Maximising Opportunities to Detect Verbal Deception: Training Police Officers to Interview Tactically

Abstract

The purpose of this research was to investigate the efficacy of a novel information gathering technique for detecting truthful and deceptive verbal accounts in interview settings. Five police officers were trained to use each of the three interview techniques, namely tactical, strategic and early. Post-training, each interviewed 30 mock suspects (five truth tellers and five liars in each interview condition) who had taken part in an immersive interactive computer game, competing as either a terrorist (deceiver) or a builder (truth tellers). Post-interview, officers completed a questionnaire designed to collect veracity judgments, confidence levels and the type of interviewee behaviour that had influenced their veracity decision. Results revealed a significant advantage for detecting both deceivers and truth tellers using our new tactical procedure (67% and 74% accuracy, respectively) versus the strategic (54% and 42%) and early (53% and 47%) interviews. Additionally, when interviewing tactically, not only were interviewers more confident in their judgments but they also reported using verbal behaviour more to inform their judgments. We introduce the new tactical approach to interviewing and discuss our findings, suggesting why tactical interviewing is effective for detecting verbal deception. Copyright © 2011 John Wiley & Sons, Ltd.

Full text

Maximising Opportunities to Detect Verbal Deception:
Training Police Officers to Interview Tactically
CORAL J. DANDO
1,
*and RAY BULL
2
1
Lancaster University, Department of Psychology, Fylde College, Lancaster, UK
2
School of Psychology, University of Leicester, Leicester, UK
Abstract
The purpose of this research was to investigate the efficacy of a novel information
gathering technique for detecting truthful and deceptive verbal accounts in interview
settings. Five police officers were trained to use each of the three interview techniques,
namely tactical, strategic and early. Post-training, each interviewed 30 mock suspects
(five truth tellers and five liars in each interview condition) who had taken part in an
immersive interactive computer game, competing as either a terrorist (deceiver) or a
builder (truth tellers). Post-interview, officers completed a questionnaire designed to
collect veracity judgments, confidence levels and the type of interviewee behaviour that
had influenced their veracity decision. Results revealed a significant advantage for
detecting both deceivers and truth tellers using our new tactical procedure (67% and 74%
accuracy, respectively) versus the strategic (54% and 42%) and early (53% and 47%)
interviews. Additionally, when interviewing tactically, not only were interviewers more
confident in their judgments but they also reported using verbal behaviour more to inform
their judgments. We introduce the new tactical approach to interviewing and discuss our
findings, suggesting why tactical interviewing is effective for detecting verbal deception.
Copyright © 2011 John Wiley & Sons, Ltd.
Key words: verbal deception; tactical interviewing; training police officers
INTRODUCTION
When investigating wrongdoing, contradistinguishing liars and truth tellers has long
exercised practitioners and researchers alike. This is particularly so during formal interviews
with those suspected of wrongdoing (from herein referred to as suspects), during which
interviewees are asked to provide an explanation of their involvement in an event and
investigators are tasked with gathering information to assist criminal justice systems to decide
whether an interviewee’s account is veridical.
* Correspondence to: Coral J. Dando, Lancaster University, Department of Psychology, Fylde College, Lancaster
LA1 4YW, UK.
E-mail: c.dando@lancaster.ac.uk
Journal of Investigative Psychology and Offender Profiling
J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jip.145
Copyright © 2011 John Wiley & Sons, Ltd.

The research reported here concerns just such situations, with three objectives. First, to
provide a further empirical evaluation of our ‘tactical’interview procedure, which employs
a drip feed approach to the revelation of information items during an interview to
maximise opportunities to detect verbal deception. Second, to investigate how well, or
otherwise, this technique is applied by professional police interviewers. Finally, to report
interviewers’performance in terms veracity judgments, each having undergone training to
use (i) our tactical technique, (ii) a strategic technique, where information is not revealed
until the end of the interview and (iii) an early technique, where information is revealed at
the beginning of an interview.
Detecting deception
The literature reveals that deciding veracity is a complex task, with most people including
professional lie catchers generally perform at around chance levels (e.g. Ekman, O’Sulivan, &
Frank, 1999; Vrij, 2004, 2008; Vrij & Mann, 2001, but also see Mann, Vrij, & Bull, 2004 for
an example of above chance performance). Although in the UK (and many other countries)
police investigators are not concerned with deciding veracity, they are tasked with maximising
the efficacy of each interview situation to uphold the rule of law and support natural justice.
Accordingly, they must consider how best to manage the flow of information between
interviewer and interviewee during these complex social interactions. Indeed, the current UK
investigative interview model PEACE (an acronym for the stages of an investigative
interview: Planning and preparation; Engage and explain; Account; Closure; Evaluate)
advocates that interviewers plan and prepare for every interview, which includes considering
how to manage ‘evidence that suggests that he/she [suspect] might have committed an
offence’and handle ‘information/evidence that emerges from the interview’(Centrex,
2004, p. 86).
There exists a significant body of empirical research suggesting how cognitive effort
and working memory can influence verbal indicators of deception in interview situations.
In brief, deceivers often provide less detailed accounts and shorter answers than truth
tellers (see Sporer & Schwandt, 2006, for a meta-analysis), behaviours thought to emanate
from the increased demands on working memory associated with constructing, verbalising
and maintaining a deceptive account during an interview (e.g. DePaulo, Lindsay, Malone,
Muhlenbruck, Charlton, & Cooper, 2003; Sporer & Schwandt, 2006; Vrij, 2000, 2008;
Vrij, Mann, & Fisher, 2006; Zuckerman, DePaulo, & Rosenthal, 1981). This is particularly
so when a liar has had little time to prepare, when providing less detail can be a deliberate
strategy (Vrij, 2008). By revealing as little information as possible, liars are able reduce the
likelihood of contradicting themselves and/or the facts known to the interviewer and hence
hope to appear truthful (Bull & Dando, 2010; Granhag, Andersson, Strömwall, & Hartwig,
2004; Hartwig, Granhag, & Strömwall, 2007).
When able to prepare, liars may formulate a lie script, believing that by planning what they
are going to say they can reduce contradictions. Liars are also aware that as an interview
progresses, they may have to create morecomplex lies in order to account for new information
presented and to ensure that new lies are consistent with any previous deception (e.g. Cody,
Marston, & Foster, 1984). By keeping it simple at the outset, deceivers believe that this
introduces opportunities for verbal manoeuvring, thereby allowing ‘flexibility’to add detail
only when requested by the interviewer, and ‘thinking time’should the inclusion of additional
detail and/or explanations of certain behaviour become necessary (Bull & Dando, 2010). It has
also been argued that because deceptive interviewees will usually be unaware of all of the
190 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

