Friends With Benefits: Behavioral and fMRI Studies on the Effect of Friendship
Reminders on Self-Control for Compulsive and Non-compulsive Buyers
Forthcoming in: International Journal of Research in Marketing
Eline de Vries
Universidad Carlos III de Madrid, Spain
Bob M. Fennis
Gert ter Horst
University of Groningen, the Netherlands
Bob Fennis: email@example.com
Eline de Vries: firstname.lastname@example.org
Does the real or imagined presence of friends invariantly drive consumers to engage in
disinhibited behavior, and give in to the “urge to splurge” in the face of consumption
temptations? Or might there be situations in which being with friends or even merely thinking of
friends or the friendships we have with them can actually improve self-control?
In five studies, using a unique combination of controlled experiments examining overt
consumer behavior and functional magnetic resonance imaging (fMRI), we propose and show
that the extent to which consumers identify a goal conflict between giving in to buying
temptations on the one hand and the perceived consequences for maintaining satisfactory
relationships with close friends on the other is a critical mediator of whether friendship
reminders decrease or increase self-control. We further show that such a goal conflict is most
likely for consumers with a chronic, compulsive tendency for uncontrolled, disinhibited
acquisition and consumption —for consumers classified as compulsive buyers. For their non-
compulsive counterparts, in contrast, acts of acquisition and consumption, even incidental
disinhibited ones, are perceived to be less problematic in light of their friendships and hence do
not induce a goal conflict to the same extent. Our findings provide insights into social influences
on self-control and identify the concept of friendship reminders as a way to reduce a common
type of dysfunctional consumer behavior. In addition to enhancing consumer well-being,
reducing compulsive buying will substantially reduce handling costs for organizations. Hence,
the findings are of academic, societal and managerial relevance.
Key words: compulsive buying, self-control, social influence, friendship, fMRI
“A faithful friend is the medicine of life.”
The notion that humans have a strong need to acquire and maintain meaningful social
relationships is widely accepted (Baumeister & Leary, 1995; Maslow, 1954). Friendship, defined
as a close, intimate, mutual relationship with a same-sex peer (Sullivan, 1953), has been
consistently proven to be beneficial for a wide range of issues, including well-being (Baumeister
& Leary, 1995) and mental and physical health (Wilkinson, 1999). But the impact of close
relationships does not stop there. Research in marketing has found that relationships exert a
strong influence on a host of consumption decisions ranging from holiday destinations to battery
choices (Argo, Dahl, & Manchanda, 2005; Ariely & Levav, 2000; Childers & Rao, 1992;
Hofmann, Baumeister, Förster, & Vohs, 2012; Rook & Fisher, 1995). Moreover, researchers
have recently begun to explore how close-relationship partners affect a specific kind of
consumption situations: those that involve self-regulation (see for example Dzhogleva &
Lamberton, 2014). Self-regulation is generally defined as the psychological and behavioral
process that moves people toward desired, and away from undesired end-states (Fitzsimons,
Finkel, & VanDellen, 2015). One aspect of self-regulation is self-control, which takes place
when a predominant response is inhibited, altered or overridden (Muraven & Baumeister, 2000).
Relationship partners may have an important impact on whether consumers exercise self-control
(Dzhogleva & Lamberton, 2014).
Work in consumer behavior generally corroborates the notion that friends may lower
consumers’ self-control when eating (DeCastro, 1994; Redd & DeCastro, 1992), consuming
alcohol (Zhang & Shrum, 2009), or going shopping (Kurt, Inman, & Argo, 2011; Luo, 2005).
Indeed, a recent survey among 2000 British women (Kirkova, 2013) reports that 62 percent of
female respondents spent more money when they shop with a friend compared to when they
shop alone. The researchers calculated that over the course of a year, those who shop with
friends spend almost £900 (approximately €1,000 or $1,200) more than what they would have
done shopping alone. Does this mean that friends invariantly drive consumers to engage in
disinhibited behavior, and give in to the “urge to splurge” in the face of consumption
temptations? Or might there be situations in which being with friends or even merely thinking of
friends or the friendships we have with them can actually improve self-control?
