Writing to your past-self can make you feel better

Abstract

Self-compassionate writing has been shown to be helpful for improving the mental state in some individuals. Here, we investigated how the writer’s attitude toward his/her past, present and future and the focus of the writing, i.e., social experience in the past versus self-experience, modulate these effects. In Experiment 1, 150 undergraduates wrote a compassionate letter to their past-self and to their future-self and responded to the Japanese version of the Adolescent Time Inventory–Time Attitudes (ATI-TA) questionnaire. Writing to past-self decreased negative feelings more than writing to future-self. Further, participants who had negative feelings toward their past, present, and future, as assessed by the ATI-TA, were more likely to be emotionally affected by writing a letter to their past-self. In Experiment 2, 31 undergraduates wrote a letter focusing on what they had experienced together with someone, and another 31 undergraduates wrote focusing on what they had experienced alone. Focusing on a social experience was more helpful for recovering from negative feelings than focusing on a self-experience. In conclusion, writing a compassionate letter to one’s past-self can improve mood, especially in individuals with a negative time attitude who focus their writing on a social connection.

Full text

Frontiers in Psychology 01 frontiersin.org
Writing to your past-self can
make youfeel better
ErikoSugimori
1*, Mayu Yamaguchi
1* and TakashiKusumi
2
1 School of Human Sciences, Faculty of Human Sciences, Waseda University, Tokyo, Japan, 2 Division
of Cognitive Psychology in Education, Graduate School of Education, Kyoto University, Kyoto, Japan
Self-compassionate writing has been shown to be helpful for improving the
mental state in some individuals. Here, weinvestigated how the writer’s attitude
toward his/her past, present and future and the focus of the writing, i.e.,
social experience in the past versus self-experience, modulate these eects.
In Experiment 1, 150 undergraduates wrote a compassionate letter to their
past-self and to their future-self and responded to the Japanese version of the
Adolescent Time Inventory–Time Attitudes (ATI-TA) questionnaire. Writing to
past-self decreased negative feelings more than writing to future-self. Further,
participants who had negative feelings toward their past, present, and future, as
assessed by the ATI-TA, were more likely to beemotionally aected by writing
a letter to their past-self. In Experiment 2, 31 undergraduates wrote a letter
focusing on what they had experienced together with someone, and another 31
undergraduates wrote focusing on what they had experienced alone. Focusing
on a social experience was more helpful for recovering from negative feelings
than focusing on a self-experience. In conclusion, writing a compassionate
letter to one’s past-self can improve mood, especially in individuals with a
negative time attitude who focus their writing on a social connection.
KEYWORDS
autobiographical memory, depression, fading aect bias, nostalgia, group
reminiscence therapy
Introduction
People are likely to have a “rosy view” of their past, a phenomenon called rosy retrospection.
Rosy retrospection is the tendency for an individual to evaluate past events more favorably than
present events (Mitchell etal., 1997; Wirtz etal., 2003). It has been shown that pleasant life events
are better recalled (ompson etal., 2013) and come to mind more readily (Levine and Bluck,
2004) than unpleasant life events. Several studies reported that unpleasant feelings about past events
fade in memory faster than pleasant feelings about past events, called the fading aect bias (Walker
etal., 1997, 2003; Skowronski etal., 2014). Further, a larger emotional-intensity drop for unpleasant
events than for pleasant events has been observed (Walker etal., 1997; Landau and Gunter, 2009;
Ritchie etal., 2009, 2015). During a rosy view, memory is distorted in a positive way. Evidence
supporting this comes from studies showing that college students remembered having more As on
their high school transcripts than they actually had (Bahrick etal., 1996, 2008), and that individuals
remembered their medical-test results as better than they really were (Christensen etal., 2003;
Croyle etal., 2006). e rosy retrospection phenomenon is common to all cultures and has many
implications for contemporary society. Above all it may serve as a coping mechanism that can
promote well-being.
Our research focuses on how rosy retrospection can beharnessed to make a person feel
better. In the current study, wedene positive memory as a memory of a past event that
OPEN ACCESS
EDITED BY
Hans Henrik Knoop,
Aarhus University, Denmark
REVIEWED BY
Antonia Csillik,
Université Paris Nanterre, France
Nasr Chalghaf,
University of Gafsa, Tunisia
*CORRESPONDENCE
Eriko Sugimori
sugimori@waseda.jp
Mayu Yamaguchi
mayuyamaguchi@fuji.waseda.jp
RECEIVED 25 October 2023
ACCEPTED 07 February 2024
PUBLISHED 27 February 2024
CITATION
Sugimori E, Yamaguchi M and
Kusumi T (2024) Writing to your past-self can
make youfeel better.
Front. Psychol. 15:1327595.
doi: 10.3389/fpsyg.2024.1327595
COPYRIGHT
© 2024 Sugimori, Yamaguchi and Kusumi.
This is an open-access article distributed
under the terms of the Creative Commons
Attribution License (CC BY). The use,
distribution or reproduction in other forums is
permitted, provided the original author(s) and
the copyright owner(s) are credited and that
the original publication in this journal is cited,
in accordance with accepted academic
practice. No use, distribution or reproduction
is permitted which does not comply with
these terms.
TYPE Original Research
PUBLISHED 27 February 2024
DOI 10.3389/fpsyg.2024.1327595

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 02 frontiersin.org
elicited an emotionally positive response as described previously
(Williams etal., 2022). When an individual experiences a situation
with predominantly negative feelings, such negative feelings can
trigger the recollection of positive memories, especially memories that
are related to a social connection, which might help to overcome the
challenging situation and enhance well-being. Such positive
autobiographical memories act as rewards in themselves (Speer etal.,
2014). Positive memories have been shown to bepowerful in their
ability to repair mood aer a negative mood induction (Joormann and
Siemer, 2004; Joormann etal., 2007). It was reported that positive
memories can beused as a buer for the eects of negative experiences
(Speer and Delgado, 2017). In their study, Speer & Delgado compared
the stress responses of individuals who recalled a positive or a neutral
memory and showed that individuals who retrieved positive memories
had a smaller cortisol response to the stressor than did individuals
who retrieved neutral memories, and also reported a less negative
aect. Later, the same group showed that recalling positive memories
with a social component could beparticularly powerful in reducing
the cortisol response following the same stressor task (Speer and
Delgado, 2020). at is, rosy retrospection has an impact on mental
health and well-being.
Negative events can beturned into a positive memory and being
listened to with compassion might behelpful in making this happen.