information available to interviewers, providing a less detailed account reduces the risk of
contradictions (e.g. Strömwall, Hartwig, & Granhag, 2006; Vrij, 2000).
In contrast, truthful interviewees typically just ‘tell the truth’, believing that as they have
nothing to hide, (e.g. Colwell, Hiscock-Anisman, Memon, Woods, & Michlik, 2006), their
innocence will be apparent (Gilovich, Savitsky, & Medvec, 1998; Lerner, 1980). This has
been found to manifest itself in a more detailed account of the event and the inclusion of
event information that is probably consistent with that known by the investigator.
Taking account of these differences, previous research in this domain has considered how
best to use potentially incriminating information during an interview, advocating the
revelation of this evidence in its entirety at the end of the interview, as follows: ‘interrogations
started with the Introduction step, followed by a free recall...after which the interrogator
posed a number of specific questions...the final specific question concerned whether the
suspect confessed to the crime. After this the evidence against the suspect was presented’
(Hartwig, Granhag, Strömwall, & Vrij, 2005, p. 475; see also Hartwig, 2005 for a more
complete description of the procedure), the intended effect is to maximise the cognitive load
experienced by the suspect and help investigators plan an interview strategy based on
assumptions concerning suspects’strategies (see Granhag & Hartwig, 2008).
Indeed, when asked to judge the veracity of mock suspects, 41 police recruits (trainee
police officers) trained to reveal information as described previously (and who had also
conducted the interviews) were remarkably accurate, obtaining an overall accuracy rate of
85.4% (truth accuracy 85%; lie accuracy 85.7%) compared with an overall accuracy rate of
56.1% (truth accuracy 57.1%; lie accuracy 55%) obtained by police officers who had not
been trained (Hartwig, Granhag, Strömwall, & Kronkvist, 2006).
The current research
This paper extends the information-gathering deception detection literature in a number of
ways. First, we compare the aforementioned ‘late’approach to revealing information with a
novel procedure that is tactical in nature and an ‘early’approach. Our tactical approach
constitutes an important evolution of that currently offered in the literature in that it differs
markedly in its approach to the actual revelation of the information and does not solely concern
itself with potentially incriminating evidence. Second, rather than using police trainees or
researchers as interviewers, we utilised experienced serving police investigators to conduct our
interviews with mock suspects, a necessary progression aimed at answering questions
pertaining to training, practicability and forensic application. Finally, that our mock suspects all
took part in an immersive, interactive computer game designed to elicit complex deceptive
verbal behaviour in a laboratory setting resulted in our interviewers being in the position of
having to contend with large amounts of information. Previous published research has
considered just a small number of items of potentially incriminating evidence (typically 3),
which we would argue falls far short of that encountered by real investigators in complex cases.
To address these concerns, we have previously introduced a tactical interview procedure
(Bull & Dando, 2010; Dando & Bull, 2009, 2010) employing a ‘drip feed, gradual’
approach to revealing information, in which revelation occurs throughout the questioning
phase of an interview rather than at the very end. Thus, it dictates that interviewers
continuously evaluate the means to optimise their objective (to conduct an effective and
ethical interview) and in doing so demands that the revelation of items of information
(whether incriminating or not) is managed both independently and incrementally
throughout the interview. Previous strategic techniques, however, appear to only consider
Tactical interviewing 191
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

