Volume 37, pages 503–520 (2011)
Adding Insult to Injury: Effects of Interpersonal Rejection
Types, Rejection Sensitivity, and Self-Regulation on
Obsessive Relational Intrusion
H. Colleen Sinclair1?, Roshni T. Ladny1,2, and Amy E. Lyndon3
1Department of Psychology, Mississippi State University, Mississippi
2Department of Sociology, Florida State University, Florida
3Department of Psychology, East Carolina University, Greenville, North Carolina
: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :
This study tested the I3model [Finkel, 2007; 2008] of intimate partner violence as applied to obsessive relational intrusion (ORI) to
assess the relation among self-regulation, rejection, rejection sensitivity (RS), and stalking-related aggression. In Study 1,
participants (N5221) read one of three vignettes: no relationship termination, an ‘‘internal’’ rejection (involves an internal
attribution to the rejected as cause of relationship ending), or an ‘‘external’’ rejection (external attributions for relationship
demise). Next, participants experienced one of two conditions manipulating self-regulation (no depletion vs. depletion). Finally,
participants rated their likelihood of engaging in ORI (e.g. unwanted pursuit and/or aggression). Consistent with predictions,
participants receiving an internal rejection reported higher aggression than participants experiencing an external rejection,
especially when depleted of self-regulation. Study 2 extended the design of Study 1 by adding in a screening survey of RS. Internal
rejections still yielded more aggression than other conditions, but this was especially so when rejection-sensitive persons were
depleted of self-regulation. In addition to providing support for the I3model of aggression, this research shows that not all types of
rejection are created equal. Aggr. Behav. 37:503–520, 2011.
: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :
rr2011 Wiley Periodicals, Inc.
Keywords: aggression; rejection; stalking; self-regulation; rejection sensitivity
The drive to attain or maintain a romantic
relationship is a fundamental aspect of our need to
belong [Baumeister and Leary, 1995]. When the
satiation of this need is thwarted—through inter-
personal rejection or other obstacles—people can
respond by engaging in acts of obsessive relational
intrusion [ORI; Cupach and Spitzberg, 2004]. ORI is
defined as ‘‘unwanted and repeated pursuit or
invasion of an individual’s sense of privacy by a
stranger or acquaintance who either desires or
presumes an intimate relationship’’ [Cupach and
Spitzberg, 1998, p. 234–235]. ORI behaviors can fall
on a continuum that ranges from unwanted pursuit
behavior (e.g., sending uninvited gifts, calling
frequently) to aggression (e.g., threats, violence).
When ORI becomes threatening to its targets, it is
termed stalking [Cupach and Spitzberg, 1998;
Spitzberg and Rhea,
approximately 3.4 million Americans each year
[Baum et al., 2009], carries significant psychological,
physical, social, and financial costs [Sheridan and
Lyndon, 2011], and has been shown to precede
68–85% of actual and attempted spousal homicides
[McFarlane et al., 1999]. In order to better
understand how ORI, and thus stalking, can form,
we need to examine how contributing and protective
factors interact to affect its occurrence.
The goal of this study is to draw on the I3model of
intimate partner violence (IPV) [Finkel, 2007, 2008;
Finkel and Eckhardt, 2011] to experimentally
examine how rejection as an instigating factor,
self-regulation as an inhibiting force, and rejection
sensitivity (RS) as an impelling factor contribute to
the perpetration of stalking behavior. There is
Published online in Wiley Online Library (wileyonlinelibrary.com).
Received 1 July 2010; Revised 8 August 2011; Accepted 18 August
?Correspondence to: H. Colleen Sinclair, Department of Psychology,
Mississippi State University, P.O. Box 6161, Mississippi State,
MS 39762. E-mail: email@example.com
rr 2011 Wiley Periodicals, Inc.
evidence that self-regulation [DeWall et al., 2007;
Finkel et al., 2009; Stucke and Baumeister, 2006],
rejection [see Blackhart et al., 2006; Buckley et al.,
2004; Gerber and Wheeler, 2009; Leary et al., 2006],
and RS [see Romero-Canyas et al., 2010a for a
review] affect aggression. The present research
examines all three of these factors within the new
I3Model [Finkel, 2007, 2008].
I3Model of IPV Perpetration
The I3model was originally conceived as an
organizing framework for the myriad literature on
predictors of IPV. Although this meta-theory was
originally applied to IPV, we believe that this model
can be extended to understand when ORI, and thus
stalking, is likely to occur. Stalking is a type of
Williams et al., 2006], and has been found to
precede, occur during, and following estrangement
in intimate relationships
Williams and Frieze, 2005].
In the I3model [Finkel, 2007, 2008; Finkel and
Eckhardt, 2011, see Fig. 1], intimate aggression is
deemed more likely if potential perpetrators have
strong instigating triggers, strong impelling factors,
and weak inhibiting factors. An instigating trigger is
an event that normatively provokes aggression in a
typical person. Impelling forces are more distal in
nature and affect the strength of the urge to respond
to an instigator with aggression (what Finkel and
Eckhardt refer to as determining the potential
perpetrator’s ‘‘urge-readiness’’). In contrast, inhibit-
ing forces are dispositional or situational factors
that may help an individual override the urge to
respond aggressively. Thus, people will commit acts
of aggression when the impelling forces and
instigating forces exceed the inhibiting forces.
For example, many people experience the instigating
factor of a relationship termination, but may not
aggress in response. However, if the rejected person is
high in RS, an aggressive response to a rejection is
more likely [Downey et al., 2000, 2004a,b]. An
aggressive response is even more likely when inhibiting
factors are weak, such as depleted self-regulation
resources [Finkel et al., 2009; Romero-Canyas et al.,
2010a]. Although depleted self-regulation resources,
RS, and interpersonal rejection are not the only types
of factors that could interact in the model [Slotter
et al., 2011], they represent the focus of the present
Instigating Factor: Rejection
Simply put, rejection hurts. Interpersonal rejection
activates the same neurological indicators as physical
pain [Eisenberger and Lieberman, 2004]. Rejection is
also threatening. Being rejected is a threat to multiple
basic needs, including a need for positive self-regard
[Baumeister and Leary, 1995], and for control
[Gerber and Wheeler, 2009]. Rejection clearly
qualifies as an instigating factor for aggression. Both
experimental and nonexperimental studies have
linked interpersonal rejection and aggression [see
Gerber and Wheeler, 2009; Leary et al., 2006 for
reviews]. Research also shows that relationship
violence escalates as a rejection looms [Ryan et al.,
1999]. Further, once dissolution occurs, threats,
physical or sexual violence can follow, sometimes
labeled ‘‘separation violence’’ [see Brownridge, 2006].
Rejection as an instigating factor for aggression is
frequently occurs when people try to obtain or reobtain
a relationship [Tjaden and Thoennes, 1998]. When it
comes to motives for stalking, revenge for rejection is a
primary motive of the most dangerous stalkers [Mullen
et al., 1999]. In contrast, the least violent, but still
invasive, ‘‘intimacy seeking’’ stalkers [Mullen et al.,
1999] appear to be those pursuing an unrequited love
[Cupach and Spitzberg, 2004]. In either case, rejection
is a situational variable that can provoke ORI.
Types of rejection.
unwanted pursuit or aggression—is triggered may
depend on the type of rejection. Few studies have
examined whether the type of rejection affects the
impact of rejection [Gerber and Wheeler, 2009;
What type of ORI—
Note: Factors in parentheses represent those of interest to the present research.
Fig. 1. I3model of IPV perpetration adapted to this study of ORI. ORI5obsessive relational intrusion; IPV5intimate partner violence.
504Sinclair et al.
Molden et al., 2009; Romero-Canyas et al., 2010b].
Yet within the stalking literature, victims are advised
that they should not let pursuers ‘‘down easy’’ by
providing indirect, passive, rejections, but rather to
explicitly reject stalkers [e.g., Carll, 1999; Cupach
and Spitzberg, 2004; de Becker, 1997]. Therefore,
another goal of this study was to examine how types
of rejection influence reactions to rejection.
Integrating attribution theory [Heider, 1958], we
operationalized an indirect rejection as one that
provides an external attribution for the rejection.
Examples of an external rejection might include ‘‘I
just don’t have any time to date anyone right now,’’
‘‘My parents don’t approve,’’ or ‘‘It’s not you, it’s
me.’’ In any case, the reasons for rejection are
factors other than the characteristics or behavior of
the rejected. According to some [e.g., Carll, 1999; de
Becker, 1997], this external rejection may only lead
to continued, or even enhanced [Cupach and
Spitzberg, 2004], unwanted pursuit as the rejected
pursuer may maintain hope for a relationship and
thus may persist to wait out (e.g. not enough time
right now) or overcome the situational obstacle (e.g.,
parental disapproval). Also, according to research,
making external attributions for breakups is linked
with greater postbreakup distress as people may see
the demise of a good relationship due to situational
obstacles as more unfair than if the members of the
relationship brought it to an end [Tashiro and
Frazier, 2003]. The greater distress may trigger more
clinging to the relationship via unwanted pursuit.
