Download Psychophysiological correlates of the misinformation effect

PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
1
Psychophysiological correlates of the misinformation effect
Katja Volz *, Rainer Leonhart **, Rudolf Stark ***, Dieter Vaitl *, Wolfgang Ambach *
* Institute for Frontier Areas of Psychology and Mental Health, Wilhelmstraße 3a, D79098 Freiburg, Germany
** Institute of Psychology, University of Freiburg, Engelbergerstraße 41, D-79085
Freiburg, Germany
*** Bender Institute of Neuroimaging, University of Giessen, Otto-Behaghel-Straße
10h, D-35394 Giessen, Germany
Correspondence:
Katja Volz
Institut für Grenzgebiete der Psychologie und Psychohygiene,
Freiburg, Germany
Wilhelmstraße 3a
D-79098 Freiburg
Germany
E-Mail: [email protected]
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
2
Abstract
The misinformation effect refers to memory impairment that arises after
exposure to misleading information (Loftus, 2005, p. 361). The present study focuses
on the peripheral psychophysiology of false memories induced in a misleading
information paradigm. On the basis of Sokolov’s orienting reflex and studies
concerning the Concealed Information Test (CIT, Lykken, 1959), the main hypothesis
assumes differences between true and false memories in terms of the accompanying
autonomic measures. It also is assumed that a cued recall of original information
preceding the recollection phase reduces misinformation effects. Seventy-five
participants watched a video that included nine randomized details. After a tenminute retention phase, the subjects read a narrative text. Six out of the nine details
were replaced by misleading details. Following this, the participants completed a
cued recall task for three of the original items. In a subsequent CIT with truthful
answering electrodermal responses, phasic heart rate, respiration, and response
behavior were measured. Finally, the level of confidence and source monitoring were
assessed. The misinformation effect was replicated with newly developed materials
in three recollection tasks. Cued recall had no influence on the misinformation effect.
Autonomic measures did not differ between true and false memories in the CIT.
Electrodermal responses reflected the subjective importance the participants
attributed to details in the source monitoring task. Therefore, electrodermal
responses are interpreted as a correlate of subjective remembering in a
misinformation paradigm.
Keywords: False
memory,
paradigm,
measures
Concealed
Electrodermal
information
activity,
Cued
test,
Misinformation
recall,
Autonomic
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
1
3
Introduction
Eyewitness testimony is crucial for concluding police investigations (Coupe &
Griffiths, 1996; Paulo, Albuquerque, & Bull, 2013). However, most police officers do
not know that eyewitness memory is often distorted (Kebbell & Milne, 1998). If
memory impairment arises after exposure to misleading information, this is called a
misinformation effect (Loftus, 2005, p. 361). The vulnerability to false memories has
been a focus of memory and forensic research for nearly four decades. However, to
date, there have been few studies investigating the physiological correlates of false
memory (for a review, see Johnson, Raye, Mitchell, & Ankudowich, 2012; Schacter &
Slotnick, 2004). In the present study, we examined peripheral physiological measures
as possible indicators of false memories in a misinformation paradigm. Additionally,
we aimed to reduce misinformation effects using a cued recall procedure.
1.1
The misinformation effect
False memory research dates to the 1970s. In a series of five experiments,
Loftus, Miller, and Burns (1978) evoked false memories of a traffic sign. The threestage procedure applied was named the misinformation paradigm and has since
been used by various research groups (e.g., Belli, Lindsay, Gales, & McCarthy, 1994;
McCloskey & Zaragoza, 1985; Tversky & Tuchin, 1989). In the misinformation
paradigm, the subjects first watch a video or slides typically showing crime or crimerelated plots. After a distractor or retention phase, the researchers introduced
misinformation hidden in a narrative or questions about the event (e.g., “How fast
was the car going when it ran the stop sign?” Loftus et al., 1978, p.19). Finally, the
subjects complete memory tasks about the event. Typically, forced-choice,
recognition, source identification, or level-of-confidence tests are used to gather
memory data (Johnson, Hashtroudi, & Lindsay, 1993; Tversky & Tuchin, 1989). If
misleading information successfully provoked a misinformation effect, this is reflected
in two ways: in reduced recall of original information and enhanced recall of
misleading information (Loftus, 2005).
To date, there is no clear explanation of how false memories emerge in a
misinformation paradigm. Initially, it was assumed that misleading information
replaces the original memory (Loftus, 1979; Loftus & Loftus, 1980). Later, an
integration of misleading and original information into one mixed memory was
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
4
considered (Loftus & Hoffman, 1989). However, the coexistence of original and
misleading information has been demonstrated by several research groups (e.g.,
Bekerian & Bowers, 1983; Belli, 1988; Wright, 1993). Lindsay and Johnson (1989)
assumed that the sources of information are confused during retrieval. In the source
monitoring framework (Johnson et al., 1993; for a summary, see Lindsay, 2008) false
memory occurs when misleading information is misattributed to the source of the
original information. Such source monitoring is driven by judgment processes that
interact with several characteristics of memories, which are typical of a specific
source that a memory could have (Johnson et al., 1993). Based on the assumptions
of the source monitoring framework, the present study employed a cued recall task to
reduce misinformation effects.
1.2
Reduction of misinformation effects
As it is still unclear which exact processes drive the misinformation effect, it is
an open question how it can be reduced reliably. Per the discrepancy detection
principle, warnings before the misleading information (Eakin, Schreiber, & SergentMarshall, 2003; Greene, Flynn, & Loftus, 1982) or misinformation received from an
unreliable source (Bodner, Musch, & Azad, 2009; Dodd & Bradshaw, 1980) can
reduce misinformation effects. However, the evidence the evidence concerning
warnings given in between the misleading information phase and the recollection
phase is still equivocal (for a review, see Blank & Launay, 2014).
The search for a procedure that can be applied after misleading information
was given and that reduces the effect of misleading information on memory, is still
ongoing. In applied forensic research, this is pursued with two approaches: the
Cognitive Interview (CI; Fisher & Geiselman, 1992) and the Self-administered
Interview (SAI; Gabbert, Hope, & Fisher, 2008). Both approaches rely mainly on a
mental reinstatement of the crime (mentally recreating the context of an event, as
well as the physiological, cognitive, and emotional states at the time of an event), that
supports correct source identification (Fisher, Geiselman, & Amador, 1989; Memon &
Bull, 1991; Paulo et al., 2013). Both interviews were investigated in applied contexts
and achieved large effect sizes (Dodson, Powers, & Lytell, 2015; Fisher et al., 1989;
Gabbert et al., 2008; Hope, Gabbert, Fisher, & Jamieson, 2014; Köhnken, Thürer, &
Zoberbier, 1994; Köhnken, Milne, Memon, & Bull, 1999). For example, a recent metaanalysis found average effect sizes of d = 1.20 for comparisons between correctly
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
5
reported details in the Cognitive Interview compared to control interviews (Memon,
Meissner, & Fraser, 2010).