how to use potentially incriminating evidence (this being information-concerning
behaviour that has the potential to reveal deception) and moreover only evidence ‘known’
to the investigator, apparently viewing it as a whole, it being presented in bulk at the end
of the interview, (e.g. ‘the evidence was disclosed right at the end of the interrogation’,
Hartwig, 2005, p. 30).
This strategic approach does not consider how best to handle information that may emerge
during the interview or how large amounts of information items might be managed in such a
manner as to assist investigators during an interview. In the UK immediate electronic assistance
(i.e. real time digital streaming of interviews for use whilst an interview is in progress) is not
available to interviewers. Hence, a strategic approach is likely to place considerable cognitive
demands upon the interviewer, in terms of necessitating several concurrent cognitive
operations. Interviews are dynamic situations that evolve quickly, and in the absence of any
immediate assistance, officers have to recall what an interviewee has stated, in both the free
account and questioning phases, and retain this information until the closing stages of the
interview process. By using the strategic approach, it is not until this point that they reveal all
the incriminating evidence (in bulk) and then challenge any discrepancies between (i) the
interviewee’s free account and questioning performance and/or (ii) an interviewee’s account
and the revealed information. In addition, during a strategic interview, investigators are also
required to construct and pose appropriate questions concerning event information/evidence,
prior to revealing that evidence, whilst also being cognizant of the information provided earlier
in order to appropriately and productively challenge any discrepancies.
The distinction between strategic and tactical cognition was introduced by Miller,
Galanter and Pribram (1960): ‘The molar units in the organization of behaviour will be
said to comprise the behavioural strategy, and the molecular units, the tactics’(p. 17). In
the context of suspect interviewing, the molar units of an interview protocol constitute a
strategic decision to withhold information for as long as possible. The molecular units
constitute tactical decisions as to the precise time and order in which to optimise the
revelation of individual pieces of evidence to the suspect. Our tactical approach inherits the
strategic objective laid down as previously described, but it introduces a tactical layer of
incremental evidence utilisation. It is our contention that this maximises opportunities to
detect verbal deception by exploiting gaps in a liar’s scripts, at the same time providing
innocent interviewees early opportunity to convey their honesty.
To date, the mock suspect paradigm used to test suspect interviewing approaches has
involved short crime scenarios, during which the event behaviours of the deceptive and
truth telling mock suspects are usually experimentally matched apart from minor crime
actions and participants’behaviour is controlled by the experimenter in terms of being
instructed how to behave or what to say (Hartwig et al., 2006, 2007; Hartwig, Granhag,
Stromwall, & Vrij, 2004). In most investigative settings, police and other agencies
encounter complex events (e.g. multiple suspects, different patterns of deceptive and
innocent behaviour, embedded lies and large quantities of information). Thus, a new mock
suspect paradigm was required that is sensitive to the demands of complex deceptive
behaviours and multiple items of information/evidence.
Accordingly, we used an approach in which participants created their own deceptions,
rather than maintaining deceptive statements and actions presented by an experimenter. In
brief, interviewees (hereafter referred to as mock suspects)first play an interactive virtual
computer game called Dodgy Builders Ltd, in groups of four. Each player competed
individually as either a ‘builder’or a ‘terrorist’and was striving to complete their role-
specific task before the other three players did so. The builder’s task was to build part of an
192 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

Olympic stadium, whereas the terrorists’was to blow up the Olympic stadium. Terrorists
were provided with a brief outline of the builders’global task in order to allow them to
devise ways of appearing to be a builder, thus masking their true identity. The first player
to complete his/her task wins the game.
The current study provides a test of the proposition that tactical interviewing increases
overall accuracy by enhancing the detection of liars and truth tellers: that limiting a deceptive
interviewee’s verbal options from the very start of the questioning phase may heighten the
investigative value of the available information, as well as signalling to innocent interviewees
the need to account for each item of evidence/information.
METHOD
Participants
One hundred and fifty graduate and post-graduate students, with a mean age of 27.3years
(standard deviation [SD]= 2.69), participated as mock suspects (78 male and 102 female
participants). Interviews were conducted by five experienced police investigators from three
large UK police forces (two women and three men, with a mean length of service of 19.2years,
ranging from 6 to 26years), each of whom were advanced investigative interviewers having
undergone extensive specialist police interview training (ranging from 6months to 9years
prior to their participation in this research). Interviewers underwent 4-day training prior to
participation. In brief, interviewers were initially sent a DVD (featuring example
interviews) and an instruction manual, outlining each of the three interview techniques
(tactical, strategic and early). Interviewers then attended 2-day face-to-face training course
run by the research team, which included numerous practice interviews and extensive
performance feedback.
Design and procedure
The study comprised three phases: (i) participants played an interactive, immersive
computer game as either truth tellers (mock builders) or deceivers (mock terrorists); (ii) 1
hour after the game had finished players were interviewed (individually) about their
gaming by one of the five interviewers; and (iii) interviewers then completed a post-
interview questionnaire, which not only asked them to make a veracity judgement but also
collected qualitative and quantitative information about their interview experience.
Phase one
Upon arrival, groups of four participants self-selected where they wished to sit (the role of
either a ‘builder’or a ‘terrorist’having previously been allocated by the experimenter to each
of the four seats positioned around a large table). Each participant was shown a training video
on their individual laptop computer, which they listened to through headphones. These video
recordings introduced the players to the software and gave instructions on how to operate it,
outlined the game objectives, explained each player’s individual role and the game rules.
Participants were also provided with instruction cards, which could be referenced during the
game. Participants competed against each other to finish their role-specific task (builder or
terrorist), taking turns to traverse the game board using dice throws to determine the number of
squares they could move. Players were free to travel anywhere on the board: to the shops
Tactical interviewing 193
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