Evidence supporting this come from studies showing that participants
who elaborated on the positive aspects of past negative events reported
increased positive emotions and memory content upon future
recollections of the same negative event, up to 2 months aer the initial
recollection (Speer etal., 2021). Another study reported that memory-
reframing helped children to remember a recent tonsillectomy more
positively than those assigned to a control condition (Pavlova etal.,
2022). Further research found that when listened to attentively, a
positive memory becomes more positive, and a negative memory
becomes less negative (Pasupathi and Oldroyd, 2015). It was also
reported that participants who were listened to with empathy had a
signicantly longer talking time than non-empathy or non-response
participants, higher nostalgia scores than non-response participants,
and higher positive emotion scores than non-empathy and
non-response participants, as well as lower negative emotion scores
than non-response participants (Sugimori etal., 2020). ese studies
collectively show that the current feeling toward an autobiographical
memory is manipulable post collection by exposure to empathy.
On the other hand, one study has reported that a substantial
proportion of individuals conceal signicant pieces of information
and that several categories of nondisclosed information exist (Farber,
2003). In addition, some cultures (East Asia compared to Western;
Chen and Danish, 2010) and some personalities (paranoid ideation;
Murphy etal., 2012) are reportedly less willing to disclose themselves
to an outsider. ese individuals might benet from self-
compassionate interventions, such as letter writing, which can bedone
at one’s own discretion without disclosing private information to
anyone. A lot of research has shown the positive impact of self-
compassion interventions on mental health and well-being, as
reviewed in Kotera and Van Gordon (2021). Previous studies have
found that self-compassionate letter writing is eective in the
treatment of anorexia nervosa (Kelly and Waring, 2018), in the
improvement of women’s body satisfaction (Stern and Engeln, 2018),
and in reducing strong shame (Swee et al., 2023) and smoking
behavior (Kelly etal., 2010). erefore, it is believed that writing a
letter to one’s past-self with compassion, without having to disclose
one’s past or gain empathy from others, can lead to improved
well-being.
In the current study, wecompared writing to past-self versus
writing to future-self. Research has shown that writing to one’s
future-self can have a positive impact on mental health and well-
being (Chishima etal., 2021; Schippers et al., 2023). However, the
mental process behind well-being achieved by writing to one’s
future-self and that achieved by writing to one’s past-self might
bedierent. According to (Chishima etal., 2021; Schippers et al.,
2023), writing a letter to one’s future-self can clarify what one needs
to work on to achieve one’s academic and career goals. In this sense,
writing a letter to one’s future-self potentially creates anxious
feelings by making oneself realize what is not sucient for now,
while writing a letter to one’s past-self comforts one’s feelings. at
is, the mental states aected by writing a letter to one’s future-self
and past-self would bedierent. In addition, the attitude toward
one’s own past, present and future, which is called time attitude,
would inuence how the mental states change by writing a letter to
one’s future and past-self. People who have a positive attitude toward
their past might feel positive by writing a letter to past-self, while
people who have a positive attitude toward their future might feel
positive by writing a letter to future-self. In a similar way, people
who have a negative attitude toward their present might bemore
likely to beinuenced by writing to past-self because their mental
state is unstable and because positive memories have an ability to
repair mood aer a negative mood induction (Joormann and
Siemer, 2004; Joormann etal., 2007).
In the present study, wealso compared writing about a social
experience in the past versus writing about an event experienced alone
in the past. Speer and Delgado (2020) found that people were likely to
choose to reminisce about highly social memories more frequently
than about less social memories of equally positive feeling and
recalling memories that included higher social context led to a greater
dampening of the physiological stress response (i.e., cortisol). On the
other hand, Pillemer etal. (2007) found that memories of positive
self-worth oen focus on achievement/mastery themes, while
memories of negative self-worth focus on interpersonal/aliation
themes. Under the condition of self-compassion, it would
beinteresting to examine whether the recall of social connections or
one’s own personal memories leads to mental stability and well-being.
Aims of the study
In the current study, we rst investigated how writing a self-
compassionate letter to past-self can aect the mental state in healthy
individuals by comparing it to writing a self-compassionate letter to
future-self in Experiment 1. Wealso investigated what kind of people
are inuenced by writing to their past−/future-selves from the viewpoint
of time attitude, i.e., the attitude toward one’s own past, present and
future. In Experiment 2, weinvestigated the eects of focusing the
writing on a social experience in the past versus focusing on an event
experienced alone. All participants were healthy undergraduates.
erefore, the results need to beinterpreted with respect to the limited
demographic characteristics of the participants. However, the ndings
conrmed our predictions and have implications for the development
of more eective self-compassionate intervention tools.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 03 frontiersin.org
Methods
Study design and study period
e study was designed as an independent measures study.
Experiments were carried out from May to August 2022 (Experiment
1) and from September to October 2022 (Experiment 2).
Participant recruitment
Participants were recruited from among undergraduate students
of Waseda University by way of yers posted around campus and by
sending out recruitment e-mails. To be included in the study,
participants had to beof Japanese nationality, raised in Japan, healthy,
and with no history of taking psychedelic medication throughout life.
Candidate participants who were currently seeking treatment for a
mood or mental disorder were excluded from participation.
Participation in the study was voluntary and bore no relation to course
credits or performance evaluation.
Participant informed consent
Written informed consent to participation in the study and
publication of the results was obtained from all participants prior to
conducting experiments.
Measures
Demographic characteristics
e age and sex of the participants were recorded.
Profile of Mood States second edition short
version
e Prole of Mood States second edition (POMS2) questionnaire
assesses the short-term mood state, which is understood to
betransient and frequently uctuating, of individuals 13 and older.
Weused the paper-based Japanese short version of the POMS2, which
takes about 5–10 min to complete (Yokoyama and Watanabe, 2015).
e POMS2 short version contains 35 items from the full-length
version of POMS2 (Heuchert and McNair, 2012). e POMS2 is
composed of six subscales: Anger-Hostility (AH), Confusion-
Bewilderment (CB), Depression-Dejection (DD), Fatigue-Inertia (FI),
Tension-Anxiety (TA), and Vigor-Activity (VA). Only the VA subscale
carries a positive valence. Items in each subscale are scored from 0
(“Not at all”) to 4 (“Extremely”) on a Likert-type scale.
Total mood disturbance
We calculated the scores for Total Mood Disturbance (TMD) from
the subscales of the POMS2 short as follows: TMD = (TA) + (FI) +
(CB) + (AH) + (DD) - (VA) (Heuchert and McNair, 2012, 2015). A
higher TMD score indicates a more negative mood.