In our study, we employed a simple task preceding the recollection phase that
aimed to strengthen correct source identification in a misinformation paradigm. We
designed the task using the mental reinstatement principle that is utilized in both
interview forms of applied forensic research. The simple procedure used the original
scene in the video as a cue to facilitate correct source identification. This process
resulted in a cued recall task, which will be described later.
1.3
Physiological correlates of false memory
The main goal of our study was to examine peripheral physiological measures
as possible indicators of false memories in a misleading information paradigm. Past
research has mainly comprised studies using functional magnetic resonance imaging
(fMRI), positron emission tomography (PET), or event-related potential (ERP) and
other false memory paradigms (for a review, see Johnson et al., 2012). To the
authors’ knowledge, only one study has used autonomic measures (Baioui, Ambach,
Walter, & Vaitl, 2012). Also, only few studies used the misinformation paradigm (e.g.,
Okado & Stark, 2005). The combination of autonomic measures and the
misinformation paradigm, however, is promising. Autonomic measures might function
as sensible indicators of false memories, which rely on the principles of the orienting
reflex (OR; Sokolov, 1963).
The OR is the physiological, cognitive, and behavioral response to a given
stimulus (Sokolov, 1963). Autonomic measures like electrodermal activity (EDA),
respiration line length (RLL), and phasic heart rate (pHR) are discussed to reflect this
basal process (Sokolov, 1963). The strength of an OR is influenced by the novelty,
intensity, and signficance of the stimulus (Sokolov, 1963). The stimulus significance
is the special importance and meaning a subject attributes to an item (see also
Ambach, Dummel, Lüer, & Vaitl, 2011), and, for this study, the stimulus significance
is crucial for examining the physiological responses to true memories compared with
false memories. The difference between less and highly significant stimuli is wellreflected, particularly by EDA (Barry, 1996). A method that uses autonomic measures
to differentiate between stimuli of different significance is the Concealed Information
Test (CIT; Lykken, 1959).
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
6
The CIT is a well-designed and valid method to detect information using
physiological measures (for a review, see Ben-Shakhar & Elaad, 2003; Meijer, klein
Selle, Elber, & Ben-Shakhar, 2014). The CIT assumes that physiological responses
differ between crime-relevant and crime-irrelevant information if a subject has
knowledge about a crime (Lykken, 1959). Besides other approaches, the OR is
discussed as the main explanation of this difference (Verschuere, Ben-Shakhar, &
Meijer, 2011). The CIT asks several questions referring to different crime-relevant
categories. Typically, a question (e.g., “Was this fruit lying on the window sill?”) is
combined with five items showing possible alternatives. Only subjects with crimerelated knowledge will recognize the right answer and react differently to crimerelated items. In a typical response pattern, test subjects respond to crime-relevant
(significant) items with greater EDA and smaller RLL and pHR (Ambach, Stark,
Peper, & Vaitl, 2008; Ben-Shakhar & Elaad, 2003; Elaad & Ben-Shakhar 2008;
Gamer, Rill, Vossel, & Gödert, 2006; Verschuere, Crombez, de Clercq, & Koster,
2004). The CIT also reflects recognition if the information is not concealed. This
outcome is especially true for EDA, which mainly reflects OR, whereas, pHR and RLL
are also discussed in the light of concealment processes (Ambach et al., 2008; klein
Selle, Verschuere, Kindt, Meijer, & Ben-Shakhar, 2015).
In misinformation paradigms, the recognition of original information is impaired
by misinformation. Based on orienting theory, we assumed that a special importance
and meaning is attributed to the original but not the misleading information; therefore,
original information is more significant to the person than misleading information.
Referring to the typical response patterns in studies dealing with concealed
information, it is assumed that the original information will be accompanied by greater
EDA as well as smaller pHR and RLL responses, compared to the misleading or
unknown information.
Baioui and his colleagues (2012) already examined data on autonomic
measures gathered in a Deese-Roediger-McDermott paradigm (DRM, Deese, 1959;
Roediger & McDermott, 1995). In a DRM paradigm, participants first learn lists of
closely related words (e.g., bed, pillow, sheet) and are asked to recognize the
learned words in an upcoming recognition phase. Often, participants then falsely
remember words related to the categories they have studied before (e.g., sleep)
(Roediger & McDermott, 1995). In contrast to the misleading information paradigm,
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
7
false memories in a DRM paradigm are not evoked by misleading information but by
the activation of a conceptual scheme of the studied items. Baioui and his colleagues
(2012) also suggested that false memories are accompanied by less subjective
importance and meaning and, thus, by a smaller OR in contrast to true memories.
Their results yielded greater EDA responses associated with true memories rather
than false memories. No significant effects of pHR or RLL were found. It is still an
open question whether this response pattern can be replicated and transferred to a
misinformation paradigm.
1.4
Aims of the present study
1.4.1 Methodologically advanced replication of the misinformation effect
A methodologically advanced version of the typical misinformation paradigm
was applied. The original information was presented in a video instead of slides
because of advances in external validity; a video presents motion sequences of
action and is thus easier for a person to perceive than slides, which only show a
series of snapshots (for a review, see Takarangi, Parker, & Garry, 2006).
Additionally, a fully randomized and balanced stimulus set with nine classes of
objects comprising five objects each was created. Regarding the approach of an
enhanced stimulus set by Takarangi and her colleagues (2006), we shot a video and
filmed every scene in five variants, each containing a different critical object of the
same object class. Each subject watched a different video, which was composed
using a randomization scheme. Subjects were misled by a narrative that was also
adjusted by the randomization scheme. Recollection tasks involved recognition, levelof-confidence, source identification, and forced-choice, thereby controlling the effects
of single objects or classes. Furthermore, by using different memory tests and
examining control categories as well as control items in each class of objects, a great
range of memory data could be collected. It was investigated whether the described
materials were appropriate to induce a misinformation effect. Also, all classes of
objects used underwent exploratory analyses to examine which of them scored
highest in evoking false memories.
1.4.2 Reducing misinformation effects by cued recall
As a further question, it is proposed that, per the source monitoring framework,
a cued recall before the recollection phase may reduce misinformation effects. The
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
8
cued recall task was adapted from the concept of mental reinstatement used in
applied forensic research. For this study, the original source of information (i.e.,
visual information from the video) was used to support the recall of the original
information. Successful recall is meant to strengthen the relation between the original
source and the original information, which should reduce source confusion in the
subsequent recollection tasks.
1.4.3 Autonomic correlates of the misinformation effect
A CIT with truthful answering is used to differentiate between the autonomic
responses of different types of items and memories. The core assumption of the
study is that misinformation and false memories are accompanied by decreased
physiological responses of recognition compared to the original information and true
memories. This should be reflected in smaller EDA and greater pHR and RLL
responses. It is further investigated whether the autonomic correlates differ between
source confusion errors (confusing an original source with misleading one) and
correct source identifications.
