Personality Disorders: Theory, Research, and
Treatment

Changes in Negative Affect Following Pain (vs.
Nonpainful) Stimulation in Individuals With and Without
a History of Nonsuicidal Self-Injury

Konrad Bresin and Kathryn H. Gordon

Online First Publication, October 10, 2011. doi: 10.1037/a0025736

CITATION

Bresin, K., & Gordon, K. H. (2011, October 10). Changes in Negative Affect Following Pain

(vs. Nonpainful) Stimulation in Individuals With and Without a History of Nonsuicidal

Self-Injury. Personality Disorders: Theory, Research, and Treatment. Advance online
publication. doi: 10.1037/a0025736

Changes in Negative Affect Following Pain (vs. Nonpainful)

Stimulation in Individuals With and Without a History of

Nonsuicidal Self-Injury

Konrad Bresin and Kathryn H. Gordon

North Dakota State University

Theoretical models of nonsuicidal self-injury (NSSI; i.e., purposeful destruction of
body tissue without suicidal intent) suggest that individuals engage in NSSI in order to
regulate negative affect (NA). One limitation of the current research is that most studies
using pain as a proxy of NSSI have failed to include a nonpainful stimuli control group.
This study attempted to address this limitation by comparing the emotional response to
painful (vs. nonpainful) heat stimulation in individuals with a history of NSSI (vs. no
history of NSSI) following a negative mood induction. It was predicted that among
individuals with a history of NSSI, pain would lead to larger reductions in NA than
warm stimulation. Furthermore, it was predicted that this reduction would be larger
than that of controls regardless of condition. The hypotheses received partial support,
such that pain led to greater reductions in negative affect for individuals with a history
of NSSI, but not compared to controls. Future research directions are discussed.

Keywords:

nonsuicidal self-injury, NSSI, self-harm, pain, negative affect

Nonsuicidal self-injury (NSSI) is defined as

self-inflicted damage to one’s own body tissue
that is performed in the absence of suicidal
intent (Nock, 2009). NSSI is a behavior that is
difficult to understand because it lies in oppo-
sition to the common principle of approach/
maximize pleasure and avoid/minimize pain
(Freud, 1929; Gray, 1982; Kahneman & Tver-
sky, 1979). In spite of this, individuals report
that they often engage in NSSI to reduce nega-
tive affect (NA; Nock & Prinstein, 2004). Em-
pirical work supports this prediction in that
NSSI proxies lead to reduced NA (e.g., Franklin
et al., 2010). Still, there are gaps in the current
literature. For example, currently it is unclear if
painful stimulation is necessary to regulate
emotion or whether any tactile stimulation is
adequate. The primary aim of this study was to

compare individuals with a recent history of
NSSI (i.e., within the last year) to individuals
with no NSSI history on their emotional re-
sponse to a painful (vs. nonpainful) heat stim-
ulation following a mood induction.

Nonsuicidal Self-injury as Affect Regulation

Nock and Prinstein (2004) proposed that one

possible function of NSSI is to reduce NA. Sub-
sequent theory has expanded upon this prediction
(e.g., Chapman, Gratz, Brown, 2006; Nock, 2009;
Selby & Joiner, 2009). For example, the experi-
ential avoidance model (EAM) proposes that in-
dividuals who have intense reactions to emotional
stimuli and difficulty regulating emotions may
desire to escape from aversive experiences (e.g.,
NA) via NSSI (Chapman et al., 2006). One pos-
sible mechanism proposed by the EAM is distrac-
tion. However, based on this model, it is unclear
why individuals would choose NSSI over other
methods of distraction (e.g., watching TV). Build-
ing on the EAM, Selby and Joiner’s emotional
cascade theory (2009) suggests that this is because
among emotionally dysregulated individuals, less
intense methods of distraction (e.g., a cold
shower) are unsuccessful in regulating intense
emotions, and only strong sensations (e.g., physi-
cal pain) are able to distract from high NA states.

Konrad Bresin and Kathryn H. Gordon, Department of

Psychology, North Dakota State University.

