Page 1 of 8

Advances in Social Sciences Research Journal – Vol. 10, No. 12

Publication Date: December 25, 2023

DOI:10.14738/assrj.1012.16027

Vrij, A., Leal, S., & Fisher, R. P. (2023). Interviewing to Detect Lies About Opinions: The Devil’s Advocate Approach. Advances in

Social Sciences Research Journal, 10(12). 245-252.

Services for Science and Education – United Kingdom

Interviewing to Detect Lies About Opinions: The Devil’s Advocate

Approach

Aldert Vrij

Department of Psychology,

University of Portsmouth,

PO1 2DY, Hants, United Kingdom

Sharon Leal

Department of Psychology,

University of Portsmouth (UK)

Ronald P. Fisher

Department of Psychology,

Florida International University

Writing this article was funded by the Centre for Research and Evidence on Security Threats

(ESRC Award: ES/N009614/1)

INTERVIEWING TO DETECT LIES ABOUT OPINIONS: THE DEVIL’S ADVOCATE APPROACH

The verbal cues lie tellers spontaneously report during interviews are weak and unreliable

(DePaulo et al., 20023; Vrij, Hartwig, & Granhag, 2019). Researchers therefore asked for

interview protocols to be developed that would elicit such verbal cues (Vrij & Granhag, 2012).

Ten years later several interview protocols have emerged (Vrij, Granhag et al., 2022), including

Cognitive Credibility Assessment (Vrij, Fisher, & Blank, 2017; Vrij, Mann et al., 2021), Reality

Interviewing (Bogaard et el., 2019; Colwell et al., 2007), the Strategic Use of Evidence (Granhag

& Harwig, 2015; Hartwig et al., 2014) and the Verifiability Approach (Nahari, 2019; Palena et

al., 2021). All these interview protocols focus on distinguishing between truthful and deceptive

statements about alleged activities. However, practitioners are also interested in detecting lies

when people discuss their opinions. The Devil’s Advocate Approach is the only interview

protocol we are aware of aimed to distinguish between truthful and deceptive opinions. In this

article we present the (scarce) Devil’s Advocate Approach research conducted to date.

THE IMPORTANCE OF DETECTING LIES ABOUT OPINIONS

Being able to distinguish between truths and lies about opinions can be important in various

security settings, for example when deciding whether (i) an informant is really against the

network of people s/he is willing to spy upon; (ii) an applicant is really committed to the aims

of the intelligence organisation s/he wants to join; (iii) the sole reason for entering the country

is indeed to study at a university; (iv) the person convicted of terrorism has really given up

terrorism and therefore can be released from prison; (v) the activist is indeed against violent

protests; (vi) and the person is really against corruption. Incorrect veracity judgements can do

a lot of harm. Take for example Humam Khalil al Balawi, a medical doctor who grew up in

Page 2 of 8

246

Advances in Social Sciences Research Journal (ASSRJ) Vol. 10, Issue 12, December-2023

Services for Science and Education – United Kingdom

Jordan. Balawi was meant to give information about Ayman al Zawahiri, al-Qaida’s number two

at that time. American and Jordanian intelligence services were keen to debrief Balawi at a

secure location in Khost, Afghanistan. At arrival, he blew himself up (30 December 2009) killing

seven CIA agents and one Jordanian officer. Balawi had posted (under a different name)

extreme anti-American views on the internet and the CIA tipped off the Jordanians about this.

Jordan intelligence services wanted to turn Balawi into a double agent to penetrate al-Qaida.

Balawi convinced the Jordanians that he was the asset they were looking for. Jordanian and

American intelligence agencies became increasingly tempted to believe that he had become a

reliable double agent who could lead them to Zawahiri. The possibility that his views remained

unchanged was discounted (Riedel, 2019).

