The Journal of General Psychology

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/vgen20

Effect of post-encoding emotion on long-term

memory: Modulation of emotion category and
memory strength

Bo Wang

To cite this article: Bo Wang (2021) Effect of post-encoding emotion on long-term memory:
Modulation of emotion category and memory strength, The Journal of General Psychology,
148:2, 192-218, DOI: 10.1080/00221309.2020.1769543

To link to this article: https://doi.org/10.1080/00221309.2020.1769543

View supplementary material

Published online: 29 May 2020.

Submit your article to this journal

Article views: 696

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=vgen20
THE JOURNAL OF GENERAL PSYCHOLOGY
2021, VOL. 148, NO. 2, 192–218
https://doi.org/10.1080/00221309.2020.1769543

Effect of post-encoding emotion on long-term memory:

Modulation of emotion category and memory strength
Bo Wang
Central University of Finance and Economics

ABSTRACT ARTICLE HISTORY

Although studies have shown that post-encoding emotion Received 2 July 2019
enhances long-term memory, it is unclear whether the effect Accepted 10 May 2020
varies depending on a specific category of emotion and
KEYWORDS
whether the effect depends on memory strength. In the cur-
Discrete emotion;
rent study, participants encoded a list of words, half of which familiarity; memory
were presented once and the remaining half of which were strength; recognition
presented four times. Afterwards participants watched a video memory; recollection
that was neutral, or induced one of the following emotions:
anger, sadness, disgust, fear, or happiness. Overall recognition
was lower in the anger-inducing condition than in the sad-
ness-inducing or happiness-inducing condition. Recollection
(i.e., adjusted rate of correctly identifying a word as old while
being able to retrieve details related to the study episode) was
lower in the anger-inducing condition than in the sadness-
inducing condition. No significant effect of emotion was found
on familiarity (i.e., adjusted rate of correctly identifying a word
as old without being able to retrieve any details related to the
study episode). Furthermore, memory strength modulated the
effect on familiarity: Familiarity for prioritized words was lower
than for non-prioritized words only in the “anger” and
“sadness” conditions but not in other emotion conditions. The
effect of post-encoding emotion differs depending on a spe-
cific category of emotion and modulation of memory strength
relies on a specific component of recognition memory.

Ample evidence has shown that emotion induced after encoding can
enhance memory (e.g., Dudai, 2004; Sch€ onauer, 2018). For instance, in a
study by Nielson, Yee, and Erickson (2005), participants encoded a list of
words and then watched either a neutral video (i.e., tooth brushing) or a
negative video (i.e., dental surgery). The researchers found that the video-
induced emotion enhanced delayed performance of free recall and
recognition memory. Such an enhancement effect has been replicated by
subsequent studies (e.g., Judde & Rickard, 2010; Liu, Graham, & Zorawski,
2008; Nielson & Lorber, 2009; Nielson & Powless, 2007; Wang & Fu, 2010;

CONTACT Bo Wang [email protected], [email protected] Department of Psychology, Central

University of Finance and Economics, Beijing 100081, China.
Supplemental data for this article is available online at https://doi.org/10.1080/00221309.2020.1769543.
ß 2020 Taylor & Francis Group, LLC
 THE JOURNAL OF GENERAL PSYCHOLOGY 193

Wang & Sun, 2015; 2017). However, two questions remain to be answered:
(1) Does the effect differ depending on a specific category of emotion (e.g.,
sadness, fear, anger)? (2) Does memory strength modulate the effect?

Effects of discrete emotions on long-term memory

Despite the findings from prior studies, it is unclear whether the effect of
post-encoding emotion depends on a specific category of emotion. Most
prior studies only used a neutral and a negative video, with the latter usu-
ally about dental surgery (e.g., Liu, Graham, & Zorawski, 2008; Nielson &
Powless, 2007; Nielson et al., 2005), which may elicit a mixture of fear, anx-
iety and disgust. Other studies used videos that primarily induced sadness
(Wang & Fu, 2010), or anger (Wang & Ren, 2017; Wang & Sun, 2015).
However, no prior studies have directly compared the effects of discrete
emotions (e.g., happiness, fear, anger, sadness, disgust), rendering unclear
whether the effect of emotion on long-term memory depends on a specific
category of emotion.
The rationale for examining the effects of discrete emotions is based on
several aspects. First, although it has been shown that the amygdala modu-
lates the memory for arousing experiences (e.g., McGaugh, 2018), under-
standing the effects of discrete emotions helps to elucidate whether arousal
per se is the factor that drives long-term memory. According to the cir-
cumplex model proposed by Russell (1980), valence and arousal are the
two dimensions of emotion. Although arousal has been emphasized as the
critical factor (e.g., Mather & Sutherland, 2009; Nielson & Arentsen, 2012),
there has been evidence that the arousal account may be inadequate. For
instance, in a study by Wang and Fu (2010), participants watched a neutral,
positive, or negative video several minutes after encoding a list of neutral
words. The researchers found that, although both the positive video (a
comic play) and negative video (a clip about the Wenchuan earthquake)
significantly increased arousal, only the negative video improved item
memory. However, a further test of the arousal account requires the exam-
ination of the effect of arousal from different sources (i.e., of different emo-
tional properties). Particularly, it is necessary to investigate whether arousal
from different negative states (e.g., anger, sadness, fear) exerts differen-
tial effects.
Second, studies have revealed distinct mechanisms underlying discrete
emotions. A study by Sinha, Lovallo, and Parson (1992) found that anger
and fear were associated with differential diastolic blood pressure and per-
ipheral vascular resistance. Furthermore, whereas joy was associated with a
slight decrease in autonomic nervous system (ANS) response, sadness was
associated with a modest increase in ANS response. The differentiation of
194 B. WANG

discrete emotions based in ANS was replicated by Collet, Vernet-Maury,

Delhomme, and Dittmar (1997), who observed that skin potential responses
could be used to distinguish surprise and fear from sadness, and to distin-
guish fear from disgust. Rainville, Bechara, Naqvi, and Antonio (2006) fur-
ther found in the states of fear and happiness a reduction in respiratory
period, which was less consistently observed in anger. There has also been
evidence from patient and neuroimaging studies that discrete emotions can
be differentiated. Damages to the amygdala and insula were found to be
respectively associated with deficits in fear and disgust (c.f., Calder,
Lawrence, & Young, 2001). In a neuroimaging study, where short movies
and mental imagery were used to induce discrete emotions (i.e., anger, fear,
surprise, sadness, disgust, happiness), Saarim€aki et al. (2016) found that,
regardless of induction method, specific activation patterns across cortical
and sub-cortical brain areas (i.e., medial and inferior lateral prefrontal cor-
tices, frontal pole, precentral and postcentral gyri, precuneus, and posterior
cingulate cortex) constituted representations of different emotional states.
Taken together, the above studies have provided evidence supporting differ-
ential mechanisms underlying discrete emotions.
Third, although there has been evidence for differential effects of
discrete emotions on memory, in most studies emotions were induced by
to-be-remembered words or pictures and as such the effects of post-
encoding emotions have been unclear. For instance, Levine and Burgess
(1997) found that, whereas happiness resulted in better memory for a nar-
rative as a whole, sadness or anger enhanced memory for information spe-
cificity. The distinct effects of discrete emotions were also demonstrated in
studies using pictorial stimuli. For instance, Wang (2013) found that overall
recognition for angry and fearful faces was significantly better than for
positive faces. However, recognition for sad faces was significantly worse
than for positive faces. The effects of discrete emotions were observed even
within the negative category in that overall recognition for angry, disgusted,
fearful, or surprised faces was significantly better than for sad faces. To
summarize, discrete emotions can have differential effects on memory.
However, the evidence has come from studies in which the stimuli for
emotion induction overlap the encoded stimuli; it is unknown whether dis-
crete emotions induced after encoding can exert differential effects on
long-term memory.
It should be noted, however, that Kaplan, Damme, and Levine (2012)
proposed a motivation-based framework under which pre-goal emotions
(e.g., anger, disgust, fear) and post-goal emotions (e.g., sadness, happiness)
have differential effects on attention and memory. Pre-goal emotions and
post-goal emotions have been shown to narrow and broaden attention,
respectively (e.g., Gable & Harmon-Jones, 2010). Furthermore, a recent
 THE JOURNAL OF GENERAL PSYCHOLOGY 195

study showed that positive emotion enhanced association memory

(Madan, Scott, & Kensinger, 2018). No study, however, has directly tested
whether pre-goal and post-goal emotions have differential effects on long-
term memory.