Konrad Bresin is now at the University of Illinois Urba-

na-Champaign. This research was supported, in part, by a
grant from ND EPSCoR to Kathryn H. Gordon (NSF Grant
EPS-081442).

Correspondence concerning this article should be addressed

to Konrad Bresin, Psychology Department, University of Illi-
nois Urbana-Champaign, 603 East Daniel St., Champaign, IL
61820. E-mail: [e-mail]

Personality Disorders: Theory, Research, and Treatment

© 2011 American Psychological Association

2011, Vol.

●●, No. ●, 000–000

1949-2715/11/$12.00

DOI: 10.1037/a0025736

1

Pain as Affect Regulation

A growing body of research has found sup-

port for affect regulation models in that proxies
of NSSI can decrease the experience of NA.
One series of studies found a reduction in NA
and heart rate following an imagined NSSI in-
cident in individuals with a history of NSSI
(e.g., Haines, Williams, Brain, & Wilson,
1995). The major limitation of these studies is
that it is unclear if these results would replicate
with the inclusion of the physical sensations
experienced during actual NSSI. Other studies
using undergraduate samples without a history
of NSSI suggest that the experience of pain may
decrease NA. For example, Bresin, Gordon,
Bender, Gordon, and Joiner (2010) measured
NA before and after the experience of pain in
two samples of college undergraduates. Results
showed that NA decreased following the expe-
rience of pain. One major limitation of this
study is that there was no manipulation of emo-
tion prior to the exposure to physical pain. It is
possible that at high levels of NA, pain may
exacerbate NA. Also, this study included indi-
viduals without a history of NSSI, which poten-
tially limits generalizability of the results to
self-injuring populations.

These limitations were addressed in a study

by Franklin et al. (2010). In this study, affective
modulation of the startle reflex was measured
during anticipation of a stressful speech and
after a cold pressor task. The results showed
that, regardless of NSSI history, participants
displayed a reduction in the startle reflex after
exposure to the cold pressor task, indicating that
the experience of pain may reduce NA. One
limitation of this study is the lack of a control
condition that consisted of nonpainful tactile
stimulation. Based on this study, it is unclear
whether an intense experience, such as pain, is
necessary for decreasing NA (as predicted by
Selby & Joiner, 2009) or if any tactile stimula-
tion would also decrease NA.

Niedtfeld et al. (2010) attempted to address

this limitation by including painful and non-
painful stimulation. In this study, a group of
individuals with borderline personality disorder
(BPD; 80% with a history of NSSI) and healthy
controls were exposed to emotion-eliciting
slides before being exposed to either painful
stimuli or warm nonpainful stimuli. Amygdala
activity was recorded using functional MRI.

Both groups showed deactivation in the
amygdala during pain and nonpainful heat
stimuli following negative slides, though this
effect was larger for individuals with BPD
than controls. These results may suggest that
any sensory stimulation, not just painful stim-
ulation, may be effective in reducing NA. In
the current context, the major limitation of
this study is that it was conducted on individ-
uals with BPD, not self-injurers per se.
Though NSSI is one possible symptom of
BPD, it is possible to meet criteria for BPD
without engaging in NSSI (American Psychi-
atric Association, 2000). It is feasible that the
relationship between pain and emotion may
differ for those who engage in NSSI but do
not necessarily meet criteria for BPD.

Current Study

The current study attempts to address limita-

tions of previous research. Individuals with a
history of NSSI in the last year and without a
history of NSSI were recruited. Participants
completed a mood induction where they wrote
about a time when they were not living up to an
important personal attribute (e.g., being intelli-
gent). Following the mood induction, partici-
pants were either exposed to painful heat stim-
ulation or nonpainful heat stimulation (based on
participant ratings; detailed below). Finally,
participants gave self-report levels of affect at
three times during the study (Time 1: before any
mood induction; Time 2: following the mood
induction; Time 3: following the painful/
nonpainful induction). Based on previous the-
ory and research (Niedtfeld et al., 2010; Selby
& Joiner, 2009), we predicted that for individ-
uals with a history of NSSI, pain would lead to
a larger reduction in NA than warm stimulation.
Furthermore, we predicted that the NSSI group
that experienced painful stimulation would have
a reduction in NA larger than that of the control
group regardless of condition.

