Dangerous Decisions: The Impact of First Impressions of Trustworthiness On The Evaluation of Legal Evidence and Defendant Culpability
Dangerous Decisions: The Impact of First Impressions of Trustworthiness On The Evaluation of Legal Evidence and Defendant Culpability
Dangerous Decisions: The Impact of First Impressions of Trustworthiness On The Evaluation of Legal Evidence and Defendant Culpability
University, 1355 Oxford Street, Halifax, Nova Scotia, B3H 4J1 Canada
(Received 19 August 2008; final version received 18 March 2009)
There is little support for the long-standing assumption that judges and jurors can
accurately assess credibility. According to Dangerous Decisions Theory (DDT;
Porter & ten Brinke, Legal and Criminological Psychology, 14, 119!134, 2009),
intuitive evaluations of trustworthiness based on the face may strongly bias the
interpretation of subsequent information about a target. In a courtroom setting,
the assessment of evidence provided by or concerning a defendant may be
fundamentally flawed if its interpretation is influenced by an initial, spontaneous
assessment of trustworthiness. In an empirical test of DDT, participants were
presented with two vignettes describing major or minor crimes, accompanied by a
photograph of the supposed defendant, previously rated as highly trustworthy or
untrustworthy in appearance. Participants evaluated culpability following the
presentation of evidence in each case. Participants required less evidence to arrive
at a guilty verdict and were more confident in this decision for untrustworthy-
appearing defendants. The current evidence supports DDT and has implications
for legal decision-making practices.
Keywords: facial trustworthiness; intuition; bias; legal decision-making; credibility
assessment
Introduction
Judges and jurors are faced with the task of weighing the available evidence against a
defendant following arguments presented by adversarial parties. In some cases, their
decisions come down to a credibility contest as observed by Judge Josephson in the
Air India mass murder case (R v. Malik & Bagri, 2005): . . . the determination of
guilt devolves to the weighing of the credibility of a number of witnesses who testified
in these proceedings. Thus, particularly in cases lacking physical evidence, the ability
of legal decision-makers to make accurate determinations of credibility through an
unbiased, objective evaluation of evidence for or against the defendants guilt is
essential. While our justice system acknowledges that decisions are not infallible,
relying on the principle of reasonable doubt, there is good reason to question the
validity of credibility assessment in the courtroom. In a Canadian inquiry into the
wrongful murder conviction of Thomas Sophonow, Justice Cory observed that
witnesses commonly use deception, and many are smooth and convincing liars who
have fooled judges and juries (Wilson, 2003). As highlighted by a recent string of
notorious wrongful convictions (in Canada, these include Donald Marshall Jr, David
Milgaard, and Guy Paul Morin), concern over credibility assessment in the
courtroom is warranted given the enormous individual and societal consequences
that can result from incorrect decisions.
Credibility is a matter within the competence of lay people. Ordinary people draw
conclusions about whether someone is lying or telling the truth on a daily basis
(p. 248). Further, in the Supreme Court case R. v. Francois (1994), Justice McLachlin
stated: In the end, the jury must decide whether it believes the witnesss story in
whole or in part. That determination turns . . . on the demeanour of the witness and
the common sense of the jury. Despite the general attitude that people are able to
make accurate determinations of credibility based on common sense, empirical
research suggests that deception detection is a flawed process with errors occurring in
approximately 45% of all assessments (e.g. Bond & DePaulo, 2006). While one might
predict that professionals who detect deception on a daily basis (e.g. judges, police
officers, etc.) would outperform laypersons, they too typically perform around the
level of chance, or even below chance, in judging the credibility of speakers (Ekman
& OSullivan, 1991; Porter, Woodworth, & Birt, 2000; see Porter & ten Brinke, 2009;
Vrij, 2008). However, judges and laypersons alike tend to be highly confident in their
assessments despite the fallibility of their judgements (e.g. Porter & ten Brinke, 2009;
Vrij, 2008). While it is not possible to know how often mistakes concerning
credibility occur in the courtroom, numerous wrongful convictions contest the belief
that such assessments and the evaluation of evidence in general are matters of mere
common sense. As such, the common sense argument advocated by the high courts,
and the confidence that is placed in these decisions, may be considered the first sign
of danger in courtroom assessments of credibility.
difficult and prone to error (Ekman, 1992; Leach, Talwar, Lee, Bala, & Lindsay,
2004; Porter & ten Brinke, 2008).