However, others argue that an explicit direct
rejection is not a definite fix either [Cupach and
Spitzberg, 2004]. Direct rejection has previously been
defined as one that clearly communicates to indivi-
duals that they were rejected because they were not
liked [Molden et al., 2009]. Paralleling our operatio-
nalization of an external rejection, we defined a
direct rejection as one that uses an internal attribu-
tion, such that the rejection is attributed to the
characteristics of the person being rejected. Internal
attributions include reasons such as the physical
appearance, intelligence, behavior, and/or personality
traits of the rejected individual. With an internal
rejection, it is made clear that ‘‘it’s not me, it’s you.’’
Social psychological research has long shown that
individuals tend to avoid making internal attribu-
tions for failures in order to protect their self-esteem
[e.g., the self-serving bias, Mezulis et al., 2004;
attributions to prejudice, Major et al., 2003]. If
unable to make external attributions for negative life
events, self-esteem drops. The desire to lash out may
then rise [Baumeister and Vohs, 2004; Baumeister
et al., 1996; Bushman and Baumeister, 1998; Twenge
and Campbell, 2003], potentially including engaging
in stalking [Cupach and Spitzberg, 2004; Meloy,
1998]. In fact, Ford and Collins  found that
heightened cortisol responses to a rejection from a
potential date only occurred when the rejected
person made self-blaming attributions. Once these
internal attributions occurred, heightened cortisol
responses followed, which led to subsequent partner
derogation. Thus, it may be that only an ‘‘internal’’
rejection leads to aggression.
Inhibiting Factor: Self-Regulation
Whether any ORI is triggered may also depend
resources to inhibit the impulse to engage in unwanted
pursuit and aggression. Self-regulation is defined
broadly as the processes by which the self alters its
responses or inner states in order to reduce perceived
discrepancies between current and desired goal states
processes are ‘‘switched on’’ so that individuals can
respond in a relatively normative way to events that
make one aware that the current situation is discrepant
from what one desires, such as when one perceives an
event as unfair [Baumeister and Vohs, 2007; van de
Bos, 2010]. Self-regulation works like an energy
source, such that it is depleted after use and
replenished after rest [Hagger et al., 2010; Muraven
and Baumeister, 2000]. If self-regulation abilities are
depleted—perhaps due to low blood glucose levels,
insufficient development of self-control and emotion
regulation skills, impairments in prefrontal cortex
that recently or repeatedly tax one’s ability to
self-regulate—people can react very strongly, and
and/or disappointing discrepancies between one’s
present reality and one’s desired goal [Gailliot and
Baumeister, 2007; van de Bos, 2010].
Emerging research has linked depleted self-regulation
with an increased likelihood of aggressive responding to
events that highlight that one’s current state is
incongruous with one’s desired state [e.g., Denson
et al., 2010; DeWall et al., 2007, 2010; Finkel
et al., 2009]. Specifically testing the link between
self-regulation and intimate aggression, Finkel et al.
 randomly assigned participants to self-regulation
depletion or no-depletion conditions. Participants then
received insulting or positive feedback from their
romantic partner about the creativity of a drawing.
Only participants both insulted and depleted chose to
have their romantic partner endure longer, painful,
yoga positions. The insult created a discrepancy
505Self-Regulation and Rejection
between what individuals would like to think of
themselves vs. what they were being told was their
partner’s low opinion. It was the lack of self-regulation
that inhibited participants’ ability to manifest calm
responses to this discrepancy. We sought to extend this
research and study the link between self-regulation
A breakup is an event that creates a discrepancy
between the desired fundamental goal—to have a
relationship—and the current state of the relationship’s
end. To reconcile the discrepancy, one can change one’s
internal state by changing the desired goal (e.g., decide
one did not want that relationship). Alternatively, the
automatic goal-directed response might be to engage in
pursuit behaviors, even if unwanted, to regain the
relationship as is potentially customary [Battaglia
et al., 1998; Langhinrichsen-Rohling et al., 2000].
Accordingly, the ability of individuals to accept the
breakup and walk away from the relational goal may
be limited if they do not have sufficient self-regulation
resources to prevent themselves from engaging in
unwanted pursuit behavior.
An internal attribution rejection may create an
additional hurdle. The internal attribution rejection
also creates a discrepancy between how one would
like to view oneself—positively—and how one is
viewed by another. Research has already documented
that when one is insulted and depleted of self-
regulation, angry retaliation is more likely [DeWall
et al., 2007; Finkel et al., 2009]. However, this ‘‘double
whammy’’ type of rejection may be depleting in and
of itself. Research has documented that grappling
with a rejection is taxing on one’s self-control
[Blackhart et al., 2006]. Thus, one might have
sufficient self-regulation resources to deal with one
discrepancy (relational goal or positive self-regard
goal), but not necessarily the other one. It is as if the
individual has to triage the injuries to the self,
determining where to allocate limited self-regulation
resources to yield the best outcomes. One may use the
resources to forgive the partner for their harsh words,
but, still faced with a loss of connection, may pursue
of the lost relationship. Reobtaining belonging would
then bolster self-esteem [Leary and Downs, 1995].
Without self-regulation resources, however, forgiveness
is unlikely and aggressive retaliation is likely.
Impelling Factor: Rejection Sensitivity
In particular, we believed that rejection and self-
regulation would have important consequences for
those who were prone to RS [Downey and Feldman,
1996]. RS is an individual difference variable
whereby individuals have a chronic tendency to
anticipate rejection in interpersonal situations and
are thus hypervigilant for cues of pending rejection
and over-reactive toward such indicators. This
defensive motivational system is directed toward
maintaining a relationship with the perceived
rejecter. Nonetheless, their over-reactions can often
be hostile [Romero-Canyas et al., 2010a] and
aggressive [Downey et al., 2000], including stalking
behavior [Kamphuis et al., 2004].
Given their anxiety about rejection, rejection-
sensitive persons may not show any differences based
on rejection type in their tendency to over-react to
rejection. Even negative relationship events that do
not result in an explicitly communicated relationship
termination could be seen as signs of a termination to
come. However, part of the reason rejection-sensitive
people may react so strongly to rejection is
thatthey take it so
attributions—assuming the worst of both themselves
and their partner—are often the automatic inference
(e.g., hostile attribution biases). On the other hand,
situational explanations for the negative relationship
event require a more controlled assessment of the
event. Thus, rejections that offer such a situational
explanation dilute the impact of the rejection. In fact,
when Romero-Canyas et al. [2010b] studied determi-
nants of rejection-sensitive people’s responses to
rejection, they specifically avoided giving participants
the opportunity to infer external attributions for why
they were not accepted because they felt it would
dampen the effect of the rejection. Further, Ayduk
et al.  argued that having rejection-sensitive
people activate the ‘‘other-perspective’’ that allows
them ‘‘to attend to situational information that
may provide alternative explanations for another’s
seemingly destructive behavior’’ (p. 778) could
inhibit an automatic retaliatory responses.
Ayduk et al.  used this rationale to build
support for their assertion that the negative con-
sequences of being rejection sensitive could be
mitigated by self-regulation. Rejection-sensitive per-
sons with strong self-regulation abilities can inhibit
affording them the opportunity to reflect on
the triggering incident and consider alternative
arguments with longitudinal research, Ayduk et al.
found that rejection-sensitive people did not exhibit
the interpersonal difficulties (e.g., aggression, low
self-worth) typically exhibited by persons high in RS
when they were also high in self-regulation. Similar
findings that self-regulation capabilities serve as a
buffer for the effects of RS were also found in
subsequent experiments [Ayduk et al., 2002].
506 Sinclair et al.
We conducted two studies to explore how self-
regulation, RS, and rejection may interact to yield
ORI. In the first study, we wanted to examine how
rejection and self-regulation might interact to lead to
either pursuit or aggressive behaviors. This first
study thus tested two aspects of the I3model and was
intended to examine whether types of rejection vary
in the types of ORI they trigger. We manipulated
self-regulation resources (e.g., depleting or not
depleting resources through a thought suppression
exercise) and then analyzed how the individual
responded to a hypothetical intimate relationship
termination. Reasons for the rejections were either
internal (internal attributions provided as reason for
termination) or external (involving external attribu-
tions for the relationship’s demise). Responses were
recorded using the ORI scale [Cupach and Spitzberg,
1998]. In Study 2, the same manipulations of
rejection type and self-regulation were employed,
but we focused on the aggressive subscale of the ORI
and used Rejection Sensitivity Questionnaire data
froma mass screening
Feldman, 1996] to examine the full I3model.
Across both studies, we expected that higher rates of
both pursuit and aggressive ORI would be present in
relationship termination conditions, especially the
internal rejection condition, than in the no-termination
condition. In particular, higher rates of ORI in the
internal rejection condition are anticipated to largely
be due to differences in rates of aggressive ORI.
Further, we expected this rejection–aggression link to
be exacerbated under certain conditions. Namely, in
both studies, we expected self-regulation depletion
among the rejected to lead to more ORI than no
depletion, particularly when it came to assessing
aggressive ORI among those internally rejected. This
depletion of self-regulation resources is especially
important for those high in RS. Thus, in Study 2,
we expected those internally rejected, depleted of
self-regulation, and high in RS to exhibit the highest
rates of aggressive ORI.