In the current paper we identify an important condition under which the frequently
observed adverse effect may be reversed. Specifically, using a unique combination of controlled
experiments examining overt consumer behavior and functional magnetic resonance imaging
(fMRI) focusing on relevant neural substrates, we propose and show that the extent to which
consumers identify a goal conflict between giving in to buying temptations on the one hand, and
the perceived consequences for maintaining satisfactory relationships with close friends on the
other, is a critical mediator of whether friendship reminders decrease or increase self-control. We
further propose that such a conflict is most likely for consumers with a chronic, compulsive
tendency for uncontrolled, disinhibited acquisition and consumption —for consumers classified
as compulsive buyers. For their non-compulsive counterparts, in contrast, acts of acquisition and
consumption, even incidental disinhibited ones, are perceived to be less problematic in light of
their friendships and hence do not induce a goal conflict to the same extent. If anything, the
safety and comfort that friends and friendships might represent may free these consumers to
occasionally ‘let go a little’ and disinhibit in the face of temptations.
Consequently, we propose that the degree of goal conflict that mediates the impact of
(reminders of) close friendships on consumer self-control in the face of temptations will be a
function of buyer compulsivity, such that reminders of friendship will decrease self-control
when little to no goal conflict is identified, i.e., for non-compulsive buyers for whom acquisition
and (disinhibited) consumption is generally unproblematic in light of their friendships. In
contrast, reminders of friendship will increase self-control when such a goal conflict is likely,
i.e., for compulsive buyers for whom acts of uncontrolled indulgence in response to temptations
will likely conflict with their goals of maintaining satisfactory friendship relationships. Hence,
we propose a crossover interaction between reminders of friendship and buyer compulsivity on
perceptions of conflict and on subsequent consumer self-control exertion, with goal conflict
perceptions mediating the impact of the friendship × compulsivity interaction on self-control.
Hence, formally, we propose a moderated mediation model for when and why friendship
reminders may hamper or aid consumer self-control.
Please note that our reasoning hinges on the salience of a goal conflict between indulgent
consumption on the one hand, and the goal of maintaining satisfactory friendship relationships
on the other. That implies that such a conflict (or lack thereof) can in principle be triggered by
any juxtaposition, real, imagined, concrete or more abstract, between the concept of indulgent
consumption on the one hand, and the concept of close friendship, on the other. Consequently, in
our studies, we will induce both notions using indulgent spending reminders and friendship
reminders. We will use a variety of methods, involving the real and/or imagined presence of
close others in temptation contexts as well as relying on more abstract reminders of the goal of
close friendships in such contexts. We contend that it adds to the robustness of our findings if we
show consistent effects in the expected directions across these various manipulations.
The next sections elaborate on our reasoning in more detail.
2. Conceptual Framework
2.1. The role of conflict identification in self-control
The exertion of self-control in the face of temptations has been conceptualized as the
outcome of a “tug-of-war” or conflict between two opposing forces: impulse strength and
inhibition strength (Hofmann, Friese & Strack, 2009). That is, self-control is exerted when there
is a conflict between a strong hedonic impulse to satisfy the short-term goal of indulgence on the
one hand, and the strength to inhibit or restrain that impulse in the service of valued longer-term
goals on the other. For example, a consumer may identify a conflict between the acute desire for
another pair of expensive, fashionable shoes and the long-term goal to save money. Or, more
germane to the present context, a consumer may experience a sense of conflict between
indulging in self-centered binge-consumption or overspending on the one hand, and maintaining
good and fulfilling social relationships with close others who might perceive this behavior as
disturbing and hence a threat to the relationship, on the other. Self-control exertion thus consists
of two critical stages: (1) the identification of a goal or response conflict between competing
behavioral options, choices, goals or preferences, which is followed by (2) the implementation of
an inhibitory response towards the temptation (Carver & Scheier, 1990; Myrseth & Fishbach,
2009; Strack & Deutsch, 2004). If this conflict identification results in inhibiting the impulse to
splurge and indulge, successful self-control exertion has been displayed. In contrast, when there
is no perception of conflict, mere exposure to buying temptations may suffice to induce
indulgence and hence the display of disinhibited acquisition and consumption (Wagner &
Next, to break up our reasoning, we will first argue why it is plausible to assume that
friendship reminders will reduce perceptions of goal conflict and subsequent self-control
exertion among non-compulsive consumers, followed by our argumentation for why it should
increase these perceptions and ensuing self-control exertion among their compulsive
2.2. Friendship reminders reduce perceptions of goal conflict and self-control among non-
Non-compulsive buyers may be viewed as ‘regular’ or ‘default’ consumers (to the extent
that there is such a phenomenon). To argue why friendship reminders might reduce perceptions
of a goal conflict and subsequent self-control exertion among these consumers, we draw on
several literatures. First, research on the facilitating effects of others on consumption has largely
focused on (over)eating, alcohol consumption and substance use and abuse (e.g., Howland,
Hunger, & Mann, 2012; Wagner & Heatherton, 2015). The most prolific literature has
accumulated in the food consumption domain (e.g., DeCastro, 1997; Clendenen, Herman, &
Polivy, 1994; Herman, 2015; Hetherington, Anderson, & Norton, 2006; Salvy et al., 2007). This
literature shows consistently that, compared to being alone, food consumption tends to be higher
when one is in the real or imagined presence of others (see DeCastro, 1997; Herman, 2015, for
reviews). Only a small subset of studies has focused on distinguishing between those others
either being friends or strangers (see Herman, 2015, for an overview). Strikingly, these studies
suggest that while in the presence of strangers people sometimes tend to eat less, they typically
eat more when in the presence of friends (e.g., Clendenen et al., 1994; Salvy et al., 2007). This
research resonates well with work by Wilcox and Stephen (2013) who found that consumers
focusing on close friends (strong ties) in an online social media environment showed lower self-
control than consumers focusing on mere acquaintances (weak ties).
