Content uploaded by Melissa B. Russano
Author content
All content in this area was uploaded by Melissa B. Russano on Oct 25, 2017
Content may be subject to copyright.
This article was downloaded by: [The University of Texas at El Paso]
On: 10 January 2012, At: 10:49
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Psychology, Crime & Law
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/gpcl20
Minimization and maximization
techniques: assessing the perceived
consequences of confessing and
confession diagnosticity
Allyson J. Horgan a , Melissa B. Russano b , Christian A. Meissner a
& Jacqueline R. Evans a
a Department of Psychology, University of Texas at El Paso, El
Paso, Texas, USA
b Roger Williams University, Bristol, Rhode Island, USA
Available online: 06 Jan 2012
To cite this article: Allyson J. Horgan, Melissa B. Russano, Christian A. Meissner & Jacqueline R.
Evans (2012): Minimization and maximization techniques: assessing the perceived consequences of
confessing and confession diagnosticity, Psychology, Crime & Law, 18:1, 65-78
To link to this article: http://dx.doi.org/10.1080/1068316X.2011.561801
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-
conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Minimization and maximization techniques: assessing the perceived
consequences of confessing and confession diagnosticity
Allyson J. Horgan
a
, Melissa B. Russano
b
, Christian A. Meissner
a
* and
Jacqueline R. Evans
a
a
Department of Psychology, University of Texas at El Paso, El Paso, Texas, USA;
b
Roger
Williams University, Bristol, Rhode Island, USA
(Received 14 December 2010; final version received 4 February 2011)
Identifying interrogation strategies that minimize the likelihood of obtaining false
information, without compromising the ability to elicit true information, is a
challenge faced by both law enforcement and scientists. Previous research suggests
that minimization and maximization techniques may be perceived by a suspect
as an expectation of leniency and a threat of harsher punishment, respectively,
and that these approaches may be associated with false confessions. The current
studies examine whether it is possible to distinguish between minimization and
maximization techniques that do or do not influence a suspect’s perceptions of the
consequences of confessing. Results indicate that techniques that manipulate the
perceived consequences of confessing influence both the decision to confess and
the diagnostic value of confession evidence.
Keywords: confession; interrogation; interviewing; social perception; police
interviewing techniques
Introduction
Increasing true confessions from the guilty and eliminating false confessions from
the innocent are two important interests of the criminal justice system. Obtaining
true confessions allows open cases to be closed quickly and promotes guilty
pleas, speeding up the conviction process (Costanzo, 2004). The importance of
reducing or eliminating false confessions has become evident given the role of
(false) confession evidence in recent DNA exonerations (Kassin et al., 2010). The
Innocence Project (www.innocenceproject.org) reports that approximately 25% of
the over 250 wrongful conviction cases were the product of, at least in part, a false
admission elicited from the defendant.
Although research has focused on the problem of false confessions and the
various factors that may lead a person to falsely implicate themselves (Gudjonsson,
2003; Kassin et al., 2010), more recent studies have begun to explore the diagnostic
value of confessions (or the ratio of true to false confessions elicited) resulting
from various police interrogation techniques (see Meissner, Russano, & Narchet,
2010; Narchet, Meissner, & Russano, in press). We believe it is of critical
*Corresponding author. Email: cmeissner@utep.edu
Psychology, Crime & Law
Vol. 18, No. 1, January 2012, 6578
ISSN 1068-316X print/ISSN 1477-2744
#2012 Taylor & Francis
http://dx.doi.org/10.1080/1068316X.2011.561801
http://www.tandfonline.com
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
importance to identify those techniques that maximize the diagnostic value of
confession evidence, not just those that produce false confessions particularly if
we are to encourage the law enforcement community to seriously consider the
results of social science research and adapt their procedures accordingly.
Furthermore, we believe it imperative that such recommendations have a strong
empirical basis, grounded in psychological theory and evaluated using appropriate
experimental rigor (see Meissner, Hartwig, & Russano, 2010; Meissner, Russano, &
Narchet, 2010).
The purpose of the current study was to further our understanding of the
diagnostic value of confessions obtained using two common approaches to
interrogation, namely minimization and maximization tactics. Previous research
has indicated that these approaches (which actually each consist of a package of
techniques) are associated with the elicitation of false confessions (Kassin &
Kiechel, 1996; Kassin & McNall, 1991; Klaver, Lee, & Rose, 2008; Narchet et al., in
press; Russano et al., 2005), suggesting perhaps that these approaches should be
avoided. For reasons that will be discussed below, this advice may be impractical.
The goal of the current study was to explore whether some forms of these common
interrogative methods used by law enforcement might prove more diagnostic than
others.
Kassin and McNall (1991) originally argued that most modern-day interrogation
techniques can be identified as either a minimization or maximization technique.