Adolescent Time Inventory–Time Attitudes
e Adolescent Time Inventory–Time Attitudes (ATI-TA)
questionnaire assesses the attitude toward one’s past, present and future
(Mello and Worrell, 2007; see Worrell etal., 2013) for the original
ATI-TA. Weused the paper-based Japanese version of the ATI-TA
(Chishima etal., 2019). e ATI-TA consists of six subscales assessing
two valences (positive and negative) for each of three time periods
(past, present, and future). (a) Past Positive (“My past is full of happy
memories”), (b) Past Negative (“My past makes me sad”), (c) Present
Positive (“I amhappy with my current life”), (d) Present Negative (“I
am not satised with my life right now”), (e) Future Positive (“I
amexcited about my future”), and (f) Future Negative (“inking
about my future makes me sad”). ATI-TA items are rated on a 5-point
Likert-type scale from 1 (“totally disagree”) to 5 (“totally agree”).
Procedure
All experiments were conducted at Waseda University campus in
a quiet experimental room. All participants were instructed by the
same experienced researcher. Before the experiment, participants were
told that they will bewriting a letter to their past-self and future-self
with compassion. No details about the purpose or aims of the study
were given. Further, to make sure they took the experiment seriously,
participants were told that their letters will be collected, and the
content checked by a researcher.
Experiment 1
e experiment was an individual laboratory experiment assigned
to 150 participants (54 males, 94 females, and 2 no-response; mean
age = 20.86 years (SD = 2.04)). e participants sat in a chair with a
desk in front of them. First, all participants were asked to answer
POMS2 short. Participants were then randomly divided into two
groups of 75 participants each. is was a simple randomization under
the condition that the numbers of males and females are the same in
the two groups. erefore, both groups were comprised of 28 male
and 47 female participants. Participants in the “past-self → future-
self ” group (mean age = 20.81 years, SD = 2.14) were asked to write a
letter to themselves 5 years ago. Participants in the “future-self → past-
self ” group (mean age = 20.93 years, SD = 1.95) were asked to write a
letter to themselves 5 years in the future. In both conditions, they were
instructed to write in a sympathetic way as the person who
understands them best. No time limit was set, but they usually nished
writing within 5–10 min. Aer nishing the letter, they lled out
POMS2 short again and they were then dismissed.
Seven days later, the participants came back and answered POMS2
short, and were then asked to write a letter to themselves 5 years in the
future if they had previously written to their past-self, and to write a
letter to themselves 5 years ago if they had previously written to their
future-self. e instructions were the same as 7 days earlier. Aer
nishing the letter, they answered POMS2 short again. Finally, they
answered the ATI-TA. Wedeliberately chose a time period of 5 years
for writing the letters to the past-and future-self because both
represented distinct stages in the lives of the participants, i.e., high
school student versus future university graduate, accompanied by
characteristic environmental changes.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 04 frontiersin.org
Experiment 2
e experiment was an individual laboratory experiment assigned
to 62 participants who had not participated in Experiment 1. e
participants sat on a chair with a desk in front of them. First,
participants answered the POMS2. Participants were then randomly
divided into 2 groups of 31 participants each. As in Experiment 1, this
was a simple randomization, therefore both groups consisted of 11
males and 20 females. Participants were asked to write a letter to
themselves 5 years ago. Participants in the “self-alone” group (mean
age = 20.77 years; SD = 2.01) were asked to focus their writing on
something they had experienced by themselves alone. Participants in
the “social connection” group (mean age = 20.13 years; SD = 1.77) were
asked to focus on something they had experienced together with
someone else. In both conditions, they were instructed to write in a
sympathetic way as the person who understands them best. Aer
nishing the letter, they answered POMS2 short again.
Participant flow diagram
Participant ow diagrams for Experiment 1 and Experiment 2 are
shown in Figure1.
Data analysis
We calculated sample size using g*power (Faul etal., 2009) under
conditions of an eect size of 0.3, α of 0.05, and power (1- β) of 0.95in
Experiment 1, and determined that the sample size in each group
should be56, while under conditions of an eect size of 0.3, α of 0.05,
and power (1- β) of 0.80in Experiment 2, and determined that the
sample size in each group should be32.
In Experiment 1, as dependent variables based on POMS2,
weused the average of [TMD (Total Mood Disturbance) score
after writing the letter] minus [TMD score before writing the
letter], and the averages of [each subscale (TA, FI, CB, AH, DD,
VA) score after writing the letter] minus [each subscale score
before writing the letter] for the writing to past-self condition
and for the future-self condition to see the change of their mood
by writing the letter. Differences between the two writing
conditions were analyzed by analysis of variance (ANOVA),
followed by the REGWQ (Ryan-Einot-Gabriel-Welsch and Quiot)
post-hoc test as a multiple comparison test using JASP software.
Wealso investigated differences in the number of people who felt
positive/no change/negative after writing a letter to their past−/
future-selves were analyzed using the χ2 test for whom (past-self,
future-self) x mood change (positive, no change, negative), with
the dependent variable being the number of people. p-values of
less than 0.05 were considered statistically significant. Then
weseparated the participants in each condition (the writing to
past-self condition and the writing to the future-self condition)
into three groups based on their mood change after writing the
letter (positive, no change, negative). The dependent variables in
that case were the scores of the six subscales of the ATI-TA
assessing two valences (positive and negative) for each of the
three time periods (past, present, and future). We also
investigated the net scores [(positive score) – (negative score)] of
ATI-TA for each of the three time periods (past, present, and
future) corresponding to mood changes after writing a letter to
past−/future-self. Differences among those conditions were
analyzed by analysis of variance (ANOVA), followed by the
REGWQ (Ryan-Einot-Gabriel-Welsch and Quiot) post-hoc test
as a multiple comparison test.
Using KH Coder, a quantitative text analysis soware developed
by Higuchi (2001), weinvestigated the dierences in the frequency of
words describing others in the letter to past-self between participants
whose mood changed positively and participants whose mood
changed negatively aer writing the letter. en we conducted
hypothesis testing for the dierence in the population proportions
between the frequency of words describing others in a letter which
aected the participants’ mood positively and those in a letter which
aected the participants’ mood negatively.
In Experiment 2, as dependent variables, based on POMS2,
weused the average of [TMD (Total Mood Disturbance) score aer
writing the letter] minus [TMD score before writing the letter],
and the averages of [each subscale (TA, FI, CB, AH, DD, VA) score
aer writing the letter] minus [each subscale score before writing
the letter] for the “self-alone” group and for the “social connection”
group to see the mood change aer writing the letter. Dierences
between the two writing conditions were analyzed by analysis of
variance (ANOVA), followed by the REGWQ (Ryan-Einot-Gabriel-
Welsch and Quiot) post-hoc test as a multiple comparison test.
Wealso investigated the dierences in the number of participants
who felt positive/no change/negative aer writing a letter to past-
self focusing on something they had experienced by themselves
versus focusing on something they had experienced together with
someone else by using the χ2 test.