2
Methods
2.1
Participants
Seventy-five healthy student participants (49 f., 26 m., age: 23.7 ± 2.7 y.) from
different faculties except psychology and neuroscience were recruited via student
services and university bulletins. Written consent was obtained prior to the
experiment, which met all the ethical requirements per the declaration of Helsinki.
Sixteen Euros were paid for participation.
2.2
Design
Three different types of categories were varied within subjects: Categories in
which no misleading information was induced (three control categories) were
compared with categories in which misleading information was introduced (six misled
categories). Whether a cued recall task reduces misinformation effects within the
mislead categories or not, was investigated in three misled categories with cued
recall and three misled categories without cued recall. Each misled category
comprised three different types of items: Original items introduced as original
information in a video, misleading items serving as misleading information in a
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
9
narrative text, and unknown information used as control items in the recollection
phase. Control categories comprised one original and four control items. All factors
were manipulated within-subject.
2.3
Materials and procedure
The memory-related focus of the study was disguised by the cover story ‘moral
sense and cognition’. The experimenter did not mention memory tests in advance.
The video that was used to introduce original information lasted approximately
6.5 minutes and showed a young woman stealing a ring and money prior to a job
interview. Nine particular items, each drawn from a different class of objects per a
randomization procedure described later on, served as the original information.
Object classes were name on a door plate, time, picture of a landscape, box, fruit,
key pendant, color of an envelope, drink, and playing card. As one class of objects
comprised five possible items, each scene showing a relevant object was filmed five
times. Finally, the participants viewed a video composed of a basic plot and nine
randomized video fragments, twice. They were asked to divide it into meaningful
sections and rate every section on a scale of 0 (“morally not reprehensible”) to 100
(“morally reprehensible”).
The subsequent distractor tasks were labeled as ‘cognitive performance
tasks’. Subjects completed a five-minute Stroop-test and a five-minute two-back-task.
Length and difficulty were adapted from former misinformation studies (e.g., Loftus,
1977; McCloskey & Zaragoza, 1985; Belli et al., 1994; Chambers & Zaragoza, 2001;
Wilford, Chan, & Tuhn, 2014).
Subsequently, a narrative text describing the video plot served as
misinformation source: Six out of nine original items were randomly replaced by
misleading items (e.g., “kiwi” replacing “apple”); the remaining three were described
by naming the object class (e.g., “fruit”) and served as control categories. Subjects
were asked to read the narrative text, first divide it into meaningful sections, then rate
every section (analog to the video task) and assign a brief title to each.
Next, the participants received the cued recall disguised as a ‘scenic
imagination’ task. Covering each misled category with cued recall, the participants
viewed a screenshot of the video with the particular item (e.g., an apple lying on the
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
10
window sill) blackened out. They were asked to imagine the video scene as precisely
and in as much detail as possible. Twenty seconds after onset of the screenshot, a
verbal question was asked to indicate the class of objects of the searched item (e.g.,
“Which fruit was lying on the window sill?”), to facilitate the participants’ active recall
of the original item. After a self-chosen delay (30 seconds minimum; no maximum),
the participants wrote down a description of the blackened object and continued with
the next screenshot.
Subsequently, the participants were led to an experimental chamber and
connected to polygraph leads to complete the CIT with truthful answering. The CIT
consisted of nine blocks referring to nine categories. In each block, a question
appeared two seconds before the presentation of a particular item (e.g., “Did she
take the money out of this envelope?”). Items were presented for 10 seconds as
textual pictures (640 x 480 pixels) on a 19-inch monitor at a distance of 90
centimeters, followed by a blank screen lasting for 8 to 10 seconds. Simultaneous
with the presentation of an item, two indication fields (yes and no) prompted the
participant to answer. If no answer was given, after two seconds, the indication fields
disappeared. If an answer was given, the respective field remained visible for the rest
of this trial. The participants were required to indicate whether the item shown was
included in the video by giving a verbal yes or no answer and pressing one of the two
response keys as quickly as possible. The key assignment was balanced between
subjects. For each misled category, one original item, one misleading item, and three
control items were shown one after the other. In control categories, one original and
four control items were shown. The main run was preceded by a training run
consisting of two blocks with five neutral items each. This resulted in a total of 55
item presentations. The first item of a category was always a control item, which was
discarded from the analysis of physiological measures.
The subjects were then detached from the leads and asked to complete three
different memory tests. In all these tests, the items were presented as textual
pictures. First, the subjects were told to indicate on a scale from -3 (“definitely not
seen”) to +3 (“definitely seen”) whether the item appeared in the video (level of
confidence). Second, the participants were instructed to decide if the shown item was
presented in the video, the text, or during the physiological measurement (source
identification). Third, all items of a class of objects were presented simultaneously
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
11
and the participants were asked to choose the item which appeared in the video
(forced choice).
Each participant received an individually cut video, an individual narrative text,
and individual item sequences in all recollection tasks per a randomization scheme.
Randomization procedures warranted that the role of each item as original,
misleading, or control was balanced across subjects. For the CIT, the sequence of
categories was balanced as well as the item sequence within categories; particularly,
the relative position of crime-relevant and misleading items was balanced.
Randomization was also applied to item sequences in the concluding three-fold
memory test.
2.4
Physiological recording
Physiological data was recorded in a dimly lit, electrically, and acoustically
shielded experimental chamber (Industrial Acoustics GmbH, Niederkrüchten,
Germany). The temperature was maintained by air conditioning and was set to
approximately 22.4°C at the beginning with a maximum increase of 0.9°C throughout
the recording. The subjects sat in an upright position; so they could easily reach the
keyboard and watch the 19’-inch’-monitor.
Skin
conductance,
electrocardiogram,
respiratory
activity,
and
finger
plethysmogram were measured. Physiological data were logged using the
Physiological Data System I 410-BCS (J&J Engineering, Poulsbo, Washington), and
converted from analog to digital at a resolution of 14 bits, allowing skin conductance
to be measured with a resolution of 0.01µS. Stimulus on-/offsets and physiological
data were sampled at a rate of 510Hz. Standard Ag/AgCl electrodes (Hellige;
diameter 0.8cm), electrode paste of 0.5% saline in a neutral base (TD 246 Skin
Resistance, Mansfield R&D, St. Albans, Vermont, UK), and a constant voltage of
0.5V were used for recording skin conductance. Electrodes were placed over the
thenar and hypothenar muscles of the non-dominant hand. Electrocardiogram (ECG)
was measured using Hellige electrodes (diameter 1.3cm) according to Einthoven II.
The thoracic and abdominal respiratory activity was registered by two PS-2
biofeedback respiration sensor belts (KarmaMatters, Berkeley, CA) with a built-in
length-dependent electrical resistance. An infrared pulse sensor in a cuff around the
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
12
end phalanx of the middle finger of the non-dominant hand was used to record the
finger plethysmogram.
2.5
Data analysis
The recorded finger plethysmogram was not analyzed because of insufficient
signal quality. Skin conductance data from eight subjects had to be discarded from
the analysis because of electrodermal non-response (> 85% of responses < 0.01µS).