Method

Participants

Undergraduates (1,612) were screened using

the Deliberate Self-Harm Inventory (DSHI;
Gratz, 2001) for participation in this study. Of
these, 136 participants (8%) indicated that they

2

BRESIN AND GORDON

had engaged in NSSI in the past 12 months, 251
participants (15%) indicated engaging in NSSI
more than 12 months ago, and the remaining
participants (76%) reported no history of NSSI.

For the NSSI group, 59 participants (34 fe-

male) were recruited. The average time since
the last NSSI episode was 5.52 months
(SD

⫽ 4.48). The most common methods in-

cluded: cutting (50%), scratching (41%) and
burning (29%). Participants reported an average
of 3.28 methods (SD

⫽ 1.84). The median num-

ber of NSSI incidents was 15 (range

⫽ 1–1000).

For the non-NSSI group, 56 (31 Females) par-
ticipants without a history of NSSI were re-
cruited. Across groups, the sample was 96%
Caucasian and the mean age was 19.48 years
(SD

⫽ 2.53).

Procedure

Upon arrival to the laboratory, participants

provided consent. Then, participants rated the
intensity of a series of warm temperature stim-
uli. Next, participants completed a preinduction
measure of affect (i.e., Time 1 NA). Then, par-
ticipants completed the mood induction and a
Time 2 NA assessment. Immediately following
the mood induction, participants were randomly
assigned to be exposed to either a nonpainful
stimulation or a painful stimulation (based on
their prior ratings) and subsequently reported
their affect (i.e., Time 3 NA). Finally, partici-
pants were thanked for their time and excused.
Participants were compensated with course
credit or 10 dollars. All procedures were ap-
proved by the university’s Institutional Review
Board.

Materials

Screening measure.

All potential partici-

pants completed the DSHI (Gratz, 2001) on a
secure website prior to participating in the
study. The DSHI is a 17-item questionnaire that
assesses the lifetime history of a variety of NSSI
methods. First, participants indicated if they had
ever intentionally injured areas of their body
without suicidal intent using a specific method
(e.g., cutting). If they responded “yes”, they
answered follow-up questions (e.g., when was
the last time that it happened?). Participants
who indicated they had engaged in any form of
NSSI within the last 12 months were sent an

email about study participation. The time frame
was chosen to be consistent with previous stud-
ies (e.g., Gratz et al., 2011). Participants who
reported no history of NSSI were contacted
about participation in the study for the control
group.

Pain stimuli.

All heat stimuli were pro-

duced by a Medoc Thermal Sensory Analyzer, a
computerized device for the quantitative assess-
ment of heat- and cold-induced pain. All stimuli
were presented to the left volar forearm. Partic-
ipants were exposed to pain stimuli twice during
the session. At the beginning of the experiment,
participants were exposed to and rated a series
of temperatures (40 °C, 43 °C, 45 °C, 48 °C, 50
°C). On each trial, the thermode began at 35 °C.
The temperature level increased at a rate of 4 °C
per second until the target temperature was
reached. Once the target temperature was
reached, the thermode maintained the tempera-
ture for seven seconds.

Following each trial, participants made a ver-

bal rating of stimulus intensity on a scale rang-
ing from 0 (no heat sensation) to 100 (intoler-
able heat pain) with 50 (heat pain threshold) as
a midpoint (Tran, Wang, Tandon, Hernandez-
Garcia, & Casey, 2010). The order of the tem-
peratures was held constant. In order to prevent
habituation, there was a 30-second time block
between trials (Tran et al., 2010). If the partic-
ipant felt they could not withstand a tempera-
ture, they were instructed to click a mouse but-
ton which decreased the temperature at a rate
of 10 °C per second until the adaptation tem-
perature was reached.

Following the mood induction, participants

were exposed to the temperature they previ-
ously rated nearest to 20 (nonpainful stimula-
tion) or the temperature they rated nearest to 60
(painful stimulation) based on random assign-
ment (cf. Niedtfeld et al., 2010). No tempera-
tures were used if the participant did not previ-
ously complete the entire seven second trial.
Procedures were the same as earlier in the
study.