Minimization generally involves a gentle, friendly approach in which the interrogator
attempts to gain the suspect’s trust and minimize the seriousness of the offense.
Examples of minimization techniques include stressing the importance of coopera-
tion, expressing sympathy, blaming the victim, and providing face-saving excuses.
Maximization, on the other hand, generally involves the use of harsher techniques or
‘scare tactics’that are confrontational in nature and are designed to emphasize the
seriousness of the situation. Examples of maximization techniques include expressing
absolute certainty in the suspect’s guilt, shutting down denials, exaggerating the
seriousness of the offense, and bluffing about evidence. During an interrogation,
an investigator may use multiple forms of minimization and/or maximization
(cf. Leo, 1996). For example, an interrogator might tell a suspect that officers found
his fingerprints on the murder weapon (a maximization technique), and then justify
the crime by saying that the suspect must have been provoked by the victim
(a minimization technique).
Previous research suggests that minimization and maximization techniques
manipulate the suspect’s perceptions of the consequences of confessing and are,
therefore, often interpreted by the suspect as the equivalent of an expectation of
leniency (if a confession is provided) or a threat of harsher punishment (if no
confession is provided), respectively (Kassin & McNall, 1991). Kassin and McNall
assessed participants’perceptions of various interrogation transcripts. When
minimization techniques were displayed, such as excuses and justifications for the
crime provided by the interrogator, participants perceived the interrogation as non-
coercive, but believed that an implicit offer of leniency was made in return for the
suspect’s confession. The maximization transcript included the use of ‘scare tactics’
66 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
by the interrogator and an overstatement of the evidence against the suspect.
Participants reading this transcript believed there was an implicit threat of harsher
punishment if the suspect did not confess. Based upon these results, Kassin and
McNall posited that the use of maximization and minimization techniques may alter
a suspect’s perception of the expected consequences of confessing, which may affect
a suspect’s decision to confess. Of course, it is particularly problematic if these
techniques lower the diagnosticity of an interrogation by increasing the likelihood of
false confession, and recent experimental works suggest that this is indeed the case
(Kassin & Kiechel, 1996; Klaver, Lee, & Rose, 2008; Narchet et al., in press; Russano
et al., 2005).
Russano et al. (2005) examined the use of minimization techniques versus
an explicit offer of leniency during an interrogation using a novel laboratory
paradigm. In the paradigm (which will be described in more detail in the method
section), participants were induced to cheat (or not) during an experimental task.
All participants were later subjected to an interrogation in which the experimenter
accused them of cheating and asked them to confess to the act. In this
interrogative context, Russano et al. found that the use of minimization techniques
and an offer of leniency increased both true and false confessions. The researchers
also examined diagnosticity, or the ratio of true to false confessions, finding that
both minimization and direct offers of leniency decreased the diagnostic value of
the interrogative evidence. Russano et al. also explored participants’perceptions of
the interrogation and found that innocent participants exposed to minimization
tactics felt more pressure to confess than those in the control condition. This
paradigm has been used in subsequent studies to examine several other factors
that impact interrogations and confessions such as investigator biases and the
influence of non-coercive techniques (Meissner, Russano, & Narchet, 2010;
Narchet et al., in press).
For example, Narchet et al. (in press) investigated whether an interrogator’s pre-
existing beliefs about the guilt or innocence of a suspect affects their interroga-
tion strategy and the likelihood of eliciting true versus false confessions. In this
study, interrogators were not scripted, but rather could apply any combination of
15 different interrogation techniques, including several forms of minimization and
maximization. The researchers manipulated the guilt/innocence of the suspect, as
well as the interrogators’expectations of guilt or innocence prior to the interroga-
tion (by providing them with information regarding the likely guilt/innocence of
participants). Narchet et al. found that the use of minimization and maximization
techniques significantly reduced the diagnostic value of the interrogative evidence,
and that this effect was exacerbated by interrogators’biases towards perceiving guilt
or innocence. Minimization and maximization tactics also influenced participants’
perceptions of the interrogation and thereby moderated the likelihood of obtaining
true versus false confessions. Specifically, these techniques influence guilty partici-
pants’perceptions of proof and feelings of guilt, leading to increased true confessions
in such cases. In contrast, minimization influenced innocent participants’perceptions
of pressure to confess and therein increased the likelihood of false confessions.
These findings are consistent with findings by Sigurdsson and Gudjonsson (1996)
Psychology, Crime & Law 67
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
and Redlich and Kulish (2009) that true confessions appear to be the product
of ‘internal’pressures, while false confessions appear to be elicited via ‘external’
pressures.
