ArticlePDF Available

Abstract

Participants (N = 189) witnessed the theft of a computer and were immersed into what they were led to believe was an actual police investigation that culminated in a live showup. After the crime, an officer responded to the scene to take witness statements. Minutes after his arrival, the officer received a radio dispatch that could be heard clearly by the witnesses. The dispatch either stated that the Sherriff had “…caught the guy…” or “…detained a suspect who matched the thief’s description…”, and instructed the officer to bring the witnesses to identify the suspect. The witnesses then met with two deputies who conducted a live showup with an innocent suspect or the actual culprit. Choosers were more confident than rejecters across all conditions. Also, overhearing the suggestion that the sheriff had caught the guy significantly increased false-identifications, and boosted witness confidence in these errors, but did not affect accurate suspect-identifications.
Running Head: PRE-ADMONITION SUGGESTION IN LIVE SHOWUPS
This article has been accepted for publication in Applied Cognitive Psychology. Copyright notice: This article
may not exactly replicate the final published version. It is not the copy of record. Please do not duplicate or
distribute any part of this article without the permission of the author.
Pre-Admonition Suggestion in Live Showups:
When Witnesses Learn that the Cops Caught the Guy
Mitchell L. Eisen Amaia Skerrit -Perta
Jennifer M. Jones Jade Owen Gabriela C. Cedré
California State University, Los Angeles
Acknowledgements: We would like to thank Chief Wall and Detective Larry Bohannon
of the California State University Police for their help with this project and Los Angeles
County Sheriff Leroy Baca and his department for their generous assistance in
conducting this research. We would also like to acknowledge Marilyn Orozco, Alma
Olaguez, Gabrielle Aroz, Satchel Pratt, Derek Eisen, Madhavi Guiot, Joseph Williams
and Jessica Pope for their work in collecting and entering the data. Finally, we want to
thank Andrew Smith for his insightful comments and suggestions in creating the
manuscript. All inquiries related to this manuscript should be directed to Dr. Mitchell
Eisen, meisen@calstatela.edu.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
2
Abstract
Participants (N = 189) witnessed the theft of a computer and were immersed into what
they were led to believe was an actual police investigation that culminated in a live
showup. After the crime, an officer responded to the scene to take witness statements.
Minutes after his arrival, the officer received a radio dispatch that could be heard clearly
by the witnesses. The dispatch either stated that the Sherriff had “…caught the guy…” or
“…detained a suspect who matched the thief’s description…”, and instructed the officer
to bring the witnesses to identify the suspect. The witnesses then met with two deputies
who conducted a live showup with an innocent suspect or the actual culprit. Choosers
were more confident than rejecters across all conditions. Also, overhearing the suggestion
that the sheriff had caught the guy significantly increased false-identifications, and
boosted witness confidence in these errors, but did not affect accurate suspect-
identifications.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
3
Pre-Admonition Suggestion in Live Showups:
When Witnesses Learn that the Cops Caught ‘the’ Guy.
On a busy street in the middle of the day, Scott W. was working as a construction
worker when he noticed a man in his truck, starting to drive away. Scott ran up to the
truck and tried to block his way, but the driver escaped with the vehicle and his property
inside. Scott told the police that he got a very good look at the culprit, and described him
as a white male in his 30s with reddish facial hair. While Scott was completing his report
to the responding officer, their conversation was interrupted by a radio dispatch
announcing that the truck was recovered, a suspect was in custody, and they wanted the
witness to come make an identification. Scott W. was then driven to the location where
the suspect was detained, and upon arrival at the scene he was given his keys and phone
(which were found in the vehicle), and he could see the stolen truck parked nearby. One
of the officers then read Scott the standard departmental pre-showup admonition, which
instructed the witness not assume the person in custody was the actual culprit. The
witness was then shown the suspect, Darwin Romero, who was handcuffed and in police
custody. Surprisingly, Mr. Romero was not a white male in his 30s with reddish facial
hair, but rather a Hispanic male with black hair and olive skin, who just happened to be in
the park near the stolen vehicle. Despite the fact that the suspect differed so significantly
from his initial description, Scott identified Romero as the thief. To sort out the obvious
discrepancy between his initial description and the suspects appearance, Scott reasoned
that he must have confused the blemishes on his pockmarked face for red facial hair.
At trial, the defense argued Scott W. likely decided Romero must have been the
thief due to the circumstances of his detention (i.e., the appearance that he was found
PREADMONITION SUGGESTION IN LIVE SHOWUPS
4
shortly after the crime with the truck and the property). To counter this argument, the
prosecutor reasoned that reading the admonishment to the witness prior to the field
identification procedure (i.e., the showup) should have provided a safeguard against the
witness assuming that the police must have caught the actual thief (People vs. Darwin
Romero, 2015, San Francisco County, 15022282).
The prosecutor’s argument in this case was certainly not without merit, as two
different meta-analyses have shown that witnesses who are admonished prior to an
identification test are less likely to falsely identify an innocent suspect (Clark, 2005;
Steblay, 1997). However, in actual cases, it is not entirely clear how robust the protective
effect of the admonition is, particularly with showups. The fluid and uncontrolled nature
of the pursuit of a suspect in the field presents many opportunities for witnesses to be
exposed to potentially suggestive information that can raise their expectation that police
have caught the actual perpetrator. Moreover, recent research indicates that suggestions
from the person administering an identification test that the actual culprit is likely present
could dramatically limit the protective qualities of unbiased pre-identification instructions
(Quinlivan, Neuschatz, Cutler, Wells, McClung, & Harker, 2012). Quinlivan and her
colleagues coined the term pre-admonition suggestion to describe this phenomenon.
Pre-admonition Suggestion
Quinlivan et al. (2012) proposed that suggestions made to a witness prior to being
read unbiased pre-lineup instructions could mitigate the prophylactic effect of the
admonition. The logic is straightforward: any suggestion made to a witness indicating
that the actual culprit is present in an identification test runs directly counter to the gist of
the admonition that he/she may or may not be there, and therefore increases the
PREADMONITION SUGGESTION IN LIVE SHOWUPS
5
likelihood that a witness will make an identification, even if recognition is weak or
absent. These investigators argued that unlike the perfunctory reading of pre-lineup
instructions, suggestions made to witnesses prior to the admonition may be viewed as
more personalized and relevant to the case at hand, thus increasing the effect of this
information relative to the admonition. Quinlivan et al. (2012) tested this proposition by
showing participants a video of a mock crime followed by the administration of a culprit-
absent photo array, in which participants were offered a suggestion that the culprit would
be present in the six-pack prior to being instructed in either a biased or unbiased manner.
In this study, the pre-admonition suggestion came in the form of telling witnesses,
“Surely you can pick the perpetrator.” These investigators found that offering this
suggestion, which directly inferred that the actual culprit was likely present in the six-
pack, essentially mitigated the protective effect of the admonition, leading to increased
choosing, and increased confidence in the selections made.
The current study was designed to build on the work done by Quinlivan and her
colleagues by comparing participants who were exposed to the suggestion that the police
believed they had caught the guy before being properly admonished, to those who were
admonished without having heard this suggestion. Also, since these investigators only
used a culprit-absent condition, they could not examine whether the pre-admonition
suggestion would also have increased choosing when the actual culprit was present in the
lineup and the suspect’s match to the witnesses memory was strong. Moreover, the pre-
admonition suggestion used by these investigators was fairly direct, indicating that the
task was easy and the culprit was likely there. The current study examined the effect of a
more subtle form of indirect suggestion that we believe is more likely to occur in actual
PREADMONITION SUGGESTION IN LIVE SHOWUPS
6
investigations (i.e., overhearing radio communications indicating that the police believed
they had caught the actual culprit). In addition, this phenomenon was examined in
showups conducted under highly realistic field conditions.
Showups in the Field
It can be argued that pre-admonition suggestion is an even greater concern in
showups than lineups. With a fair lineup, if a witness does not recognize any of the
suspects, but concludes that the police must have the actual culprit and consequently
decides to make a selection anyway, their errors should be evenly distributed across the
lineup members. However, with a showup, because there is only one person being
presented, the decision to choose will always result in the suspect being picked.
