C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

Crisis 2012; Vol. 33(2):106–112

© 2012 Hogrefe Publishing

Research Trends

Self-Injurious Behavior vs.

Nonsuicidal Self-Injury

The CNS Stimulant Pemoline as a

Model of Self-Destructive Behavior

Christopher M. Bloom

1

, Shareen Holly

1

, and Adam M. P. Miller

2

1

Department of Psychology, Providence College, Providence, RI, USA

2

Department of Psychology, Cornell University, Ithaca, NY, USA

Abstract. Background: Historically, the field of self-injury has distinguished between the behaviors exhibited among individuals with a
developmental disability (self-injurious behaviors; SIB) and those present within a normative population (nonsuicidal self-injury;
NSSI),which typically result as a response to perceived stress. More recently, however, conclusions about NSSI have been drawn from
lines of animal research aimed at examining the neurobiological mechanisms of SIB. Despite some functional similarity between SIB
and NSSI, no empirical investigation has provided precedent for the application of SIB-targeted animal research as justification for
pharmacological interventions in populations demonstrating NSSI. Aims: The present study examined this question directly, by simulating
an animal model of SIB in rodents injected with pemoline and systematically manipulating stress conditions in order to monitor rates of
self-injury. Methods: Sham controls and experimental animals injected with pemoline (200 mg/kg) were assigned to either a low stress
(discriminated positive reinforcement) or high stress (discriminated avoidance) group and compared on the dependent measures of
self-inflicted injury prevalence and severity. Results: The manipulation of stress conditions did not impact the rate of self-injury demon-
strated by the rats. The results do not support a model of stress-induced SIB in rodents. Conclusions: Current findings provide evidence
for caution in the development of pharmacotherapies of NSSI in human populations based on CNS stimulant models. Theoretical impli-
cations are discussed with respect to antecedent factors such as preinjury arousal level and environmental stress.

Keywords: nonsuicidal self-injury, self-injurious behavior, pemoline, animal model

Self-injury has become a phenomenon of great interest for
many clinicians and researchers. Traditionally, the field has
distinguished between those self-harming behaviors occur-
ring among individuals with cognitive and developmental
disabilities (i.e., self-injurious behavior; SIB), and those
occurring in normative populations (i.e., nonsuicidal self-
injury; NSSI). It is to be expected, then, that these two con-
ditions present in very different ways: SIB often includes
stereotypic and repetitive self-harming behaviors, long hy-
pothesized to serve a self-stimulating function for the indi-
vidual (Carr, 1977). Typical manifestations include head-
banging, hitting or striking oneself, biting, scratching, hair-
pulling, or hitting arms or legs against other objects
(Luiselli, 2009). The specific behavior observed is largely
associated with the diagnosis, with head-banging and face-
hitting more commonly linked with autism or intellectual
disability, skin-picking typically linked with Prader-Willi
syndrome, and lip-, tongue-, and digit-biting seen primarily
in Lesch-Nyhan syndrome (Kies & Devine, 2004). In con-
trast, NSSI has been applied to a wide range of behaviors
that result in the immediate damage of one’s own body

tissue in the absence of intent to die (Nixon & Heath, 2009).
A recent study found that the most common methods of
NSSI cited by university students were cutting and scratch-
ing (Laye-Gindhu & Schonert-Reichl, 2005; Nixon, Clou-
tier, & Aggarwal, 2002). The specific behavior observed is
likely one of convenience to the individual, as the majority
of episodes of self-injury are thought to occur in private
(Walsh, 2006).

Despite the differences in presentation, the same behav-

ioral models exist to explain the functional morphologies
of both SIB and NSSI. In 1977, E. G. Carr established four
functions of SIB: (1) social-positive functions are used
to obtain a tangible reward or social attention, (2) social-
negative functions are used to escape or avoid an un-
desired task or demand, (3) automatic-positive functions
serve to obtain a desired sensation, and (4) automatic-
negative functions are used to alleviate pain or discomfort.
In the field of developmental disabilities, automatic rein-
forcement is typically understood as a response maintained
by sensory stimulation within the individual (Luiselli,
2009). For example, individuals with developmental delays

DOI: 10.1027/0227-5910/a000127

Crisis 2012; Vol. 33(2):106–112

© 2012 Hogrefe Publishing

often engage in SIB to resolve internal states of both over-
and understimulation. Relative to SIB used for attention-
seeking, escaping or avoiding a task demand, tangible-elic-
iting, or even easing of medical ailments, this automatic
reinforcement category of SIB often occurs in the absence
of any identifiable environmental stressors (Luiselli, 2009).
As it is maintained primarily by intrapersonal stimulation
(as opposed to external or interpersonal conditions), SIB
motivated by this type of automatic reinforcement is much
more intransigent than its socially maintained counterparts.