Mood induction.

As stimuli for the mood

induction, participants completed the computer-
ized selves interview (Shah & Higgins, 2001).
First, participants listed four “attributes of the
person you would ideally like to be” and four
“attributes of the person you feel you should
be.” After providing all eight attributes, partic-
ipants rated each attribute on how far they were

3

CHANGES IN AFFECT FOLLOWING PAIN

from possessing it (Far), and how important it
was for them to possess (Important) on a 7-point
Likert scale (1

⫽ not at all, 7 ⫽ extremely). A

score was calculated to decipher the attribute
each participant rated most important and cur-
rently furthest from (i.e., Far

⫹ Important).

Then, participants were asked to “choose a

specific time in your life in which you were
supposed to act with or be attribute and were
not” to write about. Participants were encour-
aged to pick a situation that caused great stress
and made them “upset, mad, nervous, sad, or
fearful.” After choosing a situation, participants
rated the situation on a scale to 1 (not stressful
at all) to 10 (the most stress I have experi-
enced). Finally, participants were given 10 min-
utes to describe the situation they chose in detail
including: where they were, who was there,
what they were doing, and the body sensations
they experienced (cf. Miller et al., 1987). Par-
ticipants were asked to try to generate the same
sensations and feelings that they felt in the
situation while they were writing. Pilot data in a
sample of 20 undergraduates showed a large
effect size for this mood induction (d

⫽ .88).

Dependent Measure

NA was measured with the Positive and Neg-

ative Affective Schedule (PANAS; Watson,
Clark, & Tellegen, 1988). In order to limit in-
terruption of affect assessments, we used five
items (upset, hostile, ashamed, nervous, afraid;
Thompson, 2007). Participants rated each item
on a 5-point Likert scale (1

⫽ not at all, 5 ⫽

extremely) with the instructions of “rate each
item to the extent you feel at this moment.”
Participants completed the PANAS three times
during the study: Time 1 (before mood induc-
tion), Time 2 (following the mood induction),
and Time 3 (following the painful/non painful
induction). Internal consistencies ranged from
.76 –.79.

Results

Manipulation Check

Inconsistent with previous research (Russ et

al., 1992), there was no effect of NSSI group on
intensity ratings of heat stimuli and group did
not interact with condition during the first ex-
posure ( p’s

⬎ .30). As expected, participants in

the nonpainful condition rated the intensity of
the stimuli at the second exposure lower
(M

⫽ 24.12, SD ⫽ 13.82) than participants in

the painful condition (M

⫽ 56.63, SD ⫽ 20.91)

and this difference was significant, F(1,
111) 83.72, p

⬍ .001, partial ␩

2

⫽ .45. The

average

temperatures

were

41.74

°C

(SD

⫽ 2.18) and 47.77 °C (SD ⫽ 2.07) for the

nonpainful and painful conditions, respectively.
Groups did not significantly differ in the stress
rating of their event for the mood induction,
t(112)

⫽ .87, p ⫽ .38, d ⫽ .18. Across groups,

the average rating was 6.83 out of 10
(SD

⫽ 1.70).

For the mood induction, there was not a sig-

nificant main effect of group, F(1, 112)

⫽ 1.53,

p

⫽ .21, partial ␩

2

⫽ .01. There was a signifi-

cant main effect of time, F(1, 112)

⫽ 234.35,

p

⬍ .001, partial ␩

2

⫽ .51. The main effect of

time indicated that the mood induction was suc-
cessful in significantly increasing NA across
groups. It is interesting that there was a signif-
icant

group-by-time

interaction,

F(1,

112)

⫽ 9.61, p ⬍ .01, partial ␩

2

⫽ .04. Contrary

to previous research (Nock & Mendes, 2008),
individuals without a history of NSSI had a
significantly larger increase in NA following
the mood induction, t(112)

⫽ 3.10, p ⬍ .01, d ⫽

.58. Groups did not significantly differ at
Time 1, t(112)

⫽ 1.61, p ⫽ .11, d ⫽ .29, but at

Time 2, the non-NSSI group had significantly
higher NA, t(112)

⫽ 2.81, p ⬍ .01 d ⫽ .51 (see

Table 1 for means).