These facilitating, disinhibiting effects of the presence of friends on food intake have
been attributed to (lack of) impression management concerns: while the presence of strangers or
acquaintances might sometimes motivate one to control food intake so as not to communicate an
undesirable impression, the safety and comfort that close friends and friendships represent
typically frees the consumer from any such concerns and hence effectively reduces any self-
control motivation. By implication, if such concerns play less of a salient role, then there is no
ground to infer that an (incidental) act of indulgence in the face of temptations will evoke a
strong goal conflict in the consumer when in the real or imagined presence of friends. Rather,
even mere exposure to consumption temptations may then promote more disinhibited indulgence
as previous research on self-control suggests (see Wagner & Heatherton, 2015, for an overview).
Hence, to the extent that perceiving such a conflict is a necessary and sufficient antecedent of
self-control, we should thus observe less self-control in non-compulsive buyers when friendship
reminders are salient.
2.3. Friendship reminders increase perceptions of goal conflict and self-control among
For compulsive consumers, the role of friendship reminders is posited to be opposite to
the former case. Compulsive buying involves an overpowering urge to indulge and systematic
failure to curb that urge, leading to chronic, disinhibited, uncontrolled, excessive buying ‘binges’
of consumer goods (Dittmar, 2005; Faber & O’Guinn, 2008; Müller, Mitchell, & De Zwaan,
2015). The tail of the compulsive buying distribution has been deemed so problematic that it has
made it as an entry as part of the obsessive-compulsive and related disorders in the fifth edition
of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V, 2013). Indeed, buying
compulsivity is frequently part and parcel of a more generalized obsessive-compulsive pathology
involving a chronic, more generic impairment in inhibiting prepotent responses across a variety
of domains including hoarding, (binge) eating and substance use (Müller et al., 2015). While
estimates vary, some studies suggest that the prevalence of consumers who are classified as
compulsive buyers may be as high as 16 per cent of the general population (Dittmar, 2005; Faber
& O’Guinn, 1992; Friese, 2000; Müller et al., 2015; Ridgway, Kukar-Kinney, & Monroe, 2008).
That such behavior is disturbing and problematic has not escaped compulsive buyers
themselves. Indeed, these consumers frequently experience post-purchase guilt (Valence,
d’Astous, & Fortier, 1988) and/or regret after engaging in acts of compulsive buying (Faber &
O’Guinn, 1992). They go to great lengths trying to keep their behavior and purchases secret from
close others whom they suspect might be able to infer that their acquisitions are the product of a
compulsive disorder (Friese, 2000; O’Guinn & Faber, 1989). These observations signal that
impression management concerns toward one’s close others may play a profound role in the
buying and consumption decisions of compulsive buyers. More importantly, to the extent that
these ‘close others’ are close friends, the strength of those concerns is likely much larger than for
non-compulsive buyers. But why would compulsive buyers’ concerns focus particularly on their
close friendships? The literature suggests several grounds to infer that the origins of these
concerns may well lie in compulsive buyers’ perceiving their own disinhibited behavior as a
direct threat to maintaining and consolidating these valued social relationships. First, in contrast
to other types of close relationships, such as those with family members, friendships are not a
given, but are optional (Chopik, 2017). Hence, the relationship exists as long as it is mutually
beneficial, but may be terminated when it ceases to be. Particularly for (young and adult)
consumers with a compulsive disorder, this is a daunting, yet realistic prospect (e.g., Borda et al.,
2013), making the display of disinhibited acquisition and/or consumption in the real or imagined
presence of friends not safe and comfortable but threatening and problematic. Second, research
shows that displays of compulsivity may indeed undermine friendship bonds as compulsive
individuals tend to experience substantial problems making and keeping friends (Borda et al.,
2013; Torres, Cruz, Vicentini, Lima, & Ramos-Cerqueria, 2016). Third, these findings converge
with additional research from the perspective of these friends, which shows that they indeed
perceive displays of disinhibited, uncontrolled behaviour as problematic, reducing perceptions of
interpersonal reliability and trustworthiness (Righetti & Finkenauer, 2011). Hence, displays of
buying compulsivity are not only an imagined, but also a real threat to maintaining close and
fulfilling social relationships.