As noted, the goal of Study 1 was to examine
whether different types of rejection led to different
types of ORI, particularly when one was depleted of
self-regulation. Participants were presented with one
of three different scenarios. Two featured a breakup,
but used either internal or external attributions to
explain the rejection. The third featured a relationship
disappointment but no termination. Next, we
chose a thought suppression manipulation of self-
regulation resources. Thought suppression tasks are
[Schmeichel and Baumeister, 2004]. To attempt to
mirror what might be real-world attempts to cope
with a breakup by attempting to avoid thinking of
the event, the thought suppression exercise was one
in which participants did just that: try not to think
of the scenario breakup. We believed that those
receiving an internal rejection who were depleted of
self-regulation resources would exhibit the most
aggressive behavior. In contrast, those receiving an
external rejection who were not depleted should be
the most likely to have the restraint to walk away.
The sample consisted of 221 (117 women, 104
men) undergraduate students at a public university
in the Southeast, of which 66.1% were Caucasian,
27.1% were African American, and 6.8% other.
Participants’ ages ranged from 17 to 24 (SD51.18)
with a mean age of 18.48 years. This final sample of
221 emerged after 14 participants were discarded
due to numerous unanswered questions.
The study design was a 3 (rejection type: internal
vs. externalvs. control/no
termination)?2 (self-regulation: depleted vs. not
depleted)?2 (ORI type: aggressive vs. pursuit -
within-groups variable) mixed factorial design.
The study was titled ‘‘Memory for Interpersonal
Events.’’ Sessions were conducted in groups of
10–15 participants. Each group was randomly
assigned to either the self-regulation depletion
condition or no-depletion control condition. After
an informed consent was obtained, all participants
read a randomly assigned vignette that manipulated
a rejection in an hypothetical intimate relationship.
recall details later on. After reading, participants
exchanged the scenario for processing questions.
Next, both groups completed a writing assignment
that served as the self-regulation manipulation.
Then participants indicated which of the behaviors
on the ORI inventory they would use to respond to
the scenario events. Finally, participants answered
507Self-Regulation and Rejection
manipulation check questions, demographic questions,
were debriefed, and given credit.
of three vignettes (703 words). Each scenario
described an individual in a 3-year long relationship
who is ready to advance the relationship. She/he has
planned a surprise romantic evening with concert
tickets. In the control condition, the romantic
partner expresses a desire to go to the concert, but
declines because of an important school assignment.
Thus, there is a negative relationship event, but no
termination. In the two experimental conditions, the
participant is rejected when the romantic partner
declines the invitation because she/he ends the
relationship. The rejecter breaks up with the
individual in one of two ways: internal rejection or
external rejection (complete scenarios are available
from the authors).
Internal rejection was manipulated by having the
rejecter make internal attributions for the breakup
following the presentation of the concert tickets:
Participants received one
‘‘Well, uhyI don’t think that’s a great idea
really. We have not been getting along
latelyy.You’ve changed a lot this past year.
I guess, to be honest, I really am not interested
in you anymore. There is nothing I find
appealing about you. Not physically, or well,
emotionally either. So, basically we don’t need
to be together. I’ve meant to tell you earlier,
but I just never got around to it. As far as the
concert goes, I am going with other people.
I think it is best that we not see each other
External rejection condition was manipulated by
having the rejecter makes external attributions for
the breakup, citing reasons outside the rejected
‘‘Well, uhyI don’t think that’s a great idea
really. You see, I have been so busy lately.
I never have time to see my friends or family. I
am always stressed out over work and school. I
need to focus more on those aspects of my life.
School and work do not allow me to have time
for a relationship. I meant to tell you earlier,
but just never got around to it. As far as the
concert goes, I don’t have time to go. It is best
for me that we not see each other anymore.’’
In the control condition, the partner expresses desire
to go to the concert, but declines the offer due to a
school assignment. Although there is an external
reason for the negative relationship event (e.g., lost
money, wasted plans for a concert), the control
condition does not express termination of the
relationship. To the contrary, the partner apologizes
and expresses affection.
‘‘Well, uhyI really want to go with you and
I’m sorry I’m in such a bad mood, but my
chemistry lab professor just dumped another
assignment on us. He wants it in by tomorrow!
I have not had time to work on it yet because
of other class work and my lab partner has
been totally unreliable. I would much rather go
to the concert with you obviously, but I cannot
afford to blow this assignment off. And
now I don’t have time to go because I have
to go track down my lab partner. Sorry.
Processing and memory questions.
ensure that participants thought about the scenario,
they were asked three open-ended questions about
the scenario (e.g., ‘‘What are five feelings/thoughts
that the scenario evokes in you?’’). Participants were
also asked the following: ‘‘Based on what you have
just read, please write a summary of the events that
happened to you and you partner. Describe what
your partner did and how you reacted. Please put all
details in first person terms such as: ‘I felty..’ and
‘My partner didy..’.’’ This recall section was
followed by four factual recall questions about the
scenario events. A sample multiple choice question
was ‘‘What were you hoping would take place at the
end of the concert?’’, with options of ‘‘(A) you would
get to meet his/her family, (B) you could find a way to
break up with him/her, (C) discussion about making a
final commitment to each other, (D) discussion about
future vacation plans.
In order to
manipulations were achieved via writing exercises
based on thought suppression techniques [e.g.,
Wegner and Pennebaker, 1993]. After writing down
some initial reactions, participants in the experi-
mental group were asked to do a ‘‘Clear Your
Head’’ writing exercise. They were instructed to
record all their thoughts for five minutes, but in
order to clear their thoughts they had to not think
about anything related to the scenario that they just
read. Self-regulation would be depleted by requiring
participants to censor their thoughts and what they
508Sinclair et al.
wrote as it requires self-control to suppress one’s
thoughts (Wegner and Pennebaker).
The control group completed a ‘‘Free Writing’’
exercise where they were instructed to write down
any thoughts that came to their mind for five
minutes. They did not have to exert any self-control
to keep from writing anything.
Dependent variable materials
ORI and aggression.
The 42-item ORI [Spitzberg
and Cupach, 2007] inventory was adapted to assess the
respondent reports that they would commit ORI in
reaction to the hypothetical scenarios. The items range
from relatively low severity (e.g., leave unwanted gifts,
leave unfriendly messages, spam, get into a public
argument) to relatively high severity (e.g., threats,
physically endangering behavior). There are eight levels
of behavior that can be categorized as either pursuit
tactics or aggressive tactics (Table I).
Participants were instructed to first rate the like-
lihood of whether they would think about each
behavior and then to rate the likelihood of whether
they would engage in the behavior. For each item,
participants responded using a Likert scale ranging
from 0 (not likely at all) to 5 (very likely) to consider/
commit the act. We included both ‘‘think’’ and
‘‘engage’’ formats as we thought participants might
feel more comfortable reporting that they would
consider a particular behavior than engage in it. In
the end, however, analyses revealed that results for the
think and engage formats showed the same pattern.
Thus, the scales were combined yielding two DVs: ORI
pursuit (including think and engage, 32 items, a5.88)
and ORI aggressive (including think and engage, 52
items, a5.95). Given the hypotheses called for
comparisons based on ORI type, pursuit vs. aggressive
ORI was treated as a within-subjects variable.
were asked questions to ensure that they felt the
relationship was over. The questions were: ‘‘How
likely or unlikely do you think it is that this
relationship will survive?’’ (05very unlikely to
55very likely), ‘‘How strong or weak do you think
this relationship is?’’ (05very weak to 55very
strong), and ‘‘Do you feel this relationship is over or
continuing?’’ (05definitely over to 55definitely
continuing), and Thus, lower scores indicated that
participants felt that the relationship was less likely
to continue. Reliability was a5.89.
We also wanted to verify that they not only
recognized that the relationship was terminated but
also felt some negative affect, particularly rejection,
in response to the situation and were asked to rate
how hurt, angry, and rejected they felt (also on a 0–5
point scale, with 05very rejected/hurt/angry and
55very accepted/unhurt/happy). Thus, lower scores
indicated a greater degree of negative affect.
Reliability was a5.78.
Further, we wanted to ensure that participants
correctly identified the reason for the relationship
termination as either due to external or due to internal
variables. Thus, participants were asked: What was the
primary reason for the relationship breakup, if there
wasany? Options included:
(B) Primarily due to factors associated with me,
(C) Primarily due to factors associated with my
partner, and (D) Primarily due to external factors (e.g.
school, work, situation, third parties). If internal rejection
participants chose ‘‘B’’ they were given a ‘‘1’’ on the
internal attribution, ‘‘A’’ responses were eliminated,
and all other responses were scored as ‘‘0.’’ If external
rejection participants chose ‘‘C’’ or ‘‘D’’ they were given
a ‘‘1’’ for making an external attribution, ‘‘B’’ was
coded as a ‘‘0’’ and ‘‘A’’ responses were excluded.