Modulation of memory strength

Understanding the modulation of memory strength helps to elucidate the
conditions under which post-encoding emotion exerts its effect. In spite of
the many studies examining the effect of emotion on long-term memory
(e.g., Judde & Rickard, 2010; Nielson & Powless, 2007; Wang & Sun, 2015),
whether memory strength can be modulatory is unclear. However, there
has been evidence suggesting that post-encoding emotion is more likely to
enhance weak memories. For instance, Dunsmoor, Murty, Davachi, and
Phelps (2015) manipulated memory strength by varying the number of pre-
sentations. In their study, after encoding a list of object pictures (from
either “animal” or “tool” category) either once or three times, participants
undertook emotional learning, during which a list of object pictures (from
either “animal” or “tool” category) were presented, with some picture trials
paired with an electronic shock. At the 6 h or 24 h delayed test, recognition
memory for object pictures was selectively enhanced if other objects from
the same category were paired with shock. Importantly, the retrospective
enhancing effect emerged only for weak memory (i.e., memory for items
that were presented once during encoding).

Modulation of memory components

Studies have shown that recollection and familiarity are the two dissociable
components that underlie recognition memory (Yonelinas, 2002). Although
there has been no direct evidence as to whether the modulation of memory
strength can vary depending on memory component, findings from prior
studies suggest that it is necessary to take into account memory compo-
nent. For instance, Yonelinas, Parks, Koen, Jorgenson, and Mendoza (2011)
found that post-encoding stress caused by skydiving enhanced familiarity-
based recognition, but not recollection for neutral pictures, indicating that
stress enhanced the underlying memory strengths rather than generation
process involved in memory tests such as free recall. A recent study by
Wang and Ren (2017), however, showed that post-encoding emotional
arousal from an anger-inducing video impaired recollection but had no
effect on familiarity for either neutral or negative pictures. Although it is
unclear what has caused such a discrepancy, the literature suggests that a
full understanding of the effect of post-encoding emotion requires
196 B. WANG

consideration of recollection and familiarity as the two components under-

lying recognition memory.

The current study

The overarching goal of the current study was to answer two questions: (1)
Does the effect of post-encoding emotion differ depending on a specific
category of emotion? (2) Is the effect contingent on memory strength?
Based on the evidence showing differential mechanisms underlying discrete
emotions (e.g., Collet et al., 1997; Saarim€aki et al., 2016), it was hypothe-
sized that the effect would differ depending on a specific category of emo-
tion. Nonetheless, according to the arousal-based model, it is likely that
arousal, rather than discrete emotions, would be the driving factor. With
regard to the second question, based on the evidence from prior studies
(e.g., Dunsmoor et al., 2015), post-encoding emotion was hypothesized to
enhance weak memory rather than strong memory. Furthermore, the
modulation of memory strength may depend on a specific component of
recognition memory given that recollection and familiarity have been
shown to be dissociable components underlying recognition memory.
To test the above hypotheses, participants encoded a list of words, with
half of them presented once and the remaining half presented four times.
Afterwards participants were randomly assigned to watch a video that was
neutral, or induced one of the following emotions: anger, sadness, disgust,
fear, or happiness. In a delayed recognition memory test, they judged
whether a word had been presented in the initial encoding. After identify-
ing a word as old, they further made “remember”/”know” (R/K) judgments
(Rajaram, 1993). An “R” judgment indicates the retrieval of specific details
associated with an encoding episode, whereas a “K” judgment indicating
only familiarity with a presented item without the retrieval of details associ-
ated with the encoding episode (Yonelinas, 2002). The R/K paradigm has
been widely used to differentiate between recollection and familiarity as the
two components underlying recognition memory (e.g., Yonelinas et al.,
2011). This memory paradigm allows the analyses on specific components
of recognition memory, making it possible to have a refined appreciation
of the effect of post-encoding emotion.

Method
Participants
A sample size estimation was calculated using G
Power Version 3.1.9.2
software. Using moderate parameters (power ¼ 0.95, effect size f ¼ 0.25),
the analysis estimated a sample size of 90. Because of the need to
 THE JOURNAL OF GENERAL PSYCHOLOGY 197

Table 1. Brief Descriptions for the Six Videos Used in the Current Study.
Video type Descriptions Validation studies
Neutral A professor showed how to repair a CD- Wang and Ren (2017)
ROM drive.
Anger A man rushed into a cake shop and Wang and Ren (2017)
brutally beat a pregnant woman.
Sadness A woman narrated her tragic experience Wang and Fu (2010)
in Wenchuan Earthquake.
Fear Part of a film that describes a curse born The pilot study described here
when someone dies in the grip of a
powerful rage or extreme sorrow.
Disgust A surgeon demonstrated a Nielson et al., (2005)
dental surgery.
Happiness A comic short play by three famous Wang and Fu (2010)
comedians in mainland China.

counterbalance stimuli across participants, a total of 96 undergraduate stu-

dents (48 males and 48 females; mean age ¼ 22.00 years, SD ¼ 0.92 year)
were recruited. All participants were native Chinese speakers. This sample
size was determined a priori and data collection was stopped right after
data of 96 participants had been collected. Informed consent was obtained
from all participants.

Stimuli
Six 3 min videos were used respectively in the emotion conditions: neutral,
anger, sadness, fear, disgust, and happiness. Table 1 presents brief descrip-
tions of these videos, which had been validated in prior studies (e.g., Wang
& Fu, 2010; Wang & Sun, 2015) except for the one used to induce fear. A
pilot study, however, showed that 8 our of the 13 participants reported fear
as the dominant emotion they experienced during the video presentation.
Furthermore, participants had a significant increase in arousal ratings
(before watching: M ¼ 4.85, SE ¼ 0.37; after watching: M ¼ 6.62, SE ¼ 0.46),
F (1, 42) ¼ 345.99, p < .001, gp2 ¼ 0.45.
Most of the words were translated versions of the English words used in
the study by Nielson et al. (2005). The original English words were chosen
from a normative database (Paivio, Yuille, & Madigan, 1968), where all the
words (e.g., apple, child, cottage) were highly imageable and concrete, thus
rendering all the words (targets and distracters) equally “memorable”. In
order to ensure that all the words consisted of two Chinese characters (e.g.,
天空), six words that result in three-character Chinese words were excluded.
The final word list consists of 30 words as targets in the study phase and
104 words as distractors in the memory test phase (see Appendix I for all
the words and their parameters). The target and distractor words did not
significantly differ in pleasantness, F (1, 132) ¼ 0.18, p ¼ .67, arousal, F (1,
132) ¼ 0.46, p ¼ .50, concreteness, F (1, 132) < 0.001, p ¼ .98, and
198 B. WANG

familiarity, F (1, 132) ¼ 0.34, p ¼ .56. The ratio of targets to distractors is

similar to that in the study by Nielson et al. (2005), in which 30 words were
used as targets and 110 words were distractors.

Design and procedure

E-prime 1.1 (Psychology Software Tools, Inc.) was used to program the
experiment. A 6  2 mixed factorial design was used, with Emotion
Category (neutral, anger, sadness, fear, disgust, or happiness) as the
between-subjects factor and presentation times (once vs. four times) as the
within-subjects factor. Words in the “four times” condition were presented
four times randomly (i.e., by setting the weight to be “4” in the Eprime
“list” and setting the order to be “random”). Participants were randomly
assigned so that there were 16 participants (8 males and 8 females) in each
emotion category.
Words used in “once” and “four times” conditions were counterbalanced
across participants. The 30 target words were evenly divided into set 1 and
set 2. For half of participants in each group, words in set 1 were presented
once and words in set 2 were presented four times; for the other half of
participants, words in set 1 were presented four times and words in set 2
were presented once.
The experiment consists of four major stages: encoding, video presenta-
tion, filler tasks and delayed test. At the encoding stage, in each trial a fix-
ation cross appeared at the center of a screen for 500 ms, followed by a
word appearing for 1000 ms. At the end of each trial was a blank screen
lasting for 500 ms. For manipulation of memory strength, half of the 30
words were presented once and the remaining half were presented four
times. All words were presented in a random sequence for each participant.
Participants were instructed to silently read each word without being
informed of any subsequent memory test. Immediately after the encoding
stage, participants rested for one minute.
At the stage of video presentation, participants rated their current mood
and arousal respectively on two 9-point scales (with 1 indicating most
unpleasant or relaxed and 9 most pleasant or aroused). Consistent with
prior studies (e.g., Wang & Ren, 2017; Wang & Sun, 2015), the paradigm
of emotion assessment, as shown in Figure 1, is based on the Self-
Assessment Manikin (SAM) (Lang, Bradley, & Cuthbert, 2008) and the
assessment instructions were adapted from their study. Then participants
were randomly assigned to watch one of the six videos and, immediately
after video presentation, rated again their current mood and arousal on the
previous scales. Participants were also asked to report the dominant emo-
tion they experienced during video presentation by choosing from the
 THE JOURNAL OF GENERAL PSYCHOLOGY 199

Figure 1. The Self-Assessment Manikin used to assess participants’ mood (A) and arousal (B)
before and after video presentation.