THE DEVIL’S ADVOCATE APPROACH RATIONALE

The Devil’s Advocate interview protocol consists of three questions (Leal et al., 2010; Mann et

al., 2022). The first question is an expression attitude question in which the interviewee

expresses to be in favour or against the opinion-topic, e.g. “Are you in favour or against Covid

vaccinations?” This question is followed by an eliciting-opinion question in which the

interviewee is invited to report the arguments in favour of the view s/he just expressed, e.g.

“OK, you say to be in favour of Covid vaccinations. Please can you tell me all the reasons you can

think of that led you to having this opinion?” This question is followed by the third and final

question, the Devil’s Advocate question, in which the interviewee is invited to report arguments

against the view s/he just expressed. “OK, now try to play the devil’s advocate and imagine that

you are against Covid vaccinations. Please tell me all the reasons you can think of that goes

against Covid vaccinations?

In the Devil’s Advocate interview protocol, the answers to the eliciting-opinion and devil’s

advocate questions are compared. Truth tellers provide honest answers to both the eliciting- opinion question (the arguments that support their opinion) and the devil’s advocate question

(the arguments that go against their opinion). When answering the eliciting-opinion question

they provide arguments they believe in. Although they present arguments in the devil’s

advocate answer they do not believe in, they are telling the truth because they do not pretend

to believe in these answers. Lie tellers try to fool interviewers and have an opinion opposite to

what they express. They lie when answering both the eliciting-opinion and devil’s advocate

questions. In the eliciting-opinion answer they provide arguments they do not believe in whilst

trying to convince the interviewer that they endorse these arguments. When providing the

Devil’s Advocate answer they are invited to give arguments against their alleged opinion, which

is an invitation to express their true opinion. Yet, they are lying because they must pretend not

to believe in the arguments they present when answering the devil’s advocate question.

Truth tellers should find it easier to respond to the eliciting-opinion question than to the devil’s

advocate question. Arguments that support someone’s attitude (eliciting-opinion question) are

typically more readily available than reasons that oppose someone’s attitude (devil’s advocate

question) (Ajzen, 2001). This is the result of a confirmation bias, people’s tendency to seek

information that confirms rather than disconfirms their views (Darley & Gross, 1983;

Nickerson, 1998; Wason, 1960). A complete search of all available information is required to

arrive at an entirely justified belief. However, this is impossible because there is far too much

information available on all sides of a topic (Sommer et al., 2023). People thus must decide

which information to search, and searching for information in alignment with their beliefs

Page 3 of 8

247

Vrij, A., Leal, S., & Fisher, R. P. (2023). Interviewing to Detect Lies About Opinions: The Devil’s Advocate Approach. Advances in Social Sciences

Research Journal, 10(12). 245-252.

URL: http://dx.doi.org/10.14738/assrj.1012.16027

makes sense. This is similar to someone who is looking for his/her keys and believes they are

in the kitchen. It makes then sense to start looking in the kitchen rather than in the bedroom

(Sommer et al., 2023). The confirmation bias makes people more practiced in thinking about

arguments that confirm their views than arguments that oppose their views and as such pro- arguments are processed at a deeper level than anti-arguments.

Another explanation why truth tellers should find it easier to respond to the eliciting-opinion

question than to the devil’s advocate question leads to the same outcome. Arguments that

confirm people’s views are more compatible with their other related views and beliefs than

arguments that disconfirm their views (Knobloch-Westerwick, 2015; Knobloch-Westerwick,

Mothes, & Polavin, 2020). Truth tellers may touch upon those compatible views and beliefs to

answer the eliciting-opinion question. The Devil’s Advocate Approach therefore expects that

for truth tellers, responses to the eliciting-opinion question sound more eloquent than

responses to the devil’s-advocate question (eloquence is the collective term for plausibility,

immediacy, and clarity).