Based on Carr’s (1977) early work in populations exhib-

iting SIB, Nock and Prinstein (2004) proposed a similar
model of NSSI functional reinforcement. This four-func-
tion model (FFM), focusing solely on NSSI in normative
populations, described the behavior as a type of reinforce-
ment that served one of four functions for self-injurers.
These functions could be either social (external) or auto-
matic (internal) reinforcement, as well as either positive or
negative in nature. At its core, the FFM associated with
NSSI described by Nock and Prinstein (2004) is conceptu-
ally identical to the earlier functional model of SIB pro-
posed by Carr (1977). As predicted by the authors, studies
examining the self-reported functions of NSSI within nor-
mative populations found support for the use of this model.
Overwhelmingly, self-injurers cite emotion regulation, or
automatic reinforcement, as the primary function of NSSI.
In these studies, self-injurers indicate that in times of high
stress NSSI is an effective means of achieving a more de-
sirable internal state (Klonsky, 2007; Nock & Prinstein,
2004; Gratz & Roemer, 2004). This means of emotional
regulation is thought to arise from a preexisting heightened
state of physiological arousal, which, along with individual
deficits in problem-solving ability, results in a suboptimal
means of responding to a stress-inducing event or stimulus
(Nock & Mendes, 2008). As evidenced by the growing
body of literature investigating functions of NSSI, an im-
portant factor influencing a person’s ability to regulate
overwhelming emotions is the presence of perceived stress.
This has led some researchers to focus on the role of envi-
ronmental stress in relation to NSSI. One recent study ex-
amined the effect of psychological characteristics and
stressful life events among a clinical sample of adolescent
girls. The researchers reported that a negative attributional
style, when combined with an overall higher number of
interpersonal stressors, predicted an increased likelihood of
NSSI engagement (Guerry & Prinstein, 2010).

Clearly, there are important similarities in the way that

NSSI and SIB are characterized with regard to function,
particularly in terms of automatic reinforcement. However,
the differences in form (e.g., frequency, type, severity) be-
tween these two conditions, as well as the differences of
primary diagnosis (or lack thereof) in the populations in
which they appear and, finally, the relevance of environ-
mental stressors still lead to questions about the similarity
of these two phenomena. A more comprehensive under-
standing about the differences between SIB and NSSI may
be found in behavioral investigations.

As discussed above, two similar behavioral models in

human populations arose independently for both NSSI and
SIB which more closely examine the function served by
self-injury (Carr, 1977; & Nock & Prinstein, 2004). While
traditionally investigated separately as a result of clear dis-
tinctions in definition, many conclusions for both NSSI and
SIB have nonetheless come out of the same line of animal
research (Osuch & Payne, 2009). Indeed, animal models of
self-harm have been conceptualized, executed, and inter-
preted with the SIB definition in mind. The application of
these animal models of SIB to NSSI populations, therefore,
may be premature.

Animal Models of Self-Injury

Ethical and safety considerations limit the nature and type
of experimental investigations of self-injury possible in hu-
man populations (Prinstein, 2008). These difficulties led to
the development of animal models of SIB, intended to elu-
cidate the neurochemical factors involved in the self-stim-
ulating behaviors. CNS stimulants, pemoline and caffeine,
were both used to introduce self-injury in rodents. Pemo-
line (2-amino-5-phenyl-1,3-oxal-4-one) was first reported
to induce “self-aggressiveness” in rodents some 50 years
ago (Genovese, Napoli, & Bolego-Zonta, 1969). Kies and
Devine (2004) provided a more systematic review of the
influence of pemoline on self-injury and also demonstrated
the ability of the stimulant caffeine (trimethylxanthine) to
produce a similar effect.

