Psychonomic Bulletin & Review (2019) 26:1001–1007

https://doi.org/10.3758/s13423-018-1541-7

BRIEF REPORT

Enlightenment beats prejudice: The reversibility of stereotype-induced

memory distortion
Hartmut Blank 1 & Lauren Rutter 1 & Rebecca Armstrong 1

Published online: 2 January 2019

# The Author(s) 2018

Abstract
Social stereotypes impact how we remember people, but how stable is this influence? Inspired by the reversibility of the
eyewitness misinformation effect through postwarnings about the planting of misinformation (‘enlightenment’), we explored
if stereotype influence on person memory can be similarly reversed. Participants read person self-descriptions and subsequently
answered memory test questions either with or without stereotype labels, establishing sizeable stereotype-induced memory
distortion. One week later, the participants answered the same questions again, but half were enlightened about the earlier
stereotype manipulation. This eliminated the stereotype effect and restored memory for the original information, whereas
memory remained distorted without enlightenment. We discuss implications for memory distortion research and for
(undermining) the self-perpetuation of stereotypes in society.

Keywords Person memory . Social stereotypes . Warning . Reversibility

What we remember about people can be biased along social mimicking situations where the group membership of a person
expectations. Numerous studies have shown that social stereo- becomes known only after getting to know the person (e.g.,
types, like other schematic expectations, can lead to distorted learning later that a person is a lesbian, or an artist; Snyder &
remembering of person-related information (Fyock & Stangor, Uranowitz, 1978; van Knippenberg & Dijksterhuis, 1996).
1994; Leichtman & Ceci, 1995; Lenton, Blair, & Hastie, 2001; This latter type of design parallels the classic eyewitness mis-
Macrae, Schloerscheidt, Bodenhausen, & Milne, 2002; Rohner information design (Loftus, Miller, & Burns, 1978), in which
& Rasmussen, 2012; Sherman & Bessenoff, 1999; Stangor & postevent misinformation is introduced to bias accounts of
McMillan, 1992). Importantly, in a societal context, such mem- previously witnessed scenes.
ory distortion affords a (pseudo-)validation of the stereotypes Of core interest for the present research, memory distortion
and serves to legitimise and perpetuate them (Fyock & Stangor, through misinformation is not inevitable, but can often be
1994; Macrae et al., 2002; Martin, Cunningham, Hutchison, limited after the fact by postwarning participants about the
Slessor, & Smith, 2017; Snyder & Uranowitz, 1978; van earlier presence of misinformation. In a meta-analysis of 25
Knippenberg & Dijksterhuis, 1996). studies, Blank and Launay (2014) found that postwarnings
While most studies of stereotype influence on remember- reduced the misinformation effect to less than half of its size
ing used designs where target persons were already introduced on average, with some warnings (e.g., the ‘enlightenment’
as members of certain social groups (leading to potential ste- technique introduced by Blank, 1998; see below) being more
reotype influence at both encoding and retrieval), there is also effective than others. Moreover, it has been demonstrated that
research investigating retroactive stereotype influence, postwarnings can not only prevent participants from falling
for misinformation but can also reverse an earlier misinforma-
Electronic supplementary material The online version of this article tion effect (Oeberst & Blank, 2012). The latter research
(https://doi.org/10.3758/s13423-018-1541-7) contains supplementary
followed the typical misinformation design (i.e., Stage 1: orig-
material, which is available to authorized users.
inal information, e.g., a video; Stage 2: misinformation; Stage
* Hartmut Blank
3: memory test) and established the presence of a misinforma-
[email protected] tion effect, but then added an enlightenment-type postwarning
and a further memory test. In three experiments, the initially
1 established misinformation effect was completely reversed (in
Department of Psychology, University of Portsmouth, King Henry
Building, King Henry I Street, Portsmouth PO1 2DY, UK the third experiment, even after 5 weeks).
1002 Psychon Bull Rev (2019) 26:1001–1007