For lie tellers, the devil’s advocate question is an invitation to give their true opinion and the

arguments should be readily available to them. Yet, lie tellers’ responses are unlikely to show

the opposite pattern to truth tellers’ responses, because lie tellers will be motivated to use at

least two counter-interrogation strategies: Preparation and consistency (Deeb et al., 2018). Lie

tellers attempt to come across as being sincere during interviews (Leins et al., 2013), and they

could achieve this is by preparing themselves for interviews (Clemens et al., 2013; Granhag &

Hartwig, 2008). This preparation typically takes place by preparing answers to questions

expected to be asked in the interview (Clemens et al., 2013). It is therefore likely that lie tellers

will think, prior to the interview, about arguments they can present that support their

pretended opinion. This should improve the eloquence of their replies to the eliciting-opinion

question. Lie tellers typically think that inconsistency is considered a sign to deceit (Strömwall

et al., 2004) and therefore strive to be consistent (Deeb et al., 2017; Vrij et al., 2021).

Consistency in the Devil’s Advocate interview would mean providing replies of similar

eloquence to the eliciting-opinion and devil’s advocate questions. The Devil’s Advocate

Approach therefore expects that lie tellers may attempt to produce responses of similar

eloquence to the eliciting-opinion and devil’s advocate questions.

THE DEVIL’S ADVOCATE APPROACH: TWO TYPES OF COMPARISON

In the Devil’s Advocate Approach two types of veracity differences can be distinguished. First,

veracity differences in eloquence when comparing answers to the eliciting-opinion and devil’s

advocate questions. It is expected that truth tellers are more eloquent when answering the

eliciting-opinion question than when answering the devil’s advocate question, whereas such

difference is less noticeable in lie tellers. This is the type of comparison we made in the previous

section and is a so-called within-subjects comparison. However, a between-subjects

comparison can also be made: Measuring veracity differences when answering the (i) eliciting- opinion question and (ii) the devil’s advocate question. It is expected that truth tellers are more

eloquent than lie tellers when answering the eliciting-opinion question; Veracity differences

may not necessarily occur in answering the devil’s advocate question.

In real life practitioners mostly have a statement of a single individual. The absence of other

individuals to compare the statement with rules out making a between-subjects comparison.

Page 4 of 8

248

Advances in Social Sciences Research Journal (ASSRJ) Vol. 10, Issue 12, December-2023

Services for Science and Education – United Kingdom

The only option available to practitioners is to compare within a single individual different

parts of a single statement (Vrij, 2016). Such within-subjects’ comparisons are rare in research

and typically between-subjects comparisons are made: In a typical deception experiment, the

responses of groups of truth tellers and groups of lie tellers are compared (Vrij, 2008). The

within-subjects comparison in the Devil’s Advocate Approach is therefore a useful exception to

this common research practice.

In real life making within-subjects comparisons is still not straightforward due to the absence

of cut-off scores (Vrij, Fisher, & Leal, 2023). That is, what is the lowest score an individual can

have and still qualify as a truth teller? In a Devil’s Advocate Approach context, the perfect cut

off score would be that truth tellers are always more eloquent in answering the eliciting- opinion question than in answering the devil’s advocate question, whereas lie tellers are always

equally eloquent when answering both types of question. However, this is not the prediction.

The prediction is that the difference is larger in truth tellers than in lie tellers, whereas no

prediction can be made how large this difference should be to judge someone as a truth teller.

The absence of cut off scores is a general problem in lie detection based on analysing speech or

observing behaviour (Vrij, Fisher, & Leal, 2023), the two types of lie detection typically available

to interviewers in interview settings.