Moreover, recent research indicates that showups conducted in the field may be even
more suggestive than previously thought, and that admonitions may have limited
protective properties in field conditions (Eisen, Smith, Olaguez, & Skerritt-Perta, 2017).
Recently, Eisen et al. (2017) conducted three controlled experiments using a field-
simulation paradigm to compare how witnesses performed at showups when they were
led to believe the showup was part of an actual investigation, compared to conditions in
which the participants were informed the showup was being done for research purposes.
These investigators had participants witness a staged crime (the theft of a laptop
computer) and were immersed in what they were led to believe was an actual police
investigation that involved the pursuit and apprehension of a suspect. In the field-
simulation condition, a uniformed officer responded to the crime scene, and the witnesses
were escorted to a location where local law enforcement conducted a live showup with a
suspect in custody. The lab condition was done without police involvement, and
PREADMONITION SUGGESTION IN LIVE SHOWUPS
7
participants were debriefed before the showup was conducted. Results of these
experiments revealed that the field conditions induced a criterion shift, in which
participants were more likely to make an identification (accurate or false) when they
thought they were involved in an actual investigation, regardless of whether the culprit
was present or not (Experiments 2 & 3).
To explain what was driving this criterion shift, these investigators argued that
when showups are conducted by police in the field, witnesses are faced with a host of
“hot” affective components related to the unique situational factors involved in the event
that can influence their decision-making. Specifically, they argued that presenting a
similarly dressed person in police custody, near the scene, soon after the crime, would
prime witness expectations that the police have likely caught the culprit, which would in
turn increase choosing.
Pressure to choose. Eisen and his colleagues proposed that heightened
expectations triggered by showups conducted in the field would result in witnesses
feeling increased pressure to identify the suspect presented to them. Indeed, data pooled
from over 800 participants across the three field-simulation experiments revealed that
46% of participants in the field condition reported feeling pressured to identify the
suspect, compared to only 28% of participants in the lab condition, who knew the showup
was being conducted solely for research purposes. Eisen and his colleagues postulated
that this increased pressure to choose played a central role in driving participants to adopt
more lenient decision criteria for making their identification decisions. The current study
employed a version of this same field-simulation paradigm to see if pre-admonition
suggestion would result in even greater pressure to identify the suspect.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
8
Instructional bias and suspect similarity. Eisen and his colleagues also
examined the protective effect of reading witnesses an admonition (i.e., may-or-may not
instruction) before conducting a showup (Eisen et al., 2107, Experiment 1). In this
experiment, suspect similarly was manipulated by having the innocent suspect be either
the same height and weight as the thief, or approximately six inches taller and 50 pounds
heavier. Otherwise, they were dressed similarly (i.e., same common baseball cap, jeans,
and black tee shirt), and were both Latino males who were about the same age. Results of
this experiment revealed that when the suspect was a poor match to the culprit in terms of
height and weight, the admonition reduced false identifications in the field-simulation
condition by 31%. However, when the suspect was a good physical match to the culprit
in terms of both clothing and build, the admonition was far less effective, and choosing in
the field condition decreased by only 8%. In essence, when the suspect was a better
match to the culprit, instructions designed to reduce bias had little effect on choosing; but
when the suspect was clearly a poor match, the instructions had a large effect.
Brewer and Wells (2006) also examined the effect of instructional bias with high
and low similar suspects. These investigators created sequential lineups in which the
foils were either high or low in similarity to the culprit. Brewer and Wells found that in
culprit absent conditions, the may-or-may instruction reduced choosing to a greater extent
when the foils were low in similarity to the culprit, compared to when similarity was
high. Taken together, the findings from these experiments suggest that it may be easier to
influence the decision criteria of witnesses when the suspect is a poor match to their
memory for the perpetrator.
Match-to-Memory and Eyewitness Decision Making
PREADMONITION SUGGESTION IN LIVE SHOWUPS
9
Steblay (2013) proposed that when eyewitnesses approach an identification test
they will likely assume the police have caught the actual culprit. If expectations that the
police have the culprit are high, but the witness is presented with a suspect who is a poor
match-to-memory, they may question themselves; doubting their own perceptions in the
face of the contextual factors that indicate that this should be the culprit. Steblay argued
that the absence of recognition in the face of expectations that the actual culprit is likely
present will often prompt witnesses to shift to a more deliberative decision making
process (i.e., secondary process), in which non-memorial information that might help
resolve this dilemma is more likely to be considered. However, when the match-to-
memory is strong (i.e., when the actual culprit or someone who is very similar in
appearance is presented), the identification will most likely be driven by the witness’
internal sense of ecphory. Since recognition memory happens quite quickly (Brewer &
Weber, 2008), identifications made under these conditions are likely to occur with a great
deal of automaticity, and as such, contextual factors that might otherwise be considered
by the witness, are not required, and are therefore less likely to play a role in the
identification decision. This account would explain why unbiased instructions have been
found to raise decision criterion to a greater extent when considering non-similar
compared to similar suspects (Brewer & Wells, 2006; Eisen et al., 2017, Experiment 1).
More recently, Smith, Wells, Lindsay, and Myerson (under review) put forth the
present/absent discrepant-criteria hypothesis to explain how identification decisions made
in culprit-absent conditions are likely different than those made when the culprit is
present. These investigators argue that when a suspect is a weak match to an eyewitness’
memory, as is the case in culprit-absent conditions, witnesses will use a more lenient
PREADMONITION SUGGESTION IN LIVE SHOWUPS
10
decision criterion than when the suspect is a stronger match-to-memory. This new
hypothesis is rooted in work by Verde and Rotello (2007), which indicates that when
match-to-memory is weaker, people set lower decision criteria than when it is stronger.
Verde and Rotello demonstrated this by manipulating the memory strength of word lists
presented in blocks (strong versus weak), and found that when participants encountered
blocks of weak words, they set a lower criterion for choosing then when they encountered
the blocks of words with greater memory strength. Smith and his colleagues noted the
importance of this finding, as it demonstrated that participants were not simply setting a
single criterion apriori, but rather adjusted their criterion at the time of the memory test
based on how well the stimuli matched their memory.
Like Steblay’s dual process account, the present/absent discrepant-criteria hypothesis
predicts that witnesses who are presented with innocent suspects at showups must explain
why they are experiencing such a weak match-to-memory: Is it because the suspect is not
the actual culprit, or is it because their memory is not very good (e.g., maybe they did not
get a good view)? If the witnesses’ expectations that the police must have the right guy
have been elevated by pre-admonition suggestion, but match-to-memory is weak, then
witnesses are more likely to lower their criterion, thus increasing the chances of
choosing. On the other hand, if match-to-memory is strong, it is more likely that the
identification decision will be driven exclusively by the witnesses’ potent experience of
ecphory, and extraneous factors such like instructions and suggestion are less likely to be
considered by the witness.
The Current Study
PREADMONITION SUGGESTION IN LIVE SHOWUPS
11
This study was designed to examine pre-admonition suggestion in showups using
a highly realistic field-simulation paradigm where witnesses were led to believe that the
theft and showup were all part of an actual investigation, and that their identification
would presumably lead to the arrest and prosecution of the suspect. Specifically, we
were interested in the potential suggestive effects of overhearing communications
between officers revealing their belief that they had “caught the guyon making both
accurate and false identifications. The current study builds directly on the work of Eisen
et al. (2017, Experiment 1); but rather than comparing participants who were and were
not admonished, this study examined the effect of pre-admonition suggestion by having
the police admonish all the participants, and then comparing those who were and were
not exposed to suggestive information indicating that the police believed they had caught
the actual culprit.