In the present research, we applied this logic to stereotype variables, stereotypical memory distortion and accuracy (see
influence on memory and explored if a parallel effect reversal the Measures section).
can be achieved by using enlightenment in a stereotype influ-
ence setting. Participants first read person information, then a
stereotype was introduced just before answering a memory Materials
test. This corresponded to a standard retroactive stereotype
influence design (Snyder & Uranowitz, 1978; van We created self-descriptions of two male characters (Alan and
Knippenberg & Dijksterhuis, 1996), and we expected to find Greg) in a real-life context, specifically, an actual dating
stereotypical memory distortion. After 1 week, participants website (Match.com), for which two of us had registered for
returned, and half of them were told that we had made up the purposes of the study. These profiles followed a standard
the stereotypical labels applied to the featured persons (i.e., preset format, a mixture of free descriptions under headings
we revealed the earlier manipulation to these participants, like such as ‘a few words about me’ and checklist-type informa-
in a debriefing, but already within the experiment). Then, all tion (e.g., relationship status, height, weight, appearance, as-
participants answered the same memory test again. If enlight- trological sign; screenshots of the profiles used for the main
enment works with stereotypes, then we should expect a sim- study are available as Supplemental Material). The profiles
ilar effect reversal as in Oeberst and Blank’s (2012) misinfor- also included a photograph each, showing Alan or Greg as
mation study. moderately attractive men.
Each profile was constructed to be relevant to a particular
social stereotype, based on the input from three pilot studies.
In the first pilot study, we asked N = 10 participants to pick the
Method
two most stereotypical out of a list of 10 UK-typical occupa-
tions; these were builder (later assigned to Greg) and vicar (a
The research included three pilot studies (see the Materials
Church of England priest, assigned to Alan), chosen by 60%
section) and a main study. The pilot studies served to (1) select
and 80% of the judges, respectively. The second pilot study
highly stereotypical occupations (in the UK) for an effective
identified plausible content for Alan’s and Greg’s Match.com
stereotype influence manipulation, (2) produce stereotypical
profiles, to be probed later in the memory tests. A further 16
information for the person descriptions, and (3) provide
participants described traits, behaviours, activities,
stereotypicality ratings for memory test answers.
possessions, values, likes and dislikes, as well as
appearance, social class, income, relationship status, and
Participants and design smoking and drinking habits typically associated with the
occupations of a builder and a vicar. The most frequently
Combining a student (n = 30) and a general population (n = mentioned content was then incorporated into the dating
38) subsample yielded an initial total of 68 participants. To profiles and/or the memory tests (as response alternatives),
counterbalance conditions within each subsample, the data along with more neutral, stereotype-unrelated information.
from four participants were not used for analysis; thus, the The profiles did not include any occupation-related informa-
final sample consisted of 64 participants (39 female and 25 tion, as this was to be used later for the stereotype induction.
male; mean age = 25.6 years, range: 18–59 years), 28 of which The memory tests for the profiles consisted of 20 questions
were introductory psychology students (16 female, 12 male; each, probing factual information from their dating profiles
mean age = 20.6 years, range: 18–30 years), and 36 were (e.g., “What does Alan do every Friday night?”—the full tests
members of the general public (23 female, 13 male; mean are available as Supplemental Material). All questions had
age = 29.5 years, range: 19–59 years). Using two different four substantial response alternatives, with varying degrees
subsamples was a matter of convenience, but it increases the of stereotypicality. The latter was assessed in a third pilot
generalisability of our findings. study. We asked another 20 participants to “imagine there is
We investigated stereotype influence through providing an a vicar called Alan [or a builder called Greg]; you know no
occupation label for one of two target persons (but not for the other information about him apart from his occupation” and to
other—counterbalanced across participants). Memory for the rate the response options “purely based on how stereotypical
target person information was assessed after 20 min and again you think they are” in relation to the respective occupation, on
1 week later. Before the second memory test, half of the par- a scale from 0 (least stereotypical) to 5 (most stereotypical).
ticipants were enlightened on the earlier occupational label For illustration, for the question “What does Alan do every
manipulation (orthogonal to the students/general public split). Friday night?”, the four alternative responses and their
These variables formed a 2 (occupational label: yes/no, with- stereotypicality ratings were “watch game shows” = 1.6, “go
in) × 2 (assessment: Time 1/Time 2, within) × 2 (enlighten- clubbing” = 0.2, “go to his local pub” = 1.8, and “community
ment: yes/no, between) design. There were two dependent work” = 4.1. These ratings across all test questions were used
Psychon Bull Rev (2019) 26:1001–1007 1003