Taken together, the implication of the previous research might be that police
should avoid the use of minimization and maximization techniques. We believe
it is unrealistic to think that law enforcement will be receptive to psychological
research in this area if our efforts are focused solely on eliminating false confessions
and encouraging the prohibition of various interrogative methods. Rather, we seek
here to take a more positive approach in which we attempt to identify evidence-
based approaches that might improve the diagnostic value of the confessions elicited
focusing here, in particular, on those methods that police use in their everyday
practice (see Meissner, Hartwig, & Russano, 2010). To that end, the purpose of this
study was to determine whether some minimization and maximization techniques
might prove more diagnostic than others. We were interested in assessing whether
only some of these techniques imply leniency or harsher punishment (thereby
lowering diagnosticity by increasing false confession rates) and whether we might be
able to distinguish between those techniques that do and do not imply consequences.
Specifically, we theorized that minimization and maximization techniques can be
divided into two types: those that appear to manipulate a suspect’s perception of
the consequences of confessing (i.e. leading the suspect to infer an offer of leniency
or a threat of harsher punishment) and those that do not. It should be noted that
the distinction between techniques that manipulate the perception of consequences
and those that do not is a relative one, as individuals could still believe that certain
consequences to confessing exist regardless of interrogation method. Table 1 provides
examples of the minimization and maximization techniques used in the current set of
studies that were hypothesized to vary the perceived consequences associated with
confessing. These techniques were chosen based on the results of an initial pilot study
in which participants were asked to sort examples of each technique based upon
the degree to which they believed the consequences of confessing (or not) were
manipulated.
We conducted a second pilot study to further assess the extent to which the
proposed minimization and maximization techniques in Table 1 might manipulate
Table 1. Minimization and maximization techniques that vary and do not vary the perceived
consequences of confession.
Minimization Maximization
Consequences stress benefit of cooperation exaggerate consequences
downplay consequences co-conspirators against each other
face-saving excuses
minimize seriousness of offense
No consequences express sympathy assume unfriendly demeanor
assume friendly demeanor firm belief in guilt
boost ego/use flattery
appeal to conscience
68 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
participants’perceptions of the consequences associated with confession. Ninety-five
participants were recruited from the University of Texas at El Paso. Participants were
asked to imagine themselves having been accused of committing a crime and finding
themselves in an interrogation room. They were then asked to read through a series of
statements that an investigator might relate to them in an interrogative context, and to
rate their perceptions of the (implied) consequences associated with providing a
confession based upon that statement using a scale from 1‘not harsh at all’to
7‘very harsh’. As minimization tactics are expected to reduce the perceived
consequences associated with confession, we reverse coded these estimates to yield a
measure of degree of ‘manipulation.’Interrogative statements were generated for each
of the approaches detailed in Table 1, and participants were randomly assigned to
consider themselves either ‘guilty’or ‘innocent’of the crime for which they were
being interviewed. A 2 (guilt vs innocence) 2 (minimization vs maximization)
2 (consequences vs no-consequences) mixed factorial ANOVA demonstrated
significant main effects for guiltinnocence, F(1,93) 32.99, pB0.001, h
p
2
0.26,
and manipulation of consequences, F(1,93)7.26, pB0.01, h
p
2
0.07. No significant
interactions were observed. As predicted, interrogative techniques believed to
manipulate subjects’perception of the consequences associated with confession
yielded significantly greater ratings (Ms5.13 vs 4.72, respectively). In addition,
guilty participants perceived a greater degree of manipulation of consequences
associated with confession than did innocent participants (Ms5.98 vs 3.86,
respectively).
Based upon these pilot data, we conducted two studies examining individuals’
perceptions of minimization and maximization tactics that were believed to
manipulate perceptions of the expected consequences associated with confession
(vs those that do not), and further to assess whether this distinction might also
influence the diagnostic value of confessions elicited by guilty and innocent
individuals. In Experiment 1, we examined the social perception of minimization
and maximization techniques that do and do not manipulate the perception of
consequences for the elicitation of confessions from others versus oneself. Previous
survey research indicates that while participants believe that false confessions occur
and that certain populations are vulnerable to providing them, they believe that they
personally would never falsely confess (Henkel, Coffman, & Dailey, 2008). We wanted
to further examine this finding using a design that asked participants to imagine
themselves in an interrogation situation. In Experiment 2, we then examined how
the use of interrogation techniques that manipulate the perceived consequences
of confessing influence the likelihood of obtaining true versus false confessions in
an experiential interrogative context (Russano et al., 2005). It was predicted that
minimization and maximization techniques that influence the perceived consequences
of confessing would elicit less diagnostic information and, in particular, increase the
likelihood of a false confession. Based on previous research, it was also hypothesized
that participants’perceptions of the interrogation (feelings of pressure, beliefs about
the consequences associated with confessing, feelings of guilt, and perceptions of
the proof against them) would be associated with participants’decision to confess
(Narchet et al., in press).
Psychology, Crime & Law 69
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
Experiment 1
Method
Participants
One hundred and thirty-eight participants were recruited from undergraduate
psychology courses at the University of Texas at El Paso. The sample was mostly
Hispanic (74.6%) and female (60%), with a mean age of 24 years.