Against this backdrop it makes sense to assume that impression management concerns
may play an important role in the behavioural regulation of compulsive buyers. Moreover, these
concerns will mainly be targeted towards one’s close friends, rather than strangers or
acquaintances with whom a relationship goal is less at stake, and for whom the display of
uncontrolled behaviour may be deemed less diagnostic of an underlying pathology. Given the
possible adverse consequences for their long-term friendships, compulsive buyers will thus
perceive a more pronounced goal conflict between indulging in self-centered binge-spending on
the one hand, and maintaining good, fulfilling social relationships with close others on the other.
In sum and in contrast to non-compulsive buyers, compared to a stranger, acquaintance or
an alone condition, a friendship reminder condition —either by being in the physical or
psychological presence of a close friend or by being reminded of a strong friendship with a close
friend— will engender increased perceptions of goal conflict in the compulsive consumer. To the
extent that such goal conflict is a necessary and sufficient antecedent of self-control, we should
thus observe more self-control in response to buying temptations in compulsive buyers when
friendship reminders are salient.
The current research aims to contribute to the literature in six ways. First, we replicate
and extend past findings by forwarding and testing a theory-based self-regulatory account of
when and why friendship reminders increase or decrease consumer self-control. In so doing, we
reconcile conflicting findings on the self-regulatory effects of (close) others, which have
sometimes pointed to their facilitating and sometimes to their inhibiting impact on product
acquisition and consumption. More specifically, while previous research has mainly focused on
conditions where friendship cues decrease self-control and increase indulgence, the present work
highlights that this is only half the story, and that there are specific conditions where such cues
may actually foster self-control and curb the desire to indulge. Moreover, we pinpoint when and
why both types of effects may occur and so not only replicate past findings in an as yet untested
“theatre of operations” —the urge to buy consumer products—but also extend previous research
by demonstrating the beneficial effects of such cues to enhance consumer self-control,
particularly among chronically vulnerable consumers, —those high in buying compulsivity.
Furthermore, the present research is the first to propose and systematically test the underlying
process responsible for the harmful vs. beneficial impact of such friendship cues on consumer
self-control, —the extent of experienced goal conflict between giving in to buying temptations
on the one hand, and the perceived consequences for maintaining satisfactory relationships with
close friends on the other.
Second, we also extend the literature on social facilitation and inhibition by adopting a
more generic lens. More in particular, while previous research has mainly focused on the role of
physical presence of other consumers, our perspective moves beyond that by proposing that such
physical presence of friends (or strangers/ mere acquaintances) is not a necessary precondition
for observing an impact on (lack of) self-control, but a mere instance of a more general
phenomenon, i.e., the notion that the postulated effects may extend to any cue, physical or
imaginary, concrete or more abstract, that may signal the superordinate goal of maintaining
satisfactory friendship relationships to the target consumer. Hence, our work bridges the gap
between “classic” studies in social facilitation (and inhibition) and more recent work on
consumer goal activation and goal pursuit. Thus, in our studies, we employ a variety of
manipulations to remind our participants of the goal of friendship. In addition to the physical
presence of one’s close friend, we identify the imagined presence of one’s close friend and
actively thinking about friendship per se to yield comparable effects and so demonstrate the
robustness of our notions across tasks and manipulations. Moreover, the extension to any cue
activating the notion of friendship opens new doors to possibly effective (marketing)
interventions to promote consumer resilience in the face of buying temptations. That is, these
interventions need not necessarily be limited to the physical presence of one’s friends, but also
interventions that focus on other cues to activate the representation of the friendship goal may be
considered, such as (online and offline) advertising depicting (symbols of) friendship, or
imagery-evoking messages about the value of friendship.
Third, we extend the literature on consumer self-control by zooming-in on a particular,
yet largely neglected form of chronically impaired consumer self-control –compulsive buying.