It was predicted that participants who overheard the suggestive radio call would
be significantly more likely to identify an innocent suspect despite being properly
admonished, and would be more confident in that false identification when compared to
those who were not exposed to the suggestive communication. Moreover, we expected
that the pre-admonition suggestion would be most potent in influencing witnesses when
the suspect differed from the thief in terms of height and weight. However, when the
suspect was a better match to the witnesses’ memory (same clothes and same height and
weight), then choosing would be driven mainly by the better match-to-memory, and the
effect of extraneous factors like the pre-admonition suggestion would be minimized.
Following this same logic, we did not expect that the suggestion would have a significant
effect on accurate identifications in culprit-present conditions, because choosing would
PREADMONITION SUGGESTION IN LIVE SHOWUPS
12
be driven primarily by the very strong match-to-memory for the culprit who was viewed
just 20-minutes earlier.
We also examined other aspects of the witnesses’ experience to see how various
internal pressures may have affected their identification decisions. Specifically, we
predicted that the pre-admonition suggestion would result in participants feeling more
pressure to choose the suspect, and that this effect would be greatest when the culprit was
absent - when the expectations were elevated but the match-to-memory was weakest.
Additionally, it was predicted that the pressure to make an identification would be related
to increased choosing overall. Finally, we examined cross-race effects in a multicultural
urban setting.
Method
Participants
Two hundred thirty-four undergraduate psychology students were recruited at a
state university in Southern California from the participant pool in exchange for course
credit. Participants were 75.1% female and 24.9% male. The sample ranged in age from
1863 years old with a mean age of 21.14 (SD = 4.2). When asked to identify their race,
61.1% identified themselves as Latino, 19.2% identified as Asian, 3.8% identified as
African American, 3.4% identified as Anglo, and 12.4% identified themselves as ‘Other’
or did not specify. This distribution reflects the general makeup of the campus
community and surrounding area.
Procedures
Participants were told this was an experiment on memory and personality. The
procedures were conducted on six separate occasions (once per quarter). Participants
PREADMONITION SUGGESTION IN LIVE SHOWUPS
13
were sequestered in a single room while they waited to participate to make sure they were
not exposed to law enforcement activity or in contact with other students who had already
participated. While in the waiting room, participants completed the NEO-PI. Participants
were escorted in groups of seven to take part in the procedures. The confederate/thief
(hereafter, “the thief”) was embedded in each group as they were escorted to the lab
where the theft was to be staged.
Exposure to the thief and the staged crime. Once they arrived at the lab,
students were seated around a table. The thief always sat at the head of the table and
interrupted the reading of the informed consent by taking a cell phone call, and he
continued talking on his phone for approximately 20 seconds while the experimenter
repeatedly asked him to end the call. This was done to draw attention to himself and
maximize exposure. After taking the call, the thief set the timer on his cell phone for
seven minutes. Immediately after completing the consent procedures, participants were
assigned to work stations around the room where they were administered the Minnesota
Multiphasic Personality Inventory (MMPI-2), via computer. The thief was always seated
at a laptop workstation by the door. Seven minutes after taking the cell phone call, the
thief’s phone timer went off (silently, on vibrate), and he ran out of the room with the
laptop computer he had been using. The experimenter gave chase, and then returned to
the room and stated aloud, “That guy just stole our laptop.” The second experimenter
then said, “You had better call Dr. Eisen.”
The crime report. The experimenter then made a phone call to report the theft in
front of the participants so they could all hear what he/she said. The experimenter
accurately described the thief as a medium-height, medium-weight, Latino male with a
PREADMONITION SUGGESTION IN LIVE SHOWUPS
14
dark shirt, jeans, and a Dodgers cap (baseball hat from the Los Angeles Dodgers
baseball team). The experimenter described the laptop as a “silver Mac.” The
experimenter was prompted on what to say by the party on the other end of the call in
order to ensure uniformity across all trials. The experimenter then ended the call and
stated aloud, “Dr. Eisen will call the police, and he said you guys should finish the
procedures to make sure you get your research credit.”
Ten minutes after the crime was reported, a campus police officer in full uniform
responded to the scene. The officer took brief reports from the research assistants first,
and the participants never got the opportunity to make statements. The research assistants
only reported what was already discussed in front of the group when the crime was called
in (e.g., reiterating that he was a medium height, medium weight, Latino male with a dark
shirt, jeans, and a Dodger cap who ran off with a silver Mac laptop computer during the
procedures).
The Experimental Manipulation
Two minutes into taking the reports from the experimenters, and before the
participants/witnesses could make their reports, the officer received a radio call. His radio
was set to high volume to ensure that all six participants heard the message clearly. In the
control condition, the voice on the radio stated, “The L.A. County Sheriff has detained an
individual who matches the description of the thief and they want you to bring the
witnesses down to the loading dock behind King Hall to make an identification.” In the
experimental condition, the voice on the radio stated, “The L.A. County Sheriff caught
the guy behind the building and they want you to bring the witnesses down to the loading
dock behind King Hall to make an identification.” The Sheriff’s Department headquarters
PREADMONITION SUGGESTION IN LIVE SHOWUPS
15
is less than a mile away, and they have a significant presence on campus. Plus, the
campus was fairly empty on the day of the experiment, adding to the plausibility of the
sheriff’s involvement in catching the thief outside the building. Groups were randomly
assigned to either control (N = 90) or suggestion conditions (N = 99), as every other
group heard either the suggestive or non-suggestive radio call.
The identification procedure. The campus police officer escorted the
participants/witnesses to the ground floor of the building (near the door to the loading
dock) where they met with a uniformed Los Angeles County Sheriff’s Deputy.
The Showup
After instructing the participants as a group, the officer escorted each witness
individually to view the suspect, who was located just outside the building in handcuffs
standing next to another sheriff’s deputy by a patrol car. The participants stood indoors,
and viewed the suspect though two full length glass doors to protect the witness from
being seen by the suspect. The officer took notes during the showup procedure to
document confidence ratings for both positive identifications and correct rejections of the
innocent suspect. Confidence was assessed by asking the witnesses to rate how confident
they were in their identification decision (yes or no) on a 0-100 scale, with 100 being
perfectly confident, and 0 being not confident at all. After making their identification
decision, each witness was escorted to a separate area about 20 yards from the witnesses
who had not yet viewed the suspect.
Similarity: The Innocent Suspect vs. The Thief
PREADMONITION SUGGESTION IN LIVE SHOWUPS
16
Participants were assigned to either culprit-present (N = 84) or absent conditions
(N = 105)
1
. Participants in the culprit-absent conditions were assigned to either similar (N
= 49) or non-similar conditions (N = 56). In the similar condition, the thief was
approximately the same height and weight as the innocent suspect. In the non-similar
condition, the thief and the innocent suspect differed by approximately six-inches and 50
pounds, one being approximately 5' 6" and weighing 150180 pounds, and the other
being approximately 6' 0 tall and weighing 240260 pounds. Otherwise, they were both
Latino males and their clothes were generally similar. Both wore blue jeans and a black
T-shirt, and both were wearing a blue Los Angeles Dodgers baseball cap. Latino males
wearing this standard style Dodger cap on campus is very common. The use of the cap
also controlled for differences in hair styles between the actors.
Groups were randomly assigned to suggestion vs. no suggestion conditions, and
within culprit-absent conditions, similarity was systematically varied. However, target-
present and target-absent trials were done at different times of day (e.g., first half of the
day TA, second half TP). This was done because of logistical issues of getting the actual
thief to the showup with the police after the theft, rather than the innocent suspect.
Stimuli: Thieves and Innocent Suspects
For culprit-absent conditions, three different matched similar/non-similar innocent
suspect/thief pairings were used to conduct the showups for the field condition. In each
case, the pairs were clean-shaven Latino males in their early 20s. One was always
approximately 5’6” and the other was approximately 6’ tall, and they differed in weight
1
Group sizes were uneven due to unpredictable attrition related to participants failing the
manipulation checks or whole groups being dropped due to procedural errors in running
the field-simulation paradigm.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
17
by approximately 50 pounds. This approach both offered the clear advantage of stimulus
variability but also introduced the potential for differences across groups run on different
dates. Although there was expected variation in performance across the different dates
within conditions, the basic pattern of results did not vary across days, and chi-square
analyses revealed no significant differences in choosing within conditions across days
when different actors were used. These same actors were used for culprit-present
conditions.