Despite peoples well-adapted ability to control certain aspects of their facial
presentation, the analysis of emotional facial expressions appears to play a critical
role in credibility assessment in the courtroom (R. v. B. (KG), 1993). Thus, judges
and jurors view the face of a witness as a rich source of presumably relevant
information. Information about an individuals character gleaned from the face
seems to be gathered extremely quickly upon seeing an individual for the first time.
Willis and Todorov (2006) examined the effect of exposure time to the face on
trait inferences, including evaluations of trustworthiness. Results suggested that
individuals infer the trustworthiness of others almost instantaneously upon
seeing the face, after only 100 ms of exposure. Further, longer exposure time
increased levels of confidence without changing judgements. Thus, their findings
indicate that assessments of trustworthiness occur instantaneously and are enduring
in nature.
Despite highlighting the speed of this process, the paradigm designed by Willis
and Todorov (2006) did not offer insights into the accuracy of such assessments, as
the actual trustworthiness of target faces was unknown. Although some research
suggests that particular attributes can be correctly inferred from physical appearance
(e.g. Roney, Hanson, Durante, & Maestripieri, 2006; Shevlin, Walker, Davies,
Banyard, & Lewis, 2003), the assessment of trustworthiness does not appear to be
among them. Porter, England, Juodis and ten Brinke (2008) presented participants
with a group of faces consisting of Americas Most Wanted criminals and
humanitarian award (e.g. Nobel Peace Prize) winners (simply asking them to
evaluate strangers faces). Although from common lore one might assume that the
most and least trustworthy members of society could be discriminated by looking at
their faces, accuracy levels were only slightly above chance, with errors occurring in
over 40% of assessments. Thus, assessments of trustworthiness based on the face
alone, while instantaneous, are highly fallible. It is a similar assessment that may
strike judges and jurors as valid intuition upon viewing a defendant, complainant, or
witness for the first time (Porter & ten Brinke, 2009).
Other facial features correlated with trustworthiness and perceived honesty
include babyfacedness, symmetry, and attractiveness (Bull, 2006; Bull & Vine, 2003;
Lundqvist, Flykt, & Ohman, 1998; Todorov, 2008; Zebrowitz, Voinescu, & Collins,
1996). For example, baby-faced individuals receive more lenient judicial outcomes
than mature-faced individuals. Reflecting the impact of attractiveness on impression
formation and the what is beautiful is good phenomenon (Dion, Berscheid, &
4 S. Porter et al.
Walster, 1972), attractive defendants are more likely to be found not guilty, dealt
shorter sentences, and considered less dangerous than their unattractive counterparts
(e.g. Bull & Rumsey, 1988; Downs & Lyons, 1991; Esses & Webster, 1988). Similarly,
particular faces are viewed as being congruent with certain criminal offences. Thus,
there are some faces that people agree look like that of a rapist, armed robber, or
murderer (Bull & McAlpine, 1998; Dumas & Teste, 2006). Further, research suggests
that if a defendant has a face that is considered to be consistent with the charged
offence, they are more likely to be convicted than an individual with an incongruent
face (Macrae & Shepherd, 1989; Shoemaker, South, & Lowe, 1973). Recent work
indicates that this effect occurs regardless of the strength of the prosecutions
evidence such that defendants whose faces are congruent with the offence are more
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
likely to be found guilty even when evidence against them is weak (Dumas & Teste,
2006). Although the impact of these facial attributions (most notably attractiveness)
in courtroom decisions has been well established, we believe that evaluations of
trustworthiness may play a central role in biased legal decision-making, given the
evolutionary roots of friend or foe judgements and the appropriate (but not
necessarily accurate) application of such evaluations to potentially dangerous
criminals.