following seven categories: neutral, happiness, disgust, sadness, fear, anger,

and surprise. Surprise was included as an option because it is one of the
six basic emotions (Ekman & Cordaro, 2011) and was also identified as a
basic emotion in the study by Jack, Garrod, and Schyns (2014).
At the stage of filler tasks, participants were instructed to fill in a series
of questionnaires, including arousal predisposition scale (Coren, 1988),
emotion appraisal and emotion suppression scales (Gross & John, 2003),
state and trait anxiety inventory (Spielberger, Gorsuch, Lushene, Vagg, &
Jacobs, 1983) and Beck Depression inventory (Beck, Ward, Mendelson,
Mock, & Erbaugh, 1961). Participants also took some rests and performed
mathematical tasks (e.g., counting backward from 2000 by 3). The purpose
of the stage of filler tasks was to create a retention interval during which
participants could be kept from rehearsing previously encoded stimuli.
Items of questionnaires were presented on the computer screen one by
one. To minimize the variation in the time that participants spent in filling
200 B. WANG

the questionnaires, all items in a questionnaire were presented for the same
period of time for all participants. After each item disappeared from the
screen, participants were instructed to press immediately a key to make a
response. All participants finished the questionnaire tasks.
At the stage of delayed test, which began after 30 minutes had elapsed
after the end of encoding, participants first rated their current mood and
arousal on the scales that were used after video presentation, and were
then presented with a list of the 30 old words and 104 new words. In each
trial, a fixation cross appeared for 500 ms, followed by a word presented on
the screen for 1500 ms1. Right after a word disappeared, participants were
instructed to press a key to determine whether the word was old or new
(“F” for old words; “J” for new words). After identifying a word as old,
participants were instructed to make a “R” (i.e., “remember”) or “K” (i.e.,
“know”) response (Mccabe & Geraci, 2009) by pressing “G” and “H”,
respectively. A “R” response indicates conscious retrieval of details related
to the encoding of a word, whereas a “K” response indicates only familiar-
ity with the word but lack of retrieval of any details related to the encoding
episode (Yonelinas, 2002). There was a practice block involving six trials
(two old words and four new words) to make sure that all participants
understand the instructions regarding “R” and “K” responses. Participants
were encouraged to be as accurate as possible. The detailed instructions are
presented in Appendix II. All words were presented randomly for each
participant.

Statistical analyses
According to the two-high threshold model (Snodgrass & Corwin, 1988)
and prior studies (e.g., Wang, 2013), overall recognition was calculated by
subtracting false alarm rates from hit rates. In order to make a comparison
with studies that used d’ as the dependent variable, we also conducted an
analysis on d’. In accordance with Sharot and Yonelinas (2008), recollection
was computed by subtracting the proportion of new items receiving “R”
(i.e., “remember”) responses from the proportion of old items receiving “R”
responses. Familiarity was computed by the formula:

K ¼ Khit=ð1－RhitÞ－Kfa=ð1－RfaÞ

where Khit, Rhit, Kfa, and Rfa respectively represent “K” hit rates, “R” hit
rates, “K” false alarm rates, and “R” false alarm rates. Post hoc tests were
based on Bonferroni adjustment.
 THE JOURNAL OF GENERAL PSYCHOLOGY 201

Table 2. Characteristics of Participants Assigned to the Six Emotion Categories As Reflected

from the Questionnaires.
Emotion category
Neutral Anger Sadness Fear Disgust Happiness
Measure M SE M SE M SE M SE M SE M SE p
Arousal predisposition 32.88 1.28 34.38 1.28 33.50 1.28 31.00 1.28 33.88 1.28 34.00 1.28 .47
Emotion reappraisal 25.13 1.04 25.00 1.04 27.31 1.04 26.06 1.04 25.38 1.04 28.00 1.04 .22
Emotion suppression 16.81 0.77 15.94 0.77 18.13 0.77 16.81 0.77 17.50 0.77 15.13 0.77 .09
State anxiety 49.38 1.57 45.00 1.57 47.56 1.57 48.06 1.57 47.63 1.57 44.31 1.57 .19
Trait anxiety 50.63 1.27 47.63 1.27 47.81 1.27 48.31 1.27 48.56 1.27 46.56 1.27 .34
Depression 10.75 1.74 8.44 1.74 9.31 1.74 11.38 1.74 11.81 1.74 7.81 1.74 .49

Results
Participants’ characteristics
Table 2 presents participants’ characteristics as revealed from the question-
naires. Participants across the six emotion conditions did not significantly
differ in arousal predisposition, emotion reappraisal, emotion suppression,
state anxiety, trait anxiety or depression scores.

Manipulation check for emotion elicitation

A 6 (Emotion Category: neutral, anger, sadness, fear, disgust, or happiness)
 2 (Time: time 1 or before video presentation vs. time 2 or after video
presentation) ANOVA conducted on mood ratings showed a significant
main effect of Time, F (1, 90) ¼ 86.47, p < .001, gp2 ¼ 0.49. The main effect
of Emotion Category was also significant, F (5, 90) ¼ 14.46, p < .001,
gp2 ¼ 0.45. However, these significant main effects were qualified by the sig-
nificant interaction between Time and Emotion Category as reflected in
Figure 2, F (5, 90) ¼ 20.36, p < .001, gp2 ¼ 0.53. Simple effect analyses indi-
cated that, in the neutral condition, there was no significant change in
mood ratings over the period of video presentation (p ¼ .11). In the condi-
tions of anger, sadness, fear, or disgust, there was significant decreases in
mood ratings (all ps < .001). In the happiness condition, there were signifi-
cant increases in mood ratings (p < .001).
Furthermore, prior to video presentation, participants across the six emo-
tion conditions had comparable mood ratings, F (5, 90) ¼ 1.72, p ¼ .14,
gp2 ¼ 0.09. Immediately after video presentation, however, there was a sig-
nificant main effect of Emotion Category, F (5, 90) ¼ 31.97, p < .001,
gp2 ¼ 0.64. Results of planned pairwise comparisons were presented in the
second row of Table 3.
The ANOVA on arousal ratings also showed a significant interaction
between Time and Emotion Category as shown in Figure 3, F (5,
90) ¼ 5.51, p < .001, gp2 ¼ 0.23. Simple effect analyses revealed that there
202 B. WANG




0RRG5DWLQJV 
7LPH

7LPH




1HXWUDOLW\ $QJHU 6DGQHVV )HDU 'LVJXVW +DSSLQHVV
(PRWLRQ&DWHJRU\
Figure 2. Mood ratings at time 1 (before video presentation) and time 2 (after video presenta-
tion) in the six emotion categories (i.e., the groups who were presented with the six types of
videos). All groups rated happiness on a scale (where 1 ¼ most unpleasant and 9 ¼ most pleas-
ant). Error bars represent standard errors.

Table 3. Pairwise Comparisons with Confidence Intervals for Mood and Arousal Ratings
Immediately after Video Presentation.
Neutrality- Neutrality- Neutrality-
Dimension Neutrality-Anger Sadness Neutrality-Fear Disgust Happiness
Mood CI ¼ [1.54, CI ¼ [0.13, CI ¼ [1.08, CI ¼ [0.22, CI ¼ [–3.28,
3.27], p < .001 1.83], p ¼ .024 2.78], p < .001 1.92], p ¼ .014 1.58], p
< .001
Arousal CI ¼ [–3.69, CI ¼ [–2.71, CI ¼ [–3.65, CI ¼ [–2.21, CI ¼ [–3.48,
1.46], p 0.50], p 1.44], p 0.009], p ¼ .052 1.27], p
< .001 ¼ .005 < .001 < .001
Note. The confidence intervals (CI) are for the differences in mood/arousal ratings between two neutrality and
other emotion conditions. Mood/arousal ratings before video presentation was included as covariates.



$URXVDO5DWLQJV



7LPH

7LPH




1HXWUDOLW\ $QJHU 6DGQHVV )HDU 'LVJXVW +DSSLQHVV
(PRWLRQ&DWHJRU\
Figure 3. Arousal ratings at time 1 (before video presentation) and time 2 (after video presen-
tation) in the six emotion categories (i.e., the groups who were presented with the six types of
videos). All groups rated arousal on a scale (where 1 ¼ least aroused and 9 ¼ most aroused).
Error bars represent standard errors.

was no significant change in arousal ratings over the period of video pres-
entation in the neutral condition (p ¼ .057). In the conditions of anger, sad-
ness, fear, or happiness, there were significant increases in arousal ratings
(p ¼ .001, p ¼.041, p < .001, and p ¼ .001, respectively). In the disgust
 THE JOURNAL OF GENERAL PSYCHOLOGY 203

condition, however, arousal ratings did not significantly change over the
period of video presentation (p ¼ .77).
Furthermore, prior to video presentation, participants across the six emo-
tion conditions had comparable arousal ratings, F (5, 90) ¼ 1.00, p ¼ .42,
gp2 ¼ 0.05. Immediately after video presentation, however, there was a sig-
nificant main effect of Emotion Category, F (5, 90) ¼ 6.80, p < .001,
gp2 ¼ 0.27. Results of planned pairwise comparisons were presented in the
second row of Table 3.
Based on participants’ retrospective reports, the percentage was calcu-
lated of participants reporting their dominant emotions during video pres-
entation. For participants who watched videos that were respectively
intended for induction of neutrality, anger, sadness, fear, disgust, and hap-
piness, there were 75.00%, 93.75%, 75.00%, 93.75%, 87.50%, and 81.25% of
them who reported the above six emotions as the dominant emotions they
experienced during video presentation. There was only one male partici-
pant (1/16, 6.25%) who reported “surprise” in the “sadness” condition.
Therefore, the videos elicited specific emotions as expected.