THE DEVIL’S ADVOCATE APPROACH EXPERIMENTS

Table 1 presents the results of the six experiments we are aware of that measured eloquence

in telling the truth or lying about opinions. Two experiments only asked eliciting-opinion

questions and are, strictly speaking, not Devil’s Advocate Approach experiments, but the other

four experiments are. The six experiments measured eloquence in terms of plausibility,

immediacy, and clarity and a variable about which the Devil’s Advocate Approach makes no

prediction: The number of arguments reported. Plausibility was defined as “Does the answer

sound reasonable and genuine and was there enough of an answer to sound convincing”;

Immediacy was defined as “Is the answer given from a personal perspective and not distanced”;

and Clarity was defined as “How clearly does the reader understand what the participant was

saying by the end of the answer”. The number of arguments refer to the number of arguments

reported in favour of the alleged opinion (pro-arguments) and against the alleged opinion (anti- arguments). In the experiments, participants told the truth or lied about controversial societal

issues such as being in favour or against (i) Covid vaccinations, (ii) euthanasia, (iii) CCTV

cameras in public areas, (iv) conspiracy theories, (v) minimum age for alcohol use, and (vi)

violent protests. Participants first expressed their true opinion about each of these issues on 7-

point Likert scales ranging from [1] very much against to [7] very much in favour. The

experimenter then chose a topic the participants expressed a strong view about (scores of 1, 2,

6 or 7 on the Likert scale) and asked the participant to either tell the truth or lie (pretend to

have the opposite view) about it in the subsequent interview. Participants were given as much

preparation time as they wanted.

Table 1 presents a summary of the results. It includes Cohen’s d-scores which reflects how large

an effect is (Cohen, 1988). Cohen stated that d-values around .20 represent small effects; d- values around .50 represent medium effects; and d-values of .80 or higher represent large

effects. Small effects are real effects but barely visible, the difference in height between 15- and

16-year-old girls. Medium effects are large enough to be visible to the naked eye, the difference

in average height between 14- and 18-year-old girls (about an inch). Large effects are grossly

Page 5 of 8

249

Vrij, A., Leal, S., & Fisher, R. P. (2023). Interviewing to Detect Lies About Opinions: The Devil’s Advocate Approach. Advances in Social Sciences

Research Journal, 10(12). 245-252.

URL: http://dx.doi.org/10.14738/assrj.1012.16027

perceptible, such as the difference in average height between 13- and 18-year-old girls (about

two inches) (Cohen, 1988, p. 81). From a real-life perspective, veracity effects around d = 0.20

are not useful, but d-effects around d = .50 or larger are useful. d-scores do not represent

accuracy rates. d = 0.50 represents a higher accuracy than 50% and d-scores can exceed d =

1.00.

Table 1: Summary of Results of the Devil’s Advocate Experiments

Leal et al.

(2010)

Mann et al.

(2022)

Leal et al.

(2023)

Mann et al.

(2023)

Mann et al.

(2024)

Vrij et al.

(2022)

Eliciting-opinion question (EO)

Pro-arguments ns t > l d = 0.37 t > l d = 0.94 ns ns

Anti-arguments ns ns ns t > l d = 0.26 l > t d = -0.50

Plausibility t > l d = 0.72 t > l d = 1.15 ns (0.28) t > l d = 1.64 t > l d = 0.58 t > l d = 0.69

Immediacy t > l d = 0.56 t > l d = 0.59 ns (0.27) t > l d = 1.11 t > l d = 0.30 t > l d = 0.75

Clarity t > l d = 0.50 ns t >l d = 1.22 ns t > l d = 0.75

Devils’ advocate question (DA)

Pro arguments ns ns t > l d = 0.79

Anti-arguments ns ns ns

Plausibility l > t d = -0.86 t > l d = 0.55 l > t d = -0.41 t > l d = 1.50

Immediacy l > t d = -1.31 ns ns t > l d = 1.18

Clarity ns l > t d = -0.34 t > l d = 1.00

EO minus DA

ProMinusAnti

arguments

t > l d = 0.45 ns t > l d = 0.99

Plausibility t > l d = 1.59 t > l d = 0.47 t > l d = 0.51 ns

Immediacy t > l d = 1.95 t > l d = 0.48 t > l d = 0.38 ns

Clarity ns t > l d = 0.48 ns

Note: t > l: higher scores for truth tellers than lie tellers; l > t: higher scores for lie tellers than for truth tellers; ns:

No significant difference between truth tellers and lie tellers; blank spaces means that no veracity comparison was

made

Table 1 presents the results of three comparisons between truth tellers and lie tellers: Veracity

effects in (i) answers to the opinion-eliciting question, (ii) answers to the devil’s advocate

question and (iii) differences in answering the opinion-eliciting and devil’s advocate questions.