Primary Analysis

To test the primary hypothesis, NA scores

were submitted to a group (NSSI vs. non-NSSI)
by condition (painful stimulation vs. nonpainful
stimulation) by time (Time 2 vs. Time 3) mixed
model ANOVA, with group and condition as
between-subjects factors and time as a within-
subject factor. The main effect of group and
condition, and the group-by-condition and the
time-by-condition interactions were not signifi-
cant (all p’s

⬎ .05). There was a significant

main effect of time, F(1, 109)

⫽ 287.11, p ⬍

.001, partial

␩

2

⫽ .55, indicating regardless of

group and condition, NA decreased over time.
There was a time-by-group interaction, F(1,
109)

⫽ 4.91, p ⬍ .05, partial ␩

2

⫽ .02. How-

ever, this interaction was further moderated by
condition, F(1, 109)

⫽ 5.90, p ⬍ .05, partial

4

BRESIN AND GORDON

␩

2

⫽ .02. To follow up the three way interac-

tion, we created a change score by subtracting
Time 2 NA from Time 3 NA, so that negative
values indicated a decrease in NA. Then, we ran
pairwise comparisons between the cells. Con-
sistent with our first prediction, among the NSSI
group, those in the nonpainful condition had a
significantly smaller change in NA than those in
the painful condition, t(108)

⫽ 2.41, p ⬍ .05,

d

⫽ .69. Inconsistent with our second predic-

tion, individuals with a history of NSSI who
experienced pain did not significantly differ
from the non-NSSI group regardless of condi-
tion ( p’s

⬎ .35; See Table 1).

Discussion

The goal of this study was to examine how

individuals with a history of NSSI differed from
individuals without a history of NSSI in their
emotional response to painful versus nonpainful
heat stimulation. Based on previous research
and theory (Niedtfeld et al., 2010; Selby &
Joiner, 2009), it was predicted that for individ-
uals with a history of NSSI, pain would lead to
a larger reduction in NA compared to nonpain-
ful stimulation, and this reduction would be
larger than that of the control group regardless
of condition. Our predictions received mixed
support. Furthermore, we found that individuals
with a history of NSSI had a smaller increase in
NA compared to controls following the mood
induction, and there was no difference between
groups in ratings of pain intensity for thermal
stimuli. These findings are inconsistent with
previous research and theory (Chapman et al.,
2006; Russ et al., 1992).

Consistent with our first prediction, for indi-

viduals with a history of NSSI, painful sensa-
tions compared to nonpainful sensations led to a
larger decrease in NA. This is consistent with
previous research that has found that proxies for
NSSI lead to a reduction in NA among those
who engage in NSSI (Franklin et al., 2010). Our
results also extend these findings by showing
that warm but not painful stimulation may not
be as effective in reducing NA among these
individuals. Additionally, this finding may be
consistent with the emotional cascade theory
(Selby & Joiner, 2009), which suggests that
among individuals who engage in NSSI and
other dysregulated behaviors, intense sensations
(e.g., pain) but not less intense sensations (e.g.,
warm stimulation) are necessary to reduce NA.
This may explain why individuals who engage
in NSSI seek out a behavior that induces pain,
in that it leads to greater reductions in NA for
them. It should be noted that the emotional
cascade theory is technically a theory of BPD.
However, as mentioned in the introduction, it is
based to some extent on theories of NSSI (e.g.,
Chapman et al., 2006) and therefore seems ap-
plicable to NSSI.