Self-regulation manipulation checks.
pants in the self-regulation depleted condition
should write fewer details about the scenario than
instructed to censor their thoughts about it. Coders
reviewed the responses and coded for how often any
details relevant to the scenario were mentioned
(A) No breakup,
TABLE I. Obsessive Relational Intrusions Categorized as
Pursuit or Aggressive Tactics
Hyper-intimacyA pattern of inappropriate and excessive face-
to-face expressions of desire to enhance
relationship (six items)
Frequent contact attempts through different
forms of communication, such as e-mail,
mail, phone (five items)
Frequent communication attempts through
interaction that is directly face-to-face or
through third-parties (five items)
SurveillanceMonitoring victim and obtaining information
on victim through discreet methods (five
Information and property theft, trespassing, or
breaking and entering (four items)
Attempt to introduce challenges into victim’s
life intended to decrease the quality of life for
victim or to imply that challenges would
dissipate if the victim complies (four items)
Implicit or explicit messages of any form (letter,
telephone, computer, verbal or nonverbal,
etc.) that imply or threaten harm will occur
unless victim complies (six items)
Actions intended to harm victim or other(s)
contextually relevant to perpetrator’s
relationship with victim (seven items)
509 Self-Regulation and Rejection
(Cohen’s k5.72). The total number of words
participants wrote was counted as well. In theory,
participants in the experimental condition would
write more than the control condition to try to avoid
thinking about aspects related to the scenario. There
was 100% agreement between coders on the number
recognized that the relationship was terminated, an
ANOVA was run with rejection type as the
independent variable and the relationship termination
manipulation check scale as the dependent variable.
Participants were significantly less likely to indicate
that the relationship was continuing in the rejection
conditions than the control condition (M53.14,
SD51.27), F(2, 214)548.65, Po.001, Z25.32. Post
hoc least squares difference tests (LSD) showed that
participants who were externally rejected (M51.75,
SD51.19) were more likely to think that the
relationship had a chance of persisting, albeit remote,
than the internally rejected (M51.29, SD51.07).
Both means still fell on the end of the continuum
indicating the relationship was over.
To test whether participants indeed felt rejected,
another ANOVA was run with rejection type as the
manipulation check scale as the dependent variable
(lower scores indicated more negative affect).
(M50.88, SD51.42) conditions reported feeling
significantly more rejected than those in the
SD51.42), F(2, 214)517.68, Po.001, Z25.14.
LSD tests showed that the internal and external
rejection means were not significantly different from
each other. Thus, both types of relationship
termination made the participant feel rejected.
Finally, we checked to make sure that participants
correctly identified the reason for the termination as
internal or external. Two w2analyses were con-
ducted comparing internal vs. external rejection
conditions on the two attribution manipulation
checks. Participants in the internal rejection condition
were significantly more likely to correctly identify
the relationship termination as ‘‘due to me’’
(w2575.69, Po.001). Likewise, those in the external
rejection condition were more likely to identify the
To determine whether participants
relationship termination as ‘‘due to my partner’’ or
‘‘due to external factors’’ (w2575.69, Po.001).
comparing self-regulation conditions (depleted vs.
not) on total number of words written and number
of references to the scenario. Based on these analyses,
participants in self-regulation depleted conditions
wrote significantly fewer details related to the scenario
(M50.41, SD50.75) than participants in nondepleted
condition (M50.80, SD51.29), F(1, 220)57.68,
SD555.39) than those in the nondepleted condition
(M5113.39, SD541.75), F(1, 220)510.79, Po.001,
Z25.05. Thus, it appears evident, consistent with
Wegner’s research, that the participants in the thought
suppression condition were actively engaged in trying
to avoid thoughts of the scenario by distracting
themselves with other thoughts.
We anticipated that rejection in the form of
relationship termination would lead to ORI. More
aggressive forms of ORI were expected when the
individual experienced a rejection that made an
internal attribution to him/her for the relationship
demise, especially when the internally rejected
individual was depleted of self-regulation. In order
to see how rejection and self-regulation affected the
types of ORI, we employed two 3?2?2 repeated
measures ANOVA with rejection type and self-
regulation conditions as between-subjects factors
and ORI type (aggressive vs. pursuit) as the within-
subjects variable. Skewness and kurtosis statistics
were within acceptable ranges for both aggressive
and pursuit ORI. LSD tests were then used to
determine where any differences existed between
The analysis yielded significant main effects of
ORI type, F (1,212)5995.18, Po.001, Z25.82, and
rejection, F (2,212)55.06, P5.007, Z25.05, as well
as a significant two-way interaction between rejec-
tion and ORI type, F (2,212)53.176, P5.044,
Z25.03. In sum, the effects were as follows:
(1) participants reported higher levels of pursuit
(M52.21, SD50.88) than aggressive behavior
(M50.70, SD50.59), (2) participants reported
higher rates of ORI in the internal rejection
condition (M51.59, SD50.94) than either of the
1An ANOVA including gender as an independent variable was run,
but no gender differences were expected or found.
510Sinclair et al.
other two conditions (external: M51.25, SD51.03;
control: M51.31, SD51.04), and (3) those intern-
ally rejected were especially more likely to report
engaging in aggressive ORI than those externally
rejected or who were not rejected (Fig. 2).
These effects were qualified, however, by the
anticipated three-way interaction of rejection type,
ORI type, and self-regulation, F (2,212)56.02,
P5.003, Z25.05. Means are presented in Figure 2.
Differences were more pronounced with regard
Consistent with our hypothesis, participants reported
a significantly higher likelihood of engaging in
aggressive ORI when internally rejected and depleted
of self-regulation. Those externally rejected and not
depleted of self-regulation showed the lowest rate of
engaging in ORI (pursuit and aggressive). With these
comparisons, it is important to keep in mind that
pursuit behaviors in the no-termination condition
are not unwanted. When one has terminated a
relationship and requested no contact, however, even
‘‘low-level’’ ORI can be intrusive.
Study 1 revealed that an internal rejection did lead
to more ORI, especially aggressive ORI, than an
external rejection or no termination, particularly
when the rejected were depleted of self-regulation. In
contrast, those externally rejected but maintaining
self-regulation reported the lowest amount of
unwanted pursuit behavior (despite reporting feeling
equally hurt as those internally rejected). In all other
instances, those externally rejected did not differ
from those experiencing no termination. In fact, in
no instance was the frequency of ORI among those
externally rejected higher than the no-termination
situation. Of course, it is important to note that ORI
in the latter context is not unwanted unless
aggressive. Thus, the same behaviors carry different
meanings in the different contexts. Nonetheless,
there was no evidence to support the assertion that
letting someone down easy led to more pursuit or
Study 1 also lent preliminary support to the I3
model with regard to two of the ‘‘I’s.’’ Results
showed that although rejection is directly linked to
aggressive responding, this link is exacerbated when
one is depleted of self-regulation. Further, when it
came to pursuit behavior, impulses to engage in an
unwanted pursuit can potentially be inhibited by
available self-control, at least when the rejection was
not so harsh as to potentially expend self-control.
Ultimately, both instigating and inhibiting forces
mattered. However, we also wanted to explore
whether impelling factors could also affect aggres-
sive reactions to rejection when self-control was
lacking. For this, we turn to Study 2.
The goal of study 2 was to test the full I3model.
We largely replicated the design of Study 1 but
included individuals who had completed an abbre-
viated form of the RS scale in a mass screening
survey. Given the relative lack of differences
between groups on pursuit behavior and our interest
in potentially predicting stalking, we focused on the
aggressive subscale of the ORI. Consistent with
Study 1 findings, we expected to find that those
exposed to an internal rejection exhibited the most
aggressive behavior. We expected this reaction to be
depleted and the individual was high in RS. However,
consistent with previous literature [see Mischel and
Ayduk, 2011, for review], the aggressive responses
completely determined by the availability of self-
rejection-sensitive persons would show the lowest
aggressive responses when self-regulation was not
Participants (N5238) completed both the screen-
ing survey and the vignette study. Nine individuals
Fig. 2. Self-reported likelihood of pursuit and aggressive Obsessive
Relational Intrusion (ORI) across rejection and self-regulation
511 Self-Regulation and Rejection
had to be deleted for missing large portions of data
or failure to accurately recall pertinent details of the
scenario (e.g., claimed there was a breakup when
there was not or vice versa). The final sample of 229
was 61.1% female and 38.9% male. Ages ranged
from 17–25 with an average of 18.72 years
(67.5%); 25.0% were African-American, and 7.5%
Materials and Procedure
in the Psychology Research Participation pool
completed a mass screening survey online. Among
the screening measures was the Rejection Sensitivity
Scale–Short Form [Downey and Feldman, 1996;
Each of the eight items featured an interpersonal
situation wherein someone would risk rejection (e.g.
‘‘You call your boyfriend/girlfriend after a bitter
argument and tell him/her you want to see him/
her’’). Participants responded by indicating first how
anxious they would feel about the situation (15very
unconcerned and 65very concerned). Next, partici-
pants indicated thelikelihood
(15very unlikely, 65very likely). Responses to the
likelihood subscale are reverse scored. The means of
the anxiety and likelihood subscales are multiplied.
High scores indicate higher RS. In the present
sample, the reliability for the anxiety subscale was
a5.87, a5.88 for the likelihood subscale, and
The design and materials of Study 2
were identical to that of Study 1 with the exception
that a continuous predictor variable (RS) was
introduced and only the aggressive ORI subscales
were used as the dependent variable. Thus, the vignette
study packets utilized a 3 (rejection type: internal,
external, control)?2 (self-regulation: depleted vs.