Mood and arousal ratings before delayed test

A one-way ANOVA (Emotion Category: neutral, anger, sadness, fear, dis-
gust, or happiness) conducted on mood ratings before delayed test showed
a significant main effect of Emotion Category, F (5, 90) ¼ 2.45, p ¼ .039,
gp2 ¼ 0.12. Post-hoc tests showed that mood ratings in the neutral condi-
tion did not significantly differ from those in all other conditions (all ps >
0.21). Mood ratings in all emotion conditions did not significantly differ
from each other (all ps > 0.13) except that mood ratings in the disgust
condition were marginally significantly lower than in the happiness condi-
tion (p ¼ 0.057).
The ANOVA on arousal ratings before delayed test showed a significant
main effect of Emotion Category, F (5, 90) ¼ 3.30, p ¼ .009, gp2 ¼ 0.16.
Post-hoc tests showed that arousal ratings in the neutral condition did not
significantly differ from those in all other conditions (all ps > 0.32).
Arousal ratings in all emotion conditions did not significantly differ from
each other (all ps > 0.19) except that arousal ratings in the disgust condi-
tion were significantly lower than in the happiness condition (p ¼ .006).

Effect on overall recognition

Table 4 shows overall hit rates and false alarm rates for words presented
once and four times in the six emotion conditions. A 6 (Emotion Category:
neutral, anger, sadness, fear, disgust, or happiness)  2 (Presentation
204 B. WANG

Table 4. Overall Hit Rates and False Alarm Rates for Words Presented Once and Four Times in
the Six Emotion Categories.
Emotion category
Neutrality Anger Sadness Fear Disgust Happiness
Presentation
times Measure M SE M SE M SE M SE M SE M SE
Once Overall Hit Rates 0.64 0.05 0.60 0.05 0.69 0.05 0.68 0.05 0.57 0.05 0.66 0.05
Overall False Alarm Rates 0.05 0.03 0.15 0.03 0.07 0.03 0.05 0.03 0.03 0.03 0.07 0.03
Four times Overall Hit Rates 0.65 0.05 0.67 0.05 0.77 0.05 0.67 0.05 0.63 0.05 0.78 0.05
Overall False Alarm Rates 0.05 0.03 0.15 0.03 0.07 0.03 0.05 0.03 0.03 0.03 0.07 0.03
Note. These hit rates and false alarm rates are not adjusted and are the same as simple proportions of hits and
false alarms.

1HXWUDO
2YHUDOO5HFRJQLWLRQ 3U


$QJHU

6DGQHVV

 )HDU
 'LVJXVW
 +DSSLQHVV



2QFH )RXUWLPHV
3UHVHQWDWLRQ7LPHV
Figure 4. Overall recognition (hit rates minus false alarm rates) as a function of presentation
times in the six emotion conditions. Overall recognition was adjusted in that it was calculated
by subtracting false alarm rates from hit rates. Error bars represent standard errors.

Times: once vs. four times) ANOVA on overall recognition showed a sig-
nificant main effect of Presentation Times, F (1, 90) ¼ 7.14, p ¼ .009,
gp2 ¼ 0.07, indicating better overall recognition for words that were pre-
sented four times (M ¼ 0.62, SE ¼ 0.02) than for words presented once
(M ¼ 0.57, SE ¼ 0.02). There was also a significant main effect of Emotion
Category, F (5, 90) ¼ 2.76, p ¼ .023, gp2 ¼ 0.13. The interaction between
Emotion Category and Presentation Times was not significant as shown in
Figure 4, F (5, 90) < 1.
With regard to the significant main effect of Emotion Category, planned
contrasts showed that overall recognition in the anger condition (M ¼ 0.48,
SE ¼ 0.04) was significantly lower than in the neutral condition (M ¼ 0.59,
SE ¼ 0.04) (p ¼ .049). There were no significant differences between all
other emotion conditions and the neutral condition (all ps > 0.26). Post
hoc tests showed that overall recognition in the anger condition was signifi-
cantly lower than in the happiness condition (M ¼ 0.65, SE ¼ 0.04)
(p ¼ .037) and the sadness condition (M ¼ 0.65, SE ¼ 0.04) (p ¼ .039). All
other comparisons were not significant (all ps > 0.20). The pattern of
results was similar when d’ was used as the dependent variable.
 THE JOURNAL OF GENERAL PSYCHOLOGY 205

Table 5. Overall Hit Rates and False Alarm Rates concerning R and K Responses for Words
Presented Once and Four Times in the Six Emotion Categories.
Emotion Category
Neutrality Anger Sadness Fear Disgust Happiness
Presentation times Measure M SE M SE M SE M SE M SE M SE
Once “R” Hit Rates 0.62 0.05 0.47 0.05 0.60 0.05 0.64 0.05 0.55 0.05 0.58 0.05
"K" Hit Rates 0.02 0.03 0.13 0.03 0.08 0.03 0.04 0.03 0.02 0.03 0.09 0.03
“R” False Alarm Rates 0.03 0.02 0.10 0.02 0.04 0.02 0.03 0.02 0.02 0.02 0.02 0.02
“K” False Alarm Rates 0.02 0.01 0.05 0.01 0.04 0.01 0.02 0.01 0.01 0.01 0.05 0.01
Four times “R” Hit Rates 0.62 0.05 0.62 0.05 0.73 0.05 0.64 0.05 0.62 0.05 0.73 0.05
“K” Hit Rates 0.03 0.02 0.05 0.02 0.03 0.02 0.03 0.02 0.02 0.02 0.05 0.02
“R” False Alarm Rates 0.03 0.02 0.10 0.02 0.04 0.02 0.03 0.02 0.02 0.02 0.02 0.02
“K” False Alarm Rates 0.02 0.01 0.05 0.01 0.04 0.01 0.02 0.01 0.01 0.01 0.05 0.01
Note. These hit rates and false alarm rates are not adjusted and are the same as simple proportions of hits and
false alarms.

Effect on recollection
Table 5 shows overall hit rates and false alarm rates for words presented
once and four times in the six emotion categories. The ANOVA on recol-
lection (i.e., adjusted hit rates for “R” responses) showed a significant main
effect of Presentation Times, F (1, 90) ¼ 11.47, p ¼ .001, gp2 ¼ 0.11, indicat-
ing that words presented four times (M ¼ 0.62, SE ¼ 0.02) were more often
recollected than words presented once (M ¼ 0.54, SE ¼ 0.02). There was
also a significant main effect of Emotion Category, F (5, 90) ¼ 2.57,
p ¼ 0.032, gp2 ¼ 0.13. The interaction between Emotion Category and
Presentation Times was not significant as shown in Figure 5, F (5,
90) ¼ 1.41, p ¼ .23, gp2 ¼ 0.07. The planned contrasts showed that recollec-
tion in the anger condition (M ¼ 0.45, SE ¼ 0.04) was significantly lower
than in the neutral condition (M ¼ 0.59, SE ¼ 0.04) (p ¼ .025). Post hoc
tests showed that participants who watched the anger-inducing video recol-
lected fewer words than those who watched the sadness-inducing video
(M ¼ 0.63, SE ¼ 0.04) (p ¼ .047). There was a trend that participants who
watched the anger-inducing video recollected fewer words than those who
watched the happiness-inducing video (M ¼ 0.63, SE ¼ 0.04) (p ¼ .053). All
other comparisons were not significant (all ps > 0.14).

Effect on familiarity
The ANOVA on familiarity (i.e., adjusted hit rates for “K” responses) also
showed a significant main effect of Presentation Times, F (1, 90) ¼ 9.04,
p ¼ 0.003, gp2 ¼ 0.09, characterized by lower familiarity for words presented
four times (M ¼ 0.004, SE ¼ 0.006) than for words presented once
(M ¼ 0.03, SE ¼ 0.009). The main effect of Emotion Category was not sig-
nificant, F (5, 90) ¼ 0.62, p ¼ 0.69, gp2 ¼ 0.03. There was a significant inter-
action between Presentation Times and Emotion Category as shown in
206 B. WANG

1HXWUDO

 $QJHU
6DGQHVV

5HFROOHFWLRQ

 )HDU
 'LVJXVW
 +DSSLQHVV



2QFH )RXUWLPHV
3UHVHQWDWLRQ7LPHV
Figure 5. Recollection as a function of presentation times in the six emotion conditions.
Recollection was adjusted in that it was derived from subtracting the proportion of new items
receiving “R” (i.e., “remember”) responses from the proportion of old items receiving “R”
responses. Error bars represent standard errors.