The initial intuitive evaluation of trustworthiness proposed in DDT is thought
to influence subsequent inferences concerning the defendant (or other witness)
by making decision-making about him/her increasingly irrational, leading to such
biased decisions (Kahneman & Tversky, 1982). Although judges and jury members
no doubt strive to maintain objectivity, they are not immune from normal human
biases and may not be aware of the influences upon their decision-making
(Kaufmann et al., 2003). Further, judges are susceptible to critical thinking errors
and a reliance on false stereotypes that may be cited to justify their initial, unreliable
assessment (e.g. Granhag & Stromwall, 2004; Vrij & Mann, 2004; Vrij, 2004, 2008).
For example, while judges and laypersons alike adhere to the belief that deceptive
individuals avert their gaze, empirical research suggests that no such relationship
exists (e.g. DePaulo et al., 2003; Porter & ten Brinke, in press; Vrij, 2008). DDT
suggests that the adherence to initial impressions and subsequent justification with
erroneous beliefs will result in a non-critical, tunnel vision assimilation of
potentially ambiguous and even contradictory evidence concerning the defendant
resembling a confirmation bias described by Carretta and Moreland (1982). Ask and
Granhag (2007) found that criminal investigators have a tendency to be more
sceptical about evidence that runs counter to their original perception than evidence
consistent with that perception. Further, Porter et al. (2007) found that an increased
level of motivation exacerbated tunnel vision in a deception detection task, leading
to decreased accuracy. Thus, Porter and ten Brinke (2009) proposed that judges
and jury members are likely to subconsciously prefer evidence confirming their
initial assessment of trustworthiness, slanting ambiguous evidence in favour of this
assessment, and creating tunnel vision in their decision-making. Particularly when
assessing evidence in the case of a severe crime where the consequences of a faulty
decision are the greatest, the strong motivation of judges and juries, coupled with the
complexity of the task, may lead to exacerbated tunnel vision and over-confidence in
the initial assessments.
Psychology, Crime & Law 5
effect would be exacerbated for severe crimes. The increased motivation felt by the
judge or jurors during a murder trial, given the weighty consequences of a faulty
decision (i.e. incarcerating an innocent individual or allowing a murderer to escape
justice) was expected to exacerbate tunnel vision, confirming initial impressions.
Specifically, if a participant initially concluded (instantaneously or intuitively)
that the defendant was untrustworthy based on his physical appearance, it was
expected that ambiguous information would be interpreted as indicative of guilt. As
such, fewer pieces of ambiguous evidence would need to be considered prior to
reaching a decision of guilt. Additionally, it was hypothesized that participants
would be highly confident in their guilty verdict for untrustworthy defendants and
would be less likely to overturn their verdict in the face of powerful exonerating
evidence, particularly for severe crimes where motivation and resulting tunnel vision
was expected to be exacerbated.
Pilot study
A pilot study was conducted in order to rate the photographs and created evidence,
allowing for selection of trustworthy and untrustworthy defendant faces as well as
increasingly incriminating evidence sets for each of four (two petty, two severe) crime
vignettes.
Method
Participants
Participants in the pilot study were 27 individuals (15 females, 12 males) recruited
from the student population of Dalhousie University.
Apparatus
Twenty photographs of Caucasian males were chosen from the Karolinska Set of
Images (Lundqvist et al., 1998) and were rated on various attributes by participants in
the pilot study. The photographed individuals were actors between 20 and 30 years
old, exhibiting neutral expressions. Pilot participants rated attractiveness, babyfaced-
ness, symmetry, kindness, and trustworthiness on a 1 (not at all) to 7 (highly) scale for
each face. Pilot participants also rated the severity of four crime vignettes describing a
robbery resulting in murder, a sexually motivated murder, a car theft and a fraud, as
6 S. Porter et al.
well as the strength of evidence pieces (e.g. eyewitness, alibi, confession, DNA, etc.)
associated with each scenario on seven-point scales. The determination of severity
was based on whether the crimes were summary (punishable by a sentence of 2 years
less a day) or indictable (felony in the US) offences, and was confirmed with pilot
study ratings. For crime severity, a rating of one denoted a very petty crime while
seven indicated an extremely serious, major crime. Similarly, when rating evidence
pieces, a one indicated a highly ambiguous piece of evidence not indicative of guilt,
whereas a seven indicated that the evidence was highly incriminating and suggestive of
guilt. Last, a subset of exonerating evidence associated with each vignette was also
scored on a 1 (not at all exonerating) to 7 (highly exonerating) scale.