FIGURE1
Participant flow diagrams for Experiment 1 and Experiment 2.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 05 frontiersin.org
Results
All data in Experiments 1 and 2 were obtained from healthy
Japanese undergraduates with no history of taking psychedelic
medication throughout life.
Mood change after writing to past-self vs.
after writing to future-self
In Experiment 1, we asked the participants to write a
compassionate letter to their past-self and 7 days later another
letter to their future-self or the other way around. In other words,
weused a cross-over design to investigate how a letter to the past-
self and to future-self would affect the mood in the same
participant. For scoring, wecalculated the change in the scores
of the Total Mood Disturbance (TMD) after writing to future−/
past-self as follows:
TMD change = Average of [TMD score after writing the
letter] minus [TMD score before writing the letter].
A negative value for the TMD change indicates that negative
feelings decreased. On the other hand, a positive value for the
TMD change means that negative feelings increased.
First, weinvestigated if the order of letter writing had any
effect on the TMD change using a two-factor ANOVA design,
with “order of letter-writing” (future-self → past-self vs. past-self
→ future-self) × gender (female vs. male) × “to whom” (future,
past). Results showed no effect of “order of letter-writing”
(F(1, 146) = 1.00, MSE = 18.40, η2 = 0.04, p = 0.26) or of “gender”
(F (1, 146) = 0.91, MSE = 16.77, η2 = 0.03, p = 0.20), while a main
effect for the factor “to whom, past vs. present” was observed
(F(1, 148) = 60.21, MSE = 8.51, η2 = 0.64, p < 0.01). Therefore,
the factors of “order of writing letter” and “gender”
were discarded.
The results in Table1 show that the number of participants
who had a more positive mood after writing a letter to past-self
was larger than that of participants who had a more positive
mood after writing a letter to future-self. In addition, the number
of participants who had a more negative mood after writing a
letter to future-self was larger than that of participants who had
a more positive mood after writing a letter to past-self (x2
(2) = 29.921, p < 0.01).
Next, weinvestigated the magnitude of change in mean TMD
(TMD after writing the letter minus TMD before writing the
letter). As can beseen from Figure2A, writing a letter to past-self
led to a negative TMD change, meaning the mood increased. In
contrast, writing a letter to future-self led to a positive TMD
change, meaning the mood decreased. To investigate how each of
the POMS2 subscales contributed to the overall TMD change,
weconducted a 2 (past vs. future) × 6 (TA, FI, CB, AH, DD, VA)
ANOVA on the data ([after minus before] subscale change). The
interaction was significant. Subscale changes were significantly
lower when writing to past-self than when writing to future-self
for the subscales TA (F(1, 149) = 80.19, p < 0.01), DD (F(1,
149) = 118.92, p < 0.01), and CB (F(1, 149) = 5.40, p < 0.05)
(Figure2B). The subscale change was significantly higher when
writing to past-self than when writing to future-self for the
subscale VA (F(1, 149) = 5.87, p < 0.05) (Figure2B). No significant
difference in subscale change was found for AH (F(1, 149) = 1.48,
ns) or FI (F(1, 149) = 1.62, ns) (Figure2B). These results show
that reduced TA and DD scores and an increased VA score were
the main drivers of the increase in mood after writing a letter to
past-self.
Average scores of ATI-TA correspond to
mood changes after writing letters to
past-self
To further investigate what kind of participants are influenced
by writing to their past-self, weemployed the ATI-AT scale to
measure the participants’ attitude toward their past, present, and
future. Then, weinvestigated how average scores of ATI-TA
correspond to mood changes after writing letters to past-self
using ANOVA, with mood after writing a letter (between: more
positive, no change, more negative) X time (within: past, present,
future) X mood score (within: positive, negative). Results showed
that the main effect of mood after writing a letter was significant
(F(2, 147) = 4.78, MSE = 0.36, p < 0.05, no change < more
negative). Further, the main effect of mood score was also
significant (F(1, 147) = 67.15, MSE = 2.69, p < 0.01, negative scores
< positive scores). In addition, the interaction between mood
after writing a letter X mood score was significant
(F(2, 147) = 5.38, MSE = 2.69, p < 0.01). Regarding a positive
score, there was no difference in mood after writing the letter
(F(2, 294) = 2.48, MSE = 1.53, p = 0.06). Regarding a negative
score, there was a significant mood difference after writing the
letter (F(2, 294) = 8.12, MSE = 1.53, p < 0.01, no change < more
positive, no change < more negative) (Figure3). Together, these
findings indicate that participants who have a less negative
attitude toward time were not affected by writing a letter to their
past-self, neither positively nor negatively.
The net score [(positive score) – (negative
score)] of ATI-TA corresponds to mood
changes after writing a letter to past-self
After revealing that the time attitude of an individual affects
the mood change after writing a compassionate letter to past-self,
we next wanted to investigate how the net score of ATI-AT
corresponds to mood changes after writing letters to past-self.
TABLE1 Number of participants by mood change after writing a letter to past- or future-self.
More positive mood No change More negative mood
Letter to past-self (n) 88 18 44
Letter to future-self (n) 44 17 89

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 06 frontiersin.org
Wefirst calculated the net score of ATI-AT by subtracting the
score for the negative time attitude subscales from the score for
the positive time attitude subscales. Then, wesubmitted the
ATI-AT net score to an ANOVA with mood after writing a letter
(between: more positive, no change, more negative) X time
(within: past, present, future). Results showed that the main
effect of mood after writing a letter was significant
(F(2,147) = 28.99, MSE = 5.39, p < 0.01, no change > more
negative, no change > more positive). However, the main effect
of mood score was not significant (F(2,294) = 1.94, MSE = 2.16,
p = 0.90). Similarly, the interaction between mood after writing a
letter X mood score was not significant (F(4,294) = 3.81,
MSE = 2.16, p = 1.76) (Figure 4). These results indicate that
although participants who became more negative after writing a
letter to past-self scored almost zero on their past attitude in the
ATI-AT score, this did not reach statistical significance.
Average scores of ATI-TA correspond to
mood changes after writing letters to
future-self
Next, weinvestigated how the mean scores of ATI-TA correspond
to mood changes aer writing a letter to future-self. ANOVA using the
factors mood aer writing a letter (between: more positive, no change,
more negative) X time (within: past, present, future) X mood score
(within: positive, negative) showed that the main eect of mood aer
writing a letter was not signicant (F(2, 147) = 0.14, MSE = 0.38, ns).
In addition, the main eect of mood score was signicant (F(1,
147) = 250.39, MSE = 3.02, p < 0.01, negative scores < positive scores).