Administrators: The Police
The teams of sheriff’s deputies who conducted the showups varied across trials
(the same pair was used in the first and second trials, but different pairs of sheriff’s
deputies were used in sessions three through six). Only the officer who responded to the
scene and escorted witnesses to the showups was the same in each trial (A University
Police Lieutenant). Again, this approach both offered the advantage of administrator
variability that one would hope for in a field experiment of this type, but also introduced
the potential for differences across groups run on different dates. As noted above, chi-
square analyses revealed no significant differences in choosing within conditions across
days when different actors and/or officers were used.
Post-Event Questionnaire
After the identification procedures were completed and participants were
debriefed about the deception, participants were asked to complete a questionnaire about
the experience related to the following issues: (1) Their identification decision and
confidence rating; (2) Their belief that the theft and showup procedures were real; (3)
Whether they heard the radio call and could recall the gist of the message; (4) If they had
PREADMONITION SUGGESTION IN LIVE SHOWUPS
18
heard the theft was staged for research; (5) If seeing the suspect in handcuffs made them
assume that that person was guilty; (6) Whether they felt pressured or obligated to
identify the suspect in custody at the showup; and (7) Why they thought the suspect was
or was not the actual culprit in their own words.
Manipulation check and screening for prior knowledge of the staged crime. As
part of the post-event questionnaire, participants were asked if they had heard the radio
call, and if so, they were to write down what they had heard. Ninety-five percent of the
participants (223/234) reported having heard the dispatch and were able to describe the
gist of the communication correctly. Since the experimental manipulation and simulation
as a whole were dependent on hearing the radio call, the 11 participants who reported
either not hearing the dispatch or recalling the details were dropped from all analyses.
Participants were also asked if they had heard from anyone else that the theft and
identification procedures were staged for research. Thirty-three participants reported
having heard about the study before participating, and they were also dropped from the
analyses, creating a sample of 189 participants.
How persuasive was the deception? The post-event questionnaire also asked if the
participants believed the identification procedures were real, or thought it was staged for
research purposes. If participants indicated that they thought the procedures were staged,
they were asked to rate how sure they were that the identification procedures were staged
for research on a scale of 1-100, where 1 meant not sure at all, and 100 meant they were
absolutely sure it was fake. Thirty percent of participants reported that they were more
than 50% sure that the identification procedures were staged (56/189).
PREADMONITION SUGGESTION IN LIVE SHOWUPS
19
Chi-square analyses revealed that participants who reported that they suspected the
procedures were staged were significantly less likely to make an identification X2 (1, 189)
= 13.29, p < .001, Φ = .27 (see Table 1). To account for self-report issues and concerns
about the participants believing it was real, all analyses were conducted twice, once for
the entire sample, and once without the participants who reported suspecting the
procedures were staged for research purposes. Also, when data on the post-event
questionnaire conflicted with the police report, data recorded by the officer was used for
identification decisions and confidence.
Pressure and Obligation to Choose
The post-event questionnaire also asked participants about pressure and/or
obligation they may have felt to identify the suspect. Specifically, they were asked,
“Some people report that when they are shown a person to be identified they feel like
they are being put on the spot, and need to identify the culprit at that moment. Did you
feel any pressure to identify the man being detained by the sheriff at that moment?” Also,
they were asked, “Did you feel obligated to make an identification of the man being
detained by the police?”
Design. A 2 (suggestion: we got the guy vs. no suggestion control) x 2 (culprit
presence: present vs. absent) design was employed. Also, within the culprit-absent
condition, suspect similarity was systematically varied (similar vs. non-similar). Since
more than half of the participants were Latino, ethnicity was dummy coded as Latino vs.
Non-Latino for all analyses to examine potential cross-race effects.
Results
PREADMONITION SUGGESTION IN LIVE SHOWUPS
20
A hierarchical binary logistic regression was conducted with identification
decision as the dependent variable, and the following predictors: suggestion (we got the
guy vs. no suggestion control), target presence (present vs. absent), race (Latino vs.
other), and gender as predictors. In the first block, all main effects were entered. See
Table 1 for choosing rates across all conditions. The predictor model was a significant
improvement over the constant-only model, X2 (4, N = 189) = 24.80, p < .001. Results
revealed that participants who overheard the suggestive radio call were 42% more likely
to identify the innocent suspect (B = -.88, SE = .32, Wald = 7.45, p < .01, eB = .42). Also,
when the actual culprit was presented at the showup, participants were 3.4 times more
likely to make an identification than when an innocent suspect was presented (B = 1.23,
SE = .32, Wald = 14.74, p < .001, eB = 3.42). However, race did not predict choosing (B
= .25, SE = .33, Wald = .54, p = .46, eB = 1.28), and neither did gender (B = .01, SE =
.36, Wald = .001, p = .98, eB = 1.01).
The second block examined two-way interactions between each of the variables
entered. These analyses showed a significant improvement in the model X2 (6, N = 189) =
27.79 p < .001 and revealed a significant interaction between culprit presence and
suggestion (B = 1.91, SE = .75, Wald = 6.47, p = .01, eB = 6.76). Supplemental chi-
square analyses revealed that when the culprit was absent, participants who heard the pre-
admonition suggestion were significantly more likely to mistakenly choose the innocent
suspect (52.0%) than those in the control condition (23.2%), X2 (1, N = 106) = 9.41, p <
.01, ϕ = .30. However, when the culprit was present, participants who heard the pre-
admonition suggestion were not significantly more likely to accurately choose the culprit
(69.4%) than those in the control condition (65.7%), X2 (1, N = 84) = 1.26, p = .72, ϕ =
PREADMONITION SUGGESTION IN LIVE SHOWUPS
21
.04 (see Table 1). These analyses also revealed a significant interaction between race and
suggestion (B = 3.06, SE = .83, Wald = 11.89, p < .01, eB = 21.23). Supplemental chi-
square analyses revealed that when the participant was a different race than the thief and
innocent suspect (Non-Latino), participants who heard the pre-admonition suggestion
were significantly more likely to choose the suspect (71.1%) than those in the control
condition (17.2%), X2 (1, N = 74) = 20.47, p < .001, ϕ =.53. However, when the
participant and innocent suspect were the same race (both Latino), those who heard the
suggestion were not significantly more likely to choose (51.9%) than those in the control
group (50.0%), X2 (1, N = 116) = .40, p = .84, ϕ = .02. In the third block of the
hierarchical regression, three-way interactions were examined. No significant three-way
interactions were observed.
The regression analyses were conducted a second time after excluding the 56
participants in the field condition who reported that they suspected the procedures were
staged for research purposes. These analyses revealed the same pattern of results.
Suspect similarity. The regression analyses were repeated with only the culprit-
absent participants to examine the potential effect of suspect similarity on identifications.
For these analyses, identification decision was the dependent variable, and the same set
of predictors were used (suggestion, race, and gender), except culprit presence was
replaced with suspect similarity (similar vs. non-similar). In the first block of the analysis
all main effects were entered, and the predictor model was found to be a significant
improvement over the constant-only model, X2 (4, N = 103) = 16.76, p = .002. The
participants were 2.4 times more likely to identify the similar suspect who was about the
same height and weight as the culprit than the non-similar suspect (B = .88, SE = .46,
PREADMONITION SUGGESTION IN LIVE SHOWUPS
22
Wald = 3.73, p = .05, eB = 2.41). Also, in this regression, which was restricted to only the
culprit-absent conditions, participants who overheard the suggestive radio call were 22%
more likely to identify the innocent suspect than the participants in the no suggestion
control condition (B = -1.53, SE = .46, Wald = 11.01, p = .001, eB = .22). However,
gender did not predict choosing (B = .34, SE = .51, Wald = .45, p = .50, eB = 1.41), and
neither did race (B = -.72, SE = .49, Wald = 2.19, p = .14, eB = .06). See Table 1 for
choosing rates across all conditions. In the second block, two-way interactions were
entered, and in the third block, three-way interactions were entered. No significant
interactions emerged.