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Procedure
Participants were presented with four hypothetical criminal cases (suggested to be
actual cases), and a set of at least 24 pieces of evidence corresponding to each crime.
They were asked to read each case, and then on seven-point scales (i) rate the severity
of each crime and (ii) evaluate each piece of corresponding evidence. After all
evidence was rated, participants were presented with a series of 20 photographs of
male actors and were asked to rate each photograph (as described above) and their
confidence in each such rating (also on seven-point scales).
Results
Pilot study ratings
Based on average trait inferences made by pilot study participants, the most
(M "4.70, SD"0.99) and least trustworthy (M "2.63, SD "1.45) individuals were
chosen from the rated images. A paired samples t-test indicated that the photographs
were rated as significantly different from one another on trustworthiness, t(26) "
6.99, pB0.01. Other attributions based on the face, including attractiveness,
babyfacedness, symmetry and kindness also were rated as significantly different
( p B0.05) and, as such, were examined for use as covariates in the primary study
analyses. Crime vignette severity also was examined; a repeated measures ANOVA
followed by Bonferroni corrected post hoc tests confirmed that the severe crimes
(M "6.60, SD"0.63) were indeed rated as significantly more severe than the petty
crimes (M"4.21, SD"0.92), F(1,25) "183.4, p B0.01. Further, the two vignettes
within each severity condition were equivalent in terms of severity ( ps !0.05).
In order to ensure that evidence sets accompanying each crime vignette were
similar, mean scores of the level of incrimination for each piece of evidence were
calculated and evidence was rank ordered. The five most ambiguous pieces of
evidence (defined as pieces with an average rating below four) corresponding to each
vignette were chosen for use in the main study. Next, five progressively incriminating
pieces were chosen with mean ratings between four and seven for each vignette.
Paired samples t-tests confirmed that the increasingly incriminating evidence was
rated as significantly more incriminating than ambiguous evidence for each vignette
( ps B0.01). Further, ratings of ambiguous and incriminating evidence were similar
across petty and severe crime vignettes ( ps!0.05). The final, exonerating piece of
evidence for each vignette was chosen on the basis of being rated as the strongest
Psychology, Crime & Law 7
Main study
Method
Participants
Eighty participants from the student population of Dalhousie University were
recruited for the main study with an offer of class credit points in return for
participation. Participants were predominantly female (12 male, 68 female) and had
a mean age of 20.05 years (SD "3.36).
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Apparatus
The ratings obtained from the pilot study led to the selection of photographs and
evidence for the primary study. The most trustworthy and untrustworthy faces
(highest and lowest mean ratings) were selected, and a smaller evidence set was
established for each crime. Final evidence sets consisted of five pieces of ambiguous,
followed by five pieces of progressively incriminating and one piece of exonerating
evidence. The two severe crime vignettes described a robbery resulting in murder and
a sexually motivated murder, while the petty crime vignettes described a car theft
and a fraud. Although the crime description and corresponding evidence was
different across cases within each severity condition, both were equivalent on pilot
ratings of ambiguity/incrimination. See the Appendix for crime vignettes and
associated evidence of one petty (car theft) and one severe (robbery resulting in
murder) case.
In addition to providing verdicts and measures of confidence, main study
participants completed a series of personality questionnaires, which served as
distracter tasks during the experiment.
Procedure
The severity of crime vignettes (severe or petty) served as a between-subjects vari-
able such that participants were either presented with two severe or two petty crime
descriptions. With facial trustworthiness as a within-subjects variable, each crime
vignette was accompanied by an image of the trustworthy or untrustworthy looking
defendant. As such, this study utilized a 2 (severity)#2 (trustworthiness) mixed
design.