Further, the interaction between mood aer writing a letter X mood
score was signicant (F(2, 147) = 0.13, MSE = 3.02, p = 0.04) (Figure5).
ese ndings show that a participant’s attitude toward time aects
the mood aer writing a letter to future-self.
FIGURE2
Mood change after writing a letter to past-self or future-self. Participants were divided into two groups of 75 participants each. One group wrote a
compassionate letter first to their past-self and 1 week later to their future-self. The other group wrote a compassionate letter first to their future-self
and 1 week later to their past-self. POMS2 was answered before and after the letter writing. Total Mood Disturbance (TMD) was calculated from the
POMS2 subscale scores. (A) Changes in TMD after writing the letter. (B) Contribution of POMS2 subscales to the change in TMD shown in panel (A).
Data shown are mean ± SEM.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 07 frontiersin.org
The net score [(positive score) – (negative
score)] of ATI-TA does not correspond to
mood changes after writing a letter to
future-self
After revealing that the time attitude of an individual affects
the mood change after writing a compassionate letter to future-
self, wenext wanted to investigate how the net score of ATI-AT
corresponds to mood changes after writing a letter to future-self.
Wefirst calculated the net score of ATI-AT by subtracting the
score for the negative time attitude subscales from the score for
the positive time attitude subscales. Then, wesubmitted the
ATI-AT net score to an ANOVA with mood after writing a letter
(between: more positive, no change, more negative) X time
(within: past, present, future). Result showed that the main effect
of mood after writing a letter to future-self was not significant
(F(2,147) = 0.04, MSE = 6.04, ns). The main effect of mood score
was also not significant (F(2,294) = 0.29, MSE = 2.11, ns). Finally,
the interaction between mood after writing a letter to future-self
X mood score was not significant (F(4,294) = 1.41, MSE = 2.11,
ns) (Figure6).
Dierences in the frequency of words
describing others in the letter to past-self
when the mood changed positively versus
when the mood changed negatively after
writing the letter
Using KH Coder, a quantitative text analysis soware
developed by Higuchi (2001), weextracted all the written words
from the text data of letters to past-selves, and examined the
dierences in the number of words describing others in the letters
by participants whose mood changed positively versus participants
FIGURE3
Mood change after writing a letter to past-self depending on time attitude. One hundred and fifty participants wrote a compassionate letter to their
past-self. POMS2 was answered before and after the letter writing and the change in total mood disturbance was calculated from the POMS2 subscale
scores. In the end, all participants answered the Adolescent Time Inventory–Time Attitudes (ATI-TA) questionnaire. Data shown are mean ± SEM.
FIGURE4
Mood change after writing a letter to past-self in relation to the Adolescent Time Inventory–Time Attitudes (ATI-AT) net score. POMS2 was answered
before and after the letter writing and the total mood disturbance was calculated from the POMS2 subscale scores. Participants answered the ATI-TA
questionnaire after finishing writing a compassionate letter to past-self. Net scores (positive minus negative) of the ATI-TA are indicated on the Y-axis.
Data shown are mean ± SEM.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 08 frontiersin.org
whose mood changed negatively aer writing the letter. Words
regarded as used to describe others were: friends, teachers,
classmates, peers, family, parents, father, mother, people around
them, seniors, juniors, etc. In the letters which created a positive
mood, there were 63 words (0.72 words per person) describing
others, and in the letters which created a negative mood, there were
21 words (0.48 words per person) describing others. Hypothesis
testing for the dierence in the population proportions revealed
that the proportion of words describing others in the letter of
participants whose mood changed positively aer writing a letter
was signicantly higher than the proportion of words describing
others in the letter of participants whose mood changed negatively
aer writing the letter (α = 0.05, g = 0.0996, p = 0.0125).
The eects of focusing the writing on a
social experience in the past versus
focusing on an event experienced alone
In Experiment 2, wewanted to see how writing about a social
connection versus writing about a self-experience affects mood.
Therefore, weasked the participants in the “self-alone” group to
focus their writing on something they had experienced by
themselves alone and participants in the “social connection”
group to focus on something they had experienced together with
someone else. For scoring, wecalculated the change in TMD
scores after writing to past-self in the same way as in Experiment
1. ANOVA with “TMD after writing to past-self” X “focus of
FIGURE5
Mood change after writing a letter to future-self depending on time attitude. One hundred and fifty participants wrote a compassionate letter to their
future-self. POMS2 was answered before and after the letter writing and the change in total mood disturbance was calculated from the POMS2
subscale scores. At the end, all participants answered the Adolescent Time Inventory–Time Attitudes (ATI-TA) questionnaire. Data shown are
mean ± SEM.
FIGURE6
Mood change after writing a letter to future-self in relation to the Adolescent Time Inventory–Time Attitudes (ATI-AT) net score. POMS2 was answered
before and after the letter writing and the total mood disturbance was calculated from the POMS2 subscale scores. Participants answered the ATI-TA
questionnaire after finishing writing a compassionate letter to future-self. Net scores (positive minus negative) of the ATI-TA are indicated on the Y-axis.
Data shown are mean ± SEM.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 09 frontiersin.org
writing” (self vs. social connection) showed that there was a
significant main effect for “focus of writing” (F(1, 60) = 4.81,
MSE = 10.51, f = 0.28, p < 0.05; x2(2) = 13.549, p < 0.01)
(Table2; Figure7A, respectively). These results show that when
participants focused their writing on what they did with
someone else, they became less negative than when they focused
on what they experienced by themselves alone. To investigate
how the POMS2 subscales contributed to this mood change,
weconducted a 2 (between self-alone vs. social connection) × 6
(within TA, FI, CB, AH, DD, VA) ANOVA on the data ([after
minus before] for each subscale of POMS2). The interaction was
not significant (F(5, 300) = 1.36, MSE =0.78, η2 = 0.15, ns).
The main effect of the factor of the way of letter was not
significant (F(1, 60) = 2.42, MSE = 1.69, η2 = 0.20, ns). The
main effect of the factor of subscales was significant
(F(1, 60) = 4.13, MSE = 0.78, η2 = 0.26, p < 0.01), with VA > FI,
VA > DD, and VA > TA (Figure7B). The DD subscale score in
the social connection group decreased significantly more
than the DD subscale score in the self-alone group (t test,
p < 0.05).
TABLE2 Number of participants by mood change after writing a letter to
past-self.
Focus of
writing
More
positive
mood
No mood
change
More
negative
mood
Self-alone 15 5 11
Social connection 28 2 1
FIGURE7
Mood change after writing a compassionate letter to past-self with a focus on a self-alone experience or on a social connection experience.