Responses in skin conductance were defined as an increase in conductance that
was initiated within a time period of one to five seconds after image onset. The
amplitude of the response was automatically evaluated as the difference between
response onset and the subsequent maximal value in the set time window (Furedy,
Posner, & Vincent, 1991). Phasic heart rate data from one subject had to be
discarded from analysis because of frequent extrasystoles. The remaining data were
notch filtered at 50 Hz and underwent an automatic R-wave peak detection prior to
visual inspection of the resulting data. The R-R intervals were transformed into heart
rate and real-time scaled (Velden & Wölk, 1987). The heart rate during the last
second before trial onset served as the pre-stimulus baseline. The phasic heart rate
was calculated by subtracting this value from each second-per-second post-stimulus
value. For extracting the trial-wise information of the phasic heart rate, the mean
change in heart rate within 15 seconds after trial onset compared with the prestimulus baseline was calculated (Bradley & Janisse, 1981; Verschuere, Crombez,
Koster, van Bockstaele, & de Clercq, 2007). Respiratory data was manually scanned
and low-pass filtered (10dB at 2.8 Hz) for eliminating the artifacts. A method from
Timm (1982), modified by Kircher and Raskin (2003) was used to process the data:
RLL was computed over a time interval of ten seconds after trial onset. It integrates
information on the frequency and depth of respiration. RLL data from both bolts were
averaged. In all analyses within-subject z-standardized physiological data were used
(Lykken & Venables, 1971; Ben-Shakhar, 1985; Gamer et al., 2006).
2.6
Statistical analysis
All statistical analyses were conducted using SPSS 23 (IBM Corp., Armonk,
NY). A sensitivity analysis was conducted using G*Power (version 3.1.9.2, Faul,
Erdfelder, Lang, & Buchner, 2007). As defined a priori, all subjects with conspicuous
values on at least three depending variables were discarded from analysis.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
13
Conspicuous values were defined as all values more than one-and-a-half interquartile
range above the seventy-fifth or below the twenty-fifth percentile (Tukey, 1977). This
resulted in the exclusion of three subjects. However, no changes in significance
occurred when the analyses were conducted with all subjects. Prior to every analysis,
the requirements were tested: If variables were not distributed normally a Wilcoxontest instead of a t-test for paired samples was used. For Wilcoxon-tests Cohen’s d
(Cohen, 1988) was calculated using the formulas r = z/
=
(Field, 2009, p. 550) and d
(Rosenthal, 1994, p. 239). If the Mauchly’s test indicated a violation of the
assumption of sphericity, the degrees of freedom were corrected according to
Greenhouse-Geisser. All tests were conducted with an alpha level of 0.05; the alpha
level of all comparisons using t-tests was corrected per the Bonferroni-Holm method
(Holm, 1979).
Regarding the analysis of the misinformation effect, the number of affirmations
in the CIT, level of confidence ratings, and the number of correct source
identifications of original items were compared between the control and misled
categories using a Wilcoxon test. It was further analyzed, if in misled categories with
cued recall, source identification errors increased or source identification accuracy
decreased, when compared with misled categories without cued recall. One-way
repeated measurement ANOVAs were used to analyze the effects of the item type
(original vs. misleading vs. control) on the physiological variables. The comparison of
physiological variables between true (affirmed original items) and false (affirmed
misleading items) memories was conducted using a t-test for paired samples. Finally,
the answers in the source identification task were subdivided into confusing (original
items attributed to text; misleading items to video), correct (original items to video;
misleading items to text), and incorrect (original or misleading items to CIT) source
identifications. For all physiological measures a one-way repeated measures ANOVA
was used to compare the source identifications (confusing vs. correct vs. incorrect).
3
Results and discussion
3.1
Replication of the misinformation effect
Regarding the behavioral data, the misinformation effect should be reflected in
three measures: The number of affirmations in the CIT, the level of confidence (LOC)
ratings, and the number of correct source identifications. Descriptive statistics of
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
14
affirmations in the CIT and LOC ratings are provided in Table 1; descriptive statistics
of source identifications are presented in figure 1. In the CIT, 9.4% of misleading
items were (falsely) affirmed (“false memory rate”). 3.0% of control items were
(falsely) affirmed (“basic error rate”). False memory rate exceeded basic error rate, z
= -3.37, p < .001, d = 0.57. In the misled compared with control categories the
original items were less often affirmed, z = -2.60, p = .009, d = 0.64, the level of
confidence for these items was lower, z = -3.19, p = .001, d = 0.81, and the correct
source identification of these items decreased, z = -5.41, p < .001, d = 0.78.
Figure 1. Answers to original items in the source identification task: percentage of correct
(“video”), incorrect (“CIT”), and confusing (“text”) answers, seperately for the three types of
categories (left). Answers to misleading items in the source identification task: percentage of
correct (“text”), incorrect (“CIT”), and confusing (“video”) answers, seperately for the two
types of categories (right). (Error bars represent 95% confidence intervals.)
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
15
Table 1.
Percentage of affirmations in the CIT and mean level of confidence (LOC) ratings, separate
for control, original, and misleading items of the three types of categories.
Category
Item
Control
Control
Misled without cued recall
Original
Control
Original
Misled with cued recall
Misleading
Control
Original
Misleading
M
SEM
M
SEM
M
SEM
M
SEM
M
SEM
M
SEM
M
SEM
M
SEM
CIT
0.030
0.006
0.838
0.026
0.029
0.007
0.773
0.030
0.088
0.020
0.030
0.058
0.743
0.032
0.100
0.021
LOC
1.527
0.066
6.285
0.107
1.505
0.074
6.000
0.125
1.861
0.115
1.491
0.066
5.778
0.147
2.014
0.129
M = mean; SEM = standard error of the mean; LOC rating scale from 1 (“definitely not seen”) to 7 (“definitely
seen”)
This indicates that a misinformation effect was successfully induced, as
reflected in the CIT answers, in the level of confidence, and in the source
identifications. The newly developed materials proved to be suitable in replicating the
misinformation effects, which were repeatable across three different recollection
tasks.
To test for differences between the employed classes of objects, the
descriptive statistics of the misinformation effect (affirmations to original items in
misled vs. control categories in the CIT) and of false memories (affirmations to
misleading items in the CIT) were computed separately, for each of the nine classes.
Adding up both criteria, drink, name on a doorplate, time, fruit, and key pendant
scored highest in distorting memory. This outcome might indicate that memory for
numeric or textual classes (e.g., time or name) may be impaired more easily than for
pictorial classes (e.g., picture). It is tempting to speculate that visual original
information is encoded better, and therefore, is less vulnerable to memory
impairment than textual or numeric information. In line with the previous literature
(Heath & Erickson, 1998; Paz-Alonso & Goodman, 2008) peripheral classes (not
focused in the video, e.g., drink or key pendant) appeared to score higher than the
central classes (e.g., box). In field applications of the CIT, it also is assumed that
memory of places of stolen items (e.g., box or envelope) are typically remembered
better than peripheral details (e.g., time) (Nakayama, 2002; Osugi, 2011).
