 2.0: Risk Factors and Recommendations’. This document contains the following sections: The Psychology of Confessions; Relevant Core Principles of Psychology; Scientific Methodologies and Consensus; Racial Bias; Third Degree Practices; Current Law Enforcement Practices in the U.S.; Practices in England and Elsewhere; What Causes False Confessions?; New Science-Based Approaches to Interrogation; Consequences of Confessions; Proposed Remedies (including Video Record All Suspect Interviews and Interrogations; Require Evidence-Based Suspicion to Conduct Interrogation; Impose Limits on the Confrontational Approach; Adopt Science-Based Investigative Interviewing Practices; Protect Youthful and Other Vulnerable Suspect Populations). Although the main focus of this document is on avoiding interrogation-induced false confessions, which are nowadays rare in countries that already follow the Méndez Principles/the PEACE method, it will be of great relevance to countries in which there are still such confessions.

To gain an understanding of the status of investigative interviewing/the adoption of the 2021 Méndez Principles in a large number of countries in which little has changed over the years (thus to better understand how change can best be achieved) Walsh et al. (2024) have compiled an edited volume.

Also, at the time of writing this article, work on the writing of our second edition of the 1999 book Investigative Interviewing: Psychology and Practice was nearing completion. The 1999 version stated that ‘…there is now a substantial body of research and theory in psychology which … can guide both training and the development of investigative work’ and this is even more the case 25 years later. So much so, that this second edition involves not just the two original authors, but also many more co-authors, the topic having substantially broadened and deepened into an edited volume (Milne and Bull, 2024).

A comment for the future

As readers will know, there can be many joys that occur when visiting other countries. For me, one of these relates to the (unexpected) questions that are asked during or just after giving a presentation/lecture/seminar on some of the above topics. Fairly recently, after I gave a lecture at a university in Malaysia a PhD student asked me a question that future researchers (and I) need to act upon – the question was ‘Does it matter that all of the “real-world” research that you have positively talked about (such as the views of jail inmates, the analyses of police interviews) has been conducted only with crime perpetrators who have actually been detected?’ – does that mean that all such research might be suspect?