Participants were divided into two groups of 31 participants each. One group wrote with a focus on a self-alone experience. The other group wrote
with a focus on a social connection experience. POMS2 was answered before and after the letter writing. Total Mood Disturbance (TMD) was
calculated from the POMS2 subscale scores. (A) Changes in TMD after writing the letter. (B) Contribution of POMS2 subscales to the change in TMD
shown in panel (A). Data shown are mean ± SEM.

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 10 frontiersin.org
Discussion
In this study weinvestigated how writing a compassionate letter
to oneself can make us feel better. e ndings of Experiment 1
showed that writing a letter to past-self is eective in decreasing
negative mood and writing a letter to future-self is eective in
increasing negative mood. is is in line with a previous report
showing that writing to the future-self can reduce delinquency (van
Gelder etal., 2013; Rutchick etal., 2018). Connecting the current-self
to the future-self might induce anxiety about the future, which in turn
prevents the current self from doing something bad, or alternatively
might promote positive future-oriented behaviors, such as self-
control, money saving, and academic performance. In contrast, a
study conducted in the early stage of the COVID-19 pandemic found
that writing to future-self led to a decrease in negative aect and an
increase in positive aect (Chishima et al., 2021). is could
beexplained as follows. When people are in a stressful situation, and
when they know the bad situation (i.e., Covid-19 pandemic) will
denitely disappear at some point in the future, writing to future-self
is eective in improving the mental state. In contrast, when people are
uncertain if the stressful situation will go away, writing to future-self
might not comfort one’s mental state.
Regarding the time attitude, Experiment 1 also showed that
people who have a less negative attitude toward their past, present, and
future are less likely to beemotionally aected by writing letters to
their past-self. at is, people who are content with their past, present,
and future would not bementally aected by writing a letter to their
past-self. In addition, people whose mood changed negatively aer
writing a letter to past-self were not positive toward their past. e
results of Experiment 1 are consistent with previous studies which
suggested that negative feelings in a current situation can trigger
positive memory, especially memory related to a social connection,
which might enhance well-being (Joormann and Siemer, 2004;
Joormann etal., 2007). is time, wefound that some participants did
not have rosy memories, although they were fewer than those who
did, and for these participants writing a letter to past-self may not lead
to mental health and well-being.
Qualitative text analysis in Experiment 1 showed that the number
of words regarded as describing others in the letter to past-self was
greater for participants whose mood changed positively compared to
participants whose mood changed negatively aer writing the letter.
e results of Experiment 2 also showed that when participants
focused their self-compassionate letter on what they did with someone
else, they became less negative than when they focused on what they
experienced by themselves alone. e dierence in TMD change was
substantial, showing that remembering the feeling of being connected
with other people is highly eective in overcoming a negative mood
compared to remembering something one did alone. Interestingly, the
TMD change for “past-self ” in Experiment 1 (Figure2) amounts to
about the average of the TMD change from “self-alone” and “social
connection” in Experiment 2 (Figure7A), most likely reecting this
mixture in the focus of the writing. Wethen looked further into
changes of POMS2 subscales in Experiment 2 and found that the DD
subscale score in the social connection group decreased signicantly
more than the DD subscale score in the self-alone group, which
contributed to the mood increase seen in the social connection group.
erefore, a lack of social connection and/or a feeling of loneliness
when writing the letter likely resulted in a more negative mood in the
self-alone group. is is in agreement with a previous report showing
that recalling positive memories with a social component could
beparticularly powerful in reducing the cortisol response following
the same stressor task (Speer and Delgado, 2020) and a meta review
about long-term care residents which concluded that good social
connection is linked to better mental health outcomes (Bethell etal.,
2021). On the other hand, some previous studies showed that
achievement themes were prominently represented in memories of
positive self-regard and interpersonal themes were prominently
represented in memories of negative self-regard (Pillemer et al., 2007,
2013). Considering those ndings, it is reasonable that remembering
interpersonal problems might bemore depressing than anything else,
whereas thinking about your achievements with lots of support from
your surroundings might bemore comforting than anything else.
The current results need to beinterpreted within the limits
of this study and considering the narrow demographic
characteristics of the participants. However, previous studies
showed that writing about their memories by themselves has
benefits for older people’s well-being, and that writing about
memories might have a therapeutic effect (Botella and Feixas,
1993; Elford etal., 2005). Robinson and Murphy-Nugen (2018)
also showed the benefits of life review writing for older adults
who are at risk of isolation and depression. Johnson and O’Brien
(2013) and Urken and LeCroy (2021) also found that self-
compassionate writing interventions led to improvements in self-
compassion and mental health. Therefore, writing letters to one’s
past-self appears to bea good method for people of various ages
and with a wide range of mental health conditions. Because
Pauley and McPherson (2010) noted that individuals with
depression or anxiety might find it challenging to practice self-
compassion, suggesting the need for tailored interventions,
compassionate letter writing to self might require additional
support or professional guidance for people with these conditions.
Future studies should therefore investigate a variety of age groups
and clinically relevant study populations. In addition, there might
bepotentially confounding variables or factors that could have
influenced the observed mood changes seen in the present study.
Factors like the participants’ current life circumstances, mental
health conditions, or other external events could have potentially
confounded the results. A more thorough exploration of these
variables will strengthen the interpretation of future findings.
Another limitation is that the participant number in Experiment
2 was slightly smaller than expected. In the future, it should beof
interest to investigate if writing to past-self could beused as a
regular treatment to improve mood and relief depression.
Similarly, it would beinteresting to study if writing a letter to
future-self on a regular basis can be effective in attaining
measurable improvements in future behavior, performance and
success in life.
Conclusion
is study revealed that writing a letter to your past-self with
compassion can be eective in transiently elevating one’s mood,
especially when the focus is on an experience with friends, family, or
acquaintances. When there are no negative feelings in a present
situation, writing a letter to past-self will not aect mood. Further,

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 11 frontiersin.org
writing a letter to your future-self can beeective in decreasing mood,
which might beuseful to induce motivation for self-improvement in
the future. Together, these ndings show how writing to past-self and
future-self can eectively alter one’s mood, which has implications for
the future design of self-compassionate intervention tools and their
application in everyday life and in therapeutic settings.
Data availability statement
e raw data supporting the conclusions of this article will
bemade available by the authors, without undue reservation.
Ethics statement
e studies involving humans were approved by Waseda Ethics
Review Procedures concerning Research with Human Subjects
(approval No. 2021-163). e studies were conducted in accordance
with the local legislation and institutional requirements. e
participants provided their written informed consent to participate in
this study.
Author contributions
ES: Conceptualization, Data curation, Formal analysis,
Investigation, Writing – original dra, Writing – review & editing.
MY: Formal analysis, Investigation, Writing-review & editing. TK:
Funding acquisition, Supervision, Writing – review & editing.