The ability of CNS stimulants to induce self-injury pro-

vided an opportunity to experimentally examine this be-
havior and to draw some conclusions regarding underlying
chemical mechanisms and potential pharmacotherapies
(Kies & Devine, 2004). Muehlmann, Brown, and Devine
(2008) used pemoline to investigate drugs previously used
as treatments for SIB in populations with developmental
disabilities. Their work demonstrated the efficacy of three
drugs in inhibiting the self-injurious effects of pemoline
injection – risperidone, valproate, and topiramate – all of
which have been used with some success in clinical popu-
lations to reduce SIB (Muehlmann et al., 2008).

Applicability of Animal Models to NSSI

Although behavioral investigations in human research ap-
pear to suggest similarities in the development and main-
tenance of self-injury, the generalizability of the animal
models of self-injury, interpreted as models of SIB, to pop-
ulations presenting with NSSI remains to date untested.
The belief that these two manifestations are in fact similar
may be presently unwarranted given the wealth of research
on the differences in behavioral presentation between NSSI
and SIB as well as the general differences between the pop-
ulations in which they present.

Indeed, while the CNS stimulant animal models provided

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

107

© 2012 Hogrefe Publishing

Crisis 2012; Vol. 33(2):106–112

a wealth of information concerning human self-injury, they
focused exclusively on SIB in populations with developmen-
tal disabilities and attempted in no way to generalize their
findings to more normative populations. As addressed in the
earlier section comparing SIB and NSSI, there is a distinct
possibility that mechanisms of action may differ between
these normative and clinical populations. Nevertheless, many
authors cited these works as reason to implement similar
pharmacotherapies in normative populations demonstrating
NSSI (Osuch & Payne, 2009). Griengl, Sendera, and Danten-
dorfer (2001) reported a case study in which sertraline, val-
proate, doxepin, and risperidone were all implemented with
some success to treat a severe case of self-injury in a 39-year-
old male patient. However, even as authors chose to general-
ize the animal model findings, they noted the relatively pre-
mature understanding of how these models address factors
such as sex, mood, or stressors in the topography of SIB
among rodents (Osuch & Payne, 2009).

This trend toward generalization may be due, in part, to

the growing belief that SIB and NSSI may be more similar
than previously thought (Nock, 2010). For example, both
appear to follow the similar behavioral models proposed
by both E. R. Carr (1977) and Nock and Prinstein (2004)
in each of their respective fields. This line of reasoning is
problematic, however, because of the differences between
the maintaining conditions reported in clinical SIB and nor-
mative NSSI populations. Animal models of SIB appear to
simulate conditions in which a heightened state of arousal
is alleviated through engaging in self-harming behaviors,
but they do not account for the effects of stress on self-in-
jury commonly reported in NSSI populations. Further, the
underlying mechanism or direction of reinforcement (i.e.,
positive or negative) in these models is still unknown. The
current experiment proposed to determine if the application
of an animal model of self-injury is appropriate for the field
of NSSI, and if CNS stimulants such as pemoline may serve
to better model NSSI, SIB, or both.

The primary goal of the present study was to simulate

an SIB condition using a supported animal model of self-
injury, and to systematically manipulate stress conditions
in order to increase arousal in rodents. Subsequently, the
animals’ stress response was monitored to establish the rel-
ative impact of environmental stress on SIB. Specifically,
we sought to determine whether a fear-evoking environ-
ment results in an increase in self-injury. In other words,
whether SIB and NSSI are indeed similar phenomena and
application of an external stress condition would likely in-
crease the frequency of SIB in rats.

Methodology

Subjects

The subjects of the present experiment included 16 female
Sprague-Dawley rats, obtained from the Charles River

Laboratory in Michigan. The rats, maintained at 90% of ad
libitum weight, weighed between 180–250 g at the time of
the experiment. Each rat was housed individually in a vi-
varium with a 12-h light/12-h dark cycle (lights on at
8 a.m.), at a constant temperature of 22–23°C, and with
water available ad libitum.

Materials

Consistent with recommendations from past animal research-
ers (Kies & Devine, 2004), the drug pemoline was used to
elicit self-injury in the rats. In accordance with standard dos-
age regulations, pemoline was suspended in peanut oil at a
concentration of 1 mg/60 ml, stirred continuously.