Design and procedure
A 2 (guilty vs innocent participant)2 (interrogation methods that manipulate
consequences vs no consequences)2 (own vs other likelihood of confession) mixed
factorial design was used in the present study. Guiltinnocence and interrogation
method were manipulated between subjects, while participants’ratings of own vs
other likelihood of confession were presented as a repeated measure.
Participants were presented with a description of the Russano et al. (2005)
paradigm in which two students (A and B) were instructed to solve several logic
problems, some of which were to be solved together and some individually. In the
guilty condition, participants were told that during one of the individual problems,
student A asked student B for help with one of the problems, and that student B
responded by providing the answer. Participants in the innocent condition were told
that the experimental session went according to the instructions given. In both
conditions, participants read that after finishing the set of logic problems, the
research assistant entered the testing room, explained that there seemed to be a
problem, and separated students A and B. Participants then read an interrogation
script in which the interrogation methods were manipulated. In the conseque-
nces condition, the research assistant (or ‘interrogator’) employed minimization and
maximization techniques that were hypothesized to manipulate the perceived
consequences of confessing. In the no manipulation of perceived consequences
condition, the research assistant employed minimization and maximization techni-
ques that were hypothesized not to manipulate the perceived consequences of
confessing. Table 1 provides examples of each of the techniques used by the
interrogator. At the end of both interrogation scenarios, the research assistant asked
student B to sign a statement admitting to sharing information on the individual
problem. After reading the introduction and interrogation scenario, participants
estimated the probability (0100% scale) that: (1) they would sign a confession
statement if they were placed in that situation and (2) other people would sign a
confession statement if they were placed in that situation.
Table 2. Rated probability of self vs others confessions as a function of interrogation method.
Self Others
Consequences Guilty 0.50 (0.51) 0.69 (0.21)
Innocent 0.09 (0.28) 0.64 (0.25)
No consequences Guilty 0.26 (0.44) 0.60 (0.24)
Innocent 0.06 (0.24) 0.47 (0.26)
Note: standard deviations are provided in parentheses.
70 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
Results and discussion
A 2 (guilty vs innocent participant)2 (interrogation methods that manipulate
consequences vs no consequences)2 (own vs other likelihood of confession)
mixed factorial ANOVA was conducted. Table 2 provides the mean estimates of
confession across cells of the design. Main effects of own versus other confession
estimates, F(1,134)23.58, pB0.001, h
p
2
0.15, guiltinnocence, F(1,134)104.13,
pB0.001, h
p
2
0.44, and interrogation method, F(1,134)10.75, pB0.01, h
p
2
0.07,
were found. Participants who read a guilty script were more likely to endorse
confession (across own vs other estimates) than did those who read an innocent
script (Ms0.51 vs 0.32, respectively), and participants believed themselves less
likely to confess than others in the same situation (Ms0.23 vs 0.60, respectively).
Importantly, participants who read a script that was thought to manipulate the
consequences associated with confession were significantly more likely to endorse
confession when compared with those reading a script that was proposed not to
manipulate perceived consequences (Ms0.48 vs 0.35, respectively).
In addition to the main effects, a significant guiltinnocence ownother
confessioninterrogation method interaction was observed, F(1,134)4.24,
pB0.05, h
p
2
0.03. To assess this interaction, separate ANOVAs were conducted
for each interrogation method. For the no manipulation of consequences script,
main effects of guiltinnocence, F(1,67)8.22, pB0.01, h
p
2
0.11, and ownother
confessions, F(1,67)62.0, pB0.001, h
p
2
0.48, were found. For the manipulation
of consequences script, these main effects for guiltinnocence, F(1,67)15.97,
pB0.001, h
p
2
0.19, and ownother confession, F(1,67) 44.75, pB0.001,
h
p
2
0.40, were found as well as a guiltinnocenceownother confession interac-
tion, F(1,67)10.86, pB0.01, h
p
2
0.14. The data suggest participants believed
that while guilty others would be more likely to confess than innocent others
when the consequences associated with confessing were not manipulated, they
believed that both guilty and innocent individuals would be equally likely to
confess under conditions in which the consequences of confessing were manipulated.
When considering whether they, themselves, would confess under these conditions,
participants believed that, if guilty, they would be more likely to confess when the
consequences were manipulated than compared with the no manipulation condition.
If innocent, however, participants rated the likelihood of themselves confessing as
low regardless of interrogation method.
