Suspect Interviewing
: A U.S. Perspective

  • Sean Patrick O'Callaghan

Student thesis: Doctoral Thesis


Interviewing a suspect is a crucial aspect of any criminal investigation. An interviewee might lie or tell the truth, be innocent or guilty, but in either case it is up to the investigator to decipher what happened and who is ultimately responsible. Despite its importance, little is still known about how law enforcement interview within the United States. Though all aspects of investigative interviewing are examined in this thesis, it is primarily concerned with the interviewing of criminal suspects. As a result, the research in this thesis began by examining the dialectical relationship between interview law and practice, focusing specifically on commonalities and contrasts among
different law jurisdictions.
The discussion within this thesis will suggest that this legal divergence must be understood in the context of how current interviewer practices are shaped. Four major areas of focus were examined: 1) What are the current views and social norms with regards to interviewing? 2) How do detectives utilize techniques
when questioning a suspect? 3) How do detectives question suspects? and 4) What effect do different question types and interview techniques have on an interviewee’s response? To answer these research questions, this thesis attempts to highlight the effectiveness of American law enforcement interviewing methods
through a series of empirical studies.
The first chapter of this thesis opens by providing a brief history of police interviews with suspects, and includes a discussion of the literature on the subject of false confessions. It then goes on to discuss the many differences between the “Reid Technique” commonly taught in the United States and the British PEACE model of “investigative interviewing.” This policy divergence must be understood in the context of differing responses to miscarriages of justice and investigative failures, as well as the research surrounding it. Drawing on the experience in the United Kingdom, where audio and video recording has been used routinely, this thesis displays the benefits of electronic recording, which all
too often is missing within the United States.
The second chapter of this thesis examines the various legal differences that exist within the United States and other jurisdictions, focusing specifically on the United Kingdom and France. This policy divergence must be understood in the context of differing responses to miscarriages of justice and legal decisions.
The first empirical study examined the current police opinion with regard to specific rules and procedures governing suspect interviewing. In almost every situation, investigators’ selfreported perceptions differed when considering the interviewing of witnesses and victims compared to suspects.
Study two analyzed sixty real-life interviews with suspects obtained from eight states and sought to see how American law enforcement utilizes both questioning and technique methods. In assessing the number of techniques and questions used by investigators as a function of time, each interview was found to vary depending on the type of crime under investigation, the type of interviewee, and the state in which the interviewer practiced. The results suggested that the majority of the interviews in the sample contained more unproductive techniques and questions. Implications of how suspects respond during these types of police interviews are then explored. These results are presented using a modified version of the Griffiths Question Map (GQM) and Technique Timeline, providing a visual representation of each interview in its entirety.
The final Chapter provides an overview of the findings, outlines the limitations, and discusses the possibility for future research. Results were discussed for their consistency with prior research, policy implications, and methodological shortcomings.
Date of Award3 Aug 2020
Original languageEnglish
Awarding Institution
  • University of Portsmouth
SupervisorBecky Milne (Supervisor)

Cite this