Procedure

Prior to the experimental trials for the present study, the 16
rats were magazine-trained through shaping during 1–2 h
periods each day. Shaping involved reinforcing lever-
pressing in the presence of a light in order to receive food
pellets (i.e., low-stress condition) or to avoid shock (i.e.,
high-stress condition). Training continued on a half-hour
program of 10 s randomly selected light/no-light discrimi-
nated-stimulus trials. Each trial was separated by variable
22.5 s intertrial intervals until performance reached asymp-
tote. Upon successful completion of magazine training,
four rats from each operant paradigm were then randomly
selected to receive pemoline injections. The remaining
eight rats were assigned to receive a peanut-oil vehicle.

Experimental procedure and data collection were sched-

uled to occur over 6 consecutive days, with pemoline injec-
tions occurring on day 5. Data for days prior to injection (days
1–4) and following injections (days 5–8) provided compari-
sons on operant training prior to and following the cessation
of injections. On days 3 and 4, each rat was injected subcuta-
neously with 200 mg/kg of either pemoline suspended in pea-
nut oil or peanut oil alone. As per protocol, the rats were
weighed and examined for injuries each morning during the
experiment. Any injuries discovered were rated separately by

Table 1. Tissue trauma rating scale (Turner et al., 1999)

Score

Severity

Description

0

No SIB

No tissue damage

1

Very Mild SIB Slight edema, pink moist skin, involves

small area

2

Mild SIB

Moderate edema, slight erythema, slightly
denuded skin, involves medium area
and/or involves multiple sites

3

Moderate SIB

Substantial edema and erythema, large ar-
ea, substantially denuded skin, and/or in-
volves multiples sites

4

Severe SIB

Clear/open lesions, and/or amputation of
digit, requires immediate euthanasia

108

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

Crisis 2012; Vol. 33(2):106–112

© 2012 Hogrefe Publishing

two observers using a criteria adapted from Kies and Devine
(2004) (see Table 1). During the course of the experiment,
35-min operant sessions occurred twice daily for each rat.
The first session occurred within 2 h 30 min after injection,
and the second session occurred approximately 70 min after
the first session had been completed. In order to prevent the
occurrence of time-related effects, group-start times were
counterbalanced. All sessions during the experimental period
were video-recorded. Rats were visually inspected for inju-
ries at 9 a.m., 5 p.m., and 12 a.m., following the inspection
and injury severity rating procedure presented in Table 1 rec-
ommended by Turner, Panksepp, Bekkedal, Borkowski, and
Burgdorf (1999).

Results

Number of Injuries

To determine the main effect of the drug and stress condi-
tions on the number of self-inflicted injuries, two Mann-
Whitney U, nonparametric tests were calculated. The first
analysis revealed that the drug condition accounted for the
number of injuries (U = 64, p < .001). Median values (with
interquartile ranges reported in parentheses) for both the
drug and control conditions were 2.00 (2.00) and 0.00
(0.00), respectively, suggesting that animals injected with
pemoline were more likely to develop SIB. A second
Mann-Whitney U test failed to reveal a statistically signif-
icant effect for the stress condition on the development of
self-injury (U = 32, p = 1.00). Median values for both the
low and high stress conditions were 0.50 (2.50) and 1.43
(2.00), respectively. Interquartile ranges for the nonsignif-
icant analysis exceeded the median values, indicating dis-
persion in the data. The development of SIB did not differ

as a function of stress condition with animals trained in fear
conditioning responding very similar to those trained in
discriminated positive reinforcement. Daily rates of self-in-
jury among pemoline-injected animals (n = 8) in both the
low- and high-stress conditions are presented in Figure 1.

Injury Severity

To determine the main effect of the drug and stress condi-
tions on the severity of self-inflicted injuries, two Mann-
Whitney U, nonparametric tests were calculated. This anal-
ysis revealed that drug condition accounted for the severity
of injuries (U = 64, p < .001), median values (with inter-
quartile ranges reported in parentheses) for both the drug
and control conditions were 4.00 (1.75) and 0.00 (0.00),
respectively, suggesting that animals injected with pemo-
line were more likely to develop SIB. A second Mann-
Whitney U test failed to reveal a statistically significant
effect for the stress condition on the severity of self-inflict-
ed injuries (U = 34, p = .819), with median values for both
the low- and high-stress conditions being 0.50 (3.75) and
1.00 (4.00), respectively. Interquartile ranges for the non-
significant analysis exceeded the median values, indicating
dispersion in the data. Results revealed that the severity of
SIB did not differ as a function of stress condition with
animals trained in fear conditioning responding very simi-
lar to those trained in discriminated positive reinforcement.