The regression analyses were conducted a second time after excluding the 56
participants in the field condition who reported that they suspected the procedures were
staged for research purposes, and the same pattern of results was revealed.
Confidence
A 2 (suggestion: suggestive radio call vs. no suggestion control) x 2 (culprit
presence: present vs. absent) x 2 (identification decision: yes vs. no) ANOVA was
conducted to examine the relative effects of the suggestion, culprit presence, and
identification decision on witness confidence. Given that suspect similarity was not
perfectly crossed (as similarity was only varied in the culprit-absent condition),
participants were collapsed across similarity for purposes of these analyses. Descriptive
data from these analyses can be found in Table 2. A main effect was revealed for
identification decision, as choosers displayed higher confidence in their identification
decisions, M = 78.99, 95% CI [72.33, 85.66], than non-choosers, M = 39.26, 95% CI
[31.87, 46.65], F(1, 184) = 62.12, p < .001, η2 = .26. However, no difference in
PREADMONITION SUGGESTION IN LIVE SHOWUPS
23
confidence was found between culprit-present or absent identifications, F(1, 184) = 0.55,
p = .46, or decisions made in the suggestive versus non-suggestive conditions, F(1, 184)
= 0.13, p = .72.
These analyses revealed a significant interaction between culprit presence and
identification decision, F(1, 184) = 15.00, p < .001., η2=.08. Not surprisingly, choosers
were far more confident in their decisions when the culprit was present (CP) than when
he was absent (CA). In essence, choosers were more confident in accurate compared to
false identifications. This difference was quite large in the control group (CP: M = 87.82;
CA: M = 62.31, Cohen’s d = 1.25), but was relatively small in the suggestive condition
(CP: M = 85.97, CA: M = 79.89, Cohen’s d = .30).
An interaction between choosing and experimental condition was also revealed,
F(1, 184) = 3.70, p < .06, η2 = .02. Close examination of Table 2 shows that when the
actual culprit was present at the showup, choosers showed comparably high levels of
confidence in both the suggestive (M = 85.97) and control conditions (M =87.82), and
non-choosers showed comparably low levels of confidence in their incorrect rejections
across conditions (Suggestion M = 29.20, Control M = 26.07). However, when an
innocent suspect was presented, participants in the suggestion condition were
substantially more confident in their false identifications (M = 79.89) than the controls
(M = 62.31), Cohen’s d = .71, and were also less confident in their correct rejections than
the controls (Suggestion M = 40.91, Control M = 60.86), Cohen’s d = .47.
These analyses were repeated after removing the 56 participants who reported that
they suspected the procedures were staged for research purposes. Again, a main effect
was revealed for identification decision, as choosers displayed higher confidence in their
PREADMONITION SUGGESTION IN LIVE SHOWUPS
24
identification decisions, M = 77.55, 95% CI [70.57, 84.53], than non-choosers, M =
40.97, 95% CI [32.08, 49.86], F(1, 131) = 41.06, p<.001, η2 = .25, and no differences in
confidence were found between culprit-present or absent identifications, F(1, 131) =
0.95, p = .76, or decisions made in the suggestive versus non-suggestive conditions , F(1,
131) = 0.17, p = .68. Also, the same interactions were observed between culprit presence
and identification decision, F(1, 131) = 6.72, p = .01, η2 = .05, and choosing and
experimental condition F(1, 131) = 5.44, p = .02, η2 = .04.
Pressure to Choose
Chi-square analyses were conducted to examine self-reports of feeling pressured to
make an identification by participants across the experimental conditions. Results of
these analyses revealed that participants in the suggestion condition were more likely to
report feeling pressured to make an identification (43.4%) than were participants in the
control condition (27.8%), X2(1, 189) = 5.02, p < .04, ϕ = .16. However, further analyses
showed that the difference was only evident when the culprit was absent X2(1, 105) =
4.84, p < .03, ϕ = .16, and no relationship was found in culprit-present conditions X2(1,
84) = .77, p <.001, ϕ = .09 (see Table 3). Moreover, participants who were who reported
feeling pressured to choose were more likely to identify the suspect (63.2%), compared to
those who reported no such pressure (43.8%), X2(1, 189) = 6.58, p = .02, ϕ = .19. Again,
further analyses revealed that this difference was only evident when the culprit was
absent X2(1, 105) = 15.32, p < .001, ϕ = .38, and was not significant when in culprit-
present conditions X2(1, 84) = .25, p = .64, ϕ = .06. These analyses were repeated after
removing the doubters, and although participants who felt pressured were still more
PREADMONITION SUGGESTION IN LIVE SHOWUPS
25
likely to choose in culprit-absent conditions, pressure was no longer significantly related
to suggestion.
Discussion
This experiment was designed to examine the effect of overhearing suggestive
information indicating that the officers thought they had “caught the guyon making an
identification at a showup conducted by police in the field. This experiment was
conducted in a highly realistic setting where witnesses were led to believe they were
involved in an actual criminal investigation, and their decision would result in the suspect
being arrested and prosecuted.
As predicted, when the culprit was absent, participants who overheard the
suggestive radio communication indicating that the officers believed they had caught
the guy were substantially more likely to mistakenly identify the innocent suspect
presented at the showup. Table 1 shows that when considering the sample as a whole,
participants who overheard the suggestive radio call falsely identified the innocent
suspect more than twice as often as those in the control condition (52.0% vs. 23.6%), and
the difference was even greater when the suspect was not a good match to the culprit in
terms of height and weight (48.1% vs. 13.8%). These results replicate and extend the
work of Quinlivan et al. (2012), and show that the pre-admonition suggestion effect
generalizes to showups conducted under highly realistic field-simulation conditions, and
is potent even when the suggestion is quite subtle (i.e., overhearing radio
communications between officers). These data highlight important limits to the protective
effect of the admonition, and show that when witnesses are exposed to suggestive
information that runs counter to instructions designed to warn them against assuming the
PREADMONITION SUGGESTION IN LIVE SHOWUPS
26
person being presented is the actual culprit, the cautionary effect of the admonition is
largely eliminated.
Culprit-present choosing. Although overhearing the suggestive radio call indicating
that the police believed they had caught the guyhad a large effect on choosing in
culprit-absent conditions, when the actual culprit was presented at the showup, the effect
was much smaller, and not statistically significant. In essence, the suggestion did not
simply result in a criterion shift that increased choosing equally across the board. Rather,
the suggestion had a differential impact on choosing dependent on the presence of the
culprit.
This predicted pattern of findings fits well with the present/absent discrepant-
criteria hypothesis described earlier (Smith et al., under review), and suggests that when
participants were faced with a suspect who was a poor match-to-memory (i.e., when
presented with an innocent suspect) they may have lowered their criterion for choosing. If
witnesses do in fact lower their criterion for choosing when faced with a suspect who is a
poor match-to-memory, then it may be easier to influence the decision criteria of
eyewitnesses who encounter innocent suspects than it is for those who encounter guilty
suspects. Some evidence of this comes from research on the post-identification feedback
effect (e.g., Bradfield, Wells, & Olson, 2002; Charman & Wells, 2012). Bradfield et al.
(2002) found that confirming feedback increased confidence in innocent suspect
identifications to a greater extent than confidence in culprit identifications. According to
their Selective Cue Integration Framework (SCIF), eyewitnesses use external information
when they do not have a strong match-to-memory experience. Because eyewitness
confidence is really just an extension of an eyewitness’ decision criterion (Wixted &
PREADMONITION SUGGESTION IN LIVE SHOWUPS
27
Gaitan, 2002), the logic of the SCIF can also be applied to an eyewitness’ identification
decision.