Participants were randomly assigned to one of these
six conditions (see Study 1 for vignette details). After
reading the scenario, participants completed scales
identical to Study 1:
More than 1,300 students
*Processing and memory questions to verify that
they understood the facts of the scenario (that a
breakup had occurred or not, the stated reason
for the termination, etc.).
*The self-regulation manipulation (e.g., thought
suppression writing exercise vs. free writing
exercise for 5min).
*ORI aggression subscales (e.g., surveillance,
intimidation, harassment, threats, violence) using
the same think/engage format (05not likely at all
to 55very likely to consider/do). Reliability for
the ORI aggressive tactics scale was a5.95 in
*Rejection manipulation checks which included
the same Study 1 items assessing negative affect
(e.g., whether the participant felt rejected or
accepted, a5.72) and understanding that the
relationship was over (a5.88). Higher scores on
these manipulation checks (range 0–5) meant
more positive affect and belief that the relationship
was continuing. An additional manipulation
check verifying that participants could identify
the type of attribution made by the hypothetical
partner was also included.
*Self-regulation manipulation checks were likewise
the same and included checks for mentioning the
scenario and counting the number of words used
in the writing exercise. Note, there was no
discrepancy in counting of words, and the
Cohen’s k for the coding of whether the
participants wrote scenario-relevant details was
rejection conditions made participants feel rejected
and that participants recognized the relationship as
over. We ran two ANOVAs including rejection type
as the independent variable and either negative
affect or ‘‘relationship over’’ items as the DVs. LSD
were used for post hoc follow-ups. As in Study 1,
reported significantly more negative affect than
(M51.35, SD50.90), F(2, 225)514.24, Po.001,
Z25.11. There were no significant differences
between internal (M50.65, SD50.80) and external
rejections (M50.80, SD50.79) on negative affect.
Likewise, participants were significantly more
likely to recognize that the relationship was continu-
ing in the control condition (M53.22, SD51.21)
than in either the external or internal rejection
conditions, F(2, 225)546.44, Po.001, Z25.29.
Again, those in the external rejection condition
(M51.86, SD51.28) were slightly more likely to
say the relationship could continue than those in the
internal rejection condition (M51.33, SD51.13),
As in Study 1, we verified that the
512Sinclair et al.
but the mean for the external condition was still in
the ‘‘terminated’’ range of the responses.
Finally, we again used two w2analyses to verify
that participants correctly identified attributions in
the internal rejection and external rejection condi-
tions, respectively. Participants in the internal
rejection condition were significantly more likely to
correctly identify the relationship termination as
‘‘due to me’’ (w2571.19, Po.001). Likewise, those
in the external rejection condition were more likely
to identify the relationship termination as ‘‘due to
my partner’’or ‘‘due
Two ANOVAs were run com-
paring self-regulation conditions (depleted vs. not)
on total number of words written and number of
references to the scenario. Based on these analyses,
participants in depleted conditions wrote signifi-
SD51.10) than participants
13.81, Po.001, Z25.08. This occurred despite the
participants in self-regulation-depleted condition
SD551.41) than those in the nondepleted condition
to external factors’’
Preliminary analyses (e.g., an ANOVA testing the
effects of rejection type and self-regulation on
aggression and a second ANOVA adding RS to
the 3?2 design by using a RS median split to create
high and low categories) revealed no significant
differences on aggressive ORI between external
rejection and no termination control across any of
the conditions. These results are consistent with the
findings of Study 1 for aggressive ORI. Thus,
rejection type was dummy coded as 15internal
rejection and 05external/control.2Consistent with
the recommendations of Aiken and West (1991) for
testing interactions between categorical and contin-
uous variables, scores on the RS scale were centered,
and hierarchical regression was used. Main effects
entered in the first step. Two-way interactions were
entered in the second step. The three-way interaction
was entered last. Tests of multicollinearity yielded
scores within acceptable ranges.
Results of the regression are presented in Table II.
Main effects of rejection type and self-regulation
emerged, as did interactions of rejection type and
self-regulation. Although there was no main effect of
RS and the main effect of self-regulation was not
predicted, the results were largely consistent with
hypotheses and mirror those of Study 1. Of primary
interest in Study 2 was significant three-way inter-
action of rejection type, self-regulation, and RS. We
expected that the highest amount of aggressive ORI
should be reported by rejection-sensitive individuals
with depleted self-regulation resources and exposed
to an internal rejection.
To explore the nature of the interaction, we used
procedures outlined by Aiken and West  and
Dawson and Richter [2006; http://www.jeremydaw
son.co.uk/slopes.htm] by taking RS scores at 1/? 1
TABLE II. Summary of Hierarchical Regression Analysis for Variables Predicting Application of the Aggressive ORI (N5252)
Step 1 Step 2 Step 3
Rejection sensitivity (RS)
1Po.08;?Po.02;??Po.005;???Po.001. Note. R25.10 for Step 1; DR25.04 for Step 2; [F(3, 202)53.05, P5.030, R25.14]; DR25.03 for
Step 3 [F(1, 202)55.69, P5.018, R25.17]. IR5internal rejection (1), external rejection/control (0); SR5self-regulation, Depleted (0), Not
Depleted (1); RS5rejection sensitivity, centered.
2The regression and ANOVAs were repeated including gender, and
no main effects or interactions with gender appeared. Additional
analyses were also run including the self-regulation manipulation
checks as covariates in an ANCOVA and as controlled variables
entered in the first step of the hierarchical regression. None of the
manipulation checks proved to be significant covariates nor were
they significant predictors in the regression. The pattern of results did
not change with their inclusion.
513Self-Regulation and Rejection
SD. As can be seen in Figure 3, consistent with
hypotheses, those high in RS and depleted of self-
regulation showed the highest amount of aggression
when internally rejected. Slope difference tests
[Dawson and Richter, 2006] revealed that the self-
regulation depleted and high RS condition stands
out as significantly different from all other condi-
tions (vs. high RS/nondepleted: t5?4.55. Po.001;
low RS/nondepleted t5?3.36. P5.001; vs. low
RS/depleted t53.37. P5.001). However, rejection-
sensitive persons who were not depleted of self-
regulation resources showed the lowest amount of
aggressive behavior even when internally rejected.
Slope difference tests showed those high in RS but
not depleted of self-regulation were no different
from those low in RS and not depleted and only
marginally different from those low in RS and
depleted (t5?1.78. P5.077) For those low in RS,
there did not appear to be a significant interaction
that rejection type matters for those low in RS such
that individuals generally become more aggressive
when subjected to an internal attribution in a
The results of Study 2 closely resemble those of
Study 1. The primary hypothesis was supported.
Individuals who were highly sensitive to rejection,
who were internally rejected, and who had depleted
amount of aggressive ORI. These findings are
consistent with the I3model and emerging research
on RS and self-regulation. Although the combina-
tion of all the three components of the model
predicted aggressive ORI, instigating and impelling
factors alone did the same. Participants expressed
more aggressive ORI with an internal rejection than
either an external or no rejection. Further, rejection
type and self-regulation interacted just as they did in
Study 1. However, as expected, the effects of RS
were contingent upon self-regulation resources such
This study extended the I3Model of IPV to
stalking-related behavior. Specifically, the impelling
factor of RS, the instigating trigger of rejection, and
the inhibitory factor of self-regulation all affected
whether participants reported potentially engaging
in ORI. We found that individuals rejected with an
internal attribution reported that they would con-
sider engaging in ORI (Study 1 and 2). However, we
also found that those who were internally rejected,
depleted of self-regulation (Study 1 and 2), and
rejection sensitive (Study 2) were more likely than
any other group to endorse aggressive responses.
In contrast, those experiencing an external rejection
rarely differed from the no termination condition
except once in Study 1 when they were not depleted
of self-regulation, and then they showed less ORI
than any other group.
Before moving to interpretations, it is important
to address caveats. For example, the sample
was limited to undergraduate students. However,
among young adults (ages 17–22), and higher
rates of ORI are often found in college samples
(e.g., 27%) than in national samples [Miller and
Ravensberg, 2003; Spitzberg and Cupach, 2007;
important to see whether our findings extend
beyond the present sample, college students are an
important population to study.