Taken together, the results of Experiment 1 indicated that participants perceived
the use of manipulative techniques in the interrogation scripts and understood that
such techniques may lead (other) people to falsely confess. However, participants
failed to recognize the impact of manipulative techniques when estimating their
own confession decisions. This finding is consistent with prior research indicating
that participants believe that false confessions occur and that people have certain
vulnerabilities to false confessions, but that they themselves are relatively immune
to such situations (Henkel, Coffman, & Dailey, 2008). It appears that participants
may be falling victim to the fundamental attribution error (Ross, 1977) by failing to
take into account situational factors, like manipulation of the perceived conse-
quences of confessing, when evaluating their own behavior in the interrogation room.
In contrast, they appear to be perceptive to situational factors when considering
how other people would react to an interrogation. This finding may also be a result
Psychology, Crime & Law 71
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
of actorobserver effects (Jones & Nisbett, 1972). When participants are in the
role of the observer, they believe other people will not be able to overcome the
manipulative interrogation techniques and will confess. However, when placed in
the actor role, participants believe they will overcome the situational effects and
not falsely confess. In Experiment 2, we moved from examining the social perception
of interrogation techniques to examining the actual behavior of participants in an
experiential interrogative context using the Russano et al. (2005) paradigm.
Experiment 2
Method
Participants
One hundred and thirty-two participants were recruited for the current experiment.
Participants were mostly female (65.2%) and Hispanic (88.6%), with a mean age
of 19 years.
Design and materials
Participants were randomly assigned to one of four experimental conditions based
on a 2 (guilty vs innocent) 2 (minimization and maximization techniques that
manipulate perception of consequences vs minimization and maximization techni-
ques that do not manipulate perception of consequences) between-participants
factorial design.
The manipulation of perceived consequences interrogation script and the no
manipulation of perceived consequences script each contained a total of six
interrogation techniques; four minimization and two maximization techniques (see
Table 1). At the end of the interrogation, participants were asked to sign a confession
statement that indicated, ‘I admit to sharing information on the triangle problem.’
After the interrogation phase, participants completed a debriefing questionnaire
that included several items regarding their perceptions of the interrogation, including
(a) the amount of pressure they felt was placed upon them by the interrogator,
(b) their assessment of the consequences associated with confessing, (c) how guilty
they were made to feel by the interrogator, (d) their perceptions of the proof of
guilt against them, and (e) how severe participants perceived the consequences would
be if they admitted to sharing (all seven-point Likert scales). These items were
selected given that they were most predictive of confessions in previous research
(see Narchet et al., in press).
Procedure
Two male and two female undergraduate research assistants were recruited to
participate as experimenters/interrogators. All research assistants were trained
extensively to ensure that the interview scripts were followed for each participant
and that the procedure was identical for each participant. Interrogations were
video recorded and assessed to ensure that experimenters adhered to the scripted
manipulations.
72 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
The procedure was consistent to that of Russano et al. (2005). The beginning of
the experimental session was consistent for all conditions. A female confederate and
a participant arrived at the lab at the scheduled experiment time. The testing room
was a small, bare room with no windows, similar to what may be used for
interrogations in a police station. After obtaining informed consent, the experi-
menter explained that the purpose of the study was to examine individual versus
group decision-making. Once the pair had completed a short rapport-building task,
they began the problem-solving phase of the experiment. The experimenter explained
that the individual problems were to be completed entirely individually, without any
discussion about answers or strategies. The team problems were to be worked on
together by sharing information about strategies and answers. Participants were
reminded several times about the importance of working together on the team
problems and working alone on the individual problems. This instruction served as
the critical rule of the experiment.
In the guilty condition, while working on the last individual problem, the
confederate feigned difficulty arriving at an answer. After waiting to ensure that
the participant had answered the problem, she asked the participant what answer
he/she calculated. This gave the participant the opportunity to break the rules of
the experiment or ‘cheat.’Participants that did not comply with the request for
information in the guilty condition or that attempted to elicit information from
the confederate in the innocent condition were excluded in the analysis (n16).
In the innocent condition, the confederate did not attempt to elicit any information
about the problem from the participant.
After the participant and confederate completed the problems and a filler
questionnaire, the experimenter explained that he or she had looked over the packets
and there appeared to be a problem (note that the experimenter did not actually
review the problems in order to remain blind to the condition). The experimenter
then asked the confederate to leave the room in order to speak to both of them
individually. After five minutes of isolation, the experimenter began questioning the
participant about breaking the rules of the experiment by sharing information on
one of the individual problems. The experimenter stated that the supervising
professor had been notified, and that the professor wanted the experimenter to
document the situation. The experimenter also informed the participant that the
professor was irritated about the situation and might consider this a case of academic
dishonesty.
At this point in the questioning, the experimenter continued with one of the
two interrogation approaches, either the manipulation of consequences script or the
no manipulation of consequences script. At the end of each of the interrogation
scripts, the experimenter asked the participant to sign a statement admitting their
participation in the cheating incident. When participants signed the statement, the
experimenter thanked them for their cooperation, asked for an explanation of
their side of the story, and then exited the room, explaining that someone would
be with the participant shortly. When participants refused to sign the statement,
the experimenter then went through as many as two abbreviated versions of the
script, repeating the same request using different phrases. For those participants
who still refused to sign the statement, the experimenter thanked them for their time
and exited the room.