Similarity, same race bias, and match-to-memory. If eyewitnesses are more
reliant on external information when match-to-memory is weak, then pre-admonition
suggestion should be most potent when the innocent suspect is clearly a poor match to the
culprit in terms of physical appearance. Indeed, as expected, pre-admonition suggestion
increased choosing significantly in the non-similar condition when the innocent suspect
was dressed similarly, but was otherwise a very poor match to the culprit in terms of
physical build, but not in the similar condition when the thief and innocent suspect were a
good match in regards to both clothing and build. This is consistent with the pattern of
results reported by Eisen et al. (2017, Experiment 1), who found that contextual factors
designed to induce response bias (i.e., not reading the admonition to participants before
the showup), significantly increased choosing in the non-similar, but not the similar
condition. In both the current study and in this previous experiment, when the physical
appearance of the innocent suspect was quite similar to the thief (i.e., same clothes and
same height and weight), contextual factors like pre-identification instructions and pre-
admonition suggestion had a more minimal effect on choosing. However, in the non-
similar condition, when the match-to-memory was designed to be poor, biased
instructions and pre-admonition suggestion significantly increased false identifications.
Cross-race effects. Previous research has shown that people are better at
recognizing faces from their own race compared to other races (Meissner & Brigham,
2001). Therefore, it should logically be more difficult for witnesses to match other race
faces to their memory for the culprit. Although this experiment was conducted in a large
PREADMONITION SUGGESTION IN LIVE SHOWUPS
28
multicultural metropolitan area, a cross-race effect emerged. Specifically, interactions
emerged between race and both suggestion and culprit presence. In culprit-absent
conditions, innocent suspect identifications were significantly higher in the cross-race
group; but, when the culprit was present, accuracy was not affected. Also, when only
considering culprit-absent conditions, pre-admonition suggestion dramatically increased
choosing for the cross-race group (5% to 55%), but had a more modest and non-
significant effect on same race choosing (33% to 50%). Taken together, the pre-
admonition suggestion clearly had its greatest effect in the conditions when match-to-
memory was presumably weakest; when the culprit was absent, when the innocent
suspect was a poor physical match to the culprit, and when the witness was from a
different race.
Pressure to Choose
The results of the current study replicate the findings of Eisen et al. (2017,
Experiment 2), who used the exact same paradigm and also found that participants who
reported feeling pressure to choose were significantly more likely to falsely identify the
innocent suspect in culprit-absent conditions; but pressure was not related to choosing
when the actual culprit was presented at the showup. Moreover, in the current
experiment, this effect was clearest in the suggestion condition. In essence, the pre-
admonition suggestion appeared to heighten the expectation that the police had caught the
actual culprit, and then, when the witness was faced with an innocent suspect who did not
match their memory of the thief, they were put in a rather difficult situation where all of
the contextual factors indicated that he must be the thief, but they were not experiencing a
strong sense of recognition. This situation appeared to induce added pressure to choose,
PREADMONITION SUGGESTION IN LIVE SHOWUPS
29
which made it more difficult to reject the innocent suspect. The confidence data shed
additional light on this pattern of findings.
Confidence
Choosers exhibited heightened confidence across conditions relative to rejecters
regardless of whether they were making accurate or false identifications. This replicates
results reported by Eisen et al. (2017), who observed this same pattern of responding
when using a very similar version of this field-simulation paradigm. Eisen and his
colleagues postulated that heightened expectations provoked by being presented with a
suspect in custody, in what they are lead to believe is an actual police investigation, both
enhanced confidence in identifications, and made witnesses doubt themselves when they
rejected the suspect. These investigators argued that pressure to choose the suspect led to
a criterion shift, resulting in more lenient criterion for choosing and a more stringent
criterion for rejection. In essence, the pressure to choose made it more difficult to reject;
and when they did reject, they did so with lower confidence.
In the current study, when the culprit was present, no difference in confidence
was found between choosers or rejecters across the suggestion and control conditions.
However, when the culprit was absent, participants who overheard the suggestive radio
call were substantially more confident in their identifications compared to those in the
control condition, both when considering the sample as a whole and after removing the
doubters. It appears that overhearing the suggestive radio call not only increased
choosing, but also bolstered the witnesses’ confidence in mistaken identifications. This
replicates the results reported by Quinlivan et al. (2012), who also found that pre-
admonition suggestion bolstered confidence in false identifications.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
30
When examining rejecters in culprit-absent conditions, participants who
overheard the suggestive radio call were less confident in their correct decision to not
identify the innocent suspect compared to controls, both when looking at all participants
and after removing the doubters. It appears that the increased pressure to choose induced
by the pre-admonition suggestion made it more difficult to reject; and when they did
reject, they did so with lower confidence. Overall, overhearing the suggestion that the
police believed they had caught the guy boosted expectations that the police had detained
the actual culprit, leading to increased confidence among choosers and suppressed
confidence among rejecters.
Summary and Conclusions
Overall, these data replicate the work of Quinlivan et al. (2012) and show that the
pre-admonition suggestion effect generalizes to showups conducted under highly realistic
conditions where the witnesses were led to believe that their identifications would lead to
the arrest and prosecution of the suspect. Moreover, these new data indicate that the
dangerous effect of this type of suggestion may be limited to culprit-absent conditions,
and is most likely to increase false identifications when the innocent suspect is a poor
physical match to the culprit and when the witness is from a different race. Under these
conditions, match-to-memory is presumably weakest, and the suspect is more likely to be
influenced by contextual factors surrounding the identification procedures that have the
potential to induce response bias.
Although these data give us a look at how witnesses perform when they learn that
the police think they have a caught the actual culprit under real-world conditions, these
findings are clearly limited to the conditions examined in this experiment. It is possible
PREADMONITION SUGGESTION IN LIVE SHOWUPS
31
that different types of pre-admonition suggestions introduced using other field-simulation
paradigms may yield different results. More field-simulation studies are needed using
different paradigms with more complete random assignment schemes to see how these
findings generalize to other conditions and different types of identification procedures.
PREADMONITION SUGGESTION IN LIVE SHOWUPS
32
References
Bradfield, A.L., Wells, G.L., & Olson, E.A. (2002). The damaging effect of confirming
feedback on the relation between eyewitness certainty and identification accuracy.
Journal of Applied Psychology, 87(1), 112-120.
doi:10.1037//0021=9010.87.1.112
Brewer, N., & Weber, N. (2008). Eyewitness confidence and latency: Indices of memory
processes, not just markers of accuracy. Applied Cognitive Psychology, 22, 827-
840. doi:10.1002/acp.146
Brewer, N. & Wells, G. L. (2006). The confidence-accuracy relationship in eyewitness
identification: Effects of lineup instructions, foil similarity, and target-absent base
rates. Journal of Experimental Psychology: Applied, 12, 11-30,
doi:10.1007/s10979-898X.12.1.11
Charman, S.D., & Wells, G.L. (2012). The moderating effect of ecphoric experience on
post-identification feedback: A critical test of the cues-based inference
conceptualization. Applied Cognitive Psychology, 26(2), 243-250.
doi:10.1002/acp.1815
Clark, S. E. (2005). A re-examination of the effects of biased lineup instructions in
eyewitness identification. Law and Human Behavior, 29(4), 395424.
doi:10.1007/s10979-005-5690-7
Eisen, M. L., Smith, A. S., Olaguez, A., & Skerritt-Perta, A. (2017). An Examination of
Showups Conducted by Law Enforcement Using a Field-Simulation Paradigm
Psychology Public Policy and the Law, 23(1) 1-22.
doi:https://doi.org/10.1037/law0000115
PREADMONITION SUGGESTION IN LIVE SHOWUPS
33
Leippe, M. R., Eisenstadt, D., & Rauch, S. M. (2009). Cueing confidence in eyewitness
identifications: Influence of biased lineup instructions and pre-identification
memory feedback under varying lineup conditions. Law and Human
Behavior, 33, 194212. doi:10.1007/s10979-008- 9135-y
Meissner, C. & Brigham, J. C. (2001). Thirty years of investigating the own-race bias in
memory for faces: A meta-analytic review. Psychology, Public Policy, and the
Law, 7(1), 3-35. doi:10.1037/1076-8971.7.1.3
People vs. Darwin Romero, 2015, San Francisco Superior Court, 15022282
Quinlivan, D. S., Neuschatz, J. S., Cutler, B. L., Wells, G. L., McClung, J., & Harker, D.