Another potential limitation of this study is that
the data were self-reported responses to a hypothe-
tical rejection. Participants might not truly feel
rejected and might respond in a socially desirable
manner such that they under-report the likelihood
1998].Although it is
Fig. 3. Three-way interaction of rejection sensitivity, self-regulation,
and rejection type on ORI aggressive tactics. ORI, obsessive relational
514 Sinclair et al.
that they would engage in stalking-related aggres-
sion. It is noteworthy that significant numbers of
participants did endorse ORI behaviors despite
However, anticipated behavior does not necessarily
match actual behavior. Nonetheless, researchers
have used anticipated behavior in hypothetical
situations to study topics that are not easily
reproducible in the laboratory, such as the effect
of alcohol on anticipated condom use [MacDonald
et al., 1996]. If we could approximate behavioral
measures of stalking in the lab, we could find even
larger effects than found employing self-report
measures. Still, we do not assume that our one time
breakup manipulation captures all the intricacies of
the demise of an invested intimate relationship, but
recreating such a dynamic in the lab is inherently
challenging. Further, certain stalking behaviors can
be nonviolent, and although we could ascertain that
they were unwanted, unwanted behaviors may not
be threatening enough to create the required level of
emotional distress in the target to qualify as
‘‘stalking.’’ Also, ORI is a ‘‘repeated course of
conduct’’ yet most studies of aggressive behavior
employ a single behavioral dependent variable.
examination of stalking perpetration difficult. Other
options of surveying past or current behaviors and
correlating with individual difference measures
would eliminate the strength of the experimental
nature of this study. Thus, self-report instrumenta-
tion is presently a limitation affecting all work on
stalking perpetration. Should these challenges be
overcome in future research it would be a welcome
Finally, we chose to have participants deplete their
self-regulation resources by suppressing information
about the scenario. Thought suppression has been
successfully used as a means of self-regulation
depletion in the past [Muraven et al., 1998].
However, with the present manipulation, it is
difficult to discern whether self-regulation depletion
alone was a sufficient disinhibitor. Thought suppres-
sion can lead to ironic processing (making the
forbidden thoughts more accessible) or even angry
rumination (repeatedly reliving the negative event)
[Denson, 2009; Wegner et al., 1987; Wegner and
Zanakos, 1994]. Ultimately, self-regulation plays a
role (as both ironic processing and rumination
deplete self-regulation), but efforts should be made
in the future to test the independent effects of
rumination, ironic processing, and self-regulation.
Keeping these limitations in mind, we move on to
discuss potential implications.
Interpretations and Implications
Not all rejection is created equal.
with previous research that showed a causal link
between rejection and aggression [Blackhart et al.,
2006; Gerber and Wheeler, 2009; Leary et al., 2006],
we found that rejection led to stalking-related
behavior. We also found that type of rejection
mattered. Experts have advised stalking victims to
be explicit and avoid letting perpetrators ‘‘down
easy’’ [Carll, 1999; de Becker, 1997]. Yet we found
that an explicit rejection led to an expressed
willingness to be more intrusive than an indirect
rejection. ‘‘Nice’’ rejections actually led to the lowest
rates of ORI when individuals had self-regulation
Note that our operationalization of rejection is
not the only potential differentiation between types
of rejection. For instance, there is a growing body of
research examining the differences between ostra-
cism, nonverbal exclusion, verbal rejection, and
receiving information that one is likely to live a
lonely life in the future [see Gerber and Wheeler,
2009]. Further, although we manipulated the one
external), there are other dimensions that may
influence potential responses to a romantic breakup,
including globality, stability, intentionality, change-
ability, and controllability [Anderson, 1983; Weiner,
1985]. There is clearly a rich source of future
rejection research integrating attribution theory.
In particular, it would be worthwhile to pursue
whether threats to control [Gerber and Wheeler,
2009], self-regard [Baumeister and Vohs, 2004;
Baumeister et al., 1996; Bushman and Baumeister,
1998; Twenge and Campbell, 2003], or both mediate
the impact of rejection. We did not directly test
whether self-regard threats were the key mediator
between internal rejections and aggression, but have
reason to believe that internal rejections do threaten
self-regard. In general, internal attributions for
negative events are threatening to self-esteem [see
Campbell and Sedikides, 1999]. Also, in general,
interpersonal rejection lowers self-esteem [Leary and
Downs, 1995]. Further, in our pilot tests of internal
vs. external rejections, we found internal attribution
rejections significantly lowered self-esteem.3However,
3Specifically, items included were: ‘‘I would feel like a failure,’’ ‘‘I
would feel I had good qualities (reversed),’’ ‘‘I would not feel
confident,’’ ‘‘I would feel bad about myself’’ (responded to on a
7-point Likert scale where 15strongly disagree, 75strongly agree)
and had a reliability of a5.82. Internal M54.52, SD51.45,
External M53.24, SD51.42, ANOVA F(1, 76)514.73, Po.001,
515 Self-Regulation and Rejection
internal attributions rejections did not result in
lower scores on indices of control.4When differences
did appear on items to assess control, those receiving
an ‘‘internal’’ rejection had higher scores, perhaps
feeling they had more control over themselves than
situational problems. Thus, we have reason to
believe that rejections based on internal attributions
contradict the rejected persons’ positive self-appraisal
and can lead to aggressive behavior as a means to
restore their sense of self.
In contrast, external rejections may have allowed
individuals to protect their self-esteem. Because their
sense of self was less threatened by critical remarks,
their self-image may not have been as damaged
when rejected with an external attribution instead of
an internal attribution.
receiving an external rejection may not have felt
the need to retaliate to regain positive self-regard.
Instead, they may have been better equipped to
forgive. Forgiveness is often operationalized as a
lack of a desire for retaliation when wronged.
Emerging research on self-regulation and forgive-
ness shows that self-regulation is required to forgive
transgressions [Molden and Finkel, 2010; Pronk
et al., 2010]. This may be why the externally rejected
reported the lowest amounts of unwanted beha-
viors—pursuit or aggression—when they were not
depleted of self-regulation. They were able to walk
away. It could be that making external attributions
for rejections—whether by the rejecter or rejected—
is actually an inhibitor.
Additional explanations for the differences be-
tween the types of rejections should also be
explored. Evidence suggests that nicer rejections
are common [Folkes, 1982]. People may expect that
they will be ‘‘let down easy.’’ When rejecters violate
this script, the rejected could feel more entitled to
violate other social norms by acting aggressively.
The role of self-regulation.
breakup one has the option to walk away, pursue, or
retaliate. Self-regulation resources also appear to
play a key role in determining which path an
individual might choose. For example, consistent
with forgiveness research, only those with resources
were able to walk away in the face of a less insulting
When faced with a
rejection, whereas the lack of these resources
enhanced the likelihood of aggressive responding.
When it came to the choice to engage in nonag-
gressive pursuit behavior, the interpretation begs
further research. What we found was that most
everyone engaged in pursuit behavior except those
let down easy with sufficient self-control to reject all
One finding that was surprising was that there was
not evidence to suggest that depletion effect was
larger in one rejection over another when predicting
aggression. In Figures 2 and 3 (when RS isn’t
present), the self-regulation depletion had a rela-
tively equivalent magnitude of impact on rejected
people for exacerbating aggressive responding, just
those rejected harshly started off higher on aggres-
sive responding. Although we had no hypotheses
about this specifically, one might have expected the
impact of self-regulation depletion to be stronger in
internal rejection conditions. It may be that
individual differences determine the magnitude of
the impact of self-regulation depletion, as the greater
reactivity to the internal rejection when depleted was
quite apparent among the rejection sensitive. We
turn to RS now.
rejection clearly mattered for those most hypervigilant
for signs of a relationship’s demise. All the scenarios
could have elicited negative reactions from those
rejection sensitive and self-regulation depleted. Instead
of rejection-sensitive people taking every negative
relationship event or type of rejection personally, they
seemed slightly more able to discern a harsh from
‘‘polite’’ rejection. Of course, when the rejection did
confirm their worst fears about a rejection of the
relationship and them, they reacted in a more extreme
fashion than those not as sensitive. However, this
extreme reaction was only elicited among the rejection
sensitive persons who were depleted of self-regulation.
Thus, we confirmed the new research arguing that the
reactions of the rejection sensitive to rejection are
completed mediated by the availability of self-
regulation resources [Ayduk et al., 2000]. In contrast,
those with intact self-control seemed to have the
wherewithal to walk away.
Implications for research and application.
Overall, we offer support for the I3Model of IPV.
Studying the impelling, instigating, and inhibiting
variables of RS, rejection, and self-regulation helped
us understand how, when, and for whom different
types of ORI may occur. Thus, the first contribution
of the present research is to expand the application
of the I3model to unwanted pursuit behavior and
not just IPV.
Our variations in types of
4We included the following items: ‘‘I would feel in control,’’ ‘‘I could
fix the situation,’’ ‘‘I believe there were things I could have done to
change the outcome,’’ ‘‘I believe there are things I could do to have
control over the future of the relationship,’’ and ‘‘If I wanted to make
it happen, I could make sure the relationship continued.’’ (a5.80 as
a combined scale). There was no significant difference on two of the
items (including, ‘‘I would feel in control’’ item) and on the items
where differences did emerge the means were consistently higher in
the internal condition than in the external condition.
516Sinclair et al.
However, further research is needed. We do not
believe that it is exclusively the combination of these
three variables that lead to ORI. Indeed, the
nonexperimental literature on stalking perpetration
has identified an array of additional impelling and
inhibiting variables to pursue. We may have only hit
upon one ‘‘right mix’’ of variables for the I3model.
Other combinations may not yield a full three-way
interaction [Fay and Sinclair, 2011; Slotter and
Finkel, 2010] whereas some do [Slotter et al., 2011].
It is an important future direction to identify which
factors do and do not elicit the three-way interac-
tion, and why or why not?
Ultimately, this study has shown that rejection
increases the likelihood of both pursuit and aggres-
sive ORI, both on its own and in conjunction with
self-regulation deficits and sensitivity to rejection.