to determine the degree of stereotypical memory distortion in memory tests, as accurately as possible, with this new in-
the main study (see the Results section). formation in mind. Participants in the no-enlightenment
group were simply told to answer the memory tests again
Procedure based on the information presented about Alan and Greg in
the first session. All instructions in Session 2 were deliv-
Participants consented to participate in a two-session study ered orally (i.e., the wordings given here are approximate,
investigating memory for person information on websites. but very close). Note that the occupational labels used in
They were tested individually or in groups up to eight in quiet Session 1 were removed from all test booklets in Session 2
university or everyday settings. Session 1 included three (leaving them there after enlightening participants that they
phases: the dating profiles, a filler task, and memory tests for were made up would have been strange, and for compara-
the profiles. Instructions and materials were presented via an bility we discarded them in the no-enlightenment group as
automatically timed PowerPoint presentation (shown on PCs/ well). Upon finishing the tests, we fully debriefed all par-
laptops or a projection screen in group sessions), augmented ticipants and asked them not to discuss the true nature of
by experimenter (L.R. or R.A.) clarifications as needed. The the study with potential future participants.
presentation started with an overview of the first session and
then explained that participants would now see two personal
profiles from a dating website (each spread across three slides) Results
and were to study them carefully; each profile would be
shown for 6 minutes (2 min per slide). After the profiles had Measures
been presented, participants worked on filler tasks (two
problem-solving tasks and a sudoku puzzle) for 20 min and Our analysis focused on two key dependent variables, (1)
then completed two memory tests (one for Alan and one for stereotypical memory distortion and (2) accuracy. We deter-
Greg); they were allowed 8 min for each test. mined stereotypical memory distortion in two steps. First,
The two memory tests were provided as separate book- for each participant, we added up the stereotypicality ratings
lets, with the order of presentation (Alan first, then Greg, or (see above) for their chosen answers to the 20 memory test
the reverse) counterbalanced. Participants were asked to questions for Alan and Greg. Then, to create more easily
circle one of four answer options for each test question interpretable values and to control for slight differences in
and also to provide confidence ratings for their choices, the stereotypicality of the response options for Alan and
on a 1 to 5 scale. On the front page of each booklet was Greg, we divided these raw scores by the maximum possible
a box containing the same photograph of the person as was stereotypicality score (60.6 for Alan and 71.1 for Greg, re-
shown on the dating profile, as “a reminder of who Alan spectively, if always the most stereotypical answer was cho-
[Gregory] is”, alongside other information (two lines stat- sen). This produced relative stereotypicality scores,
ing, e.g., “username: Alan_d_509” and “gender: male”) in expressed as percentages of the maximum possible stereo-
the right-hand part of the box. The sole purpose of this was typical memory distortion for Alan and Greg. The second
to provide credibility and context for the crucial stereotype dependent variable (accuracy) was the percentage of correct
manipulation: The box of one of the booklets contained a answers across the 20 memory test questions for Alan and
third line stating “occupation: vicar” (or “occupation: Greg.
builder”). This was always provided in the second booklet; Note that our dependent variables are not redundant—a
our concern was that providing occupation information in correct answer was not always low in stereotypicality.
the first booklet would lead participants to expect occupa- Specifically, when ranking the answer options for each of
tion information in the second booklet as well, and encour- the 40 memory test questions by stereotypicality, the cor-
age speculations that might influence their test answers in rect answer occupied Rank 1 (highest stereotypicality)
unpredictable ways (e.g., based on imagined stereotypical eight times, 13 times Rank 2, 11 times Rank 3, and eight
occupations). times Rank 4 (lowest stereotypicality). In this sense, ste-
Session 2 took place exactly 7 days after Session 1. reotypical memory distortion and accuracy were separate
Participants filled out the memory tests once again (again, aspects of memory performance. Still, the effects of our
8 min for each), but, crucially, half of them (orthogonal to stereotype manipulation on these measures were connect-
the counterbalancing split) were enlightened on the earlier ed. If occupational labels produce stereotypical memory
stereotype manipulation before answering the tests. distortion, then this must come at the expense of accuracy.
Specifically, the experimenter explained that the occupa- O n t h e o t h e r h a n d , a n y re d u c t i o n i n m e m o r y
tion information about Alan (or Greg) on the front page of stereotypicality after enlightenment need not necessarily
the memory test in the first session was incorrect and we reinstate memory accuracy (the memory might have been
had made it up. We then asked them to complete the permanently impaired); this is an open question.
1004 Psychon Bull Rev (2019) 26:1001–1007