The first two comparisons are between-subjects comparisons, whereas the third comparison is

a within-subjects comparison. The Devil’s Advocate Approach predicts differences in the first

and third comparison, but not necessarily in the second comparison. In both the first and third

comparisons, truth tellers are expected to obtain higher eloquence scores than lie tellers.

First, Table 1 shows that in five out of six experiments, truth tellers were more eloquent than

lie tellers when answering the eliciting-opinion question, particularly in terms of being more

plausible and more immediate. The effect sizes ranged from d = 0.58 to d = 1.64 for plausibility

and from d = 0.30 to d = 1.11 for immediacy.1 This supports the hypothesis. In addition, truth

tellers presented their arguments more clearly than lie tellers in three out of five experiments.

The findings for pro- and anti-arguments do not present a clear picture.

1

In Leal et al. (2023) the effect sizes were d = 0.28 for plausibility and d = 0.27 for immediacy.

Page 6 of 8

250

Advances in Social Sciences Research Journal (ASSRJ) Vol. 10, Issue 12, December-2023

Services for Science and Education – United Kingdom

Second, although several significant veracity effects emerged in the devil’s advocate question,

the results were conflicting and do not present a clear picture. Third, regarding the within- subjects comparison, in truth tellers more substantial differences emerged in being plausible

and immediate when answering the eliciting-opinion than devil’s advocate question than in lie

tellers, supporting the hypothesis. No clear picture emerged for number of arguments and

clarity.

One more devil’s advocate experiment was carried out which is not presented in Table 1

because it measured consistency rather than eloquence (Deeb et al., 2017). In all experiments

listed in Table 1 participants took part individually, whereas in Deeb et al. (2017) the

participants were pairs. Pairs were formed from participants who shared the same opinion

about a topic. Truth tellers were asked to express their true opinion in the interview whereas

lie tellers were asked to present in the interview the opposite view than they in fact have. The

pairs were given time to prepare themselves and were interviewed individually. The overlap in

their answers (consistency) was measured. Confirming what the researchers had predicted,

truth telling pairs were more consistent with each other in response to the opinion-eliciting

question than to the devil’s advocate question, whereas deceptive pairs were equally consistent

with each when answering the eliciting-opinion and devil’s advocate questions.

CONCLUSION

The experiments carried out to date showed that lying about opinions can be detected by

paying attention to how plausible and immediate the responses are. A between-subjects

comparison revealed that truth tellers sounded more plausible and more immediate than lie

tellers when answering the eliciting-opinion question. A within-subjects comparison revealed

that truth tellers, more than lie tellers, sounded more plausible and more immediate when

answering the eliciting-opinion question than when answering the devil’s advocate question.

The effect sizes showed that these differences are large enough to be detected with the naked

eye during interviews. However, the absence of cut-off scores could still create problems when

making veracity judgements in real life.

References

Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of Psychology, 52, 27-58.

https://doi.org/10.1146/annurev.psych.52.1.27

Bogaard, G., Colwell, K., Crans, S. (2019). Using the Reality Interview improves the accuracy of the Criteria‐ Based

Content Analysis and Reality Monitoring. Applied Cognitive Psychology, 33, 1018–1031.

https://doi.org/10.1002/acp.3537

Clemens, F., Granhag, P. A., & Strömwall, L. A. (2013). Counter-interrogation strategies when anticipating

requests on intentions. Journal of Investigative Psychology and Offender Profiling, 10, 125–138.

https://doi.org/10.1002/jip.1387

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labelling effects. Journal of Personality and Social