Psychology, Crime & Law 73
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
Once the experimenter left the room, a lab manager immediately entered the
testing room to begin debriefing the participant. The lab manager explained the true
purpose and set up of the experiment, explained that the participant was not in any
trouble, and that there was no angry professor to face. The main focus of the
debriefing was to ensure that the participant understood why the use of deception
was necessary and that the participant understood he/she was not in any trouble.
Participants provided a self-report rating of how much pressure they felt to sign the
confession statement on a scale from 1 to 7, and then completed the debriefing
questionnaire described previously.
Results and discussion
Manipulation check
As previously described, the debriefing questionnaire assessed, among other items,
how severe participants believed the consequences would be to admitting to sharing
information on the triangle problem on a scale from 1 (extremely severe) to 7
(not at all severe). Participants in the consequences condition perceived that the
consequences of admitting to sharing information would be less severe than those
participants in the no manipulation of consequences condition (Ms4.00 vs 3.41,
respectively), t(130)1.96, p0.05, d0.34, thereby confirming our successful
manipulation of the perceived consequences of confessing.
True vs false confessions
A 2 (interrogation method: consequences vs no consequences)2 (guilt vs
innocence)2 (interrogator gender: male vs female) hierarchical loglinear anal-
ysis was conducted on participants’decision to confess (sign vs no sign). While
interrogator gender was included as a control variable, no main effects or interactions
involving this variable were observed. Confession rates for guilty and innocent
participants across the interrogation manipulation are presented in Table 3, along with
diagnosticity ratios computed across the interrogation conditions.
Consistent with previous research (Narchet et al., in press; Russano et al., 2005), a
significant main effect of guilt was found, x
2
(1)50.53, pB0.001, L2.89, such that
guilty participants were more likely to confess (89.4%) than innocent participants
(31.8%). This main effect, however, was qualified by a significant interrogation
methodguiltinnocence interaction, x
2
(1)8.48, pB0.01. To assess this interac-
tion, the effect of the two interrogation methods on guilty and innocent partici-
pants was investigated separately. Pairwise comparisons demonstrated that true
confessions significantly decreased when the interrogators used techniques that
manipulated the suspect’s perception of consequences as compared to when
Table 3. True and false confession rates and diagnosticity ratio by interrogation method.
Condition True confessions False confessions Diagnosticity ratio
Consequences 81.8% (27) 42.4% (14) 1.93
No consequences 97.0% (32) 21.2% (7) 4.58
Note:N33 for each cell. Frequencies are provided in parentheses.
74 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
interrogators used techniques that did not manipulate the perception of consequ-
ences, x
2
(1)4.50, pB0.05, L1.96. In addition, false confessions significantly
increased when techniques that manipulated perception of consequences were used
as compared with when techniques that did not manipulate the perceived con-
sequences were used, x
2
(1)3.60, p0.05, L1.20. It appears that manipulating the
perceived consequences of confessing in an interrogation context works to
elicit confessions by raising the expected consequences of not confessing but then
decreasing these consequences in exchange for a confession. The perceptual contrast
effect created by the process of ramping up the consequences and then providing
an incentive for confessing appears to have significantly increased the likelihood of
a false confession while reducing true confessions. In the next section, we further
examine the factors that influence the decision to confess for both innocent and guilty
participants.
Diagnosticity was computed as the ratio of true-to-false confessions elicited,
with higher ratios indicating a greater likelihood of true (versus false) evidence
being elicited in each interrogation condition. Consistent with the trade-off in
true versus false confessions across the interrogation manipulation noted above,
it appears that techniques that do not manipulate the perceived consequences
associated with confessing were 2.37 times more diagnostic than techniques that
manipulated participants’perceptions of the consequences.
Decision making in the interrogation room predicting true vs false confessions
In order to investigate participants’decision making process, the debriefing
questionnaire assessed participant’s perceptions of the interrogation, including: (a)
the amount of pressure they felt was placed upon them by the interrogator, (b) their
assessment of the consequences associated with confessing, (c) how guilty they were
made to feel by the interrogator, and (d) their perceptions of the proof of guilt
against them. Logistic regressions were conducted separately for guilty and innocent
participants to assess the associations between these measures and the likelihood of
true versus false confessions. Significant regression models were observed for both
guilty, x
2
(4)16.51, pB0.01, and innocent, x
2
(4)25.49, pB0.001, participants.