Stereotypical memory distortion (A) No Enlightenment

We conducted a three-way mixed ANOVA to analyse partic- No Label Label

ipants’ relative stereotypicality scores as a function of occu-
90

Relave Stereotypicality (%)

pational label (yes/no, within), assessment (Time 1/Time 2,
within) and enlightenment (yes/no, between).1 This yielded 80
main effects of all three variables: occupational label, F(1,
62) = 21.71, p < .001, ηp2 = .26; assessment: F(1, 62) =
11.41, p = .001, ηp2 = .16; enlightenment: F(1, 62) = 17.16, 70
p < .001, ηp2 = .22; an interaction between assessment and
enlightenment, F(1, 62) = 8.83, p = .004, ηp2 = .12, and, most 60
important, a three-way interaction, F(1, 62) = 21.37, p < .001,
ηp2 = .26 (shown in Fig. 1). Further exploration of this inter-
action revealed that the stereotype/occupational label effect 50
persisted (and descriptively increased; see Fig. 1) in the no- Time 1 Time 2
enlightenment group, Time 1, F(1, 31) = 10.42, p = .003, ηp2 =
.25; Time 2: F(1, 31) = 20.49, p < .001, ηp2 = .40, while in the
enlightenment group it was completely eliminated after the (B) Enlightenment
enlightenment, Time 1, F(1, 31) = 12.76, p = .001, ηp2 =
.29; Time 2, F(1, 31) = 0.02, p = .90, ηp2 = .00. No Label Label
To provide some context for the scores shown in Fig. 1, Relave Stereotypicality (%) 90
correct test answers (for Alan and Greg) had a relative
stereotypicality score of 64.4% on average. Interestingly, this 80
benchmark was surpassed to some degree in all conditions
(even the no-label control conditions), which might reflect
weak and unintended spreading stereotype effects due to the 70
mere inclusion of stereotype-related material in Alan’s and
Greg’s dating profiles. More important, the stereotype/
occupational label conditions clearly revealed memory distor-
60
tion above and beyond this level (as shown by one-sample t
tests, df = 31, against the 64.4% benchmark; smallest t = 5.72), 50
except of course after enlightenment (t = 1.51). Time 1 Time 2
Fig. 1 Stereotypical memory distortion as a function of enlightenment. a
Accuracy Standard effect of the stereotype/occupational label at both measurement
times, b Stereotype effect disappears after participants are enlightened on
the earlier stereotype manipulation. Error bars represent standard errors of
A corresponding three-way ANOVA of participants’ accuracy
the mean
scores revealed a complementary pattern of results. Besides
main effects of occupational label, F(1, 62) = 38.44, p < .001,
interaction showed that the stereotype/occupational label ef-
ηp2 = .38, and enlightenment, F(1, 62) = 26.98, p < .001, ηp2 =
fect on accuracy persisted (and descriptively increased; see
.30, as well as interactions between enlightenment and assess-
Fig. 2) in the no-enlightenment group, Time 1, F(1, 31) =
ment, F(1, 62) = 15.78, p < .001, ηp2 = .20, and enlightenment
8.13, p = .008, ηp2 = .21; Time 2: F(1, 31) = 19.73, p <
and occupational label, F(1, 62) = 8.64, p = .005, ηp2 = .12,
.001, ηp2 = .39, while it was—as for stereotypical memory
there was again a three-way interaction, F(1, 62) = 17.70, p <
distortion—completely eliminated in the enlightenment
.001, ηp2 = .22 (displayed in Fig. 2). Unpacking of the
group, Time 1, F(1, 31) = 29.87, p < .001, ηp2 = .49; Time
1
We also conducted——as a check of the robustness of our findings—— 2, F(1, 31) = 0.01, p = .92, ηp2 = .00.2
three-way ANOVAs on the basis of the whole initial sample of 68 participants
(i.e., with imperfect counterbalancing), and using three different random
2
realisations of a counterbalanced 64-participants sample (i.e., dropping differ- It is worth noting that both our accuracy and stereotypical memory distortion
ent sets of participants from some cells to achieve counterbalancing). These findings did not depend on whether Alan or Greg was the target of the stereo-
analyses led to highly similar results. The version we report here is the one that type induction (i.e., on the counterbalancing condition) or on the subsample of
was in between the two others for most of the main statistics. Further, parallel participants (students or general public). In respective extended ANOVAs, the
analyses to the ones we report here, but incorporating the confidence ratings critical three-way interactions shown in Figs. 1 and 2 were not compromised
from the memory tests (e.g., accuracy scores qualified by confidence), led to by higher-order interactions involving these additional variables, suggesting
highly similar findings and are therefore not reported. some generalisability of our findings.
Psychon Bull Rev (2019) 26:1001–1007 1005