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
3.2
16
Influence of the cued recall on the misinformation effect
In misled categories with cued recall, the original and misleading items were
not significantly better attributed to the correct source compared to the misled
categories without cued recall, z = -.726, p = .468, d = 0.16. There is no difference
between categories with or without cued recall with respect to the occurrence of
source-confusing errors (original items attributed to the text or misleading items
attributed to the video), z = -.061, p = .952, d = 0.01. This outcome indicates that
there is no influence of the cued recall task on the misinformation effect.
Evidence from former attempts made to decrease misinformation effects is equivocal.
Priming of original information, but not of neutral or misleading information alters
misinformation effects (Gordon & Shapiro, 2012). In some studies, warnings before
the recollection phase reduced misinformation effects (Bodner et al., 2009; Eakin et
al., 2003; Echterhoff, Hirst, & Hussy, 2005), in others they did not (Meade &
Roediger, 2002; for a review see Blank & Launay, 2014). Echterhoff and his
colleagues (2005) concluded that there were cognitive processes that could be
influenced prior to the recollection to decrease false memories. The cued recall task
may not have influenced the cognitive processes in a way that neither improved nor
impaired memory recollection. The instructions and visual cues used (i.e.,
screenshots of the video) were apparently not sufficient to support the strengthening
of the original source memory. Rather, it may have acted as a first recollection task.
On summarizing, there is no reliable method yet, to reduce misinformation effects
prior to the recollection phase. In the current state of research, the mechanisms
leading to false memory are still unclear. No cognitive processes are known that may
explain the majority of the mixed study results. The results of our study, therefore,
numbers among in the mixed state of research concerning the mechanisms and
decrease of the misinformation effect.
3.3
Autonomic correlates of the misinformation effect
Table 2 provides an overview of the raw physiological data separated by item
type. Figure 2 shows the temporal course of the phasic heart rate separately for each
item type. In all following analyses, within-subject z-standardized physiological data
were used (Lykken & Venables, 1971; Ben-Shakhar, 1985; Gamer et al., 2006). A
significant effect of item type was found for EDA, F (1.76, 66) = 80.89, p < .001,
=
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
17
0.55, but it was not found for pHR, F (1.81, 70) = 0.98, p = .371,
= 0.01, or for
RLL, F (1.63, 70) = 0.02, p = .976,
= 0.00. The original items were accompanied
by greater EDA responses when compared with the control items, t (65) = 12.06, p <
.001, d = 2.63, and when compared with the misleading items, t (65) = 8.46, p < .001,
d = 1.69. The misleading items were associated with greater EDA responses in
contrast to the control items, t (65) = 2.61, p = .011, d = 0.50.
Figure 2. Temporal course of the phasic heart rate separately for each item type.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
18
Table 2.
EDA, pHR, and RLL separated for each item type
Item type
Original
M
EDA [nS]
pHR [BPM]
SEM
Misleading
M
SEM
Control
M
SEM
229.21 29.78 137.67 23.94 115.64 19.50
0.49
0.23
0.79
0.21
0.63
0.18
RLL [arb. units] 371.99 15.78 386.75 19.54 380.32 17.29
M = mean; SEM = standard error of the mean
This indicates that the typically observed CIT effect (i.e., response differences
between the known original and unknown control items) is replicated for EDA, but not
for pHR or for RLL. These findings are in line with the former evidence from CIT
studies: When subjects answer truthfully instead of deceptively, the RLL and pHR
typically show much smaller effects, whereas, EDA shows equally large effects
(Ambach et al., 2008; klein Selle et al., 2015). In those studies, it was argued that
pHR and RLL do not reflect orienting, but rather inhibition or concealment-related
processes. As a result, EDA can be regarded as the most sensible parameter in this
study.
As a new finding, EDA responses to the misleading items were greater than
those to the control items, but smaller than those to the original items. In line with our
hypothesis, the participants showed greater ORs to the original items than to the
misleading ones. The original items must have gained special importance and
meaning for the individual participant but not the misleading ones. A possible
explanation is that the visual encoding of the original items might have been better
than the textual encoding of the misleading items. It was found earlier that visual
encoding exceeds textual encoding in depth, although textual encoded items were
treated as familiar (Dodson & Markham, 1993). Furthermore, the answer given in the
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
19
CIT is an important confound of this finding: More original than misleading items were
affirmed. Consequently, a direct comparison of the affirmed original items and
affirmed misleading items was conducted.
For the following analysis, false memories in the CIT refer to trials in which a
misleading item was affirmed, whereas, correct memories refer to affirmations of
original items. We selected participants who affirmed at least one misleading item
and had valid physiological data; this reduced the sample size to 26 (EDA) and 31
(pHR, RLL), respectively. False memories compared to correct memories were not
accompanied by smaller EDA, t (25) = 0.54, p = .298, d = 0.16 (see Figure 3), or by
greater pHR, t (30) = 0.52, p = .305, d = 0.14, or by greater RLL responses, t (30) =
0.46, p = .328, d = 0.14. A sensitivity analysis was conducted using G*Power (version
3.1.9.2). For a two-tailed t-test for paired samples, assuming a power of 95 %, an
alpha level of 5 %, and a sample size of 26, only effects of d = 0.74 or greater could
be found, with a probability of 95 %. Hence, more participants showing false
memories and more false memories per participant would have yielded more
meaningful results.
Figure 3. Z-transformed EDA of false (affirmed misleading items) and true memories
(affirmed original items) in the CIT. (Error bars represent 95% confidence intervals.)
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
20
The lack of differences in the physiological correlates between false and
correct memories is contrary to the findings of Baioui and his colleagues (2012) who
found greater EDA responses associated with true recognition than with false
recognition in a Deese-Roediger-McDermott (DRM) paradigm. It was concluded that
a physiological differentiation of falsely recognized and truly recognized items is
possible. If so, the EDA would reflect the objective knowledge rather than the
subjective belief of having it. Our findings, however, indicate that no differentiation
between false and correct memories is possible by EDA in the misinformation
paradigm. It has to be considered that false memories in the DRM paradigm are
different from false memories induced by misinformation: False memories in the
misinformation studies refer to previously encoded items for which a memory trace
was created. In contrast, false memories in DRM studies occur without such
encoding and memory trace. Assuming that EDA reflects not only the subjective
feeling of recognition but also the objective existence of a memory trace would help
to resolve the apparent contradiction between the findings of both studies. In
misinformation studies, it is debated whether source confusion is involved in the
formation of false memories.