(electricians and builders merchant) to purchase materials as necessary, the Olympic site and
their own depots to deliver purchased materials when they saw fit. Games took approximately
1hour. Players received a financial incentive of £24 for this phase.
Phase two
Having completed the game (when any one of the players had successfully completed their
tasks), participants were individually interviewed about their gaming behaviour. Interviews
commenced approximately 1hour after the game had concluded, thereby allowing sufficient
time (i) for participants to consider the post-game/pre-interview instructions and (ii) for the
interviewer to plan and prepare in readiness for the interviews. Prior to interview, all
participants were instructed verbatim ‘Your task is to convince the interviewer that you are a
builder and that you were not involved in any terrorist activity during the game’. No guidance
was given to participants as to how they should carry out the aforementioned instructions in
terms of how they might act, what strategies they might use or what they might say. Players
received an additional financial incentive of £8 for this phase of the research.
All interviews were digitally video and audio recorded. Each interviewer conducted a
total of 30 (counterbalanced for both interview condition and group) interviews over a 1-
week period, 10 from each interview condition (tactical, strategic and early), with five
participants from each group (deceiver and truth teller). Interviewers were blind to the
experimental hypotheses and were given no base rate information about the number of
deceptive or truthful interviewees they would encounter.
Interview conditions
Interviews in each of the three conditions comprised the same number of phases in the same
order, differing only when (during which phase) and how information about participant’s
gaming (known to the investigator) was presented and where appropriately challenged.
Control (early disclosure of evidence)
An early interview condition was included because prior research (Milne & Bull, 1999)
had revealed that many investigators often reveal all of the incriminating evidence near the
beginning of an interview (e.g. in the hope that a suspect will confess). Interviews
commenced with the introduction and explain phase (contact first author for full interview
protocols) and moved (seamlessly) through each of the phases, as follows. In the free
recall phase, the interviewer first disclosed all of the available ‘gaming’information,
listing them one piece at a time. The interviewee was instructed not to respond at this point
but was instead asked (using an open-ended invitation—see Oxburgh, Myklebust, &
Grant, 2010 for more on question types) subsequently to provide a free recall account of
his/her gaming behaviour in as much detail as possible (uninterrupted by the interviewer).
Having provided a free recall account, the interviewer explained that some questions
concerning each of the information items presented earlier in the interview would be
asked. Each item/behaviour was presented (one after another), following which the
interviewee was then asked to explain/account for each. Where appropriate, their replies/
explanations were either (i) challenged (i.e. any discrepancies were pointed out and an
explanation was invited) in order to clarify their version of events or (ii) accepted as being
consistent with the information known to the investigator. Interviews concluded with the
closure phase.
194 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

Strategic (late disclosure of information)
The introduction and explain phases of the interviews in the strategic condition are as
previously described. However, in this condition, interviewees were asked to provide a
freely recalled account of their game playing (the game information was not presented at
this point). The investigator then commenced the questioning phase by asking one
question concerning each of the information items (one after another) without revealing
the nature of that information. For example, the builder’s task dictated that participants
might follow a certain buying pattern: that because of the nature of the task demands, it
appears sensible to visit the electrical shop first. If the interviewee had not visited the
electrical shop, the investigator might ask the question, ‘Which shop did you visit first?’,
thereby questioning the interviewee concerning known information without revealing what
that information was or that it was known by the interviewer. The interviewer then
provided the participant with an opportunity to add/alter anything he/she has stated. At this
point (towards the end of the questioning phase), the interviewer disclosed all of the
information known to him/her, together in bulk. The interviewee was then invited to
explain/account for each. Where appropriate, the participant’s account was then
challenged, and the interviewee was invited to explain all discrepancies/contradictions.
The closure phase concluded the interview.
Tactical (gradual disclosure of information)
The introduction, explain and free recall phases of the interviews in the tactical condition
were as previously described in the strategic condition. The interviewer then commenced
the questioning phase by explaining that some questions would now be asked about the
information he/she knew concerning the manner in which the game had been played, and
in addition, some of the information offered by the interviewee in the previous free recall
phase (which may not have been known to the interviewer prior to the interview) might
also be questioned. The information was then revealed one piece at a time. Each individual
revelation was followed by an invitation to the interviewee to account for that particular
information. If appropriate (i.e. when the account provided by the interviewee contradicted
what the investigator knew or what the interviewee had previously stated in the free
account), the interviewer challenged separately the participant’s explanation/account. Each
piece of evidence was similarly presented, incrementally, piece-by-piece and challenged/
accepted accordingly until the interviewee had addressed each in turn. The closure phase
concluded the interview.
Information selection
Prior to interview, investigators were presented with a printout that documented a limited
amount of game information (we refer to these as case files). Games were sub-divided into a
number of separate phases, (every fourth dice role signalled the end of a phase and the
commencement of a new one), at which point each participant’s game-playing information
was recorded by the game administrator. The game phase information, included in the case
files, was strictly limited to the following: (i) where an individual player had been during each
of the game phases; for example, it was listed the places that he/she may have visited on the
virtual game board (i.e. shops, building sites, builder’s depots, etc.); (ii) the shop stock sold
within that phase, allowing the investigator to calculate how many items had been purchased
Tactical interviewing 195
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