(A) No Enlightenment combine in the stereotype condition to equal control

performance.
No Label Label To test the contrasting explanations, we analysed accuracy
90 after enlightenment at Time 2 as a function of the
stereotypicality rank of the correct answer. In a corresponding
80 4 (stereotypicality rank) × 2 (occupational label) ANOVA,
Accuracy (%)

hypercorrection translates into a predicted interaction between

70 these factors (more specifically, a linear trend: Accuracy
should increase from most to least stereotypical, relative to
60 the no-label control). By contrast, our interpretation that en-
lightenment just removes bias and restores access to original
50 information is independent of the stereotypicality of this orig-
inal information and does not predict an interaction.
40 The data did not support the hypercorrection idea: There
Time 1 Time 2 was little evidence for a stereotypicality Rank × Occupational
Label interaction, F(3, 93) = 1.60, p = .19, ηp2 = .05, let alone
a linear trend, F(1, 31) = 0.09, p = .76, ηp2 = .003.4 Instead,
(B) Enlightenment accuracy was more or less flat across stereotypicality rank (see
Fig. 3), in line with the idea that enlightenment corrects (but
No Label Label
not overcorrects) the stereotype influence and restores access
90 to original person information.

80
Accuracy (%)

Discussion
70
To summarise, our participants’ memory for Alan and Greg
60 was strongly biased in the direction of the stereotypic occupa-
tional labels provided for them at the first assessment, at the
50 expense of accuracy. The size of both effects was large (ηp2 >
.20). In the no-enlightenment group, the effects persisted (and
40 increased in size) at the second assessment 1 week later, even
Time 1 Time 2 without a reminder of Alan’s or Greg’s occupation. In the
Fig. 2 Accuracy as a function of enlightenment. a Standard effect of the
enlightenment group, however, when participants learned that
stereotype/occupational label at both measurement times. b Stereotype the occupations were made up and did not actually apply to
effect disappears after participants are enlightened on the earlier Alan and Greg, the stereotypical bias disappeared, and mem-
stereotype manipulation. Error bars represent standard errors of the mean ory accuracy was restored to the level of the control condition
(in fact, both effects were reduced to ηp2 = .00),5 and this held
across all levels of stereotypicality of the original person
Accuracy as a function of information stereotypicality information.
The main purpose of our study was to empirically demon-
So far, we found that stereotypical memory distortion decreased strate the reversibility of stereotype-induced memory
and accuracy increased after enlightenment. This might mean
that enlightenment restores unbiased access to the original per- 4
We also conducted separate t tests for each stereotypicality rank, comparing
son information. Alternatively, a hypercorrection mechanism accuracy in the label and no-label conditions. Even without Bonferroni cor-
rection, none of these comparisons were significant (lowest p = .07, for Rank
(i.e., simply reversing the stereotypical bias) could produce a 2).
similar overall pattern.3 That is, after enlightenment, people 5
In the context of the current replication debate (e.g., Zwaan, Etz, Lucas &
would pick answers that are less stereotypical, which would Donnellan, 2018), it is also worth mentioning that the present findings broadly
replicate——and partly bring out more convincingly——a similar pattern of
boost memory performance (perhaps even beyond the level of
findings obtained 15 years earlier in a different culture (Germany), but with a
the no-stereotype control condition) for low-stereotypical cor- largely identical design (Blank, 2001, Chap. 6). The findings paralleled the
rect answers but produce an opposite effect for correct answers present study for stereotypical memory distortion, but, for accuracy, the critical
that are highly stereotypical. Theoretically, these effects might interaction was weaker (p = .13). This was likely owed to a smaller effect size,
resulting from different and less reliable measurement (only a few items mea-
sured accuracy). In this sense, the present study (using more reliable accuracy
3
We thank reviewer Bob Belli for proposing this alternative. measurement) is a higher-powered replication of Blank (2001).
1006 Psychon Bull Rev (2019) 26:1001–1007