Discussion

The present study sought to establish the validity of the
field’s long history of distinguishing between self-injury

Figure 1. Daily rates of self-injury
among pemoline-injected rodents (n =
8) during first 3 days postinjection for
both low-stress and high-stress condi-
tions.

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

109

© 2012 Hogrefe Publishing

Crisis 2012; Vol. 33(2):106–112

that is expressed in normative populations as a response to
stress (i.e., NSSI) and self-injury observed in populations
of individuals with developmental disabilities (i.e., SIB;
Nixon & Heath, 2009; Nock & Favazza, 2009). Overall,
the results suggest that stress was not a factor in SIB ob-
served in rodents. Manipulation of environmental stress
was not found to significantly impact the frequency or se-
verity of injury demonstrated in a pemoline-induced model
of SIB. Given the distinction in the literature between SIB
and NSSI with regard to the role of environmental stress,
a more detailed examination of the present results in con-
junction with theoretical and empirical literature is war-
ranted.

Traditionally, the behaviors defined as SIB and NSSI are

examined separately. The reason for this differentiation is
readily apparent: The frequency, rate, variability, and
means by which the injuries take place are often quite dif-
ferent between the two populations. For both populations,
a more comprehensive understanding of the self-harming
behaviors is often limited to self-report or observational
studies. Animal models of self-injury provide a valuable
means to study this often-devastating behavior by allowing
for improved experimental control as well as investigations
into the proposed neurochemical mechanisms such as sero-
tonin and dopamine (Kies & Devine, 2004; Muehlmann et
al., 2008). A review of the literature in these animal models,
however, makes it clear that the intention of the researchers
was specifically to investigate populations of developmen-
tally delayed individuals such as those with autism or
Lesch-Nyan syndrome, and very little consideration is giv-
en to self-injury in normative populations (Kies & Devine,
2004; Muehlmann et al., 2008). Despite this limited scope,
many researchers applied the results of CNS-stimulant an-
imal models of self-injury to populations engaging in NSSI
(Osuch & Payne, 2009). Some have gone so far as to cite
this work to support the use of pharmacotherapies such as
olanzapine or risperidone in the treatment of self-injury in
normative populations (Kerr, Muelhenkamp, & Turner,
2010.).

Consistent with previous studies using pemoline in rats

to induce SIB (Kies & Devine, 2004), seven of the eight
rodents in the present study injected with the stimulant en-
gaged in self-injury. This behavior was not observed in the
rats injected with a peanut-oil vehicle representing the con-
trol group. As expected, the results from the present study
supported the use of pemoline as an effective means of es-
tablishing an animal model of SIB in rodents. However, in
addition to confirming the effects of pemoline, the experi-
mental component of environmental stress was included in
order to establish the relative applicability of the SIB ani-
mal model to theoretical conditions stemming from NSSI
research.

One such theory, that of emotion regulation, suggests

that self-injurers have difficulty regulating their emotions
because of an elevated baseline of physiological arousal;
in response to environmental stressors they use self-injury
to escape and avoid elevated, aversive emotions (Guerry &

Prinstein, 2010). In order to simulate an environmental
stress condition, we exposed the rats to a conditioned fear
paradigm in which a warning light signaled shock. If the
emotion regulation function of NSSI held true for rats ex-
hibiting SIB, the expected result would have shown a no-
table increase in SIB during times of high stress, as com-
pared to the low-stress conditions. Though self-injury was
clearly demonstrated throughout the course of the experi-
ment, our results fail to support the notion that self-injury
serves to relieve the effects of environmental stress as pro-
posed in models of NSSI. Indeed, self-injury onset and se-
verity were nearly identical between low-stress and high-
stress conditions. This result suggests that the proposed
emotional regulation function of NSSI may not be well
modeled using CNS stimulants such as pemoline, and that
although NSSI and SIB may seem similar under particular
conditions, they may be operating via different mecha-
nisms behaviorally and neurochemically.