Hence, differences in physiological data between different source identification
types were of special interest. In the source identification task, answers to original
and misleading items were assigned to three categories: Correct (original item
attributed to video, misleading item to text), incorrect (original or misleading item
attributed to CIT), and confusing source identifications (original items attributed to
text; misleading items to video). The physiological data gathered in the CIT then was
assigned to the different types of source identification and compared. A significant
effect of source identification type was found for EDA, F (1.19, 25) = 7.84, p = .007,
= 0.25 (see Figure 4), but not for pHR, F (1.34, 28) = 1.30, p = .275,
for RLL, F (1.70, 29) = 1.10, p = .334,
= 0.05, or
= 0.04. Compared with incorrect source
identifications, greater EDA responses were found for correct source identifications, t
(62) = 10.57, p < .001, d = 1.97, as well as for confusing source identifications, t (22)
= -2.41, p = .025, d = 0.77. Confusing source identifications did not differ from correct
source identifications, t (23) = 0.37, p = .718, d = 0.11.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
21
Figure 4. Z-transformed EDA of correct, incorrect, and confusing source identifications. (Error
bars represent 95% confidence intervals.)
In summary, greater EDA responses were observed when items, be they
original or misleading, were judged as presented in the video or text, compared to
when items were reported as first seen in the CIT. If, however, original or misleading
items were attributed as first seen in the CIT, they were not recognized as being
significant. Although these items supposedly had gained special meaning during
encoding, the orienting response, which was reflected in the EDA responses,
significantly decreased with this misattribution. This indicates that EDA reflects
subjective identification of items as important rather than their objective importance.
3.4
Limitations
One main limitation restricts the amount of conclusions drawn by this study:
For analyzing physiological correlates of false memories, the number of falsely
affirmed misleading details and the number of source confusions between original
and misleading details was relatively small, resulting in a limited test power. In further
studies, more false memories should be induced so that differences in the
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
22
accompanying autonomic measures can be tested with more statistical power. First,
this may be accomplished by optimizing the retention interval. Extending the duration
of the retention interval to one or two weeks might be preferable over distractor tasks,
which were commonly used and also employed in this study. With increasing
retention intervals, higher false memory rates were observed (Pansky, Tenenboim, &
Bar, 2011; Wilford et al., 2014). Additionally, the number of original details presented
in the video should be increased, as meta-analytic studies regarding the CIT showed
that physiological effects increase in power with increasing the number of stimulus
categories (Ben-Shakhar & Elaad, 2003). Furthermore, the details used as original
information should be presented in the video as peripheral instead of central details
(Heath & Erickson, 1998; Paz-Alonso & Goodman, 2008). In this context, it remains
unanswered whether textual or numeric details might be superior to pictorial
information. A replication study is needed to face the main limitation of low statistical
power. Considering the above suggestions might particularly allow differentiating the
EDA correlates of subjective versus objective components of recognition.
Another limitation of the study is the usage of more control than memorized
items. Similar to other studies using the CIT, memorized items (i.e., original or
misleading items) constitute a relative oddball among the more frequent control
items. Differences in physiological responding between control items and memorized
items can be due to any combination of these effects. Yet, this possible confound is
limited to comparisons including control items.
3.5
Conclusion
The present study replicated the misinformation effect with newly developed
study materials. Three different recollection tasks showed similar misinformation
effects. A randomization and balancing scheme methodologically improved the
standard misinformation procedure. With regard to the expected decrease of
misinformation effects by a cued recall task, the procedure employed in this study did
not facilitate the correct source identification to a sufficient degree. It remains
questionable whether cued recall is principally able to reduce misinformation effects
or not. The typical CIT effect was replicated for EDA but not for pHR or RLL, which is
in line with the current understanding of sub-processes ongoing in the CIT: EDA is
discussed as reflecting orienting, whereas, pHR and RLL are related to concealment
and inhibitory processes. As the source identification task revealed, EDA response
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
23
relates to the subjective importance a participant attributes to an item rather than its
actual importance. In summing up, combining the misinformation paradigm with the
CIT proved suitable to examine the psychophysiology of false memories.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
24
References
Ambach, W., Stark, R., Peper, M., & Vaitl, D. (2008). Separating deceptive and
orienting components in a Concealed Information Test. International Journal of
Psychophysiology, 70 (2), 95–104. doi: 10.1016/j.ijpsycho.2008.07.002
Ambach, W., Dummel, S., Lüer, T., & Vaitl, D. (2011). Physiological responses in a
Concealed Information Test are determinded interactively by encoding.
International Journal of Psychophysiology, 81 (3), 275-282. doi:
10.1016/j.ijpsycho.2011.07.010
Baioui, A., Ambach, W., Walter, B., & Vaitl, D. (2012). Psychophysiology of false
memories in a Deese-Roediger-McDermott paradigm with visual scenes. PloS
one, 7 (1). Online Publication. doi: 10.1371/journal.pone.0030416
Barry, R. J. (1996). Preliminary process theory: Towards an integrated account of the
psychophysiology of cognitive processes. Acta Neurobiologiae Experimentalis,
56 (1), 469–484.
Bekerian, D. A., & Bowers, J. M. (1983). Eyewitness testimony: Were we misled?
Journal of Experimental Psychology: Learning, Memory and Cognition, 9 (1),
139-145. doi: 10.1037/0278-7393.9.1.139
Belli, R. F. (1988). Color blend retrievals: Compromise memories or deliberate
compromise responses? Memory & Cognition, 16 (4), 314–326
Belli, R. F., Lindsay, D. S., Gales, M. S., & McCarthy, T. T. (1994). Memory
impairment and source misattribution in postevent misinformation experiments
with short retention intervals. Memory & Cognition, 22 (1), 40–54. doi:
10.3758/BF03197042
Ben-Shakhar, G. (1985). Standardization within individuals: A simple method to
neutralize individual differences in skin conductance. Psychophysiology, 22
(3), 292-299. doi: 10.1111/j.1469-8986.1985.tb01603.x
Ben-Shakhar, G., & Elaad, E. (2003). The validity of psychophysiological detection of
information with the Guilty Knowledge Test: A meta-analytic review. Journal of
Applied Psychology, 88 (1), 131-151. doi: 10.1037/0021-9010.88.1.131
Blank, H., & Launay, C. (2014). How to protect eyewitness memory against the
misinformation effect. A meta-analysis of post-warning studies. Journal of
Applied Research in Memory and Cognition, 3 (2), 77–88. doi:
10.1016/j.jarmac.2014.03.005
Bodner, G. E., Musch, E., & Azad, T. (2009). Reevaluating the potency of the
memory conformity effect. Memory & Cognition, 37 (8), 1069-1076. doi:
10.3758/MC.37.8.1069
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
25
Bradley, M. T., & Janisse, M. P. (1981). Accuracy demonstrations, threat, and the
detection of deception: Cardiovascular, electrodermal, and pupillary measures.
Psychophysiology, 18 (3), 307-315. doi: 10.1111/j.1469-8986.1981.tb03040.x
Chambers, K. L., & Zaragoza, M. S. (2001). Intended and unintended effects of
explicit warnings on eyewitness suggestibility: Evidence from source
identification tests. Memory & Cognition, 29 (8), 1120-1129. doi:
10.3758/BF03206381
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. Hillsdale:
Erlbaum.