100
metaphor of memory as ‘memory metal’—a metal that can be
90 hammered out of shape but will spring back into its original
shape under suitable conditions (see, e.g., Kauffman & Mayo,
80
Accuracy (%)

1993, for an overview). Likewise, it seems to be possible for a

70 memory to regain—under suitable conditions such as enlight-
enment—its original form after having been distorted. Of
60
No label course these two metaphors complement, rather than contra-
50 dict, each other. But certainly, metaphors guide our views of
Label phenomena and how we research them, and in this sense the
40
alternative metaphor proposed here, with its emphasis on
30 memory resilience, is a welcome antidote to the current focus
1 (most 2 3 4 (least on memory distortion and malleability (see, e.g., Nash & Ost,
stereotypical) stereotypical) 2017).
Fig. 3 Accuracy after enlightenment as a function of (1) the Regarding practical consequences of the reversibility of
stereotypicality rank of the correct answer to a memory test question stereotype-conserving memory distortion, the message is per-
and (2) the previous presence or absence of stereotype influence
(occupational label vs. no label) haps that attention needs to be paid to detail. Firstly, we ac-
knowledge that the present research applies to only one kind
of real-life situations—namely, when stereotypes are intro-
distortion; it was not designed to isolate specific processes duced after the fact. We cannot be sure that the present find-
underlying this effect reversal. Nevertheless, a few comments ings would extend to situations where stereotypes already in-
are in order. Principally, the reversal of the effects would be fluence the encoding of person information. Moreover, our
due to the disruption of any processes that initially produced enlightenment manipulation undermined the applicability of
the memory distortions. Candidate mechanisms discussed in a stereotype to a particular case, not the stereotype itself.
the literature include (a) stereotype-consistent response biases While there are certainly real-life equivalents of this (e.g.,
(in itself a heterogeneous category including, e.g., implicit cases of mistaken identity or group membership), it would
associations, fluency misattribution, heuristic source monitor- seem much harder to undermine stereotype influence when
ing, or intentional guessing; see coverage by Lenton et al., such a move is not possible (but still feasible in principle;
2001; Payne, Jacoby, & Lambert, 2004; Rohner & e.g., Todd, Galinsky, & Bodenhausen, 2012, showed that
Rasmussen, 2012; Sherman & Bessenoff, 1999; Stangor & memory distortion due to an unmistakable stereotype—
McMillan, 1992), and (b) biased memory retrieval (i.e., using race—can be limited through asking people to adopt the per-
the stereotype as a cue for retrieving person information, spective of the targeted protagonist). In any case, the current
which may facilitate access to stereotype-consistent informa- demonstration that it is possible to completely undo stereotype
tion and/or inhibit access to stereotype-inconsistent informa- influence on memory under some circumstances may encour-
tion; e.g., Snyder & Uranowitz, 1978; van Knippenberg & age future research to explore the circumstances under which
Dijksterhuis, 1996). Along these lines, unbiased memory per- the preservation of stereotypes can be disrupted more
formance in our study may have been restored through generally.
disrupting (but not overcorrecting; see our accuracy-by-
stereotypicality analyses) stereotype-consistent biases and/or Author note We are grateful to Ryan Fitzgerald, Jonathan
stereotype-guided retrieval. Koppel and Beatriz López for helpful comments on a draft.