 1HXWUDO
$QJHU
 6DGQHVV
)HDU
 'LVJXVW
)DPLOLDULW\

+DSSLQHVV





2QFH )RXUWLPHV



3UHVHQWDWLRQ7LPHV
Figure 6. Familiarity as a function of presentation times in the six emotion conditions.
Familiarity was adjusted in that it was computed from “K” (i.e., “know”) responses (K ¼ Khit/
(1Rhit))(Kfa/(1Rfa)), where Khit, Rhit, Kfa, and Rfa respectively represent “K” hit rates, “R”
hit rates, “K” false alarm rates, and “R” false alarm rates. Error bars represent standard errors.

Figure 6, F (5, 90) ¼ 3.15, p ¼ 0.012, gp2 ¼ 0.15. For words presented four
times, planned contrasts showed no significant differences among emotion
conditions and the neutral condition (all ps > 0.25). For words presented
once, planned contrasts showed that familiarity in the anger condition
(M ¼ 0.07, SE ¼ 0.02) was significantly higher than in the neutral condition
(M ¼ .002, SE ¼ 0.02) (p ¼ 0.012). No significant differences were found
between other emotion conditions and the neutral condition (all ps > 0.10).
Whether for words presented four times or once, post-hoc tests showed no
significant differences among the emotion conditions (all ps > 0.26).
Post hoc tests were also made between the “once” and “four times” con-
ditions in each emotion condition. The results showed that familiarity for
prioritized words was lower than for non-prioritized words in both the
 THE JOURNAL OF GENERAL PSYCHOLOGY 207

anger and sadness conditions (p < .001 and p ¼ .019, respectively). In other
emotion conditions, no significant differences were observed (ps > 0.11).
Therefore, the source of interaction comes from the differences between
one and four presentations in some (but not all) emotion categories.

Discussion
Built on past studies that examined the enhancing effect of post-encoding
emotion on long-term memory, the current study sought to answer two
questions: (1) Does the enhancing effect vary depending on a specific cat-
egory of emotion? (2) Is the enhancing effect contingent on memory
strength? Based on differential mechanisms underlying discrete emotions
(e.g., Sinha et al., 1992) and distinct effects of discrete emotions on mem-
ory (e.g., Levine & Burgess, 1997), the effect of post-encoding emotion was
expected to vary depending on a specific category of emotion. In addition,
it was hypothesized that memory strength would be modulatory in that
weak memory rather than strong memory would be enhanced.
The current finding shows that participants who watched the anger-
inducing video had lower overall recognition than those who watched the
sadness-inducing or happiness-inducing video. Furthermore, this pattern
was generally replicated when recollection was used as the dependent vari-
able although it should be noted that the difference between the anger and
happiness conditions was marginal. Regardless of whether overall recogni-
tion or recollection is used, no significant differences were observed
between the neutral and all other emotion conditions (i.e., sadness, fear,
disgust, and happiness). Furthermore, no significant differences were found
for all other pairwise comparisons (i.e., anger vs. fear, anger vs. disgust,
sadness vs. fear, sadness vs. disgust, sadness vs. happiness, fear vs. disgust,
fear vs. happiness, disgust vs. happiness). Taken together, the current
results provide partial support to the hypothesis that the effect of post-
encoding emotion would vary depending on a specific category of emotion.
It must be noted, however, that the above statement is true in comparing
one emotion condition to another, but not when comparing one emotion
condition to the neutral condition.
With regard to the second question, the current study shows that mem-
ory strength does not modulate the effect of post-encoding emotion on
overall recognition and recollection. With familiarity as the dependent
variable, however, there was an interaction between memory strength and
emotion category (i.e., familiarity was lower for prioritized words than for
non-prioritized words in the “anger” and “sadness” categories). Taken
together, the current finding provides preliminary support to the hypothesis
208 B. WANG

that memory strength would modulate the effect of post-encoding emotion

depending on specific components of long-term recognition memory.

Effects of discrete emotions

Although a number of studies have suggested the importance of discrete
emotions (e.g., Wang, 2013), in most previous studies emotions were
induced during encoding. The present investigation contributes to the lit-
erature by showing that, even when discrete emotions are temporally sepa-
rated from encoded stimuli, discrete emotions still have differential effects
on memory. The importance of examining discrete emotions is particularly
illustrated by the fact that, although valence and arousal ratings were highly
similar in the anger-inducing and fear-inducing conditions, only partici-
pants in the anger-inducing condition, but not the fear-inducing condition,
had lower overall recognition than those in the condition that induced sad-
ness or happiness.
Some findings from the current study are inconsistent with those from
prior research. First, it has been found that disgust as induced by the dental
surgery video can enhance long-term memory (e.g., Nielson et al., 2005),
but the current study does not find any effect of disgust. One reason may
be that, for participants in the current study, the video about dental surgery
was not effective in increasing arousal, as indicated by the result that
arousal ratings immediately after video presentation did not significantly
differ from those immediately before video presentation. Another possibil-
ity is that, with a short retention interval of 30 minutes, disgust is more
likely to enhance memory as measured by free recall. In fact, the study by
Nielson et al. (2005) showed that disgust enhanced recognition memory in
the 24 h delayed test, but it is unknown whether the enhancement effect
would be observed in the 30 min delayed test.
Another finding inconsistent with the literature is that the sadness-induc-
ing video, although effectively enhancing arousal ratings and reducing
mood ratings, did not enhance memory relative to the neutral condition.
The study by Wang and Fu (2010), however, showed that the same sad-
ness-inducing video enhanced long-term recognition memory for females.
It is unclear what has caused such a discrepancy, but it should be noted
that in their study there was a baseline test after encoding. As can be seen
from Figures 4 and 5, in the current study both overall recognition and
recollection for words presented four times in the sadness-inducing condi-
tion were numerically higher than in the neutral condition. It is therefore
likely that the presence of a baseline test, which serves as a retrieval prac-
tice and enhances memory strength (Schoch, Cordi, & Rasch, 2017), can
 THE JOURNAL OF GENERAL PSYCHOLOGY 209

augment the extent to which sadness enhance recognition memory. This

possibility should be tested in future studies.
Although a number of prior studies have examined the effect of post-
encoding emotion, the current study is the first to investigate the influence
of fear induced after encoding on long-term memory. Because damages to
the amygdala and insula have been found to be associated with deficits in
fear and disgust respectively (c.f., Calder et al., 2001) and the amygdala
plays a critical role in consolidating memory stored in other brain regions
such as the hippocampus (McGaugh, 2004), these two emotions could be
expected to have differential effects on long-term memory, which, surpris-
ingly, were not observed in the current study. One explanation is that
long-term memory, although benefiting from the amygdala-hippocampus
interaction, can also benefit from the release of a series of neurotransmit-
ters such as the epinephrine (McGaugh, 2018).
The current study makes a theoretical contribution by pinpointing the
inadequacy of the arousal-based account. In most past studies, after encod-
ing participants were typically presented with one of three types of videos:
a neutral video (e.g., tooth brushing), a comic video, or a negative video.
Although some researchers have suggested that arousal is the crucial factor
given the effectiveness of both positive and negative arousal in enhancing
long-term memory (Nielson & Powless, 2007), it is apparent that a solid
conclusion would require the manipulation of different categories of nega-
tive emotions. The current study poses a challenge to the view that arousal
per se drives the effect of post-encoding emotion by showing lower overall
recognition in the anger-inducing condition than in the happiness-inducing
condition even participants in the two conditions had similar
arousal ratings.
Some researchers have pointed out that the motivational component of
emotion is important. Specifically, they have distinguished between the
effects of pre-goal (i.e., anger, fear, disgust) and post-goal emotions (i.e.,
happiness, sadness) (Kaplan et al., 2012). Whereas pre-goal emotions
expand the range of information attended to and encoded in memory,
post-goal emotions reduce it. Therefore, relative to pre-goal emotions (e.g.,
anger, fear and disgust), post-goal emotions (e.g., sadness and happiness)
can be expected to enhance recollection (i.e., processing of details associ-
ated with an encoding episode). The current finding provides partial sup-
port to this prediction because recollection in the anger-inducing
condition, but not in the fear-inducing or disgust-inducing condition, was
lower than recollection in the sadness-inducing and marginally lower than
recollection in the happiness-inducing condition. Therefore, although the
motivational account sheds light on understanding the effect of discrete
emotions, it is important to recognize that not all pre-goal emotions are
210 B. WANG

created equal when it comes to the effect on long-term recogni-

tion memory.