Half of the participants (n "40) were randomly assigned to complete seven-point
ratings of perceived attractiveness, babyfacedness, symmetry, and kindness on the
two selected faces prior to vignette presentation. This served to confirm the pilot
study finding that these two individuals differed in perceived trustworthiness,
allowed an examination of the impact of making explicit ratings on subsequent
decision-making, and allowed for an examination of these facial traits as potential
covariates in subsequent analyses. After these photographs were rated, participants
completed a distracter personality questionnaire. Subsequently, the participant was
presented with the first of two counterbalanced crime vignettes. Participants assigned
8 S. Porter et al.
to the severe crime condition were presented with vignettes describing (i) a robbery
resulting in murder and (ii) a sexually motivated murder. Petty crime condition
participants were presented with (i) a fraud and (ii) a car theft. The defendants
photographs were counterbalanced to the case file such that trustworthy and
untrustworthy images were associated with each crime vignette equally. After reading
each vignette, participants were presented with evidence associated with the crime
including five ambiguous, five increasingly incriminating, and one exonerating
piece of evidence ! in that order. After each piece of evidence was presented, the
participants, based on what they knew up to that point, were instructed to indicate
their verdict (guilty or not guilty beyond a reasonable doubt) and rate their
confidence in that decision from one to seven. After participants provided their final
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Results
To ensure that the selected photographs were rated similarly on trustworthiness in
the main and pilot studies, a paired samples t-test was conducted. Assessments of
trustworthiness by those in the rating condition replicated the results of the pilot,
and the two photographed individuals again were shown to be significantly different
from one another, t(39) "2.82, pB0.01.
10
9
Trustworthy
8 Untrustworthy
7
Latency to Guilty Verdict
5
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
0
Petty Severe
Confidence in verdicts
A 2 (crime severity) #2 (trustworthiness) mixed ANOVA2 was conducted to examine
participants confidence in their (guilty) verdicts following the final piece of
incriminating evidence. As such, only those participants who returned a guilty
verdict after evidence piece 10 were included in this analysis (severe: n "30; petty:
n "29). The analysis revealed a significant interaction, F(1,57) "9.44, pB0.01
(see Figure 2), but no significant main effects, ps !0.05. Follow-up t-tests revealed
that participants were similarly convinced in the guilt of trustworthy individuals in
petty (M "5.90, SD "1.59) and severe crimes (M "5.47, SD "1.36), p!0.05.
However, participants were significantly more confident that untrustworthy defen-
dants had committed a severe crime (M "6.13, SD "0.86) compared to a petty crime
(M "5.30, SD "1.75), t(58) "2.35, p B0.05. Further, participants were more
confident that untrustworthy defendants (M "6.13, SD "0.86) were guilty of
a severe crime than trustworthy defendants (M "5.47, SD"1.36), t(29) "2.61,
p B0.05, while no such difference emerged for confidence associated with petty crime
verdicts, p!0.05.
Trustworthy
Untrustworthy
6
Confidence in Guilty Verdict
4
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
0
Petty Severe
Figure 2. Confidence in guilty verdict, rated on a 1!7 scale, as a function of crime severity
and defendant trustworthiness. Error bars represent standard deviation.
Discussion
The predictions based on DDT were supported in this study, suggesting that this
model may elucidate the process of biased legal decision-making that can lead to
wrongful convictions. As predicted by DDT, less evidence was necessary to convict
an untrustworthy looking defendant of (the same) crime compared to a perceived
trustworthy person, particularly in cases of murder (severe crimes). Further
supporting DDT, untrustworthy defendants were considered guilty based on an
average of fewer than five pieces of ambiguous evidence, while more incriminating
evidence was necessary to convict the trustworthy defendant of the same (severe)
crime. On the other hand, there was no such difference in evidence necessary to
convict trustworthy and untrustworthy individuals of minor crimes. Thus, the
suggestion that increased crime severity might increase motivation and exacerbate
Psychology, Crime & Law 11
the effect of initial impressions appears to be supported and, at least in a mock jury
study context, the presentation of severe crimes is necessary to reveal dangerous
decision-making based on initial impressions.
DDT also suggested that the effect of trustworthiness impressions will be
reflected in confidence for verdicts. In accordance with this tenet, confidence levels
were highest when participants were delivering a guilty verdict to an untrustworthy
looking individual accused of murder. Thus, it appears that perceived trustworthi-
ness not only affects verdicts in general but also jurors confidence in the types of
crimes that an individual could have committed. This finding is in line with legal
decision-making research examining face/offence congruency, wherein participants
were more likely to find a photographed individual guilty if they looked like the
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
type of individual who would engage in this criminal activity (Dumas & Teste, 2006).