Contrary to our second prediction, the expe-

rience of pain did not lead to larger reductions
in NA for the NSSI group compared to controls
in either pain condition. Though the fact that
individuals without a history of NSSI showed a
decrease in NA following pain seems unex-
pected, it is consistent with previous research
(Bresin et al., 2010). Furthermore, Franklin et
al. (2010) did not find a difference between
NSSI groups on decrease in NA following pain.
Our results may help explain this finding by

Table 1
Means and Standard Deviations for Negative Affect by Group, Condition, and Time

Time 1

Time 2

Time 3

Time 3

⫺ Time 2

History of Nonsuicidal Self-injury

Average across conditions

1.54 (.61)

2.63 (.66)

1.46 (.57)

Painful Stimulation

1.45 (.45)

2.76 (.71)

1.31 (.52)

⫺1.48 (.81)

Nonpainful Stimulation

1.61 (.70)

2.52 (.60)

1.59 (.58)

⫺.92 (.73)

No History of Nonsuicidal Self-injury

Average across conditions

1.37 (.52)

3.02 (.78)

1.44 (.63)

Painful Stimulation

1.31 (.49)

2.89 (.74)

1.41 (.56)

⫺1.49 (.89)

Nonpainful Stimulation

1.44 (.56)

3.14 (.81)

1.47 (.70)

⫺1.66 (1.04)

Note.

For History of Nonsuicidal Self-injury, n

⫽ 32 for nonpainful stimulation and n ⫽ 27 painful stimulation. For No

History of Nonsuicidal Self-injury, n

⫽ 28 for nonpainful stimulation and n ⫽ 28 for painful stimulation. Possible range

for all variables is 1 to 5, aside from Time 3

⫺ Time 2 (⫺4 to 0).

5

CHANGES IN AFFECT FOLLOWING PAIN

suggesting that it is not pain that functions
differently between groups, but nonpainful
stimulation. Therefore, whatever mechanism is
operating to reduce NA (e.g., distraction, en-
dogenous opioids) may be more sensitive in
individuals without a history of NSSI. Conse-
quently, these individuals may not need to seek
out intense stimuli (e.g., pain) to regulate affect,
since less intense stimuli (e.g., cold shower) are
sufficient. On the other hand, among those who
engage in NSSI, intense stimulation is required
to achieve the same effects. It should be noted
that these results stand in contrast to Niedtfeld
et al. (2010), who found no interaction between
group and thermal stimulation type. This may
be due to multiple differences between their
study and the current study (e.g., the measure-
ment of physiological vs. experiential aspects of
affect). Regardless, future research should focus
on attempts to elucidate the mechanisms by
which pain and NSSI reduce NA.

An interesting yet unpredicted finding was

that individuals with a history of NSSI had
smaller reactions to the mood induction com-
pared to individuals with no NSSI history. This
finding is inconsistent with previous work
showing that individuals with a history of NSSI
had increased skin conductance during a dis-
tressing task (Nock & Mendes, 2008) and did
not differ from controls following a mood in-
duction (Glenn, Blumenthal, Klonsky, &
Hajack, 2011; Gratz et al., 2011). Moreover,
this is inconsistent with theories of NSSI that
suggest such individuals are sensitive to emo-
tional stimuli (Chapman et al., 2006). These
inconsistent results suggest that more research
into emotional reactivity and NSSI is necessary.
Nevertheless, our induction significantly in-
creased NA from baseline for both groups and
thus this finding does not necessarily affect the
interpretation of our main hypotheses.

Another unexpected finding is that there were

no group differences in pain intensity ratings of
thermal stimuli. This contradicts Russ et al.
(1992) who found that individuals with BPD
who reported no pain during NSSI gave lower
pain intensity ratings during a cold pressor task
than both healthy controls and individuals with
BPD who reported pain during NSSI, while the
latter two groups did not differ. It is possible
such differences could be found in our study if
participants were divided on whether or not they
experienced pain during NSSI. Hooley, Ho,

Slater and Lockshin (2010) did find that indi-
viduals with a history of NSSI (experience of
pain during NSSI was not reported) withstood
pressure pain past their pain tolerance for a
longer time than controls. However, the meth-
ods of this study differed from the current study
(endurance of increasing pressure pain vs. ex-
posure to a constant temperature). Furthermore,
this finding is orthogonal to our main predic-
tions, since the design of our study was to
control for individual differences in pain sensi-
tivity in order to find a painful temperature for
each participant.