Modulation of memory strength

It has been documented that post-encoding emotional learning (created by
pairing object pictures with electronic shock) enhances weak memory (for
pictures presented once) rather than strong memory (for pictures encoded
three times) (Dunsmoor et al., 2015). The current study is the first to dis-
cover that the modulation of memory strength depends on specific memory
components. Although memory strength does not modulate the effect on
overall recognition or recollection, there is a significant interaction between
emotion category and memory strength with familiarity as the dependent
variable. Particularly, there is a trend that anger, relative to the neutral con-
dition, enhances consolidation of weak familiarity-based memory.
Although the current study is not in the position to provide an exact
account for why the effect only occurs for familiarity-based memory, a
study by Yonelinas et al. (2011) has shown that post-encoding stress caused
by skydiving enhanced familiarity-based recognition, but not free recall or
recollection for neutral pictures, indicating that stress enhanced the under-
lying memory strength rather than generation process involved in recogni-
tion memory. According to their speculation, post-encoding stress did not
enhance free recall because there might be lingering effect of stress at
retrieval. In the current study, it is likely that the anger-inducing video also
created too much stress in the participants, rendering it more difficult for
them to retrieve details related to the learning episode. Taken together, the
current results suggest that whether memory strength can be modulatory
might depend on the category of negative emotion and component of rec-
ognition memory. The current study thus contributes to the literature by
pinpointing the boundary conditions under which memory strength modu-
lates the effects of post-encoding discrete emotions.

Limitations and future directions

Despite the contributions of the current study, several limitations merit
mentioning. First, the effects revealed by the current study have been lim-
ited. Particularly, there were less recognition and recollection when anger
was the post-encoding emotion, which may be explained from several
aspects. First, although the impairing effect of anger was not expected,
there has been evidence that anger can lead to a decrease in cortisol, which
is inversely related to cognitive performance and explicit memory (Kazen,
Kuenne, Heiko Frankenberg, & Quirin, 2012). The current impairing effect
 THE JOURNAL OF GENERAL PSYCHOLOGY 211

of anger, therefore, might be attributed to a decrease in cortisol. Second, it

is likely that post-encoding emotions often have a facilitating effect but not
(significantly) here because the anger-inducing video might have created
more interferences than the other videos. Although the current study is not
in a position to rule out completely this possibility, it should be noted that,
if interference per se were the driving factor, the fear-inducing video would
have also impaired memory. But this is not the case observed. Third, par-
ticipants who watched the anger-inducing video were of lower memory
ability. This account, however, may not hold because, when it comes to
familiarity-based recognition memory, participants in the anger-inducing
condition did not significantly differ from those in other emo-
tion conditions.
Second, the current results have not clearly demonstrated that any one
emotion condition produces better or poorer memory relative to the con-
trol condition. For instance, overall recognition in the happiness and sad-
ness conditions were better than the anger condition. It is difficult to state,
however, whether anger reduced recognition or happiness increased it.
Future studies should further explore the effects of discrete emotions rela-
tive to the effect of a neutral condition.
Third, although the “remember-know” (RK) procedure has been widely
used to examine recollection and familiarity (e.g., Yonelinas, 2002), some
researchers have pointed out that the RK procedure may have the draw-
back of simplify measuring memory confidence or strength (e.g., Wais,
Mickes, & Wixted, 2008). In future studies, it is important to use improved
procedures such as the receiver operating characteristics (ROC) (Yonelinas,
1999) to investigate whether discrete emotions have differential effects on
the two components of recognition memory.
Fourth, the values of familiarity in the current study are very low and
data points may not be sufficient to allow meaningful conclusions about
“know” judgments. One reason for the low values of familiarity is that the
retention interval is so short that, at test, participants were still able to
retrieve details associated with the learning episode and thus make fewer
“know” than “remember” responses. Indeed, overall, the number of
“remember” responses (M ¼ 73.92) is significantly greater than that of
“know”responses (M ¼ 10.29). In future studies, a longer retention interval
should be used to induce a larger number of “know” responses.
Finally, the present study shows that, with the overall recognition and
recollection as the dependent measures, anger has an impairing effect rela-
tive to the neutral condition as well as the sadness and happiness condi-
tions. However, no effect on familiarity was found, which appears to be
inconsistent with prior studies demonstrating the facilitating effect of post-
encoding stress on familiarity-based recognition memory (McCullough &
212 B. WANG

Yonelinas, 2013). Given the range of emotions, it is unclear whether the

current null findings can be generalized or replicated. Given that the reten-
tion interval was longer (i.e., more than 60 mins) in the study by
McCullough and Yonelinas (2013), future investigations need to examine
whether extending retention interval will enhance familiarity-based recogni-
tion memory. Increasing retention interval is also important because post-
encoding emotional learning has been found to enhance weak memory
only if the memory test occurs at a 6 h or 24 h delay (Dunsmoor
et al., 2015).

Note
1. Test words on screen for only 1500 ms rather than remaining on screen until a
response was given because this ensures that all participants spent exactly the same
amount of time viewing each word. This practice is in accordance with Wang and
Ren (472017).

Funding
This study was supported by a Grant from the National Natural Science Foundation of
China [No. 31100736].

References
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for
measuring depression. Archives of General Psychiatry, 4, 561–571.
Calder, A. J., Lawrence, A. D., & Young, A.W. (2001). Neuropsychology of fear and loath-
ing. Nature Reviews Neuroscience, 2(5), 352–363. doi:10.1038/35072584
Collet, C., Vernet-Maury, E., Delhomme, G., & Dittmar, A. (1997). Autonomic system
response patterns specificity to basic emotions. Journal of the Autonomic Nervous System,
62(1–2), 45–57. doi:10.1016/S0165-1838(96)00108-7
Coren, S. (1988). Prediction of insomnia from arousability predisposition scores: Scale
development and cross-validation. Behaviour Research and Therapy, 26(5), 415–420. doi:
10.1016/0005-7967(88)90076-9
Dudai, Y. (2004). The Neurobiology of Consolidations, Or, How Stable is the Engram?
Annual Review of Psychology, 55, 51–86. doi:10.1146/annurev.psych.55.090902.142050
Dunsmoor, J. E., Murty, V. P., Davachi, L., & Phelps, E. A. (2015). Emotional learning
selectively and retroactively strengthens memories for related events. Nature, 520(7547),
345–348. doi:10.1038/nature14106
Ekman, P., & Cordaro, D. (2011). What is meant by calling emotions basic. Emotion
Review, 3(4), 364–370. doi:10.1177/1754073911410740
Gable, P., & Harmon-Jones, E. (2010). The blues broaden, but the nasty narrows.
Psychological Science, 21(2), 211–215. doi:10.1177/0956797609359622
Gross, J. J., & John, O. P. (2003). Individual differences in two emotional regulations proc-
esses: Implications for affect, relationships and well-being. Journal of Personality and
Social Psychology, 85(2), 348–362. doi:10.1037/0022-3514.85.2.348
 THE JOURNAL OF GENERAL PSYCHOLOGY 213

Jack, R., Garrod, O., & Schyns, P. (2014). Dynamic facial expressions of emotion transmit
an evolving hierarchy of signals over time. Current Biology: CB, 24(2), 187–192. doi:10.
1016/j.cub.2013.11.064
Judde, S., & Rickard, N. (2010). The effect of postlearning presentation of music on long-
term word-list retention. Neurobiology of Learning and Memory, 94(1), 13–20. doi:10.
1016/j.nlm.2010.03.002
Kaplan, R. L., Damme, I. V., & Levine, L. J. (2012). Motivation matters: Differing effects of
pre-goal and post-goal emotions on attention and memory. Frontiers in Psychology, 3,
404. doi:10.3389/fpsyg.2012.00404
Kazen, M., Kuenne, T., Heiko Frankenberg, H., & Quirin, M. (2012). Inverse relation
between cortisol and anger and their relation to performance and explicit memory.
Biological Psychology, 91(1), 28–35. doi:10.1016/j.biopsycho.2012.05.006
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008). International affective picture system
(IAPS): Affective ratings of pictures and instruction manual. Technical Report A-7.
University of Florida, Gainesville, FL.
Levine, L., & Burgess, S. L. (1997). Beyond general arousal: Effects of specific emotions on
memory. Social Cognition, 15(3), 157–181. doi:10.1521/soco.1997.15.3.157
Liu, D. L. J., Graham, S., & Zorawski, M. (2008). Enhanced selective memory consolidation
following post-learning pleasant and aversive arousal. Neurobiology of Learning and
Memory, 89(1), 36–46. doi:10.1016/j.nlm.2007.09.001
Madan, C. R., Scott, S., & Kensinger, E. (2018). Positive emotion enhances association-
memory. Emotion, 19(4), 733–740.
Mather, M., & Sutherland, M. (2009). Disentangling the effects of arousal and valence on
memory for intrinsic details. Emotion review: Journal of the International Society for
Research on Emotion, 1(2), 118–119. doi:10.1177/1754073908100435
McCabe, D. P., & Geraci, L. D. (2009). The influence of instructions and terminology on
the accuracy of remember–know judgments. Consciousness and Cognition, 18(2),
401–413. doi:10.1016/j.concog.2009.02.010
McCullough, A. M., & Yonelinas, A. P. (2013). Cold-pressor stress after learning enhances
familiarity-based recognition memory in men. Neurobiology of Learning and Memory,
106, 11–17. doi:10.1016/j.nlm.2013.06.011
McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories of emo-
tionally arousing experiences. Annual Review of Neuroscience, 27, 1–28. doi:10.1146/
annurev.neuro.27.070203.144157
McGaugh, J. L. (2018). Emotional arousal regulation of memory consolidation. Current
Opinion in Behavioral Sciences, 19, 55–60. doi:10.1016/j.cobeha.2017.10.003
Nielson, K. A., & Arentsen, T. J. (2012). Memory modulation in the classroom:
Selective enhancement of college examination performance by arousal induced after
lecture. Neurobiology of Learning and Memory, 98(1), 12–16. doi:10.1016/j.nlm.2012.04.
002
Nielson, K. A., & Lorber, W. (2009). Enhanced post-learning memory consolidation
is influenced by arousal predisposition and emotion regulation but not by stimulus
valence or arousal. Neurobiology of Learning and Memory, 92(1), 70–79. doi:10.1016/j.
nlm.2009.03.002
Nielson, K. A., & Powless, M. (2007). Positive and negative sources of emotional arousal
enhance long-term word-list retention when induced as long as thirty minutes after
learning. Neurobiology of Learning and Memory, 88(1), 40–47. doi:10.1016/j.nlm.2007.03.
005
214 B. WANG