Funding
e author(s) declare nancial support was received for the
research, authorship, and/or publication of this article. is study was
supported by Grants-in-Aid for Scientic Research from the Japan
Society for the Promotion of Science (No. 16H02837).
Acknowledgments
We would like to thank Kenji Tomita for excellent experimental
support. Writing and editing assistance was provided by Dr. Katrin
Ishii-Schrade of Dmed (http://www.dmed.co.jp) under contract with
the author.
Conflict of interest
e authors declare that the research was conducted in the
absence of any commercial or nancial relationships that could
beconstrued as a potential conict of interest.
Publisher’s note
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
References
Bahrick, H. P., Hall, L. K., and Berger, S. A. (1996). Accuracy and distortion in
memory for high school grades. Psychol. Sci. 7, 265–271. doi: 10.1111/j.1467-9280.1996.
tb00372.x
Bahrick, H. P., Hall, L. K., and Da Costa, L. A. (2008). Fiy years of memory of college
grades: accuracy and distortions. Emotion 8, 13–22. doi: 10.1037/1528-3542.8.1.13
Bethell, J., Aelick, K., Babineau, J., Bretzla, M., Edwards, C., Gibson, J. L., et al. (2021).
Social connection in long-term care homes: A scoping review of published research on
the mental health impacts and potential strategies during COVID-19. J. Am. Med. Dir.
Assoc. 22, 228–237.e25. doi: 10.1016/j.jamda.2020.11.025
Botella, L., and Feixas, G. (1993). e autobiographical group: A tool for the
reconstruction of past life experience with the aged. e International Journal of Aging
and Human Development 36, 303–319.
Chen, J. C., and Danish, S. J. (2010). Acculturation, distress disclosure, and emotional
self-disclosure within Asian populations. Asian Am. J. Psychol. 1, 200–211. doi: 10.1037/
a0020943
Chishima, Y., Huai-Ching Liu, I. T., and E Wilson, A. (2021). Temporal distancing
during the COVID-19 pandemic: letter writing with future self can mitigate negative
aect. Appl. Psychol. Health Well Being 13, 406–418. doi: 10.1111/aphw.12256
Chishima, Y., Murakami, T., Worrell, F. C., and Mello, Z. R. (2019). e Japanese
version of the adolescent time inventory-time attitudes (ATI-TA) scale: internal
consistency, structural validity, and convergent validity. Assessment 26, 181–192. doi:
10.1177/1073191116683800
Christensen, T. C., Wood, J. V., and Barrett, L. F. (2003). Remembering everyday
experience through the prism of self-esteem. Personal. Soc. Psychol. Bull. 29, 51–62. doi:
10.1177/0146167202238371
Croyle, R. T., Lous, E. F., Barger, S. D., Sun, Y. C., Hart, M., and Gettig, J. (2006). How
well do people recall risk factor test results? Accuracy and bias among cholesterol
screening participants. Health Psychol. 25, 425–432. doi: 10.1037/0278-6133.25.3.425
Elford, H., Wilson, F., McKee, K. J., Chung, M. C., Bolton, G., and Goudie, F. (2005).
Psychosocial benets of solitary reminiscence writing: an exploratory study. Aging Ment.
Health 9, 305–314. doi: 10.1080/13607860500131492
Farber, B. A. (2003). Patient self-disclosure: a review of the research. J. Clin. Psychol.
59, 589–600. doi: 10.1002/jclp.10161
Faul, F., Erdfelder, E., Buchner, A., and Lang, A. G. (2009). Statistical power analyses
using G*power 3.1: tests for correlation and regression analyses. Behav. Res. Methods 41,
1149–1160. doi: 10.3758/BRM.41.4.1149
Heuchert, J. P., and McNair, D. M. (2012). POMS 2 ®: Prole of mood states (2nd)
Tonawanda: Multi-Health Systems Inc.
Heuchert, J. P., and McNair, D. M. (2015) in POMS 2 Japanese version manual. eds. K.
Yokoyama and K. Watanabe (Tokyo: Kanekoshobo Ltd)
Higuchi, K. (2001). KH Coder index page. Availble at: http://khcoder.net/en/ Accessed
May 24, 2019.
Johnson, E. A., and O’Brien, K. A. (2013). Self-compassion soothes the savage ego-
threat system: eects on negative aect, shame, rumination, and depressive symptoms.
J. Soc. Clin. Psychol. 32, 939–963. doi: 10.1521/jscp.2013.32.9.939
Joormann, J., and Siemer, M. (2004). Memory accessibility, mood regulation, and
dysphoria: diculties in repairing sad mood with happy memories? J. Abnorm. Psychol.
113, 179–188. doi: 10.1037/0021-843X.113.2.179
Joormann, J., Siemer, M., and Gotlib, I. H. (2007). Mood regulation in depression:
dierential eects of distraction and recall of happy memories on sad mood. J. Abnorm.
Psychol. 116, 484–490. doi: 10.1037/0021-843X.116.3.484
Kelly, A. C., and Waring, S. V. (2018). A feasibility study of a 2-week self-
compassionate letter-writing intervention for nontreatment seeking individuals
with typical and atypical anorexia nervosa. Int. J. Eat. Disord. 51, 1005–1009. doi:
10.1002/eat.22930

Sugimori et al. 10.3389/fpsyg.2024.1327595
Frontiers in Psychology 12 frontiersin.org
Kelly, A. C., Zuro, D. C., Foa, C. L., and Gilbert, P. (2010). Who benets from training
in self-compassionate self-regulation? A study of smoking reduction. J. Soc. Clin.
Psychol. 29, 727–755. doi: 10.1521/jscp.2010.29.7.727
Kotera, Y., and Van Gordon, W. (2021). Eects of self-compassion training on work-
related well-being: A systematic review. Front. Psychol. 12:630798. doi: 10.3389/
fpsyg.2021.630798
Landau, J. D., and Gunter, B. C. (2009). "Don't worry; youreally will get over it":
methodological investigations of the fading aect bias. Am. J. Psychol. 122, 209–217. doi:
10.2307/27784392
Levine, L. J., and Bluck, S. (2004). Painting with broad strokes: happiness and the
malleability of event memory. Cognit. Emot. 18, 559–574. doi:
10.1080/02699930341000446
Mello, Z. R., and Worrell, F. C. (2007). e adolescent time inventory-English. Berkeley:
University of California.