L. (2012). Do pre-admonition suggestions moderate the effect of unbiased lineup
instructions? Legal and Criminological Psychology, 17(1) 165176.
doi:10.1348/135532510x533554
Smith, A. M., Wells, G. L., Lindsay, R. C. L., & Myerson, T. (Under review). The
Present/Absent Criterion Discrepancy Hypothesis: Improving the Cost-Benefit
Ratio in Showups with an Additional-Opportunities Instruction.
Steblay, N. (1997). Social influence in eyewitness recall: A meta-analytic review of
lineup instruction effects. Law and Human Behavior, 21(3), 283-297.
doi:10.1023/a:1024890732059
Steblay, N. K. (2013). Lineup instructions. In Cutler, B. L. (Ed.), Reform of Eyewitness
Identification Procedures. (pp. 65 - 86). Washington, DC: American Psychological
Association. doi:10.1037/14094-010
PREADMONITION SUGGESTION IN LIVE SHOWUPS
34
Verde, M.F., & Rotello, C.M. (2007). Memory strength and the decision process in
recognition memory. Memory & Cognition, 35(2), 254-262.
doi:10.3758/bf0319344
Wixted, J.T., & Gaitan, S.C. (2002). Cognitive theories as reinforcement history
surrogates: The case of likelihood ratio models of human recognition memory.
Animal Learning & Behavior, 30(4), 289-305. doi:10.3758/bf03195955
PREADMONITION SUGGESTION IN LIVE SHOWUPS
35
Table 1. Rates of identifications across all conditions.
Suggestion: ‘We Got the Guy’ No Suggestion Control
Target Absent Target Present Target Absent Target Present
% N % N d’ C % N % N d’ C
All participants 52.0% (26/50) 69.4% (34/49) .45 -.27 23.6% (13/55) 65.7% (23/35) .99 .11
Non similar 48.1% (13/27) 69.4% .55 -.23 13.8% (4/29) 65.7% 1.49 .34
Similar 56.5% (13/23) 69.4% .34 -.34 34.6% (9/26) 65.7% .80 -.004
After removing the 56 doubters who suspected the procedures were being done for research purposes
Suggestion: We Got the Guy No Suggestion Control
Target Absent Target Present Target Absent Target Present
% N % N d’ C % N % N d’ C
All participants 56.3% (24/40) 75.0% (29/39) .51 -.42 43.7% (11/31) 65.2% (15/23) .55 .-.12
Non similar 55.0% (11/20) 75.0% .55 -.40 26.7%% (4/17) 65.2% .55 .12
Similar 65.0% (13/20) 75.0% .29 -.53 50.0% (7/14) 65.2% .39 -.20
PREADMONITION SUGGESTION IN LIVE SHOWUPS
36
Table 2. Confidence across all conditions
No Suggestion Control Suggestion: We Got the Guy
Identification Rejection Identification Rejection
M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI
All Participants 78.34 (22.48) 70.62-86.07 54.77 (42.86) 42.84-66.70 83.33 (20.07) 78.15-88.52 35.05 (37.14) 22.84-47.23
Culprit Absent 62.31 (25.30) 47.02-77.60 60.86 (43.38) 47.34-74.38 79.89 (24.34) 70.05-89.72 40.91 (42.08) 22.71-59.10
Culprit Present 87.82 (14.11) 81.56-94.08 29.20 (30.57) 7.33-51.07 85.97 (15.96) 80.40-91.54 26.07 (26.84) 11.20-40.93
No doubters: After removing the 56 participants who suspected the procedures were staged for research
No Suggestion Control Suggestion: We Got the Guy
Identification Rejection Identification Rejection
M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI M (SD) 95% CI
All Participants 73.92 (22.79) 64.72-83.13 52.73 (40.58) 36.35-69.12 83.42 (20.63) 77.73-89.10 34.34 (37.19) 19.32-49.36
Culprit Absent 60.00 (24.57) 43.48-76.52 56.10 (43.04) 35.96-76.24 79.04 (25.02) 68.48-89.61 38.37 (41.77) 16.12-60.62
Culprit Present 84.13 (15.21) 75.71-92.56 41.50 (31.63) 8.031-74.69 87.03 (15.70) 81.06-93.01 27.90 (29.34) 6.92-48.89
PREADMONITION SUGGESTION IN LIVE SHOWUPS
37
Table 3. Percentage of Participants who reported feeling pressured to make an identification
Suggestion: We Got the Guy No Suggestion Control
Identification Rejection All Identification Rejection All
% N % N % N % N % N % N
Culprit Absent All 57.7% (15/26) 33.3% (8/24) 46.0% (23/50) 61.5% (8/13) 14.3% (6/42) 25.5% (14/55)
Absent Similar 69.2% (9/13) 50.0% (5/10) 60.9% (14/23) 77.8% (7/9) 23.5% (4/17) 42.3% (11/26)
Absent Non-Similar 46.2% (6/13) 21.4% (3/14) 33.3% (9/27) 25.0% (1/4) 8.0% (2/25) 10.3% (3/29)
Culprit Present All 35.3% (12/34) 53.3% (8/15) 40.8% (20/49) 34/8% (8/23) 25% (3/12) 31/4% (11/35)
No doubters: After removing the 56 participants who suspected the procedures were staged for research
Suggestion: We Got the Guy No Suggestion Control
Identification Rejection All Identification Rejection All
% N % N % N % N % N % N
Culprit Absent All 58.3% (14/24) 31.3% (5/16) 47.5% (19/40) 63.6% (7/11) 30.0% (6/20) 41.9% (13/31)
Absent Similar 69.2% (9/13) 42.9% (3/7) 60.0% (12/20) 85.7% (6/7) 57.1% (4/7) 71.4% (10/14)
Absent Non-Similar 45.5% (5/11) 22.2% (2/9) 35.0% (7/20) 25.0% (1/4) 15.4% (2/13) 17.6% (3/17)
Culprit Present All 31.0% (9/29) 70.0% (7/10) 41.0% (16/39) 46.7% (7/15) 25% (2/8) 39.1% (9/23)
Article
Full-text available
Objective: The Executive Committee of the American Psychology-Law Society (Division 41 of the American Psychological Association) appointed a subcommittee to update the influential 1998 scientific review paper on guidelines for eyewitness identification procedures. Method: This was a collaborative effort by six senior eyewitness researchers, who all participated in the writing process. Feedback from members of AP-LS and the legal communities was solicited over an 18-month period. Results: The results yielded nine recommendations for planning, designing, and conducting eyewitness identification procedures. Four of the recommendations were from the 1998 article and concerned the selection of lineup fillers, prelineup instructions to witnesses, the use of double-blind procedures, and collection of a confidence statement. The additional five recommendations concern the need for law enforcement to conduct a prelineup interview of the witness, the need for evidence-based suspicion before conducting an identification procedure, video-recording of the entire procedure, avoiding repeated identification attempts with the same witness and same suspect, and avoiding the use of showups when possible and improving how showups are conducted when they are necessary. Conclusions: The reliability and integrity of eyewitness identification evidence is highly dependent on the procedures used by law enforcement for collecting and preserving the eyewitness evidence. These nine recommendations can advance the reliability and integrity of the evidence. Public Significance Statement Mistaken eyewitness identification is a primary contributor to criminal convictions of the innocent. Pristine procedures for collecting and documenting eyewitness identification evidence can help prevent these mistakes. This scientific review paper makes nine system variable recommendations concerning eyewitness identification procedures that should be implemented by crime investigators in eyewitness identification cases.
Article
Full-text available
A general review of how social media searches can affect eyewitness performance. This general review is appropriate for students, lawyers and law enforcement.