We used attribution and basic needs theories to
hypothesize why different types of rejection might
make a difference in aggressive responses. We may
learn more about how impelling, instigating, and
inhibiting forces produce aggressive responses by
empirically testing other operationalizations of such
factors, but also by integrating relevant theories to
better encompass the factors that could lead to
aggression. This could then produce a more in-depth
picture as to how aggression may be elicited by
different types of instigators.
In particular, a valuable future avenue to pursue is
the question of when interpersonal rejection leads to
aggression. The results of both the current and
previous research suggest that the relationship
between rejection and aggression may be further
understood when threats to one’s need for connec-
tion are paired with other threats, such as need for
positive self-regard or need for control. In fact, this
assertion is consistent with relational goals theory
[Cupach and Spitzberg, 2004], which argues rejec-
tion alone does not trigger stalking, but rather the
extent to which that relationship goal is linked to
higher order goals (such as self-worth) determines
the likelihood to engage in stalking. Research is
needed to specifically test relational goals theory as
well as to see if it is possible to parse the independent
effects of threats to these basic needs vs. their
combined effects. It would be worthwhile to
integrate the research of others who have developed
basic needs models [Fiske, 2003], argued for links
between exclusion and basic needs [Williams et al.,
2005], and claimed that aggression results from basic
need threats [Staub, 2004].
Additional research is also needed to explore types
of rejection as it pertains to stalking behavior. As
this is just two studies, we would not go so far as to
recommend that everyone start letting romantic
pursuers down easy to lessen one’s risk of stalking.
Nor can we rule out other operationalizations of an
explicit rejection that avoid making an internal
attribution for the relationship failure as potential
means to stem stalking behavior. However, we
would advise equal caution to victim advocates in
claiming there is one right way to reject or to claim
that ‘‘the whole situation could have been avoided
ify’’ [Carll, 1999, p. 68] only there was a certain type
of rejection. These data do not support the idea that
letting the partner down easily will only perpetuate
stalking. In fact, although pursuit behavior may
follow an external rejection among those lacking
self-control, aggressive ORI was significantly less
likely among those ‘‘nicely’’ rejected, even for those
predisposed to be sensitive to rejections.
Ultimately, when a relationship fails it is going to
hurt. Whether people hurt others in retaliation
seems to depend on how much rejecters hurt vs.
spare the rejected, how sensitive people are to being
hurt, and the extent to which they can control the
reflex to lash out when injured. There may be no
good way to tell someone a relationship is over, but
there may be ways to lessen the impact for all those
The authors thank the Advanced Social Psychology
lab for their assistance on this project, especially
Katherine Collier, Benjamin Fay, and Lawrence
Perko. Also thanks to Dr. Kevin Armstrong and
Dr. Martin Giesen for their feedback on earlier
drafts of the project proposal and manuscript.
Further, we thank Dr. Grainne Fitzsimons and
Dr. Charles Carver for their valuable input on
the intricacies of self-regulation theory. Finally,
we thank Dr. Craig Anderson, Dr. Eli Finkel,
and the anonymous reviewers for the insightful
feedback that contributed to the strengthening of
Aiken L, West S. 1991. Multiple regression: Testing and interpreting
interactions. Tfhousand Oaks, CA: Sage Publications, Inc.
Anderson C. 1983. The causal structure of situations: The generation
of plausible causal attributions as a function of type of event
situation. J Exp Soc Psychol 19:185–203.
Ayduk O, Mendoza-Denton R, Mischel W, Downey G. 2000.
Regulating the interpersonal self: Strategic self-regulation for
coping with rejection sensitivity. J Pers Soc Psychol 79:776–792.
517 Self-Regulation and Rejection
Ayduk O, Mischel W, Downey G. 2002. Attentional mechanisms
linking rejection to hostile reactivity: The role of ‘‘hot’’ versus
‘‘cool’’ focus. Psychol Sci 13:443–448.
Battaglia DM, Richard FD, Datteri DL, Lord CG. 1998. Breaking
up is (relatively) easy to do: A script for the dissolution of close
relationships. J Soc Pers Relat 15:829–845.
Baum K, Catalano S, Rand M, Rose K. 2009. Stalking Victimization
in the United States. Washington, DC: The United States
Department of Justice. Office of Justice Programs, Bureau of
Baumeister RF, Leary MR. 1995. The need to belong: Desire for
interpersonal attachments as a fundamental human motivation.
Psychol Bull 117:497–529.
Baumeister RF, Vohs KD. 2004. Four roots of evil. In: Miller AG
(editor). The Social Psychology of Good and Evil. New York:
Guilford Press, pp 85–101.
Baumeister RF, Vohs KD. 2007. Handbook of Self-Regulation:
Research, Theory, and Applications. New York: Guilford Press.
Baumeister RF, Smart L, Boden J. 1996. Relation of threatened
egotism to violence and aggression: The dark side of high self-
esteem. Psychol Rev 103:5–33.
Blackhart GC, Baumeister RF, Twenge JM. 2006. Rejection’s impact
on self- defeating, prosocial, antisocial, and self-regulatory
behaviors. In: Vohs KD, Finkel EJ (editors). Self and Relation-
ships: Connecting Interpersonal and Intrapersonal Processes.
New York: Guilford, pp 237–253.
Brownridge DA. 2006. Violence against women post-separation.
Aggress Violent Behav 11:514–530.
Buckley KE, Winkel RE, Leary MR. 2004. Reactance to acceptance
and rejection: Effects of level and sequence relational evaluation.
J Exp Soc Psychol 40:14–28.
narcissism, self-esteem, and direct and displaced aggression:
Does self-love or self-hate lead to violence? J Pers Soc Psychol
Campbell WK, Sedikides C. 1999. Self-threat magnifies the self-
serving bias: A meta-analytic integration. Rev Gen Psychol
Carll EK. 1999. Violence in our Lives: Impact on Workplace, Home,
and Community. Needham Heights, MA: Allyn & Bacon.
Coleman FL. 1997. Stalking behavior and the cycle of domestic
violence. J Interpers Violence 12:420–432.
Cupach WR, Spitzberg BH. 1998. Obsessive relational intrusion
and stalking. In: Spitzberg BH, Cupach WR (editors). The Dark
Sideof Close Relationships.
Cupach WR, Spitzberg BH. 2004. The Dark Side of Relationship
Pursuit: From Attraction to Obsession and Stalking. Mahwah,
NJ: Lawrence Erlbaum Associates.
Dawson J, Richter A. 2006. Probing three-way interactions in
moderated multiple regression: Development and application of a
slope difference test. J Appl Psychol 91:917–926.
De Becker G. 1997. The Gift of Fear: Survival Signals That Protect
Us From Violence. New York, NY: Little-Brown.
DeWall CN, Baumeister RF, Stillman TF, Gailliot MT. 2007.
Violence restrained: Effects of self-regulatory capacity and its
depletion on aggressive behavior. J Exp Soc Psychol 43:62–76.
DeWall CN, Baumeister RF, Schurtz DR, Gailliot MT. 2010.
Acting on limited resources: The interactive effects of self-
regulatory depletion and individual differences. In: Hoyle RH
New York: Wiley-Blackwell, pp 243–262.
DeWall CN, Deckman T, Gailliot MT, Bushman BJ. 2011.
Sweetened blood cools hot tempers: Physiological self-control
and aggression. Aggr Behav 37:73–80.
Dennison SM, Stewart A. 2006. Facing rejection: New relationships,
broken relationships, shame, and stalking. Int J Offender Ther
Comp Criminol 50:324–337.
Denson TF. 2009. Angry rumination and the self-regulation of
aggression. In: Forgas JP, Baumeister RF, Tice DM (editors).
The Psychology of Self-Regulation. New York, NY, US:
Psychology Press, pp 233–248.
Denson TF, von Hippel W, Kemp RI, Teo LS. 2010. Glucose
consumption decreases impulsive aggression in response to
provocation in aggressive individuals. J Exp Soc Psychol
Downey G, Feldman S. 1996. Implications of rejection sensitivity for
intimate relationships. J Pers Soc Psychol 70:1327–1343.
Downey G, Feldman S, Ayduk O. 2000. Rejection sensitivity and
male violence in romantic relationships. Pers Relat 7:45–61.
Downey G, Irwin I, Ramsay M, Ayduk O. 2004a. Rejection
sensitivity and girls’ aggression. Girls and Aggression: Contributing
Factors and Intervention Principles. New York, NY: Kluwer
Downey G, Mougios V, Ayduk O, London BE, Shoda Y. 2004b.
Rejection sensitivity and the defensive motivational system:
Insights from the startle response to rejection cues. Psychol Sci
Eisenberger NI, Lieberman MD. 2004. Why rejection hurts: A
common neural alarm system for physical and social pain. Trends
Cogn Sci 8:294–300.
Fay BA, Sinclair HC. 2011. The I3model and stalking: Examining
effects of rejection, self-regulation, and narcissism on stalking
perpetration. Poster presented at the 2011 Society for Personality
and Social Psychology Conference, San Antonio, TX.