from each of the shops during that phase, although not by whom; (iii) the weight of any one
participant’s virtual van, selected to be weighed during that phase by the administrator and (iv)
a subset of information pertaining to what was being carried in two of the four player’svirtual
van, where that player themselves selected two items to reveal during a van inspection. In
preparation for each interviews, the investigator considered the information presented in the
case files, making interview planning notes where appropriate, according to condition.
Following each interview, investigators immediately completed a questionnaire (see
materials), were paid for their participation and then debriefed about the aims of the project.
MATERIALS
The investigator (interviewer) post-interview questionnaire collected quantitative data and
comprised one dichotomous veracity question and four Likert style questions inviting
interviewers to provide fixed answers on a scale ranging from 1 (e.g. definitely not telling
the truth/not at all confident/what the interviewee said/not difficult etc.) to 7 (e.g. definitely
telling the truth/very confident/how the interviewer behaved/extremely difficult).
RESULTS
Interview training check
To examine interviewers’implementation of their training across the three conditions (tactical,
strategic and early), performance was rated using a scale (ranging from 1 to 5, where
1=revealed information according to condition and training/managed responses
according to condition and training; 5=did not reveal information/manage responses
according to condition and training). The scale was developed to assess the manner in
which the gaming information was used (i.e. when it was disclosed and whether it was
disclosed appropriately according to condition: either early in the interview, individually and
incrementally throughout the questioning phase or inbulk at the end of the questioning phase),
and how participants’replies to questions pertaining to that information was eitheraccepted or
challenged (i.e. whether replies managed according to interview condition: immediately
accepted/challenged individually and incrementally or not accepted/challenged until the end of
the interview, in bulk). From herein, for the purposes of this analysis, ‘use’of gaming
information is referred to as information revelation,and‘how’participant’s replies were
accepted or challenged is referred to as response management (the scoring rubricand scales are
available from the first author).
Two independent raters, who were naive to the experimental hypotheses, scored each
interview for information revelation and response management, using the aforemen-
tioned rating scale. Analysis revealed a substantial level of agreement between raters,
Kappa=0.81, p=0.003. Overall mean scores as a function of interviewer, for both the
aforementioned measures, indicated that all had applied their training satisfactorily (see
Table 1).
To further examine interviewers’performance, a mixed analysis of variance (ANOVA),
with interviewer as the between subjects factor (interviewers 1, 2, 3, 4 and 5) and interview
condition (tactical, strategic and early) and veracity (deceiver and truth teller) as the within
subjects factors, was conducted on ratings for (i) information revelation and (ii) response
196 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

management. A significant main effect of information revelation emerged for veracity,
F(1, 20) = 7.624, p=0.012, 
2
=0.16 (M
revelation deceivers
=1.73; M
revelation truth tellers
=1.97). As
expected, there was no significant main effect of interview condition. There were no significant
interview/veracity/interviewer interactions (all Fs<0.810, all ps>0.455). Hence, irrespective
of interview condition and/or interviewer, the manner in which information was revealed
during interviews with deceivers was less in accordance with training than during interviews
with truth tellers.
1
Therewerenosignificant main effects or interview/veracity/interviewer
interactions for response management (all Fs<5.51, all ps>0.089).
Information items
The number of information items in each case file represents individual mock suspects’
game playing behaviour; therefore, the number of information items was dictated by the
manner in which each participant played the virtual game. Nonetheless, it may have been
that interviewers’performance was influenced by the number of information items they
used from the case file during each interview: it being sensible to assume that the fewer the
information items interviewers were asked to manage, the less cognitively demanding their
task and/or that if participants across the groups had played the game differently (in terms
of number of movements, etc.), they might reveal their group membership (truth teller and
deceiver) by virtue of marked differences in information items.
Analysis of the number of information items used during interviews as a function of (i)
interview (M
tactical info items
=11.42, SD=2.07; M
strategic info items
=10.74, SD=2.15; M
control info
items
=10.54, SD=2.24), (ii) interviewer (M
interviewer 1 items
=11.03, SD=2.76; M
interviewer 2 items
=
10.89, SD=2.04; M
interviewer 3 items
=10.09, SD=1.96; M
interviewer 4 items
=11.24, SD=1.71;
M
interviewer 5 items
=10.80, SD=2.76) and (iii) veracity (M
deceiver items
=11.64, SD=1.59; M
truth
teller items
=10.01, SD=1.82) revealed no significant main effects or interactions (all Fs<0.920,
all ps>0.439).
Veracity analysis
Using the dichotomous truth/lie judgement, when conducting interviews using the new
tactical technique, interviewers obtained an overall accuracy rate of 67% for deceivers and
74% for truth tellers. Having conducted interviews using the strategic and early
Table 1. Interviewers’mean overall information revelation and
response management ratings
Information revelation Response management
Mean (SD)
Interviewer 1 1.90 (0.712) 1.43 (0.679)
Interviewer 2 1.80 (0.801) 1.50 (0.679)
Interviewer 3 2.01 (0.788) 1.40 (0.874)
Interviewer 4 1.98 (0.812) 1.67 (0.711)
Interviewer 5 1.67 (0.768) 1.89 (0.788)
1
Despite a significant main effect of veracity, all interviewers conducted interviews across each of the conditions
and as a function of group in line with the training, as indicated by overall ratings <2.01 on the training scale.
Hence, veracity was excluded as a covariate.
Tactical interviewing 197
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