Coupe, T., & Griffiths, M. (1996). Solving residential burglary (Crime Detection and
Preventions Series No 77). London: Home Office.
Deese, J. (1959). On the prediction of occurrence of particular verbal intrusions in
immediate recall. Journal of Experimental Psychology, 58 (1), 17-22. doi:
10.1037/h0046671
Dodd, D. H., & Bradshaw, J. M. (1980). Leading Questions and Memory: Pragmatic
Constraints. Journal of Verbal Learning and Verbal Behavior, 19 (6), 695-704.
doi: 10.1016/S0022-5371(80)90379-5
Dodson, M., & Markham, R. (1993). Imagery ability and source monitoring:
Implications for eyewitness memory. British Journal of Psychology, 84 (1),
111-118. doi: 10.1111/j.2044-8295.1993.tb02466.x
Dodson, C. S., Powers, E., & Lytell, M. (2015). Aging, confidence, and
Misinformation: recalling information with the Cognitive Interview. Psychology
and Aging, 30 (1), 46-61. doi: 10.1037/a0038492
Eakin, D. K., Schreiber, T. A., & Sergent-Marshall, S. (2003). Misinformation effects
in eyewitness memory: The presence and absence of memory impairment as
a function of warning and misinformation accessibility. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 29 (5), 813–825. doi:
10.1037/0278-7393.29.5.813
Echterhoff, G., Hirst, W., & Hussy, W. (2005). How eyewitnesses resist
misinformation: Social postwarnings and the monitoring of memory
characteristics.
Memory
&
Cognition,
33
(5),
770-782.
doi:
10.3758/BF03193073
Elaad, E., & Ben-Shakhar, G. (2008). Covert respiration measures for the detection
of concealed information. Biological Psychology, 77 (3), 284-291. doi:
10.1016/j.biopsycho.2007.11.001
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power: A flexible
statistical power analysis program for the social, behavioral, and biomedical
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
sciences. Behavior
10.3758/BF03193146
Research
Methods,
29
(2),
26
175-191.
doi:
Field, A. (2009). Discovering statistics using SPSS. London: SAGE.
Fisher, R. P., & Geiselman, R. E. (1992). Memory-enhancing techniques for
investigative interviewing: The Cognitive Interview. Springfield: Charles C.
Thomas.
Fisher, R. P., Geiselman, R. E., & Amador, M. (1989). Field test of the Cognitive
Interview: Enhancing the recollection of actual victims and witnesses of crime.
Journal of Applied Psychology, 74 (5), 722-727. doi: 10.1037/00219010.74.5.722
Furedy, J.J., Posner, R. T., & Vincent, A. (1991). Electrodermal differentiation of
deception: Perceived accuracy and perceived memorial content. International
Journal of Psychophysiology, 11 (1), 91-97. doi: 10.1016/01678760(91)90376-9
Gabbert, F., Hope, L., & Fisher, R. P. (2008). Protecting eyewitness evidence:
Examining the efficacy of a Self-Administered Interview tool. Law and Human
Behavior, 33 (4), 298-307. doi: 10.1007/s10979-008-9146-8
Gamer, M., Rill, H., Vossel, G., & Gödert, H. W. (2006). Psychophysiological and
vocal measures in the detection of guilty knowledge. International Journal of
Psychophysiology, 60 (1), 76-87. doi: 10.1016/j.ijpsycho.2005.05.006
Gordon, L. T., & Shapiro, A. M. (2012). Priming correct information reduces the
misinformation effect. Memory & Cognition, 40 (5), 717-726. doi:
10.3758/s13421-012-0191-7
Greene, E., Flynn, M. S., & Loftus, E. F. (1982). Inducing resistence to misleading
information. Journal of verbal Learning and verbal Behavior, 21 (2), 207–219.
doi: 10.1037/0096-3445.114.1.1
Heath, W. P., & Erickson, J. R. (1998). Memory for central and peripheral actions and
props after varied post-event presentation. Legal and Criminological
Psychology, 3 (2), 321-346. doi: 10.1111/j.2044-8333.1998.tb00369.x
Holm, S. (1979). A simple sequentially rejective multiple test procedure.
Scandinavian Journal of Statistics, 6 (2), 65-70.
Hope, L., Gabbert, F., Fisher, R. P., & Jamieson, K. (2014). Protecting and
enhancing eyewitness memory: The impact of an initial recall attempt on
performance in an investigative interview. Applied Cognitive Psychology, 28
(3), 304-313. doi: 10.1002/acp.2984
Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source Monitoring.
Psychological bulletin, 114 (1), 3–28. doi: 10.1037/0033-2909.114.1.3
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
27
Johnson, M. K., Raye, C. L., Mitchell, K. J., & Ankudowich, E. (2012). The Cognitive
neuroscience of true and false memories. In Belli, R. F. (Ed.) Nebraska
symposium on motivation (p. 15-52). doi: 10.1007/978-1-4614-1195-6_2
Kebbell, M. R., & Milne, R. (1998). Police officers‘ perceptions of eyewitness
performance in forensic investigations. The Journal of Social Psychology, 138
(3), 323-330. doi: 10.1080/00224549809600384
Kircher, J. C., & Raskin, D. C. (2003). The Computerized Polygraph System II
(Software Version 4.01). Salt Lake City, UT: Scientific Assessment
Technologies.
klein Selle, N., Verschuere, B., Kindt, M., Meijer, E. H., & Ben-Shakhar, G. (2015).
Orienting versus inhibition in the Concealed Information Test: Different
cognitive processes drive different physiological measures. Psychophysiology,
53 (4), 579-590. doi: 10.1111/psyp.12583
Köhnken, G., Thürer, C., & Zoberbier, D. (1994). The Cognitive Interview: Are the
interviewers‘ memories enhanced, too? Applied Cognitive Psychology, 8 (1),
13-24. doi: 10.1002/acp.2350080103
Köhnken, G., Milne, R., Memon, A., & Bull, R. (1999). The Cognitive Interview: A
meta-analysis. Psychology, Crime and Law, 5 (1-2), 3-27. doi:
10.1080/10683169908414991
Lindsay, D. S., & Johnson, M. K. (1989). The eyewitness suggestibility effect and
memory for source. Memory & Cognition, 17 (3), 349–358. doi:
10.3758/BF03198473
Lindsay, D. S. (2008). Source Monitoring. In H. L. Roediger, III (Ed.), Cognitive
psychology of memory. Vol. 2 of Learning and Memory: A comprehensive
reference
(J.
Byrne,
Ed.)
(p.
325-347).
Retrieved
from:
http://web.uvic.ca/~dslind/?q=node/128
Loftus, E. F. (1977). Shifting human color memory. Memory & Cognition, 5 (6), 696–
699. doi: 10.3758/BF03197418
Loftus, E. F. (1979). Eyewitness Testimony. Cambridge: University Press.