Mitchell, T. R., ompson, L., Peterson, E., and Cronk, R. (1997). Temporal
adjustments in the evaluation of events: the "rosy view". J. Exp. Soc. Psychol. 33, 421–448.
doi: 10.1006/jesp.1997.1333
Murphy, J., Shevlin, M., Adamson, G., Cruddas, S., and Houston, J. (2012). Memories
of childhood threat, fear of disclosure and paranoid ideation: a mediation analysis using
a nonclinical sample. J. Aggress. Maltreat. Trauma 21, 459–476. doi:
10.1080/10926771.2012.667521
Pasupathi, M., and Oldroyd, K. (2015). Telling and remembering: complexities in
long-term eects of listeners on autobiographical memory. Appl. Cogn. Psychol. 29,
835–842. doi: 10.1002/acp.3193
Pauley, G., and McPherson, S. (2010). e experience and meaning of compassion and
self-compassion for individuals with depression or anxiety. Psychol. Psychother. eory
Res. Pract. 83, 129–143. doi: 10.1348/147608309X471000
Pavlova, M., Lund, T., Nania, C., Kennedy, M., Graham, S., and Noel, M. (2022).
Reframe the pain: A randomized controlled trial of a parent-led memory-reframing
intervention. J. Pain 23, 263–275. doi: 10.1016/j.jpain.2021.08.002
Pillemer, D. B., Ivcevic, Z., Gooze, R. A., and Collins, K. A. (2007). Self-esteem
memories: feeling good about achievement success, feeling bad about relationship
distress. Personal. Soc. Psychol. Bull. 33, 1292–1305. doi: 10.1177/0146167207303952
Pillemer, D. B., omsen, D., Kuwabara, K. J., and Ivcevic, Z. (2013). Feeling good and
bad about the past and future self. Memory 21, 210–218. doi:
10.1080/09658211.2012.720263
Ritchie, T. D., Batteson, T. J., Bohn, A., Crawford, M. T., Ferguson, G. V., Schrauf, R. W.,
et al. (2015). A pancultural perspective on the fading aect bias in autobiographical
memory. Memory 23, 278–290. doi: 10.1080/09658211.2014.884138
Ritchie, T., Skowronski, J. J., Hartnett, J., Wells, B., and Walker, W. R. (2009). The
fading affect bias in the context of emotion activation level, mood, and personal
theories of emotion change. Memory 17, 428–444. doi: 10.1080/09658210902791665
Robinson, J. T., and Murphy-Nugen, A. B. (2018). It makes youkeep trying: life review
writing for older adults. J. Gerontol. Soc. Work. 61, 171–192. doi:
10.1080/01634372.2018.1427645
Rutchick, A. M., Slepian, M. L., Reyes, M. O., Pleskus, L. N., and Hersheld, H. E.
(2018). Future self-continuity is associated with improved health and increases exercise
behavior. J. Exp. Psychol. Appl. 24, 72–80. doi: 10.1037/xap0000153
Schippers, M., de Jong, E. M., Rus, D., and Rommers, H. (2023). Shaping ideal futures:
Writing a letter to the future. Available at: SSRN 4372540.
Skowronski, J. J., Walker, W. R., Henderson, D. X., and Bond, G. D. (2014). e fading
aect bias: its history, its implications, and its future. Adv. Exp. Soc. Psychol. 49, 163–218.
doi: 10.1016/B978-0-12-800052-6.00003-2
Speer, M. E., Bhanji, J. P., and Delgado, M. R. (2014). Savoring the past: positive
memories evoke value representations in the striatum. Neuron 84, 847–856. doi:
10.1016/j.neuron.2014.09.028
Speer, M. E., and Delgado, M. R. (2017). Reminiscing about positive memories buers
acute stress responses. Nat. Hum. Behav. 1:0093. doi: 10.1038/s41562-017-0093
Speer, M. E., and Delgado, M. R. (2020). e social value of positive autobiographical
memory retrieval. J. Exp. Psychol. Gen. 149, 790–799. doi: 10.1037/xge0000671
Speer, M. E., Ibrahim, S., Schiller, D., and Delgado, M. R. (2021). Finding positive
meaning in memories of negative events adaptively updates memory. Nat. Commun.
12:6601. doi: 10.1038/s41467-021-26906-4
Stern, N. G., and Engeln, R. (2018). Self-compassionate writing exercises increase college
women’s body satisfaction. Psychol. Women Q. 42, 326–341. doi: 10.1177/0361684318773356
Sugimori, E., Shimokawa, K., Aoyama, Y., Kita, T., and Kusumi, T. (2020). Empathetic
listening boosts nostalgia levels and positive emotions in autobiographical narrators.
Heliyon 6:e04536. doi: 10.1016/j.heliyon.2020.e04536
Swee, M. B., Klein, K., Murray, S., and Heimberg, R. G. (2023). A brief self-
compassionate letter-writing intervention for individuals with high shame. Mindfulness
14, 854–867. doi: 10.1007/s12671-023-02097-5
ompson, C. P., Skowronski, J. J., Larsen, S. F., and Betz, A. L. (2013). Autobiographical
memory: Remembering what and remembering when. New York: Psychology Press.
Urken, D., and LeCroy, C. W. (2021). A randomized controlled trial of a self-
compassion writing intervention for adults with mental illness. Res. Soc. Work. Pract.
31, 254–266. doi: 10.1177/1049731520972779
van Gelder, J. L., Hershfield, H. E., and Nordgren, L. F. (2013). Vividness of the
future self predicts delinquency. Psychol. Sci. 24, 974–980. doi:
10.1177/0956797612465197
Walker, W. R., Skowronski, J. J., and Thompson, C. P. (2003). Life is pleasant--and
memory helps to keep it that way! Rev. Gen. Psychol. 7, 203–210. doi:
10.1037/1089-2680.7.2.203
Walker, W. R., Vogl, R. J., and ompson, C. P. (1997). Autobiographical memory:
unpleasantness fades faster than pleasantness over time. Appl. Cogn. Psychol. 11,
399–413. doi: 10.1002/(SICI)1099-0720(199710)11:5<399::AID-ACP462>3.0.CO;2-E
Williams, S. E., Ford, J. H., and Kensinger, E. A. (2022). e power of negative and
positive episodic memories. Cogn. Aect. Behav. Neurosci. 22, 869–903. doi: 10.3758/
s13415-022-01013-z
Wirtz, D., Kruger, J., Scollon, C. N., and Diener, E. (2003). What to do on spring
break? e role of predicted, on-line, and remembered experience in future choice.
Psychol. Sci. 14, 520–524. doi: 10.1111/1467-9280.03455
Worrell, F. C., Mello, Z. R., and Buhl, M. (2013). Introducing English and German
versions of the adolescent time attitude scale. Assessment 20, 496–510. doi:
10.1177/1073191110396202
Yokoyama, K., and Watanabe, K. (2015). Japanese translation of POMS 2: Prole of
mood states 2nd. Tokyo: KanekoShobo.