Implications for Self-Injury in Human
Populations

Analysis of the severity and prevalence of self-injury
showed no difference between animals trained in condi-
tioned fear compared to those trained under an appetitive
paradigm. This result lends support to the hypothesis that
current rodent models of self-injury may be more useful in
understanding SIB than NSSI, and that attempts to draw
conclusions regarding the neurochemical and physiologi-
cal underpinnings of NSSI from these animal models may
be misguided. The results from the present study clearly
showed that a pemoline model of self-injury more appro-
priately represented the type of self-harm that is character-
ized as SIB, as the manipulation of environmental stressors
had no significant impact on the animals’ behavior. This
result directly contradicts what is predicted by the function-
al model of NSSI, where evidence suggests environmental
stressors play a strong role in the occurrence of NSSI
(Guerry & Prinstein, 2010). Although both SIB and NSSI
have been documented in both observational and self-re-
port studies to serve automatic functions (Carr, 1977; Nock,
2010; Nock & Prinstein, 2004), current animal models of
self-injury may be too limited in scope when it comes to
simulating self-injury in normative populations.

These conclusions are important for clinicians working

with youth in normative populations, as the current pharma-
cological interventions for nonnormative populations and
normative populations alike are heavily dependent on the re-
sults from animal studies (Kerr et al., 2010; Osuch & Payne,
2009). The current work provides ample evidence for caution
in the development of pharmacotherapies of NSSI based on
CNS stimulant models. Indeed, Griengl and colleagues
(2001) reported inconsistencies in the treatment of self-injury
of a patient presenting with borderline personality disorder
using sertraline, valproate, doxepin, and risperidone. Finally,

110

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

Crisis 2012; Vol. 33(2):106–112

© 2012 Hogrefe Publishing

in addition to these pharmacological treatments, the 39-year-
old patient was also prescribed naltrexone with some success.
Naltrexone, an opiate antagonist, has also been reported as an
option for the treatment of self-injury. Its use stems from a
long-held hypothesis that self-injury may be linked to dys-
function in the endogenous opioid system (Kerr et al., 2010).
It has been suggested that self-injuries serve to potentiate
endogenous opioid release, serving a homeostatic mecha-
nism in those who have an abnormally low level of opiate
function. The potentiation of endogenous opioids is thought
to serve a self-reinforcing function. Though a potentially in-
teresting avenue, little data exists to support this possibility
outside of a smattering of case studies using naltrexone (Kerr
et al., 2010). It is also worth noting that this hypothesis ap-
pears to be limited to those individuals who self-injure for
automatic and internal functions.

Theoretical Implications and Future
Directions

Given the reported efficacy of pharmacological interventions
with samples of normative self-injuring adolescents (Kerr et
al., 2010) as well as the recent speculation about the similar-
ities with regard to function between SIB and NSSI (Nock,
2010), the results of the present study should be interpreted
with some caution. A more comprehensive understanding of
how the two behaviors differ (or not) might be more appro-
priately conceptualized at the functional level, as a recent
examination of the functions of self-injury in each population
has prompted some to speculate about their relative similarity
(Carr, 1977; Nock, 2010; Nock & Prinstein, 2004). This spec-
ulation, however, belies subtle differences in the prominent
correlates and features of the two populations. Approaching
these phenomena from a functional perspective rather than
syndromal approach (Nock & Cha, 2010) may clarify the
differences that underlie the behaviors. Perhaps, as the cur-
rent results and literature suggest, the differences are more
readily apparent in the antecedents, rather than in the conse-
quences of self-harm. The role of both arousal level and the
impact of environmental stressors may prove to be important
modifiers in the likelihood of the behavior serving a particular
function.

To further investigate this possibility, we propose repeat-

ing the experiment with the second most commonly used
CNS stimulant in models of SIB: caffeine. The ability of
caffeine to elicit self-injury is well established, albeit less
powerful than that of pemoline (Kies & Devine, 2004;
Muehlmann et al., 2008). This leaves open the possibility
that, at a lower level of stimulation, self-injury may still be
susceptible to the effects of environmental stress. Failure
to show the effect of stress on the ability of caffeine to elicit
self-injury would strengthen the proposition that animal
models of self-injury may more appropriately represent
SIB than NSSI – and would further argue that attempts to

draw conclusions about NSSI from animal demonstrations
of SIB are premature.