Psychology, 44, 20-33. Doi: 10.1037/0022-3514.44.1.20

Deeb, H., Vrij, A., Hope, L., Mann, S., Granhag, P. A., & Lancaster, G. (2017). Suspects’ consistency in statements

concerning two events when different request formats are used. Journal of Investigative Psychology and Offender

Profiling, 14, 74-87. Doi: 10.1002/jip.1464

Page 7 of 8

251

Vrij, A., Leal, S., & Fisher, R. P. (2023). Interviewing to Detect Lies About Opinions: The Devil’s Advocate Approach. Advances in Social Sciences

Research Journal, 10(12). 245-252.

URL: http://dx.doi.org/10.14738/assrj.1012.16027

Deeb, H., Vrij, A., Hope, L., Mann, S., Leal, S., Granhag, P. A., & Strömwall, L. A. (2018). The Devil’s Advocate

approach: An interview technique for assessing consistency among deceptive and truth-telling pairs of suspects.

Legal and Criminological Psychology, 23, 37-52. Doi: 10.1111/lcrp.12114

DePaulo, B. M., Lindsay, J. L., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H. (2003). Cues to deception.

Psychological Bulletin, 129, 74-118. Doi: 10.1037/0033-2909.129.1.74

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. Doi: 10.1080/10683160701645181

Granhag, P. A., & Hartwig, M. (2015). The Strategic Use of Evidence (SUE) technique: A conceptual overview. In P.

A. Granhag, A. Vrij, & B. Verschuere (Eds.), Deception detection: Current challenges and new approaches (pp. 231-

251). Chichester, England: Wiley.

Hartwig, M., Granhag, P. A., & Luke, T. (2014). Strategic use of evidence during investigative interviews: The state

of the science. In: Raskin, D.C., Honts, C.R., Kircher, J.C. (Eds.), Credibility Assessment: Scientific Research and

Applications (pp. 1-36). Oxford, UK: Academic Press.

Knobloch-Westerwick, S. (2015). The selective exposure self- and affect-management (SESAM) model:

Applications in the realms of race, politics, and health. Communication Research, 42, 959-985.

Doi:10.1177/0093650214539173

Knobloch-Westerwick, Mothes, C., & Polavin, N. (2020). Confirmation bias, ingroup bias, and negativity bias in

selective exposure to political information. Communication Research, 47(1), 104-124.

https://doi.org/10.1177/0093650217719596

Leal, S., Vrij, A., Deeb, H., & Dabrowna, O. (2023). Combining the Devil’s Advocate Approach and Verifiability

Approach to assess veracity in opinion statements. The European Journal of Psychology Applied to Legal Context,

15(2), 53 - 61. https://doi.org/10.5093/ejpalc2023a6

Leal, S., Vrij, A., Mann, S., & Fisher, R. (2010). Detecting true and false opinions: The Devil’s Advocate approach as

a lie detection aid. Acta Psychologica, 134, 323-329. Doi: 10.1016/j.actpsy.2010.03.005

Leins, D., Fisher, R. P., & Ross, S. J. (2013). Exploring liars’ strategies for creating deceptive reports. Legal and

Criminological Psychology, 18, 141-151. Doi: 10.1111/j.2044-8333.2011. 02041.x

Mann, S., Vrij, A., Deeb, H., & Leal, S. (2022). Actions speak louder than words: The Devil’s Advocate questioning

protocol in opinions about protester actions. Applied Cognitive Psychology, 36, 905-918. Doi: 10.1002/acp.3979

Mann, S., Vrij, A., Deeb, H., & Leal, S. (2023). All Mouth and Trousers? Use of the Devil’s Advocate questioning

protocol to determine authenticity of opinions about protester actions. Psychiatry, Psychology, & Law.

https://doi.org/10.1080/13218719.2023.2242433.