Article
Full-text available
Objective: The Executive Committee of the American Psychology-Law Society (Division 41 of the American Psychological Association) appointed a subcommittee to update the influential 1998 scientific review paper on guidelines for eyewitness identification procedures. Method: This was a collaborative effort by six senior eyewitness researchers, who all participated in the writing process. Feedback from members of AP-LS and the legal communities was solicited over an 18-month period. Results: The results yielded nine recommendations for planning, designing, and conducting eyewitness identification procedures. Four of the recommendations were from the 1998 article and concerned the selection of lineup fillers, prelineup instructions to witnesses, the use of double-blind procedures, and collection of a confidence statement. The additional five recommendations concern the need for law enforcement to conduct a prelineup interview of the witness, the need for evidence-based suspicion before conducting an identification procedure, video-recording of the entire procedure, avoiding repeated identification attempts with the same witness and same suspect, and avoiding the use of showups when possible and improving how showups are conducted when they are necessary. Conclusions: The reliability and integrity of eyewitness identification evidence is highly dependent on the procedures used by law enforcement for collecting and preserving the eyewitness evidence. These nine recommendations can advance the reliability and integrity of the evidence. (PsycINFO Database Record (c) 2020 APA, all rights reserved).
Article
Full-text available
While there is a growing body of research examining the relatively “cold,” cognitive decision-making components of showups, few attempts have been made to capture the “hot” affective components of showups that are thought to exacerbate the suggestiveness of the procedure. In 3 simulated-field experiments, we partnered with law enforcement to examine how participants who were led to believe they were involved in an actual criminal investigation (Field-simulation condition) differed from participants who knew they were not part of an actual investigation (Lab-simulation condition). We staged crimes for both conditions, but in the field-simulation condition, law enforcement personnel carried out mock investigations that culminated with a live showup. In Experiment 1 (N = 321), which did not include a culprit-present condition, the field-simulation condition increased innocent suspect identifications. The standard showup admonition decreased innocent suspect identifications, but only for dissimilar innocent suspects. Experiment 2 (N = 196) added a culprit-present condition and found that the field-simulation condition increased innocent suspect and culprit identifications to a similar extent. Experiment 3 (N = 367) replicated the findings of Experiment 2 and examined the impact of admonishing eyewitnesses that if they did not believe the suspect was the culprit, they might have additional opportunities to make an identification. Confidence-accuracy calibration analyses revealed that confidence discriminated accurate from inaccurate identifications in the field, but not in the lab; however, eyewitnesses who made identifications in the field were overconfident and across all levels of confidence were less likely to be correct than eyewitnesses who made identifications in the lab.
Article
Full-text available
We examined the additive and interactive effects of pre‐admonition suggestion and lineup instructions (biased or unbiased) on eyewitness identification rates. Participants watched a mock crime video, completed a target‐absent lineup identification, and completed a retrospective memory questionnaire. Prior to attempting an identification, participants were either exposed or not exposed to pre‐admonition suggestions and received biased or unbiased lineup instructions. The pre‐admonition suggestion indicated that it was likely that the perpetrator was in the lineup (surely, you can pick the perpetrator). The pre‐admonition suggestion increased false identification in the unbiased lineup condition. Furthermore, those who received the pre‐admonition suggestion were more certain in their identifications as well as other testimony‐relevant judgments than were those who did not receive the pre‐admonition suggestion. These results suggest that pre‐lineup suggestion can mitigate the beneficial effects of unbiased lineup instructions.
Article
Full-text available
Confidence inflation from confirming post-identification feedback is greater when the eyewitness is inaccurate than when the eyewitness is accurate, which is evidence that witnesses infer their confidence from feedback only to the extent that their internal cues are weak. But the accurate/inaccurate asymmetry has alternative interpretations. A critical test between these inter-pretations was conducted by including disconfirming feedback conditions. Student participants (n=404) witnessed a mock crime, had either a strong or weak ecphoric experience when making their line-up identifications, and subsequently received no feedback, confirming feedback, or disconfirming feedback. Consistent with a cues-based conceptualization of the feedback effect, disconfirm-ing feedback influenced witnesses with weak ecphoric experiences more than witnesses with strong ecphoric experiences, ironi-cally increasing the confidence-accuracy relation.
Article
Full-text available
The current article reviews the own-race bias (ORB) phenomenon in memory for human faces, the finding that own-race faces are better remembered when compared with memory for faces of another, less familiar race. Data were analyzed from 39 research articles, involving 91 independent samples and nearly 5,000 participants. Measures of hit and false alarm rates, and aggregate measures of discrimination accuracy and response criterion were examined, including an analysis of 8 study moderators. Several theoretical relationships were also assessed (i.e., the influence of racial attitudes and interracial contact). Overall, results indicated a "mirror effect" pattern in which own-race faces yielded a higher proportion of hits and a lower proportion of false alarms compared with other-race faces. Consistent with this effect, a significant ORB was also found in aggregate measures of discrimination accuracy and response criterion. The influence of perceptual learning and differentiation processes in the ORB are discussed, in addition to the practical implications of this phenomenon. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Past research has considered the impact of biased police lineup instructions upon eyewitness lineup performance. Biased instructions either suggest to the eyewitness that the perpetrator is in the lineup or otherwise discourage a no choice response. A meta-analysis of 18 studies was employed to review the hypothesis that biased instructions lead to greater willingness to choose and less accurate lineup identifications than do unbiased instructions. The role of moderating variables in the instruction procedure was also considered. In support of the hypothesis, a significantly higher level of choosing followed biased instructions. Lineup type moderated performance accuracy, however. For target-absent lineups the increased level of choosing following biased instructions resulted in reduced identification accuracy. Biased instructions within a target-present lineup generated a higher level of confidence, but had minimal impact on accuracy. Implications for police practice are discussed.
Article
Full-text available
The authors investigated eyewitnesses' retrospective certainty (see G. L. Wells & A. L. Bradfield, 1999). The authors hypothesized that extemal influence from the lineup administrator would damage the certainty-accuracy relation by inflating the retrospective certainty of inaccurate eyewitnesses more than that of accurate eyewitnesses (N = 245). Two variables were manipulated: eyewitness accuracy (through the presence or absence of the culprit in the lineup) and feedback (confirming vs. control). Confirming feedback inflated retrospective certainty more for inaccurate eyewitnesses than for accurate eyewitnesses, significantly reducing the certainty-accuracy relation (from r = .58 in the control condition to r = .37 in the confirming feedback condition). Double-blind testing is recommended for lineups to prevent these external influences on eyewitnesses.
Article
A meta-analytic review of research comparing biased and unbiased instructions in eyewitness identification experiments showed an asymmetry, specifically that biased instructions led to a large and consistent decrease in accuracy in target-absent lineups, but produced inconsistent results for target-present lineups, with an average effect size near zero (N. M. Steblay, 1997). The results for target-present lineups are surprising, and are inconsistent with statistical decision theories (i.e., D. M. Green & J. A. Swets, 1966). A re-examination of the relevant studies and the meta-analysis of those studies shows clear evidence that correct identification rates do increase with biased lineup instructions, and that biased witnesses make correct identifications at a rate considerably above chance. Implications for theory, as well as police procedure and policy, are discussed.
Article
Despite the well-documented fallibility of eyewitness memories, they influence a variety of judgments and decisions by police, lawyers and the courts. Accordingly, psychologists have tried to identify variables that may help ‘corroborate’ the accuracy of witnesses' testimony. Variables considered by memory researchers to be reliable markers of memory (e.g. confidence and decision latency) have been examined, occasionally endorsed, but more commonly rejected as useful indices of the integrity of witnesses' memories. We examine the history of these attempts to isolate reliable eyewitness memory markers, and why it is that researchers interested in exactly the same issues have reached quite different conclusions about the utility of these memorial markers. We argue that, perhaps in their continuing attempts to find decisive practical solutions, researchers and legal practitioners may have focused too much on techniques or solutions that suggest unreasonable expectations of human memory systems, and too little on refining, and then exploiting, the faithful markers of memory. Copyright © 2008 John Wiley & Sons, Ltd.