Summary and Conclusions

The present study shows that neurochemical information
gained from animal studies of self-injury may not translate
to NSSI. Although the similarities between the two behav-
iors are undeniable, current results support some distinction
between the neurochemical and behavioral manifestations
of NSSI and SIB at the level of animal model. The results
from the present study suggest that animal models of self-
injury may more closely resemble SIB in terms of type,
function, frequency, and severity. Until further research is
completed using animal models, it is premature to draw
conclusions about NSSI from animal work in self-injury.
However, further investigations using caffeine will be crit-
ical in developing a more comprehensive understanding of
how arousal levels play a precipitating role in self-injury,
specifically with regard to automatic reinforcement and the
valence of environmental stress. The development of an
animal model of NSSI that reflects the proposed differenc-
es between self-injury between normative and nonnorma-
tive populations seems essential in developing a better un-
derstanding of both phenomena.

Acknowledgments

This research was made possible by RI-INBRE Grant
P20RR016457 from the National Center for Research Re-
sources (NCRR), a component of the National Institutes of
Health (NIH). Its contents are solely the responsibility of
the authors and do not necessarily represent the official
views of NCRR or NIH.

References

Carr, E. G. (1977). The motivation of self-injurious behavior: A

review of some hypotheses. Psychological Bulletin, 84,
800–816. Retrieved from http://www.accesspointkids.com/up-
loads/Carr_1977_-_Self_Injurious_Behavior.pdf

Genovese, E., Napoli, P. A., & Bolega-Zonta, N. (1969). Self-ag-

gressiveness: A new type of behavioral change induced by pe-
moline. Life Sciences, 8, 513–515.

Gratz, K., & Roemer, L. (2004). Multidimensional assessment of

emotional regulation and dysregulation: Development, factor
structure, and initial validation of the Difficulties in Emotion
Regulation Scale. Journal of Psychopathology and Behavioral
Assessment, 26, 41–54.

Griengl, H., Sendera, A., & Dantendorfer, K. (2001). Naltrexone

as a treatment of self-injurious behavior – A case report. Acta
Psychiatrica Scandinavica, 103, 234–236.

Guerry, J. D., & Prinstein, M. J. (2010). Longitudinal prediction of

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

111

© 2012 Hogrefe Publishing

Crisis 2012; Vol. 33(2):106–112

adolescent nonsuicidal self-injury: Examination of a cognitive
vulnerability-stress model. Journal of Clinical Child and Adoles-
cent Psychiatry, 39, 77–89. doi 10.1080/15374410903401195

Kerr, P. L., Muelhenkamp, J. L., & Turner, J. M. (2010). Nonsuicidal

self-injury: A review of current research for family medicine and
primary care physicians. Journal of American Board Family
Medicine, 23, 240–259. doi 10.3122/jabfm.2010.02.090110

Kies, S. D., & Devine, D. P. (2004). Self-injurious behavior: A

comparison of caffeine and pemoline models in rats. Pharma-
cology, Biochemistry and Behavior 79, 587–598. doi 10.1016/
j.pbb.2004.09.010

Klonsky, E. D. (2007). The functions of deliberate self-injury: A

review of the empirical evidence. Clinical Psychology Review,
27, 226–239. doi 10.1016/j.cpr.2006.08.002

Laye-Gindhu, A., & Schonert-Reichl, K. A. (2005). Nonsuicidal

self-harm among community adolescents: Understanding the
“whats” and “whys” of self-harm. Journal of Youth and Ado-
lescence, 34, 445–457. doi 10.1007/s10964-005-7262-z

Luiselli, J. K. (2009). Nonsuicidal self-injury among people with

developmental disabilities. In M. K. Nock (Ed.), Understand-
ing nonsuicidal self-injury: Origins, assessment, and treatment
(pp. 157–179). Washington, DC: American Psychological As-
sociation.