techniques, interviewers obtained an accuracy rate of 54% and 53% respectively for
deceivers and 42% and 47% respectively for truth tellers. To fully analyse interviewer
performance, a series of mixed ANOVAs, with interviewer as the between subjects factor
(interviewers 1, 2, 3, 4 and 5) and interview condition (tactical, strategic and early) and
veracity (deceiver and truth teller) as the within subjects factors, were conducted on
interviewers’post-interview questionnaire responses (employing Bonferroni’s correction).
Because of the exploratory nature of this research, significant findings were investigated
employing Games-Howell post hoc tests.
Lie scale
Interviewers were asked to indicate on a lie scale of 1 to 7 whether they believed each mock
suspect was telling the truth or being deceptive during the interview (where 1=definitely
telling the truth and 7 = definitely not telling the truth). A significant main effect of veracity
emerged, F(1, 20)= 15.615, p=0.001, 
2
=0.43, and interviewers rated deceivers as more
deceptive (M
deceivers truth scale
=4.48) than truth tellers (M
truth tellers truth scale
=3.45). There was
also a significant veracity/interview interaction, F(1, 20)=18.87, p<0.001, 
2
=0.43.
Interviewers were more accurate in their ratings for both deceivers (M
tactical deceivers
=5.88)
and truth tellers (M
tactical truth tellers
=2.48) in the tactical interview conditions than in both the
strategic (M
strategic deceivers
=3.40; M
strategic truth tellers
=4.08) and early (M
early deceivers
=4.14;
M
early truth tellers
=3.80) conditions, with no significant difference between the latter two
conditions. There were no significant interview or interviewer main effects or interactions (all
Fs<0.820, all ps>0.139).
Confidence
Interviewers indicated how confident they were that their truth scale ratings were correct, on
Likert style confidence scale (where 1= absolutely confident and 7 = not at all confident). A
significant main effect of interview type emerged, F(1, 20)=15.458, p<0.001, 
2
=0.47;
interviewers were more confident having conducted tactical interviews (M
tactical confidence
=2.08)
than in both the strategic (M
strategic confidence
=3.78) and early (M
early confidence
=4.01) interview
conditions with no difference between the latter two. There was a significant interviewer/veracity
interaction, F(4, 20)= 3.895, p=0.016, 
2
=0.43. Interviewer three was less confident when
rating deceivers (M
deceiver confidence 3
=3.89) versus the other four interviewers (M
deceiver confidence 1
=
2.80; M
deceiver confidence 2
=3.08; M
deceiver confidence 4
=2.40; M
deceiver confidence 5
=2.69), with no
significant differences between the latter four interviewers. No additional significant main
effects or interactions emerged (all Fs<4.358, all ps>0.139).
Behaviour
Interviewers were asked to indicate what type of interviewee behaviour they had used to
inform their lie scale decision, again using a Likert type scale ranging from 1 to 7 (1 = only
verbal behaviour; 4=both verbal and physical behaviour; 7 = only physical behaviour). A
significant main effect of interview emerged, F(2, 20) = 6.595, p= 0.003, 
2
=0.23. When
interviewing tactically, interviewers indicated that they used verbal behaviour more when
making their lie scale decisions (M
behaviour scale
=2.81) compared with when interviewing
strategically (M
behaviour scale
=4.89) and when interviewing using the early approach
(M
behaviour scale
=4.04). There were no further significant main effects or interactions (Fs<
2.018, all ps>0.131).
198 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

Difficulty
Interviewers were asked to indicate how difficult (demanding) they had found it to conduct each
interview on a Likert type scale ranging from one to seven (1=not at all difficult; 7=extremely
difficult).Two significant main effects emerged, that of veracity and interview type, F(1, 20) =
19.615, p=0.007, 
2
=0.33 and F(2, 20)= 7.873, p=0.011, 
2
=0.24, respectively. Interviewers
rated deceivers as more demanding to interview than truth tellers (M
difficulty deceiver
=4.30;
M
difficulty truth teller
=3.10) and rated the tactical and early interviews (M
tactical difficulty
=3.80;M
early
difficulty
=4.20) as less demanding than the strategic (M
strategic difficulty
=6.10),withnodifferences
between the former two conditions. No further significant main effects or interactions emerged
(Fs<3.719, all ps>0.231).
DISCUSSION
This study investigated the efficacy of a new tactical approach to conducting investigative
interviews with those suspected of wrongdoing versus a strategic and an early technique.
Of our three objectives, the first and the third were to provide an empirical evaluation of
our ‘tactical’technique with professional investigators and to report their performance
when deciding whether a mock suspect was being veridical.
Our results provide empirical support for our contention that interviewing tactically,
where the revelation of event information occurs during the questioning phase using a drip
feed gradual approach, could enhance trained professional investigators’performance,
both in terms of increasing their detection of deceitful senders and assisting them to
appreciate when a sender was being truthful.
The dichotomous truth/lie data revealed superior percentage accuracy performance in the
tactical condition for both deceivers and truth tellers versus the strategic and early conditions.
Additionally, the lie scale and confidence data revealed that tactical interviewing not only
enhanced veracity performance per se but also strengthened lie judgments and interviewers’
confidence in these judgments, for both deceivers and truth tellers alike. Conversely, in the
strategic and early conditions, interviewers’strength of veracity judgments for both deceivers
and truth tellers were reduced. Moreover, they were much less confident, and their judgments
apparently not sufficiently polarised to allow them to discriminate reliably between deceivers
and truth tellers (as indicated by the mean lie scale and confidence scale scores that hovered
around the midpoint of the scale). That there were no interviewer interactions for strength of
veracity judgement indicates that the lie scale task was more pronounced for all of our
investigators in these latter two conditions.
An interviewer interaction did emerge when analysing the confidence data. However,
this interviewer was significantly less confident when rating deceivers only, irrespective of
interview condition. This particular police officer was the least experienced investigator
(with just 6years service), who had only completed specialist interview training 6 months
previously. Nonetheless, despite finding the task of judging deceivers more demanding,
this appeared not to affect her veracity performance, suggesting that this lack of confidence
may emanate from less investigative expertise and reduced levels of post-training
interviewing experience versus the other interviewers.
The question arises as to why the current pattern of results may have emerged when
there exists empirical support for the notion that a strategic approach to conducting
interviews is an effective method of enhancing veracity performance (e.g. Hartwig et al.,
Tactical interviewing 199
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