The effect of exonerating evidence on verdict decision-making also was
examined. Although the effect of facial trustworthiness was not significant, it should
be noted (for future research) that the hypotheses were supported by the pattern
of means. Specifically, participants were less likely to accept the innocence of
untrustworthy defendants after powerful exonerating evidence ! particularly for
severe crime vignettes. In general, participants appeared to have been unable to
overcome their initial impression of the untrustworthiness of the accused, and
interpreted evidence in such a way as to reinforce their initial impression. Future
research should further examine the impact of facial trustworthiness on the
assimilation of, not only incriminating, but also subsequently presented exonerating
evidence.
The current study provides support for DDT and suggests that this model could
have implications for understanding wrongful convictions, and perhaps for improv-
ing legal decision-making. The impact of trustworthiness, discovered here, was found
to have considerable implications for decisions of defendant culpability over and
above that associated with the well-documented what is beautiful is good (and
honest) phenomenon (Downs & Lyons, 1991). Given that the impact of attractive-
ness (among other potentially related facial features) was considered as a covariate
during statistical analyses, it appears that trustworthiness holds unique biasing
effects in the courtroom. Future studies might expand the present design to better
understand the effect of trustworthiness assessments on information assimilation,
and subsequent decision-making. To fully explicate the impact of trustworthiness
biases, this study could be extended by including a third case file in each severity
condition with no associated photograph. Further, the current study might be
modified to include photographs of truly wrongly convicted individuals to examine if
the individuals untrustworthy physical appearance may have been a factor in their
conviction. DDT also suggests more complex interactions between intuitive facial
assessments and subsequent legal decision-making. Further research may examine
the impact of the victim or complainants perceived trustworthiness on assimilation
of evidence against the defendant. Additionally, the trustworthiness of the source of
information may also play a role in legal decision-making. For example, information
concerning a defendant from a source perceived as trustworthy may be favoured in
the mind of a judge or juror over information derived from similarly knowledgeable
but untrustworthy-looking witnesses.
The results of this study could have important implications for the justice system.
Particularly for severe crimes such as murder ! transgressions that also carry the
12 S. Porter et al.
Acknowledgements
This project was supported by the Social Sciences and Humanities Research Council of
Canada (SSHRC) and the Natural Sciences and Engineering Research Council of Canada
(NSERC) through grants awarded to the first author. Graduate scholarships from NSERC
and Killam Trusts awarded to the third author also contributed to the successful completion
of this project.
Notes
1. Counterbalancing was successful, as the order of faces had no effect on confidence ratings
or latency to a guilty verdict, ps !0.05.
2. A series of bivariate correlations did not reveal any significant (ps!0.05) relationships
between ratings of perceived facial characteristics and confidence in verdicts; therefore, no
covariates were included in subsequent analyses.
References
Ask, K., & Granhag, P.A. (2007). Motivational bias in criminal investigators judgments of
witness reliability. Journal of Applied Social Psychology, 37, 561!591.
Bond, C.F., & DePaulo, B.M. (2006). Accuracy of deception judgments. Personality and Social
Psychology Review, 10, 214!234.
Bull, R. (2006). Detecting lies and deceit: The psychology of lying and the implications for
professional practice. Journal of Community & Applied Social Psychology, 16, 166!167.
Bull, R., & McAlpine, S. (1998). Facial appearance and criminality. In A. Memon, A. Vrij, &
R. Bull (Eds.), Psychology and law. Truthfulness, accuracy and credibility (pp. 59!76).
London: McGraw-Hill.
Psychology, Crime & Law 13
Bull, R., & Rumsey, N. (1988). The social psychology of facial appearance. Berlin: Springer.
Bull, R., & Vine, M. (2003). Attractive people tell the truth: Can you believe it? Poster presented
at the Annual Conference of the European Association of Psychology and Law, Edinburgh,
UK.
Carretta, T.R., & Moreland, R.L. (1982). Nixon and Watergate. Personality and Social
Psychology Bulletin, 8, 446!453.
DePaulo, B.M., Lindsay, J.J., Malone, B.E., Muhlenbruck, L., Charlton, K., & Cooper, H.