Muehlmann, A. M., Brown, B. D., & Devine, D. P. (2008). Pemo-

line (2-Amino-5-phenyl-1,3-oxazol-4-one)-induced self-inju-
rious behavior: A rodent model of pharmacotherapeutic effi-
cacy. The Journal of Pharmacology and Experimental
Therapeutics, 324, 214–223. doi 10.1124/jpet.107.128207

Nixon, M. K., & Heath, N. L. (2009). Introduction to nonsuicidal

self-injury in adolescents. In M. K. Nixon & N. L. Heath
(Eds.), Self-injury in youth: The essential guide to assessment
and intervention (pp. 1–6). New York: Routledge.

Nixon, M. K., Cloutier, P. F., & Agarwal, S. (2002). Affect regu-

lation and addictive aspects of repetitive self-injury in hospi-
talized adolescents. Journal of the American Academy of Child
and Adolescent Psychiatry, 41, 1333–1341.

Nock, M. K. (2010). Self-injury. Annual Review of Clinical Psychol-

ogy, 6, 339–363. doi 10.1146/annurev.clinpsy.121208.131258

Nock, M. L., & Cha, C. B. (2009). Psychological models of non-

suicidal self-injury. In M. K. Nock (Ed.), Understanding non-
suicidal self-injury: Origins, assessment, and treatment
(pp. 65–77). Washington, DC: American Psychological Asso-
ciation.

Nock, M. K., & Favazza, A. R. (2009). Nonsuicidal self-injury:

Definition and classification. In M. K. Nock (Ed.), Under-
standing nonsuicidal self-injury: Origins, assessment, and
treatment (pp. 9–18). Washington, DC: American Psycholog-
ical Association.

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

tress tolerance, and social problem-solving deficits among ad-
olescent self-injurers. Journal of Consulting and Clinical Psy-
chology, 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 behavior. Journal of Consult-
ing and Clinical Psychology, 72, 885–890. doi 10.1037/0022-
006X.72.5.885

Osuch, E. A., & Payne, G. W. (2009). Neurobiology of nonsuicid-

al self-injury. In M. K. Nixon & N. L. Heath (Eds.), Self-injury

in youth: The essential guide to assessment and intervention
(pp. 79–110). New York: Routledge.

Prinstein, M. J. (2008). Introduction to the special section on suicide

and nonsuicidal self-injury: A review of unique challenges and
important directions for self-injury science. Journal of Consulta-
tion and Clinical Psychology, 76, 1–8. doi 10.1037/0022-
006X.76.1.1

Turner, C. A., Panksepp, J., Bekkedal, M., Borkowski, C., &

Burgdorf, J. (1999). Paradoxical effects of serotonin and
opioids in pemoline-induced self-injurious behavior. Pharma-
cology, Biochemistry & Behavior, 63, 361–366. doi 10.1016/
S0091-3057(99)00018-0

Walsh, B. (2006). Treating self-injury: A practical guide. New

York: Guilford.

Received January 10, 2011
Revision received August 31, 2011
Accepted August 31, 2011
Published online February 15, 2012

About the authors

Christopher Bloom is an Associate Professor of Psychology at
Providence College, Providence, RI, USA. As director of the
Providence Affective Neuroscience (PAN) laboratory, his work
focuses on the development of animal models of psychopatholo-
gies of fear, stress, and anxiety, including nonsuicidal self-injury.

Shareen Holly completed her graduate training at McGill Univer-
sity, Montreal, Canada, in the Department of Educational and
Counseling Psychology and is currently working as a Visiting As-
sistant Professor at Providence College, Providence, RI, USA.
Her research interests include the emotional, social, and function-
al contributors in the development of nonsuicidal self-injury.

Adam M. P. Miller graduated from Providence College, Provi-
dence, RI, USA, with a BA in psychology and is currently a PhD
student in the psychology department at Cornell University, Ith-
aca, NY, USA. He is broadly interested in the neurobiology of
learning and memory with a focus on hippocampal-cortical inter-
actions and memory-system competition throughout the learning
process.

Christopher M. Bloom

Department of Psychology
Providence College
1 Cunningham Square
Providence, RI 02918
USA
Tel. [phone]
Fax [phone]
E-mail [e-mail]

112

C. M. Bloom et al.: Self-Injurious Behavior vs. Nonsuicidal Self-Injury

Crisis 2012; Vol. 33(2):106–112

© 2012 Hogrefe Publishing