Panel 11: Detecting deception & suspicious behaviour

Chair: Heather Flowe

LT 5&6, Thursday 21st, 0900-1100

Christina Winters

Postdoctoral Researcher | Tilburg University

Do referee-reports increase self-disclosure in vetting interviews?

Session: Panel 11: Detecting deception and suspicious behaviour (LT5)

Pre-employment security vetting often requires candidates to self-disclose risk-relevant information during an interview and provide character references. Recent research has demonstrated that the ‘illusion of knowing it all’ can be effectively applied via interviewer feedback of an interviewee’s self-generated report (i.e., alleged emergent psychographic ‘profiling’), resulting in increased self-disclosure in interviewees who receive this feedback. This study sought to determine whether similar effects can be found for applying the illusion using referee-generated report. Prior to interviewing, all interviewees (N = 76) responded to a web-based pre-screening questionnaire composed of questions analogous to the interview questions. Half the sample were randomly selected to nominate a close person (‘referee’; N = 38) to submit a personality assessment on their behalf. Prior to the virtual interview, the interviewer provided referee personality feedback to the nominator interviewees. Nominator interviewees perceived the interviewer to be more familiar with them (d = 0.53) and more knowledgeable about them (d = 0.70) compared to controls. The manipulation failed to contribute to increased self-disclosure in the nominator group, however nominators endorsed questions with greater inconsistency (d = 0.66) compared to controls, suggesting an effect of the anticipated use of their information. Relationship closeness and type (e.g., parent, partner) did not significantly influence the quantity of self-disclosure for nominators.

Co-authors: Professor Paul J. Taylor, Lancaster University; Dr Kirk Luther, Carelton University


Urszula Matthews

PhD | University of Nottingham

Strategies used to detect suspicious behaviours in public places

Session: Panel 11: Detecting deception and suspicious behaviour (LT5)

The detection of suspicious public behaviour is recognised as an important task in surveillance but, an understanding of the strategies and skills that underpin it remains elusive. This may be remiss, as such information would be critical in supporting the development of training and assistive technologies. We examined the use of strategy and expertise using Cognitive Task Analysis techniques that elicit a fine-grained account of the cues and judgements used to detect behaviour. We contrasted professionals across four security-related contexts with a balanced sample of novices to explore the contributions of expertise to task performance. We found that both novices and experts utilised strategies that shared a recognisable and common structure. However, we observed two distinct approaches in decision making shaped by the level of familiarity with the type of event. When faced with an unfamiliar, novel situation, both experts and novices went through multiple cycles of hypothesis development and testing. However, in situations featuring a familiar context, participants generated singular hypotheses early and referred to those hypotheses throughout the period of observation. This suggests that the analysis of suspicious behaviour is based primarily on understanding of familiar context. We discuss the consequences of this for training and technology development.

Co-authors: Dr Brendan Ryan and Dr Robert Houghton, University of Nottingham


Aldert Vrij

Professor | University of Portsmouth

Actions speak louder than words: The Devil’s Advocate questioning protocol in opinions about protester actions

Session: Panel 11: Detecting deception and suspicious behaviour (LT5)

The Devil’s Advocate protocol has been developed to assist making veracity assessments when someone discusses their opinion (Leal et al., 2010. The present experiment focused on protester actions rather than controversial issues (Leal et al., 2010) and also included an adapted version of the Verifiability Approach.

Participants told the truth or lied about protester actions and the participants’ answers to the eliciting opinion and Devil’s Advocate questions were compared. The Devil’s Advocate approach predicts the difference in answers (residue scores) to be more pronounced in truth tellers than in lie tellers in terms of quantity of the answers (number of words, details and arguments) and quality of the answers (plausibility, immediacy and clarity).

The hypothesis was supported but only in terms of quality: Truth tellers’ answers sounded more plausible and immediate and somewhat clearer than lie tellers’ answers. Truth tellers also reported more digital verifiable sources than lie tellers.

Co-authors: Samantha Mann, Haneen Deeb, Sharon Leal, University of Portsmouth


Aldert Vrij

Professor | University of Portsmouth

Verbal Cues to Deceit when Lying through Omitting Information: Examining the Effect of a Model Statement Interview Protocol

Session: Panel 11: Detecting deception and suspicious behaviour (LT5)

Practitioners frequently inform us that lying through omitting information is relevant to them, yet this topic has been largely ignored by verbal lie detection researchers. In the present experiment participants watched a videotape of a secret meeting between three people. Truth tellers were instructed to recall the meeting truthfully, and lie tellers were instructed to pretend that one particular person was not there. Participants were or were not exposed to a Model Statement during the interview. The dependent variables were ‘total details’ and ‘complications’. Truth tellers reported more complications than lie tellers but lie tellers reported more details than truth tellers. The Model Statement resulted in more complications and details being reported. In terms of self-reported strategies, the main veracity difference was that truth tellers were more inclined to ‘be detailed” than lie tellers. We discuss the atypical finding (most details reported by lie tellers) and ideas for future research.

Co-authors: Sharon Leal, Jennifer Burkhardt, Haneen Deeb, Oliwia Dabrowna and Ronald P Fisher, Florida Internationasl University

Back to top