Strengths and Limitations

There are limitations of this study that should

be noted. First, this study did not include a
group of individuals who had engaged in NSSI
but not in the last year. It is possible individuals
with a more recent/severe history of NSSI
would have larger reductions in NA following
pain than those with minor history (e.g., a few
incidents more than a year ago; Gordon et al.,
2010). Such comparisons could be the focus of
future studies. Second, we only measured self-
report aspects of emotion. At the same time, we
did measure affect in the moment which limits
retrospective bias, and our results conform to
similar studies using physiological measures of
emotion (Franklin et al., 2010). Still, future
research could benefit from replicating these
results with other measures of emotion. Finally,
laboratory pain is an imperfect proxy for NSSI,
in that the situation in the laboratory differs
significantly from genuine NSSI incidents. Such
differences (e.g., self-inflicted pain vs. other-
inflicted pain) may affect pain’s ability to reg-
ulate emotion. Future research may benefit by
using ecological momentary assessments,
which allow for repeated measurements in the
participants’ environment.

In addition to the limitations, it is also worth

noting the strengths of the current study. First,
compared to previous research, this study had a
relatively large sample of individuals who en-
gaged in NSSI recently. Also, we were able to
recruit an equal amount of males and females
for the NSSI group, where previous research
has used primarily female samples. Second, this
study provided a stringently controlled test of
whether pain functions differently from non-
painful stimulation in individuals with a recent

6

BRESIN AND GORDON

history of NSSI compared to those with no
history of NSSI. Third, compared to previous
research, the mood induction was one that may be
more related to real-world triggers of NSSI. Fi-
nally, this study used a method of pain induction
(i.e., heat) that is more similar to a method NSSI
(burning) than some previous studies which used
no pain induction (e.g., Haines et al., 1995) or a
cold pressor task (Franklin et al., 2010).

References

American Psychiatric Association. (2000). Diagnos-

tic and statistical manual of mental disorders (4th
ed., text revision). Washington, DC: Author.

Bresin, K., Gordon, K. H., Bender, T. W., Gordon,

J. J., & Joiner, T. E. (2010). No pain, no change:
Reductions in prior negative affect following phys-
ical pain. Motivation and Emotion, 34, 280 –287.
doi:10.1007/s11031-010-9168-7

Chapman, A. L., Gratz, K. L., & Brown, M. Z.

(2006). Solving the puzzle of deliberate self-harm:
The experiential avoidance model. Behaviour Re-
search and Therapy, 44, 371–394. doi:10.1016/
j.brat.2005.03.005

Franklin, J. C., Hessel, E. T., Aaron, R. V., Arthus,

M. S., Heilbron, N., & Prinstein, M. J. (2010). The
functions of nonsuicidal self-injury: Support for
cognitive-affective regulation and opponent pro-
cesses from a novel psychophysiological para-
digm. Journal of Abnormal Psychology, 119, 850 –
862. doi:10.1037/a0020896

Freud, S. (1929). Beyond the pleasure principle. New

York: Liveright.

Glenn, C. R., Blumenthal, T. D., Klonsky, E. D., &

Hajcak (2011). Emotional reactivity in nonsuicidal
self-injury: Divergence between self-report and
startle measures. International Journal of Psycho-
physiology, 80, 166 –170. doi:10.1016/j.ijpsycho
.2011.02.016

Gordon, K. H., Selby, E. A., Anestis, M. D., Bender,

T. W., Witte, T. K., Braithwaite, S., . . . Joiner,
T. E. (2010). The reinforcing properties of re-
peated deliberate self-harm in adults. Archives of
Suicide Research, 14, 329 –341. doi:10.1080/
13811118.2010.524059

Gratz, K. L. (2001). Measurement of deliberate self-

harm: Preliminary data on the Deliberate Self-
Harm Inventory. Journal of Psychopathology and
Behavioral Assessment, 23, 253–263. doi:10.1023/
A:1012779403943

Gratz, K. L., Hepworth, C., Tull, M. T., Paulson, A.,

Clarke, S., Remington, B., & Lejuez, C. W.
(2011). An experimental investigation of emo-
tional willingness and physical pain tolerance in
deliberate self-harm: The moderating role of inter-

personal distress. Comprehensive Psychiatry, 52,
63–74. doi:10.1016/j.comppsych.2010.04.009

Gray, J. A. (1982). The neuropsychology of anxiety:

An enquiry into the functions of the septo-
hippocampal system. New York: Oxford Univer-
sity Press.