(2003). Cues to deception. Psychological Bulletin, 129, 74!118.
Dion, K.K., Berscheid, E., & Walster, E. (1972). What is beautiful is good. Journal of
Personality and Social Psychology, 24, 285!290.
Downs, A.C., & Lyons, P.M. (1991). Natural observations of the links between attractiveness
and initial legal judgments. Personality and Social Psychology Bulletin, 17, 541!547.
Dumas, R., & Teste, B. (2006). The influence of criminal facial stereotypes on juridic
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Porter, S., England, L., Juodis, M., & ten Brinke, L. (2008). Is the face a window to the soul?
Investigation of the validity of intuitive judgments of the trustworthiness of human faces.
Canadian Journal of Behavioural Science, 40, 171!177.
R. v. B. (K.G.), [1993] 1 S.C.R. 740.
R. v. Francois [1994] 2 S.C.R. 827.
R. v. Malik & Bagri, 2005 BCSC 350.
R. v. Marquard [1993] 4 S.C.R. 223.
Roney, J.R., Hanson, K.N., Durante, K.M., & Maestripieri, D. (2006). Reading mens faces:
womens mate attractiveness judgments track mens testosterone and interest in infants.
Proceedings of the Royal Society of London B, 273, 2169!2175.
Shevlin, M., Walker, S., Davies, M., Banyard, P., & Lewis, C.A. (2003). Can you judge a book
by its cover? Evidence of self!stranger agreement on personality at zero acquaintance.
Personality and Individual Differences, 35, 1373!1383.
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Shoemaker, D.J., South, D.R., & Lowe, J. (1973). Facial stereotypes of deviants and judgments
of guilt or innocence. Social Forces, 51, 427!433.
Todorov, A. (2008). Evaluating faces on trustworthiness: An extension of systems for
recognition of emotions signaling approach/avoidance behaviors. Annals of the New York
Academy of Sciences, 1124, 208!224.
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 lies and deceit: Pitfalls and opportunities. Chichester, UK: Wiley.
Vrij, A., & Mann, S. (2004). Detecting deception: The benefit of looking at a combination of
behavioral, auditory and speech content related cues in a systematic manner. Group Decision
and Negotiation, 13, 61!79.
Willis, J., & Todorov, A. (2006). First impressions: Making up your mind after a 100-ms
exposure to a face. Psychological Science, 17, 592!598.
Wilson, P.J. (2003). Wrongful convictions: Lessons learned from the Sophonow Public Inquiry.
Canadian Police College.
Zebrowitz, L.A., Voinescu, L., & Collins, M.A. (1996). Wide-eyed and crooked-faced:
Determinants of perceived and real honesty across the life span. Personality and Social
Psychology Bulletin, 22, 1258!1269.
Appendix
Evidence:
(1) The accused is relatively new to the neighbourhood, having moved three times in the
past year.
(2) The accused matches the physical description (white, average height and weight).
(3) The accused is very interested in cars.
(4) The accused was very short on cash at the time of the crime.
(5) The accused says on the night of the theft he was home alone.
(6) The accused works in the garage near the alley where the cars were found.
(7) The accused does not own a car. A friend testified that once when he had locked his
keys in his car, the accused was able to pick his lock in under a minute.
(8) The accused tried to run from the police when they attempted to arrest him.
(9) The garage where the accused works has been investigating previously for being
Downloaded By: [Brinke, Leanne ten][University Of British Columbia] At: 16:44 10 May 2010
Evidence:
(1) The accused admits to using marijuana.
(2) The accused is from a Toronto suburb.
(3) The accused smokes cigarettes.
(4) The accused has older friends who are unemployed.
(5) The accused is unemployed.
(6) The footprint of the accused was found in the convenience store.
(7) Upon apprehension, a stocking mask was found in the pocket of the accused.
(8) Acquaintances of the accused testified that he had previously mentioned robbing
a store as a source of revenue.
(9) Another suspect in the robbery has identified the accused as his partner.
(10) When police showed the knife to the mother of the accused, she identified it as one
that she owned.
(11) A young man came forward and confessed to the robbery. The other man he
identified as his partner subsequently confessed.