Nielson, K. A., Yee, D., & Erickson, K. I. (2005). Memory enhancement by a semantically
unrelated emotional arousal source induced after learning. Neurobiology of learning and
memory, 84(1), 49–56. doi:10.1016/j.nlm.2005.04.001
Paivio, A., Yuille, J. C., & Madigan, S. A. (1968). Concreteness, imagery, [Database] and
meaningfulness values for 925 nouns. Journal of Experimental Psychology, 76(1, Pt.2),
1–25. doi:10.1037/h0025327
Rainville, P., Bechara, A., Naqvi, N., & Antonio, R. (2006). Damasio basic emotions are
associated with distinct patterns of cardiorespiratory activity. International Journal of
Psychophysiology, 61(1), 5–18. doi:10.1016/j.ijpsycho.2005.10.024
Rajaram, S. (1993). Remembering and knowing: Two means of access to the personal past.
Memory & Cognition, 21(1), 89–102. doi:10.3758/bf03211168
Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social
Psychology, 39(6), 1161–1178. doi:10.1037/h0077714
Saarim€aki, H., Gotsopoulos, A., J€a€askel€ainen, I. P., Lampinen, J., Vuilleumier, P., Hari, R.,
Sams, M., & … ummenmaa, L. (2016). Discrete Neural Signatures of Basic Emotions.
Cerebral Cortex, 26(6), 2563–2573. doi:10.1093/cercor/bhv086
Schoch, S. F., Cordi, M. J., & Rasch, B. (2017). Modulating influences of memory strength
and sensitivity of the retrieval test on the detectability of the sleep consolidation effect.
Neurobiology of Learning and Memory, 145, 181–189. doi:10.1016/j.nlm.2017.10.009
Sch€onauer, M. (2018). Sleep spindles: Timed for memory consolidation. Current Biology:
CB, 28(11), R656–R658. doi:10.1016/j.cub.2018.03.046
Sharot, T., & Yonelinas, A. P. (2008). Differential time-dependent effects of emotion on
recollective experience and memory for contextual information. Cognition, 106(1),
538–547. doi:10.1016/j.cognition.2007.03.002
Sinha, R., Lovallo, W. R., & Parson, O. A. (1992). Cardiovascular differentiation of emo-
tions. Psychosomatic Medicine, 54(4), 422–435. doi:10.1097/00006842-199207000-00005
Snodgrass, J. G., & Corwin, J. (1988). Pragmatics of measuring recognition memory:
Applications to dementia and amnesia. Journal of Experimental Psychology: General,
117(1), 34–50. doi:10.1037/0096-3445.117.1.34
Spielberger, C. D., Gorsuch, R. L., Lushene, P. R., Vagg, P. R., & Jacobs, A. G. (1983).
Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto: Consulting
Psychologists Press, Inc.
Wais, P. E., Mickes, L., & Wixted, J. T. (2008). Remember/know judgments probe degrees
of recollection. Journal of Cognitive Neuroscience, 20(3), 400–405. doi:10.1162/jocn.2008.
20041
Wang, B. (2013). Facial expression influences recognition memory for faces: Robust
enhancement effect of fearful expression. Memory (Hove, England), 21(3), 301–314. doi:
10.1080/09658211.2012.725740
Wang, B. (2015). Negative emotion elicited in high school students enhances consolidation
of item memory, but not source memory. Consciousness and Cognition, 33, 185–195. doi:
10.1016/j.concog.2014.12.015
Wang, B., & Fu, X. (2010). Gender differences in the effects of post-learning emotion on
consolidation of item memory and source memory. Neurobiology of Learning and
Memory, 93(4), 572–580. doi:10.1016/j.nlm.2010.02.005
Wang, B., & Ren, Y. (2017). Effect of post-encoding emotion on recollection and familiarity
for pictures. Quarterly Journal of Experimental Psychology (2006), 70(7), 1236–1253. doi:
10.1080/17470218.2016.1178311
 THE JOURNAL OF GENERAL PSYCHOLOGY 215

Wang, B., & Sun, B. (2015). Timing matters: Negative emotion elicited 5 min but not
30 min or 45 min after learning enhances consolidation of internal-monitoring source
memory. Acta Psychologica, 157, 56–64. doi:10.1016/j.actpsy.2015.02.006
Wang, B., & Sun, B. (2017). Post-encoding emotional arousal enhances consolidation of
item memory, but not reality-monitoring source memory. The Quarterly Journal of
Experimental Psychology, 70(3), 461–472. doi:10.1080/17470218.2015.1134604
Yonelinas, A. (1999). The contribution of recollection and familiarity to recognition and
source-memory judgments: A formal dual-process model and an analysis of receiver
operating characteristics. Journal of Experimental Psychology: Learning Memory and
Cognition, 25(6), 1415–1434. doi:10.1037/0278-7393.25.6.1415
Yonelinas, A. P. (2002). The nature of recollection and familiarity: A review of 30 years of
research. Journal of Memory and Language, 46(3), 441–517. doi:10.1006/jmla.2002.2864
Yonelinas, A. P., Parks, C. M., Koen, J. D., Jorgenson, J., & Mendoza, S. P. (2011). The
effects of post-encoding stress on recognition memory: Examining the impact of skydiv-
ing in young men and women. Stress (Amsterdam, Netherlands), 14(2), 136–144. doi:10.
3109/10253890.2010.520376

Appendix I
The Chinese words (with their English words and parameters) used in the study.

English Chinese Type Pleasantness Arousal Familiarity Concreteness

Baby 宝贝 Target 6.79 6.58 6.79 6.47
Winter 冬天 Target 6.11 6.37 6.68 6.89
House 房子 Target 6.79 6.68 7.11 7.11
Wine 果酒 Target 6.05 6.05 6.11 6.32
Child 孩子 Target 6.63 6.84 7.05 7.37
Butterfly 蝴蝶 Target 6.42 6.53 6.74 7.11
Blossom 花朵 Target 6.21 6.26 6.89 6.89
Gold 黄金 Target 6 5.68 5.95 6.26
Butterfly 黄油 Target 4.89 5.16 5.21 5.89
Arrow 箭头 Target 5.53 5.58 6.53 6.79
Ankle 脚踝 Target 5.16 5.53 7.32 7.11
Army 军队 Target 6.74 6.74 6.68 7.63
Dress 礼服 Target 6.42 6.42 6.32 6.21
Shotgun 猎枪 Target 5.32 5.42 4.95 6.11
Lobster 龙虾 Target 6.58 6.47 6.89 7.37
Ship 轮船 Target 6.37 5.95 6.11 6.84
Tweezers 镊子 Target 5.47 5.37 6.32 6.74
Queen 女王 Target 6.21 6.21 6.58 6.21
Apple 苹果 Target 6.79 6.68 8 8.11
Wife 妻子 Target 6.42 6.21 5.58 6.89
Grass 青草 Target 5.95 5.84 6.79 6.79
Salad 沙拉 Target 6.26 6.47 6.58 6.89
Corpse 尸体 Target 2.63 3.11 1.79 3.74
Sugar 食糖 Target 5.84 5.89 6.79 7.11
Woods 树林 Target 6.47 6.32 6.79 7.11
Blister 水疱 Target 3.37 3.42 4.42 4.95
Mosquito 蚊子 Target 4.68 5.42 6.84 7.21
Nail 指甲 Target 6 5.74 7.05 7.47
Home 住所 Target 6.47 6.63 7.21 7.32
(continued)
216 B. WANG