Results of the guilty model suggested that true confessions were significantly
associated with participants’feelings of guilt, b0.75, Wald4.13, pB0.05, and
their perceptions of the proof of guilt, b0.66, Wald4.15, pB0.05. In contrast,
the innocent model demonstrated that false confessions were significantly associated
with participants’perceptions of pressure, b0.55, Wald 6.44, p0.01, and the
expected consequences associated with confessing, b0.54, Wald8.39, pB0.01.
Overall, true confessions were more likely when participants experienced internal
pressure (feelings of guilt and perceptions of proof), while false confessions were
more likely when participants experienced external pressure (interrogative pressure
and severity of consequences).
General discussion
Previous research has found that the use of minimization and maximization
techniques in interrogations can manipulate a suspect’s perceptions of the
consequences associated with confessing (Kassin & McNall, 1991). Given the
Psychology, Crime & Law 75
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
association of these techniques with false confessions and reduced diagnosticity of
confession evidence (Narchet et al., in press; Russano et al., 2005), the current
study sought to determine whether some minimization and maximization
techniques might prove more diagnostic than others by investigating whether
those techniques that manipulate the suspect’s perception of the consequences of
confession are lowering diagnosticity by increasing the false confession rate.
Because minimization and maximization techniques represent the most pervasive
techniques used by law enforcement, it was important to examine whether certain
techniques might prove more diagnostic than others in order to provide evidence-
based alternatives to law enforcement agencies.
Across two studies, the use of certain minimization and maximization
techniques succeeded in manipulating the perceived consequences of confessing.
This manipulation of perceived consequences influenced both participants’beliefs
about whether they and others would confess (Experiment 1) and also influenced
the diagnostic value of confession evidence elicited from participants (Experiment
2). Participants believed that other people would be more susceptible to falsely
confessing when the expected consequences associated with confessing were
manipulated, but did not believe this manipulation would influence their own
decisions to confess if innocent. However, when actually engaging in an
experiential interrogative context (Russano et al., 2005), participants were
vulnerable to the manipulation of consequences and were thereby more likely to
provide a false confession. A comparison of predicted (self) confession rates
(Experiment 1) to actual confession rates (Experiment 2) reveals that while the
pattern of likelihood of true and false confessions is similar, participants
in Experiment 1 greatly underestimated the likelihood of confession. This suggests
that participants underestimated the power of the interrogation context. Addi-
tionally, tactics that manipulate the perception of consequences proved less
diagnostic, as they also reduced the likelihood of obtaining a true confession.
Our assessment of participants’decision processes suggested that the decision
to confess was based on different mechanisms for guilty and innocent participants,
and that this may have been related to the shifts in true and false confessions
observed across our manipulation of interrogative approaches. Specifically, guilty
participants were driven to confess based on the perceived amount of proof the
interrogator had against them and how guilty they felt about their actions
factors that were emphasized to a greater extent in the ‘no consequences’
condition. In contrast, innocent participants were more influenced by their
perception of the expected consequences of confession and the degree of pressure
placed upon them to confess factors that were exaggerated to a greater extent in
the ‘consequences’condition. Thus, it appears that the current manipulation of no
consequences versus consequences techniques may have distinguished the place-
ment of pressure on internal versus external mechanisms leading to variation in
true versus false confessions, respectively (see also Narchet et al., in press; Redlich
& Kulish, 2009). These findings suggest that techniques focusing on the strength
of the (true) evidence against a suspect and emphasizing the morality of
confession may be more productive for eliciting true confessions, limiting the
vulnerability of the innocent and increasing the diagnostic value of the confession
evidence.
76 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
We believe that further research should similarly seek to identify techniques that
will increase the diagnostic value of confession evidence, and importantly those
that reduce or eliminate the likelihood of eliciting a false confession. In this study
we collapsed across minimization and maximization techniques that do and do
not manipulate perception of consequence; it is possible, however, that minimiza-
tion techniques that do not manipulate perception of consequences may be more
diagnostic when used in isolation as compared to when these tactics are paired with
maximization techniques. Moreover, there may be other important dimensions
within minimization and maximization techniques (e.g. level of rapport established
or tone of the interview) that researchers could further assess. It is important that
we develop a better theoretical and practical understanding of these techniques,
including the various psychological mechanisms (e.g. internal vs external sources of
pressure) that result in true versus false confessions.
Given the importance of presenting diagnostic evidence in criminal trials and
the significant costs associated with false confessions, there are practice and policy
considerations that relate to the present findings. Under current US interrogative
practice, all of the techniques tested in this study are legally permissible. Based
on our findings, however, the continuing use of some of these techniques may be
detrimental to the goal of eliciting true confessions while at the same time placing
innocent persons at greater risk. Fortunately, based on the results of this study,
for the first time we can now offer more precise recommendations about which
minimization and maximization techniques might be best to avoid and which might
be advocated. By avoiding the subset of techniques that manipulate the perception of
consequences and increase ‘external’pressure to confess (see Table 1), and focusing
instead on other minimization and maximization techniques that place ‘internal’
pressure on suspects, law enforcement could significantly improve the diagnostic
value of their interrogations. We encourage additional research along these lines,
such that we might ultimately provide law enforcement with positive, evidence-based
practices that will improve the outcomes of most interest to the criminal justice
system (see Meissner, Hartwig, & Russano, 2010; Meissner, Russano, & Narchet,
2010; Narchet et al., in press).