Haines, J., Williams, C. L., Brain, K. L., & Wilson,

G. V. (1995). The psychophysiology of self-
mutilation. Journal of Abnormal Psychology, 104,
471– 489. doi:10.1037/0021-843X.104.3.471

Hooley, J. M., Ho, D. T., Slater, J., & Lockshin, A.

(2010). Pain perception and nonsuicidal self-
injury: A laboratory investigation. Personality
Disorders: Theory, Research, and Treatment, 1,
170 –179. doi:10.1037/a0020106

Kahneman, D., & Tversky, A. (1979). Prospect the-

ory: An analysis of decision under risk. Economet-
rica, 47, 263–291. doi:10.2307/1914185

Miller, G. D., Levin, D. N., Kozak, M. J., Cook,

E. W., McLean, A., & Lang, P. J. (1987). Individ-
ual differences in imagery and the psychophysiol-
ogy of emotion. Cognition and Emotion, 1, 367–
390. doi:10.1080/02699938708408058

Niedtfeld, I., Schulze, L., Kirsh, P., Herpertz, S. C.,

Bohus, M., & Schmahl, C. (2010). Affect regula-
tion and pain in borderline personality disorder: A
possible link to the understanding of self-injury.
Biological Psychiatry, 68, 383–391. doi:10.1016/
j.biopsych.2010.04.015

Nock, M. K. (2009). Why do people hurt them-

selves? New insights into the nature and func-
tions of self-injury. Current Directions in Psy-
chological Science, 18, 78 – 83. doi:10.1111/
j.1467-8721.2009.01613.x

Nock, M. K., & Mendes, W. B. (2008). Physiological

arousal, distress tolerance, and social problem-
solving deficits among adolescent self-injurers.
Journal of Consulting and Clinical Psychol-
ogy, 76, 28 –38. doi:10.1037/0022-006X.76.1.28

Nock, M. K., & Prinstein, M. J. (2004). A functional

approach to the assessment of self-mutilative behav-
ior. Journal of Consulting and Clinical Psychol-
ogy, 72, 885– 890. doi:10.1037/0022-006X.72.5.885

Russ, M. J., Roth, S. D., Lerman, A., Kakuma, T.,

Harrison, K., Shindledecker, R. D., & Mattis, S.
(1992). Pain perception in self-injurious patients
with borderline personality disorder. Biological
Psychiatry, 32, 501–511. doi:10.1016/0006-
3223(92)90218-O

Selby, E. A., & Joiner, T. E. (2009). Cascades of

emotion: The emergence of borderline personality
disorder from emotional and behavioral dysregu-
lation. Review of General Psychology, 13, 219 –
229. doi:10.1037/a0015687

Shah, J., & Higgins, E. T. (2001). Regulatory con-

cerns and appraisal efficiency: The general impact
of promotion and prevention. Journal of Person-

7

CHANGES IN AFFECT FOLLOWING PAIN

ality and Social Psychology, 80, 693–705. doi:
10.1037/0022-3514.80.5.693

Thompson, E. R. (2007). Development and valida-

tion of internationally reliable short-form of the
Positive and Negative Affect Schedule (PANAS).
Journal of Cross-Cultural Psychology, 38, 227–
242. doi:10.1177/0022022106297301

Tran, T. D., Wang, H., Tandon, A., Hernandez-

Garcia, L., & Casey, K. L. (2010). Temporal sum-

mation of heat pain in humans: Evidence support-
ing thalmocortical modulation. Pain, 150, 93–102.
doi:10.1016/j.pain.2010.04.001

Watson, D., Clark, L. A., & Tellegen, A. (1988).

Development and validation of brief measures of
positive and negative affect: The PANAS scales.
Journal of Personality and Social Psychol-
ogy,

54,

1063–1070.

doi:10.1037/0022-

3514.54.6.1063

8

BRESIN AND GORDON