2005, 2006, 2004). We have previously argued that the tactical approach serves to limit a
deceptive interviewee’s verbal options from the very start of the questioning phase.
Consequently, the investigative value of the available information is heightened because of
the manner in which evidence was revealed and challenged incrementally, thereby
disrupting the ability of deceptive interviewees to construct and verbalise a coherent and
unchallenged account of the evidence/information set. Participants interviewed using the
tactical procedure are asked to account for each piece of information/evidence prior to
being alerted to the nature of that evidence and where appropriate are immediately
challenged. Hence, deceivers are unable to remain true to any consciously created lie
scripts (Hines et al., 2010; Porter & Yuille, 1996), becoming encircled in such a manner
that weaknesses/discrepancies in their verbal armature are highlighted. Indeed, previous
research conducted using the aforementioned paradigm, with multiple information/
evidence items, revealed that mock suspects who had devised a lie script prior to being
interviewed were less able to implement that script when interviewed using our tactical
approach versus a strategic and early technique. Furthermore, deceivers reported finding
the tactical technique significantly more cognitively demanding than both the strategic and
early interviews (Dando & Bull, 2010).
Furnishing interviewees with all of the evidence/information at the start affords deceivers
time to consider and then create an account to ‘fit’the evidence, which they are able to present
unchallenged until the closing stages of the interview. The strategic technique differs in that
the evidence is not revealed early and the participants are asked to account for each behaviour
during questioning without the interviewer revealing what the evidence/information is.
Nonetheless, the technique does not advocate challenging an account until very late in the
procedure. We argue that in common with the early condition, interviewees are again able to
create and maintain what might appear to be a coherent account. Moreover, where there exists
large amountsof information (be it incriminatingor otherwise),a strategic approach is likely to
make the interviewers’task more difficult; as was evidenced here, interviewers having rated
the strategic and early techniques as being more demanding.
The incremental nature of our tactical approach, in contrast, severely limits deceptive
interviewee’s opportunities to verbally outmanoeuvre investigators. Therefore, deceptive
interviewees are forced to relinquish control over verbal performance, resulting in
increased cognitive load that we suggest maximises interviewers’deception detection
opportunities in terms of inconsistent and contradictory verbal accounts.
In the current research, each interviewer had to manage an average of 11 items of
information, a significant increase versus that utilised in the previous literature. In contrast
to our tactical technique, the strategic and early approaches are such that the interviewers’
primary task elements are not independent, instead having to process multiple evidence
explanations simultaneously rather than sequentially, demanding high levels of
interactivity that may have served to reduce task performance (see Beckmann, 2010). In
addition, a tactical approach to interviewing signals to innocent interviewees the need to
account for each item of information presented. Thus far, researchers have paid scant
attention to truth teller detection performance. We would argue that this aspect of
managing investigative interviews in forensic settings is of equal importance to that of
detecting deceivers.
Turning to the type of behaviour, interviewers used to make their veracity judgments. Our
data revealed that when interviewing tactically, participants’reported using verbal behaviour
to inform their veracity judgments, whereas in the other interview conditions, interviewers
indicated that their judgments were at least equally informed by physical behaviour. This may
200 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

also account, in part, for reduced performance in the strategic and early conditions, it being
accepted that physical, non-verbal behaviour is not a reliable indicator of deception (see Vrij,
2008). Conversely, there is much to suggest that monitoring verbal behaviour, by which we
mean what the interviewee says throughout the interview, provides some of the more reliable
cues to deception (see Bull, 2011; Vrij, 2008).
Finally, we were interested in whether professional investigators could be trained to
interview tactically, in terms of unlearning old behaviour quickly. Our findings indicate that
each of our investigators did implement all of the newly introduced interview techniques
satisfactorily, after fairly brief training (far less than is currently afforded police officers in the
UK). That stated, all had undergone previous professional interview training, which does
comprise some of the elements we introduced across all of the conditions (also see Vander
Sleen, 2010). Certainly, our protocols were based on the current UK investigative interview
model and were delivered by a researcher who has much experience of training UK police
officers at an advanced level. However, that officers were able to learn and apply our technique
quickly leads us to offer its suitability for all experienced investigators.
As with all research of this nature, there are a number of limitations that should be addressed
in the future. In addition to further independent replications of our findings, a larger sample of
police investigators of varying levels of experience are required as participants to progress
further the suitability of the technique in applied settings. A more representative sample of
mock suspects would allow us to infer the efficacy of our tactical technique across wider
population. As we have argued elsewhere, it is important that we assume able opponents when
conducting research of this nature. Hence, despite the obvious ethical considerations, in our
opinion, it is now important to conduct research using practised deceivers.
In sum, our findings indicate the efficacy of a strategic approach to conducting suspect
interviews. The criminal justice system is poorly served by less than effective interviewing; it
is of little value to anyone. Interviews are complex social interaction, which are pivotal to the
process of upholding the rule of law and perceptions of natural justice. Accordingly, they
should always be conducted with integrity and sound judgement, and it is important that we
provide investigators with the tools to do just that. We would suggest that our tactical approach
is one such tool, which our research clearly indicates has value as one of a repertoire of
techniques suitable for use.
REFERENCES
Beckmann, Jens F. (2010). Taming a beast of burden–On some issues with the conceptualisation and
operationalisation of cognitive load. Learning and Instruction,20, 250–264.
Bull, R. (2011). Interviewing to detect deception. Invited lecture at the Western Australian Police
Academy, Perth, May.
Bull, R., & Dando, C. J. (2010). Detecting verbal deception: Strategy versus tactics. Paper presented
at the Annual American Psychology and Law Society, 18
th
–20
th
March 2010, Vancouver, Canada.
Centrex. (2004). Core investigative doctrine. Cambridgeshire: ACPO.
Cody, M. J., Marston, P. J., & Foster, M. (1984). Deception: Paralinguistic and verbal leakage. In
R. Bostrom (Ed.), Communication Yearbook 8. Beverly Hills, CA: Sage.
Colwell, K., Hiscock-Anisman, C., Memon, A., Woods, D., & Michlik, P. M. (2006). Strategies of
impression management among deceivers and truth-tellers: How liars attempt to convince.
American Journal of Forensic Psychology,24,31–38.
Dando, C. J., & Bull, R. (2009). Interviewing strategically to detect verbal deception. Paper
presented at the 19th Annual Conference of the European Association of Psychology and
Law, 2–5 September 2009, Sorrento.
Tactical interviewing 201
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip

Dando, C. J., & Bull, R. (2010). Detecting verbal deception: Strategy versus tactics. Paper presented
at the Annual American Psychology and Law Society, 18–20 March 2010, Vancouver, Canada.
DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003).
Cues to deception. Psychological Bulletin,129,74–118.
Ekman, P., O’Sulivan, M., & Frank, G. M. (1999). A few can catch a liar. Psychological Science,10,
263–266.
Gilovich, T., Savitsky, K., & Medvec, V. H. (1998). The illusion of transparency: Biased assessments of
others’ability to read our emotional states. Journal of Personality and Social Psychology,75,332–346.
Granhag, P. A., & Hartwig, M. (2008). A new theoretical perspective on deception detection: On the
psychology of instrumental mind reading. Psychology, Crime, & Law,14, 189–200.
Granhag, P. A., Andersson, L. O., Strömwall, L. A., & Hartwig, M. (2004). Imprisoned knowledge:
Criminal’s beliefs about deception. Legal and Criminological Psychology,9, 103–119.
Hartwig, M. (2005). Interrogation to detect deception and truth. PhD Thesis, University of
Gothenburg, Department of Psychology.
Hartwig, M., Granhag, P. A., Strömwall, L. A., & Vrij, A. (2005). Deception detection via strategic
disclosure of evidence. Law and Human Behavior,29, 469–484.
Hartwig, M., Granhag, P. A., Strömwall, L. A., & Vrij, A. (2004). Police officers’lie detection
accuracy: Interrogating freely versus observing. Police Quarterly,7, 429–456.
Hartwig, M., Granhag, P. A., Strömwall, L. A., & Kronkvist, O. (2006). Strategic use of evidence
during police interrogations: When training to detect deception works. Law and Human Behavior,
30, 603–619.
Hartwig, M., Granhag, P. A., & Strömwall, L. A. (2007). Guilty and innocent suspects’strategies
during interrogations. Psychology, Crime, & Law,13, 213–227.
Hines, A., Colwell, K., Hiscock-Anisman, C. L., Garrett, E., Ansarra, R. L., & Montalvo, L. (2010).
Impression management strategies of deceivers and honest reporters in an investigative interview.
The European Journal of Psychology Applied to legal Context,2,73–90.
Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. New York: Plenum Press.
Mann, S., Vrij, A., & Bull, R. (2004). Detecting true lies: Police officers’ability to detect deceit.
Journal of Applied Psychology,89, 137–149.
Miller, G. A., Galanter, E., & Pribram, K. H. (1960). Plans and the structure of behavior. NY: Holt,
Rinehart & Winston.
Milne, R., & Bull, R.. (1999). Investigative interviewing, psychology and practice. Chichester: Wiley.
Oxburgh, G., Myklebust, T., & Grant, T. (2010). The question of question types in police interviews:
A review of the literature from a psychological and linguistic perspective. The International
Journal of Speech Language and the Law,17,45–66.
Porter, S., & Yuille, J. C. (1996). The language of deceit: An investigation of the verbal cues to
deception in the interrogation context. Law and Human Behaviour,20, 443–459.
Sporer, S. L., & Schwandt, B. (2006). Moderators of nonverbal indicators of deception: A meta
analytic synthesis. Psychology, Public Policy, and Law,13,1–34.
Strömwall, L. A., Hartwig, M., & Granhag, P. A. (2006). To act truthfully: Non-verbal behavior and
strategies during a police interrogation. Psychology, Crime, & Law,12, 207–219.
Vander Sleen, J. (2010). A structured model for the investigative interviewing of suspects. In R. Bull,
T. Valentine, & T. Williamson (Eds.), Handbook of psychology of investigative interviewing.
Chichester: Wiley.
Vrij, A. (2000). Telling and detecting lies as a function of raising the stakes. In C. M. Bruer, M. M.
Crommer, J. F. Nijboer, & J. M. Reintjes (Eds.), New Trends in criminal investigation and
evidence, volume 2. Antwerp: Intersentia.
Vrij, A. (2004). Why professionals fail to catch liars and how they can improve. Legal and
Criminological Psychology,9, 159–181.
Vrij, A. (2008). Detecting deceit and lies: Pitfalls and opportunities. West Sussex: Wiley.
Vrij, A., & Mann, S. (2001). Who killed my relative? Police officers’ability to detect real-life high-
stakes lies. Psychology, Crime, & Law,7, 119–132.
Vrij, A., Mann, S., & Fisher, R. (2006). An empirical test of the behaviour analysis interview. Law
and Human Behavior,30, 329–345.
Zuckerman, M., DePaulo, B. M., & Rosenthal, R. (1981). Verbal and nonverbal communication of
deception. In L. Berkowitz (Ed.), Advances in experimental social psychology. New York:
Academic Press.
202 C. J Dando and R. Bull
Copyright © 2011 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. 8: 189–202 (2011)
DOI: 10.1002/jip