Mann, S., Vrij, A., Leal, S., & Deeb, H. (2024). Use of the Model Statement in determining the veracity of opinions.

Manuscript in preparation.

Nahari, G. (2019). Verifiability approach: Applications in different judgmental settings. In T. Docan-Morgan (Ed.),

The Palgrave Handbook of Deceptive Communication (pp. 213-225). New York, NY: United States: Palgrave

Macmillan. https://doi.org/10.1007/978-3-319-96334-1

Nickerson, R. S. (1998). Confirmation Bias: A ubiquitous phenomenon in many guises. Review of General

Psychology, 2(2) 175-220 https://doi.org/10.1037/1089-2680.2.2.175

Palena, N., Caso, L., Vrij, A., & Nahari, G. (2021). The Verifiability Approach: A meta-analysis. Journal of Applied

Research in Memory and Cognition, 10, 1, 155-166. Doi: 10.1016/j.jarmac.2020.09.001

Page 8 of 8

252

Advances in Social Sciences Research Journal (ASSRJ) Vol. 10, Issue 12, December-2023

Services for Science and Education – United Kingdom

Riedel, B. (2019). The remarkable case of the triple agent and the bombing in Kost, Afghanistan.

https://www.brookings.edu/articles/the-remarkable-case-of-the-triple-agent-and-the-bombing-in-khost- afghanistan/

Sommer, J., Musolino, J., & Hemmer, P. (2023). Updating, evidence evaluation, and operator availability: A theoretical

framework for understanding belief. Psychological Review, August 17, Doi: 10.1037/rev0000444

Strömwall, L. A., Bengtsson, L., Leander, L., & Granhag, P. A. (2004). Assessing children’s statements: The impact

of a repeated experience on CBCA and RM ratings. Applied Cognitive Psychology, 18(6), 653-668.

Doi:10.1002/acp.1021

Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities, second edition. Chichester: John Wiley and Sons.

ISBN: 978-0-470-51624-9

Vrij, A. (2016). Baselining as a lie detection method. Applied Cognitive Psychology, 30, 1112-1119. Doi:

10.1002/acp.3288

Vrij, A., Deeb, H., Leal, S., & Fisher, R. P. (2022). The effects of a secondary task on true and false opinion

statements. International Journal of Psychology & Behavior Analysis, 7: 185. https://doi.org/10.15344/2455-

3867/2022/185

Vrij, A., Fisher, R., Blank, H. (2017). A cognitive approach to lie detection: A meta-analysis. Legal and

Criminological Psychology, 22, 1-21. Doi: 10.1111/lcrp.12088

Vrij, A., Fisher, R. P., & Leal, S. (2023). How researchers can make verbal lie detection more attractive for

practitioners. Psychiatry, Psychology, & Law, 30(3), 383-396. https://doi.org/10.1080/13218719.2022.2035842.

Vrij, A., & Granhag, P. A. (2012). Eliciting cues to deception and truth: What matters are the questions asked.

Journal of Applied Research in Memory and Cognition, 1, 110-117. Doi: 10.1016/j.jarmac.2012.02.004

Vrij, A., Granhag, P. A., Ashkenazi, T., Ganis, G., Leal, S., & Fisher, R. P. (2022). Verbal lie detection: Its past, present

and future. Brain Sciences, 12, 1644. https://doi.org/ 10.3390/brainsci12121644

Vrij, A., Hartwig, M., & Granhag, P. A. (2019). Reading lies: Nonverbal communication and deception. Annual

Review of Psychology, 70, 295-317. Doi: 10.1146/annurev-psych-010418-103135

Vrij, A., Mann, S., Leal, S., & Fisher, R. P. (2021). Combining verbal veracity assessment techniques to distinguish

truth tellers from lie tellers. European Journal of Psychology Applied to Legal Context, 13, 9-19. Doi:

10.5093/ejpalc2021a2

Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of

Experimental Psychology, 12, 129-140. Doi: 10.1080/17470216008416717.