Theoretical and practical implications

Open Access This article is distributed under the terms of the Creative
Commons Attribution 4.0 International License (http://
The present findings are generally good news both in terms of creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
the reliability of memory and the implications for stereotype distribution, and reproduction in any medium, provided you give
perseverance. Theoretically, the present findings, combined appropriate credit to the original author(s) and the source, provide a link
to the Creative Commons license, and indicate if changes were made.
with Oeberst and Blank’s (2012) parallel findings on the re-
versibility of the eyewitness misinformation effect, suggest
that previous views of the suggestibility of memory may have
been too extreme. In a popular metaphor, memory has been
References
described as malleable (Loftus, 1979), just like a piece of
metal hammered on the anvil. While this is (to a considerable Blank, H. (1998). Memory states and memory tasks: An integrative
degree, in the light of the massive body of research supporting framework for eyewitness memory and suggestibility. Memory, 6,
it) an apt metaphor, the present research invokes an alternative 481–529.
Psychon Bull Rev (2019) 26:1001–1007 1007

Blank, H. (2001). Soziales Gedächtnis und sozialer Einfluß auf Nash, R. A., & Ost, J. (Eds.). (2017). False and distorted memories, Hove,
Erinnerungen: Theoretische und empirische Beiträge [Social mem- UK: Psychology Press.
ory and social influence on remembering: Theoretical and empirical Oeberst, A., & Blank, H. (2012). Undoing suggestive influence on mem-
contributions] (Habilitation thesis, University of Leipzig). ory: The reversibility of the eyewitness misinformation effect.
Blank, H., & Launay, C. (2014). How to protect eyewitness memory Cognition, 125, 141–159.
against the misinformation effect: A meta-analysis of post-warning Payne, B. K., Jacoby, L. L., & Lambert, A. J. (2004). Memory monitoring
studies. Journal of Applied Research in Memory and Cognition, 3, and the control of memory distortion. Journal of Experimental
77–88. Social Psychology, 40, 52–64.
Fyock, J., & Stangor, C. (1994). The role of memory biases in stereotype Rohner, J.-C., & Rasmussen, A. (2012). Recognition bias and the phys-
maintenance. British Journal of Social Psychology, 33, 331–343. ical attractiveness stereotype. Scandinavian Journal of Psychology,
Kauffman, G., & Mayo, I. (1993). Memory metal. ChemMatters, 4–7. 53, 239–246.
Leichtman, M. D., & Ceci, S. J. (1995). The effects of stereotypes and Sherman, J. W., & Bessenoff, G. R. (1999). Stereotypes as source-
suggestions on preschoolers’ reports. Developmental Psychology, monitoring cues: On the interaction between episodic and semantic
31, 568–578. memory. Psychological Science, 10, 106–110.
Lenton, A. P., Blair, I. V., & Hastie, R. (2001). Illusions of gender: Snyder, M.. & Uranowitz, S. W. (1978). Reconstructing the past: Some
Stereotypes evoke false memories. Journal of Experimental Social cognitive consequences of person perception. Journal of Personality
Psychology, 37, 3–14. and Social Psychology, 36, 941–950.
Loftus, E. F. (1979). The malleability of human memory: Information Stangor, C., & McMillan, D. (1992). Memory for expectancy-congruent
introduced after we view an incident can transform memory. and expectancy-incongruent information: A review of the social and
American Scientist, 67, 312320. social-developmental literatures. Psychological Bulletin, 111, 42–61.
Loftus, E. F., Miller, D. G., & Burns, H. J. (1978). Semantic integration of Todd, A. R., Galinsky, A. D., & Bodenhausen, G. V. (2012). Perspective
verbal information into a visual memory. Journal of Experimental taking undermines stereotype maintenance processes: Evidence
Psychology: Human Learning and Memory, 4, 19–31. from social memory, behaviour explanation, and information solic-
Macrae, C. N., Schloerscheidt, A. M., Bodenhausen, G. V., & Milne, A. itation. Social Cognition, 30, 94–108.
B. (2002). Creating memory illusions: Expectancy-based processing van Knippenberg, A., & Dijksterhuis, A. (1996). A posteriori stereotype
and the generation of false memories. Memory, 10, 63–80. activation: The preservation of stereotypes through memory distor-
Martin, D., Cunningham, S. J., Hutchison, J., Slessor, G., & Smith, K. tion. Social Cognition, 14, 21–53.
(2017). How societal stereotypes might form and evolve via cumu- Zwaan, R. A., Etz, A., Lucas, R. E. & Donnellan, M. B. (2018). Making
lative cultural evolution. Social & Personality Psychology replication mainstream. Behavioral and Brain Sciences, 41, E120.
Compass, 11, e12338. doi:https://doi.org/10.1017/S0140525X17001972