FinkelEJ. 2007. Impelling
perpetration of intimate partner violence. Rev Gen Psychol
Finkel EJ. 2008. Intimate partner violence perpetration: Insights
from the science of self regulation. In: Forgas JP, Fitness J
(editors). Social Relationships: Cognitive, Affective, and Motiva-
tional Processes. New York: Psychology Press, pp 272–288.
Finkel EJ, Eckhardt CI. 2011. Intimate partner violence. In: Simpson JA,
Campbell L (editors). The Oxford Handbook of Close Relationships.
New York: Oxford Press (in press).
Finkel EJ, DeWall CN, Slotter EB, Oaten M, Foshee VA. 2009. Self-
regulatory failure and intimate partner violence perpetration.
J Pers Soc Psychol 97:83–499.
Fiske S. 2003. Five core social motives, plus or minus five. In:
Spencer SJ, Fein S, Zanna MP, Olson JM (editors). Motivated
Social Perception: The Ontario Symposium.Mahwah, NJ, US:
Lawrence Erlbaum Associates, pp. 233–246.
Fitzsimons GM, Bargh JA.2004.Automatic self-regulation. In:
Baumeister RF, Vohs KD (editors). Handbook of Self-Regulation:
Research, Theory, and Applications. New York: Guilford Press,
Folkes VS. 1982. Communicating the reasons for social rejection.
Journal of Experimental Social Psychology 18:235–252.
Ford MB, Collins NL. 2010. Self-esteem moderates neuroendocrine
and psychological responses to interpersonal rejection. J Pers Soc
Gailliot MT, Baumeister RF. 2007. The physiology of willpower:
Linking blood glucose to self-control. Pers Soc Psychol Rev
518Sinclair et al.
Gerber J, Wheeler L. 2009. On being rejected: A meta-analysis of
Hagger MS, Wood C, Stiff C, Chatzisarantis NLD. 2010. Ego
depletion and the strength model of self-control: A meta-analysis.
Psychological Bulletin 136:495–525.
Heider F. 1958. The Psychology of Interpersonal Relationships.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Kamphuis J, Emmelkamp P, de Vries V. 2004. Informant personality
descriptions of postintimate stalkers using the five factor profile.
J Pers Assess 82:169–178.
Langhinrichsen-Rohling J, Palarea RE, Cohen J, Rohling ML. 2000.
Breaking up is hard to do: Unwanted pursuit behaviors following
the dissolution of a romantic relationship. Violence Vict
Leary MR, Downs DL, 1995. Interpersonal functions of the
self-esteem motive: The self-esteem system as a sociometer.
In: Kernis MH (editor). Efficacy, Agency, and Self-Esteem.
New York: Plenum Press, pp 123–144.
Leary MR, Twenge JM, Quinlivan E. 2006. Interpersonal rejection
as a determinant of anger and aggression. Pers Soc Psychol Rev
MacDonald TK, Zanna MP, Fong GT. 1996. Why common sense
goes out the window: Effects of alcohol on intentions to use
condoms. Pers Soc Psychol Bull 22:75.
Major B, Kaiser CR, McCoy SK. 2003. It’s not my fault: When and
why attributions to prejudice protect self-esteem. Pers Soc
Psychol Bull 29:772–781.
McFarlane JM, Campbell JC, Wilt S, Sachs SJ, Ulrich Y, Xu X.
1999. Stalking and intimate partner femicide. Homicide Stud
Interdiscip Int J 34:300–316.
Meloy JR. 1998. The Psychology of Stalking: Clinical and Forensic
Perspectives. San Diego, CA, US: Academic Press.
Mezulis AH, Abramson LY, Hyde JS, Hankin BL. 2004. Is there a
universal positivity bias in attributions? A meta-analytic review
of individual, developmental, and cultural differences in the self-
serving attributional bias. Psychol Bull 130:711–747.
Miller C, Ravensberg V. 2003. Stalking among young adults: A
review of the preliminary research. Aggress Violent Behav
Mischel W, Ayduk O. 2011. Willpower and the cognitive-affective
processing system: The dynamics of delay of gratification. In:
Baumeister RF, Vohs KD (editors). Handbook of Self-Regulation:
Research,Theory, and Applications. New York: Guilford Press,
Molden DC, Finkel FJ. 2010. Motivations for promotion and
prevention and the role of trust and commitment in interpersonal
forgiveness. J Exp Soc Psychol 46:255–268.
Molden DC, LucasG M, Gardner WL, Dean K, Knowles ML. 2009.
Motivation for prevention or promotion following social exclu-
sion: Being rejected versus being ignored. J Pers Soc Psychol
Mullen PE, Pathe ´ M, Purcell R. 1999. Study of stalkers. Am J
Muraven M, Baumeister RF. 2000. Self-regulation and depletion of
limited resources: Does self-control resemble a muscle? Psychol
Muraven M, Tice DM, Baumeister RF. 1998. Self-control as a
limited resource: Regulatory depletion patterns. J Pers Soc
Perspect Psychol Sci
Pronk TM, Karremans JC, Overbeek G, Vermulst AA, Wigboldus D.
2010. What it takes to forgive: When and why executive
functioning facilitates forgiveness. J Pers Soc Psychol 98:
Romero-Canyas R, Downey G, Berenson K, Ayduk O, Kang NJ.
2010a. Rejection sensitivity and the rejection-hostility link in
romantic relationships. J Pers 78:119–148.
When does rejection trigger ingratiation? J Pers Soc Psychol
Ryan KM, Frieze IH, Sinclair HC. 1999. Physical violence in dating
relationships. In: Paludi MA (editor). The Psychology of
Sexual Victimization: A Handbook. Westport, CT: Greenwood
Publishing, pp 33–54.
Schmeichel BJ, Baumeister RF. 2004. Self-regulatory strength. In:
Baumeister RF, Vohs KD (editors). Handbook of Self-Regulation:
Research, Theory, and Applications.New York: Guilford Press,
Sheridan L, Lyndon AE. 2011. The influence of prior relationship,
gender, and fear on the consequences of stalking victimization.
Sinclair HC, Frieze IH. 2000. Initial courtship behavior and
stalking: How should we draw the line? Violence Vict 15:
Slotter EB, Finkel EJ. 2010. The I3theory: instigating, impelling,
Miklunicer M (editors). Human Aggression and Violence:
Causes, Manifestations, and Consequences. Washington, DC:
Slotter EB, Finkel EJ, DeWall CN, Pond RS, Lambert NM,
Bodenhausen GV, Fincham FD. (2011, August 8). Putting the
Brakes on Aggression Toward a Romantic Partner: The
Inhibitory Influence of Relationship Commitment. Journal of
Personality and Social Psychology. Advance online publication.
Spitzberg BH, Cupach WR. 2007. The state of the art of stalking:
Taking stock of the emerging literature. Aggr Violent Behav
Spitzberg BH, Rhea J. 1999. Obsessive relational intrusion
and sexual coercion of victimization. J Interpers Violence 14:
Staub, E. 2004. Basic human needs, altruism, and aggression. In:
Miller AG (editor). The Social Psychology of Good and Evil.
New York, NY: Guilford Press, pp 51–84.
Stucke TS, Baumeister RF. 2006. Ego depletion and aggressive
behavior: Is the inhibition of aggression a limited resource? Eur J
Soc Psychol 36:1–13.
Tashiro T, Frazier P. 2003. ‘‘I’ll never be in a relationship like
that again’’: Personal growth following romantic relationship
breakups. Pers Relat 10:113–128.
Tjaden P, Thoennes N. 1998. Stalking in America: Findings From
the National Violence Against Women Survey. Washington, DC:
National Criminal Justice Reference Service.
Twenge JM, Campbell WK. 2003. ‘‘Isn’t it fun to get the respect that
we’e going to deserve?’’ Narcissism, social rejection, and
aggression. Pers Soc Psychol Bull 29:261–273.
Vande BosK. 2010.Self-regulation,
behavioral disinhibition in normative judgments. In: Ramona DR,
Kay AC, Zanna MP, Olson JM (editors). The Psychology of
Justice and Legitimacy. New York: Psychology Press, pp
aggression. In:Shaver PR,
519 Self-Regulation and Rejection
Wegner DM, Schneider DJ, Carter SR, White TL. 1987. Paradoxical Download full-text
effects of thought suppression. Journal of Personality and Social
Wegner DM, Pennebaker JW. 1993. Changing our minds: An
introduction to mental control. In: Wegner DM, Pennebaker JW
(editors). Handbook of Mental Control. Englewood Cliffs, NJ:
Prentice Hall, pp 1–12.
Wegner D, Zanakos S. 1994. Chronic thought suppression.
J Pers 62:615–640.
Weiner B. 1985. A n attributional theory of achievement motivation
and emotion. Psychol Rev 92:548–573.
Williams SL, Frieze IH. 2005. Courtship behaviors, relationship
violence, and breakup persistence in college men and women.
Psychol Women Q 29:248–257.
Williams KD, Forgas JP, von Hippel W. 2005. The Social Outcast:
Ostracism, Social Exclusion, Rejection, and Bullying. London,
UK: Psychology Press.
WilliamsSL, FriezeIH, Sinclair
intimate violence. In: Hamel J, Nicholls T (editors). Family
to Gender-Inclusive Research and Treatment. New York:
HC. 2006. Stalkingand
520 Sinclair et al.