Loftus, E. F. (2005). Planting misinformation in the human mind: A 30-year
investigation of the malleability of memory. Learning & Memory, 12 (4), 361–
366. doi: 10.1101/lm.94705
Loftus, E. F., & Hoffman, H. G. (1989). Misinformation and memory: The creation of
new memories. Journal of Experimental Psychology: General, 118 (1), 100104. doi: 10.1037/0096-3445.118.1.100
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
28
Loftus, E. F., & Loftus, G. R. (1980). On the permanence of stored information in the
human brain. American Psychologist, 35 (5), 409–420. doi: 10.1037/0003066X.35.5.409
Loftus, E. F., Miller, D. G., & Burns, H. J. (1978). Semantic integration of verbal
information into a visual memory. Journal of Experimental Psychology: Human
Learning and Memory, 4 (1), 19-31. doi: 10.1037/0278-7393.4.1.19
Lykken, D. T. (1959). The GSR in the detection of guilt. Journal of Applied
Psychology, 43 (6), 385-388. doi: 10.1037/h0046060
Lykken, D. T., & Venables, P. H. (1971). Direct measurement of skin conductance: A
proposal for standardization. Psychophysiology, 8 (5), 656-672. doi:
10.1111/j.1469-8986.1971.tb00501.x
McCloskey, M., & Zaragoza, M. (1985). Misleading postevent information and
memory for events: Arguments and evidence against memory impairment
hypotheses. Journal of Experimental Psychology: General, 114 (1), 1–16. doi:
10.1037/0278-7393.13.1.36
Meade, M. L., & Roediger, H. L. (2002). Explorations in the social contagion of
memory. Memory & Cognition, 30 (7), 995-1009. doi: 10.3758/BF03194318
Meijer, E. H., klein Selle, N., Elber, L., & Ben-Shakhar, G. (2014). Memory detection
with the Concealed Information Test: A meta-analysis of skin conductance,
heart rate, and P300 data. Psychophysiology, 51 (9), 879-904. doi:
10.1111/psyp.12239
Memon, A., & Bull, R. (1991). The Cognitive Interview: Its origins, empirical support,
evaluation and practical implications. Journal of Community & Applied Social
Psychology, 1 (4), 291-307. doi: 10.1002/casp.2450010405
Memon, A., Meissner, C. A., & Fraser, J. (2010). The Cognitive Interview: A metaanalytic review and study space analysis of the past 25 years. Psychology,
Public Policy, and Law, 16 (4), 340-372. doi: 10.1037/a0020518
Nakayama, M. (2002). Practical use of the Concealed Information Test for criminal
investigation in Japan. In Kleiner, M. (ed.), Handbook of polygraph testing (p.
49-86). San Diego: Academic Press.
Okado, Y., & Stark, C. E. L. (2005). Neural activity during encoding predicts false
memories created by misinformation. Learning & Memory, 12 (1), 3-11. doi:
10.1101/lm.87605
Osugi, A. (2011). Daily application of the Concealed Information Test: Japan. In
Verschuere, B., Ben-Shakhar, G., & Meijer, E. (eds.), Memory Detection:
Theory and application of the Concealed Information Test (p. 253-275).
Cambridge: University Press.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
29
Pansky, A., Tenenboim, E., & Bar, S. K. (2011). The misinformation effect revisited:
Interactions between spontaneous memory processes and misleading
suggestions. Journal of Memory and Language, 64 (3), 270-287. doi:
10.1016/j.jml.2010.12.003
Paulo, R. M., Albuquerque, P. B., & Bull, R. (2013). The enhanced Cognitive
Interview. Towards a better use and understanding of this procedure.
International Journal of Police Science & Management, 15 (3), 190-199. doi:
10.1350/ijps.2013.15.3.311
Paz-Alonso, P. M., & Goodman, G. S. (2008). Trauma and memory: Effects of postevent misinformation, retrieval order, and retention interval. Memory, 16 (1),
58-75. doi: 10.1080/09658210701363146
Roediger, H. L. & McDermott, K. B. (1995). Creating False Memories: Remembering
words not presented in lists. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 21 (4), 803-814. doi: 10.1111/1467-9280.00004
Rosenthal, R. (1994). Parametric measures of effect size. In Cooper, H. & Hedges,
L.V. (Eds.), The handbook of research synthesis (S. 232-244). New York:
Russell Sage Foundation.
Schacter, D. L., & Slotnick, S. D. (2004). The cognitive neuroscience of memory
distortion. Neuron, 44 (1), 149-160. doi: 10.1016/j.neuron.2004.08.017
Sokolov, E. N. (1963). Higher nervous functions: The Orienting Reflex. Annual
Review
of
Psychology,
25,
545–580.
doi:
10.1146/annurev.ph.25.030163.002553
Takarangi, M. K. T., Parker, S., & Garry, M. (2006). Modernising the misinformation
effect: Development of a new stimulus set. Applied Cognitive Psychology, 20
(5), 583-590. doi: 10.1002/acp.1209
Timm, H. W. (1982). Effect of altered outcome expectancies stemming from placebo
and feedback treatments on the validity of the guilty knowledge technique.
Journal of Applied Psychology, 67 (4), 391-400. doi: 10.1037/00219010.67.4.391
Tukey, J. W. (1977). Exploratory data analysis. Reading, MA: Addison Wesley.
Tversky, B., & Tuchin, M. (1989). A reconciliation of the evidence on eyewitness
testimony: Comments on McCloskey & Zaragoza. Journal of Experimental
Psychology: General, 118 (1), 86-91. doi: 10.1037/0096-3445.118.1.86
Velden, M., & Wölk, C. (1987). Depicting cardiac activity over real time: A proposal
for standardization. Journal of Psychophysiology, 1 (2), 173-175.
Verschuere, B., Ben-Shakhar, G., & Meijer, E. (2011). Memory detection. Theory and
application of the Concealed Information Test. Cambridge: University Press.
PSYCHOPHYSIOLOGICAL CORRELATES OF THE MISINFORMATION EFFECT
30
Verschuere, B., Crombez, G., de Clercq, A., & Koster, E. H. W. (2004). Autonomic
and behavioral responding to concealed information: Differentiating orienting
and defensive responses. Psychophysiology, 41 (3), 461-466. doi:
10.1111/j.1469-8986.00167.x
Verschuere, B., Crombez, G., Koster, E. H. W., Van Bockstaele, B., & de Cleruq, A.
(2007). Startling secrets: Startle eye blink modulation by concealed crime
information.
Biological
Psychology,
76
(1-2),
52-60.
doi:
10.1016/j.biopsycho.2007.06.001
Wilford, M. M., Chan, J. C. K., & Tuhn, S. J. (2014). Retrieval enhances eyewitness
suggestibility to misinformation in free and cued recall. Journal of Experimental
Psychology: Applied, 20 (1), 81–93. doi: 10.1037/xap0000001
Wright, D. B. (1993). Misinformation and warnings in eyewitness testimony: A new
testing procedure to differentiate explanations. Memory, 1 (2), 153-166. doi:
10.1080/09658219308258229