The recent DNA exonerations have revealed the important role of false confession
evidence in leading to wrongful conviction. Researchers have begun to examine
factors that lead to false confessions and have found that the use of minimization
and maximization techniques contribute to the false confession phenomenon (Kassin
& McNall, 1991; Kassin & Kiechel, 1996; Klaver et al., 2008; Narchet et al.,
in press; Russano et al., 2005). The current studies were the first to directly distinguish
between the various forms of minimization and maximization that are commonly
used separating those that do and do not manipulate a suspect’s perception of
the consequences associated with confession. Importantly, our studies found that
such a distinction between minimization and maximization techniques can signifi-
cantly influence the diagnostic value of interrogative evidence. Although these
techniques do not eliminate the likelihood of eliciting a false confession, we believe
that our results begin to provide an evidence-based perspective that could assist law
enforcement in determining what really works in an interrogation room. We urge
researchers to focus on identifying interrogative methods that improve the diagnostic
value of confession evidence when designing future research studies.
Psychology, Crime & Law 77
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012
Acknowledgement
The research reported in this paper was funded by the US Department of Defence and the
National Science Foundation.
References
Costanzo, M. (2004). Psychology applied to law. Belmont, CA: Thompson/Wadsworth
Learning.
Gudjonsson, G.H. (2003). The psychology of interrogations and confessions: A handbook.
Chichester: John Wiley & Sons.
Henkel, L.A., Coffman, K.A.J., & Dailey, E.M. (2008). A survey of people’s attitudes and
beliefs about false confessions. Behavioral Sciences & the Law,26, 555584.
Jones, E.E., & Nisbett, R.E. (1972). The actor and observer: Divergent perceptions of the
causes of behavior. In E.E. Jones, D.E. Kanouse, H.H. Kelley, R.E. Nisbett, S. Valins, &
B. Weiner (Eds.), Attribution: Perceiving the causes of behavior. Morristown, NJ: General
Learning Press.
Kassin, S.M., Drizin, S.A., Grisso, T., Gudjonsson, G.H., Leo, R.A., & Redlich, A.D. (2010).
Police-induced confessions: Risk factors and recommendations. Law & Human Behavior,34,
338.
Kassin, S.M., & Kiechel, K.L. (1996). The social psychology of false confessions: Compliance,
internalization, and confabulation. Psychological Science,7, 125128.
Kassin, S.M., & McNall, K. (1991). Police interrogations and confessions: Communicating
promises and threats by pragmatic implication. Law & Human Behavior,15, 233251.
Klaver, J.R., Lee, Z., & Rose, G. (2008). Effect of personality, interrogation techniques and
plausibility in an experimental false confession paradigm. Legal & Criminological
Psychology,13,7188.
Leo, R.A. (1996). Inside the interrogation room. Journal of Criminal Law & Criminology,86,
266303.
Meissner, C.A., Hartwig, M., & Russano, M.B. (2010). The need for a positive psychological
approach and collaborative effort for improving practice in the interrogation room. Law &
Human Behavior,34,338.
Meissner, C.A., Russano, M.B., & Narchet, F.M. (2010). The importance of a laboratory
science for improving the diagnostic value of confession evidence. In G.D. Lassiter &
C. Meissner (Eds.), Police interrogations and false confessions: Current research, practice, and
policy recommendations. Washington, DC: APA.
Narchet, F.M., Meissner, C.A., & Russano, M.B. (in press). Modeling the influence of
investigator bias on the elicitation of true and false confessions. Law & Human Behavior.
Redlich, A.D., & Kulish, R. (2009, September). True and false confessions: Self-reported
experiences of offenders with serious mental illness. Paper presented at the HUMINT:
Integration of Science and Practice Conference, Ft Huachuca, AZ.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution
process. In L. Berkowitz (Ed.), Advances in experimental social pyschology. New York, NY:
Free Press.
Russano, M.B., Meissner, C.A., Narchet, F.M., & Kassin, S.M. (2005). Investigating true and
false confessions within a novel experimental paradigm. Psychological Science,16, 481486.
Sigurdsson, J.F., & Gudjonsson, G.H. (1996). The relationship between types of claimed false
confession made and the reasons why suspects confess to the police according to the
Gudjonsson Confession Questionnaire (GCQ). Legal & Criminological Psychology,1,
259269.
78 A.J. Horgan et al.
Downloaded by [The University of Texas at El Paso] at 10:49 10 January 2012