Continued.
English Chinese Type Pleasantness Arousal Familiarity Concreteness
Lip 嘴唇 Target 5.79 5.89 6.74 7.74
Plank 板条 Distractor 5.32 5.21 5.63 5.58
Gem 宝石 Distractor 6.58 6.37 5.89 6.16
Newspaper 报纸 Distractor 5.79 5.95 6.53 7.16
Leopard 豹子 Distractor 6.16 6 5.79 6.79
Refrigerator 冰箱 Distractor 6.32 5.95 7.05 6.79
Spinach 菠菜 Distractor 5.47 6.11 6.95 7.32
Meadow 草甸 Distractor 5.84 5.84 5.53 5.63
Lawn 草坪 Distractor 6.11 5.89 6.58 6.95
Fork 叉子 Distractor 5.42 5.74 6.63 6.53
Hammer 锤子 Distractor 5.74 5.47 6.79 7.11
Hall 大厅 Distractor 6.16 6.26 7 7.63
University 大学 Distractor 5.89 5.68 7.11 7
Geese 鹅肉 Distractor 5.26 4.89 5.16 5.05
Bowl 饭碗 Distractor 6.68 6.16 7.32 7.47
Tomb 坟墓 Distractor 4.53 4.47 5.37 5.74
Cane 甘蔗 Distractor 6.63 6.37 7.05 7
Piano 钢琴 Distractor 6.74 6.26 6.21 7
Dove 鸽子 Distractor 6.42 6.32 6.68 6.95
Factory 工厂 Distractor 5.37 5.21 5.79 6.53
Palace 宫殿 Distractor 5.53 6.11 5.42 6.84
Shore 海岸 Distractor 6.79 6.37 7 7.37
Beaver 海狸 Distractor 5.63 5.32 4.58 4.63
Ocean 海洋 Distractor 6.21 6 6.79 6.84
River 河流 Distractor 6 5.74 7 6.84
Flood 洪水 Distractor 4.95 4.95 5.37 6.32
Fox 狐狸 Distractor 5.32 5.63 6.05 6.16
Lake 湖泊 Distractor 6.26 6 6.21 6.95
Bouquet 花束 Distractor 7.11 7.11 7.37 7.26
Garden 花园 Distractor 6.26 6.58 6.95 6.89
Fire 火焰 Distractor 5.53 5.42 6.95 7.21
Jail 监狱 Distractor 4.42 4.63 5.05 5.95
Claw 脚爪 Distractor 5.42 5.32 5.79 5.79
Saloon 轿车 Distractor 6.05 5.89 6.05 6.79
Church 教堂 Distractor 6.37 6.05 6.16 6.53
Money 金钱 Distractor 6.95 6.53 7.42 7.37
Policeman 警察 Distractor 5.42 5.68 6.16 7.16
Bar 酒吧 Distractor 5.68 5.63 5.47 6
Hotel 酒店 Distractor 5.79 5.74 6.58 6.63
Alcohol 酒精 Distractor 5.16 5 5.79 6.11
Coffee 咖啡 Distractor 5.74 5.79 6.21 6.05
Truck 卡车 Distractor 5.74 5.32 6.63 6.95
Oven 烤箱 Distractor 6.16 6.21 6.68 6.89
Dawn 黎明 Distractor 6.74 6.89 7.32 7.11
Mule 骡子 Distractor 5.68 5.74 5.53 5.84
Green 绿色 Distractor 6.58 6.21 7.16 7.16
Vest 马甲 Distractor 5.84 5.84 5.47 6.58
Hoof 马蹄 Distractor 5.05 5.37 5.63 6.32
Caterpillar 毛虫 Distractor 4.47 5.26 6.16 6.53
Rose 玫瑰 Distractor 7 6.68 7.37 7.37
Dollar 美元 Distractor 6.68 6.84 6.21 6.42
Flour 面粉 Distractor 6.32 6 6.58 7.26
Star 明星 Distractor 6.16 5.89 6.26 7.16
Ink 墨水 Distractor 5.63 5.79 6.32 6.79
Barrel 木桶 Distractor 5.53 5.32 6.32 7.16
Priest 牧师 Distractor 5.95 5.89 5.68 5.79
Boy 男孩 Distractor 6.32 6.32 7.42 7.26
Nun 尼姑 Distractor 5 5.05 5.37 5.63
Lemonade 柠檬 Distractor 6.37 6.53 7.16 7.42
Woman 女人 Distractor 5.74 5.79 6.47 6.79
Flask 瓶子 Distractor 6.13 6.11 6.95 7.5
(continued)
 THE JOURNAL OF GENERAL PSYCHOLOGY 217

Continued.
English Chinese Type Pleasantness Arousal Familiarity Concreteness
Cattle 肉牛 Distractor 6.21 5.89 6.16 6.74
Meat 肉片 Distractor 6.53 6.26 6.89 7.16
Forest 森林 Distractor 6.42 5.95 6.47 7.11
Mountain 山峰 Distractor 6.58 6.26 7.16 7.11
Valley 山谷 Distractor 5.84 5.95 6.16 6.47
Cottage 山寨 Distractor 5.89 5.47 6.16 6.11
Clock 时钟 Distractor 5.63 5.79 6.53 6.32
Arm 手臂 Distractor 5.58 5.53 7.11 7.63
Shotgun 手枪 Distractor 5.42 5.21 5.89 6.95
Book 书本 Distractor 6.53 6.21 7.79 7.84
Leaf 树叶 Distractor 6.63 6.26 7.16 7.53
Harp 竖琴 Distractor 6.11 5.79 5.68 6.32
Water 水域 Distractor 5.42 5.05 5.74 5.58
Tower 塔楼 Distractor 5.84 5.47 5.74 6.79
Tablespoon 汤匙 Distractor 6.42 6.05 7 7.21
Candy 糖果 Distractor 6.58 6.11 7 7.21
Peach 桃子 Distractor 6.58 6.53 7.26 7.53
Pupil 瞳孔 Distractor 5.84 5.42 6.26 6.89
Skull 头骨 Distractor 5.16 4.95 5.37 6
Toast 吐司 Distractor 7 6.84 7 7.21
Rattle 响板 Distractor 5.16 4.58 5.11 4.84
Trumpet 小号 Distractor 5.32 5.21 5.05 6.16
Wheat 小麦 Distractor 5.58 5.05 5.47 5.84
Bird 小鸟 Distractor 6.37 6.05 6.58 7.26
Hut 小屋 Distractor 6.11 6 6.26 6.37
Shoes 鞋子 Distractor 6.11 5.95 6.68 6.58
Letter 信件 Distractor 6.16 6.05 6.63 6.79
Blood 血液 Distractor 4.63 4.58 6.58 6.21
Chair 椅子 Distractor 6.37 6.16 7.37 7.47
Engine 引擎 Distractor 5.42 5.11 5.95 5.63
Infant 婴儿 Distractor 6.37 6.16 6.79 6.63
Camp 营地 Distractor 6.21 6.11 6.11 6.05
Coin 硬币 Distractor 6.68 6.74 7.37 7.74
Fisherman 渔夫 Distractor 5.79 5.37 5.84 6.16
Umbrella 雨伞 Distractor 5.79 5.79 6.37 6.95
Tomahawk 战斧 Distractor 5.05 5.11 5.37 5.68
Steamer 蒸笼 Distractor 6.26 6.53 6.89 6.95
Plant 植物 Distractor 5.95 5.79 6.68 6.42
Paper 纸张 Distractor 5.47 6.11 6.53 7.42
Elbow 肘部 Distractor 6.11 5.95 6.79 6.89
Bullet 子弹 Distractor 5.74 5.74 5.53 6.16
Soccer 足球 Distractor 5.68 5.58 5.95 6.74
Diamond 钻石 Distractor 6.95 6.68 6.21 6.32
Note. (1) A total of 19 participants (mean age ¼ 20.21 years, SD ¼ 0.98) were recruited to rate the words. During
the rating, each word was presented at the screen center, and participants were instructed to rate the pleas-
antness, arousal, familiarity and concreteness of each word. The ratings were made using 9-point likert scales
ranging from 1 to 9, with 1 indicating the least degree and 9 the highest degree. For instance, a rating of 1
for concreteness indicates that a word is extremely abstract (not concrete), whereas a rating of 9 for concrete-
ness indicates that a word is extremely concrete. (2) The following three words were not used in the study of
Nielson et al. (2005): 足球(soccer), 面粉(flour), and 树叶(leaf).
218 B. WANG

Appendix II
The instructions for the delayed memory test.
Now we are going to test your memory for the words previously presented. Please first
focus your attention on”þ” at the screen center, and then pay close attention to the word at
the screen center. If you think you saw the word during previous study phase, please press the
key that corresponds to the letter above “saw”; If you think you did not see the word during
previous study phase, please press the key that corresponds to the letter above “did not see”.
After you press the key to indicate that you saw a word, you will be further asked to make
the following judgments. If you can consciously recollect the episode related to the study of
the word (e.g., something that happened during the encoding, or a thought occurring to you
during the study phase) , please press the key that corresponds to the letter above
“remember”. If you felt that you were just familiar with the word but could not retrieve any
details related to the study phase and could not state the reason why you felt a word was
familiar to you, please press the key that corresponds to the letter above “know”.