That is, buying compulsivity, while sometimes being part of a wider spectrum of obsessive
compulsive disorders (Müller et al., 2015), has received some —mainly phenomenological—
research attention (e.g., Black, 2011; O’Guinn & Faber, 1989), but has to our knowledge not
figured prominently in the burgeoning field of consumer self-regulation, which has focused
predominantly on the role of consumer goal pursuit and state self-control. That is all the more
striking given the high prevalence of the phenomenon in the general population (Ridgway et al.,
2008). However, the specific dimensionality of the construct, comprising both experiencing
sudden excessive urges to engage in binge spending, combined with a chronic impairment in
inhibiting these impulses, yields a “unique testing ground” for examining the dynamics of self-
regulation in consumer behavior.
Fourth, our focus on consumers classified as compulsive buyers yields insights with clear
practical and managerial implications. That is, our research highlights being reminded of
friendship as a condition under which compulsive buyers are better able to respond with self-
control to buying temptations, which may provide tools to therapists, public policy makers and
marketing practitioners concerned with consumer well-being. For instance, interventions based
on fairly simple friendship reminders (e.g., a photo of a close friend in one’s wallet, an imagery
provoking message about the value of friendship) may suffice to improve the self-regulatory
performance of consumers who are typically viewed as being chronically challenged when it
comes to self-control. Indeed, our work shows the potential for these relatively easy-to-
implement and cost-efficient approaches to reduce this type of dysfunctional consumption
behavior and as such could complement more intensive and structural forms of treatment, such
as those that rely on psychotherapy. On the other hand, our work also shows the power of such
simple friendship reminders in marketing efforts aimed at ‘regular’, non-compulsive consumers,
where such cues may actually foster an increased purchase and consumption tendency.
Fifth, we add to the consumer behavior literature by employing a triangulation approach.
That is, we aim to seek support for our main hypothesis —the extent to which consumers
identify a goal conflict underlies when and why reminders of friendship help or hurt self-
control—by demonstrating, then replicating and subsequently also explaining, the proposed
crossover interaction between friendship reminders and buyer compulsivity, while ruling out
alternate explanations (i.e., the role of social relationships per se and non-social yet positive
circumstances). In so doing, we rely on self-report, overt behavior, and fMRI indices of self-
control exertion and we use two distinct measures of buyer compulsivity, four different
indicators of goal conflict identification, and four different friendship reminder procedures.
Finally, we rely on a scenario procedure, a Stroop task and Go/No-go tasks as prevalidated
paradigms to test our propositions. Using this plethora of studies, measures and tasks contributes
to the robustness of our findings.
Finally, a major contribution of the present work is the focus on behavioral data on the
one hand and their neural underpinnings using fMRI on the other. The use of fMRI is relatively
new in consumer science (see for recent applications Berns & Moore, 2012; Couwenberg et al.,
2017; Hedgcock & Rao, 2009; Hedgcock, Vohs, & Rao, 2012; Yoon et al., 2006, 2012) and
allows the identification of cerebral activation in areas that have been reliably associated with the
identification of conflict (the anterior congulate cortex; ACC; Kerns et al., 2004; MacDonald et
al., 2000) and actual self-control exertion (the dorsolateral prefrontal cortex; DLPFC;
MacDonald et al., 2000). By combining more traditional behavioral methods with fMRI we
respond to calls for more research on the neural substrates of consumer self-control. Indeed, to
the extent that self-report data, behavioral data and neural data converge to support our notions,
the present work contributes to gaining a fuller understanding of the phenomenon of consumer
self-regulation. More specifically, our joint approach pinpoints the critical role of conflict
identification (both using self-report and its neural –ACC–manifestation) as a necessary
precursor to actual self-control exertion (witnessed in self-report indices, overt inhibition
behavior and –again– its neural substrate, DLPFC activation). Such conflict identification has
been all too often assumed, yet only rarely assessed in research on (consumer) self-regulation,
and to the extent that it has received research attention, typically has been limited to self-reports,
rather than the unique combination of self-report, behavioral and fMRI indices used in the
present work. Moreover, our approach also adds to the literature on consumer self-regulation by
highlighting “what we can do to become better at it” (Wagner & Heatherton 2011, p. 55). More
specifically, our results indicate how simple contextual cues reminding consumers of their
friendships can boost self-control, even among chronically challenged consumer segments. As
such our findings resonate well with recent “situated cognition” approaches to fostering self-
control (see Schwarz, 2006; Fennis, 2017, for an overview). That is, rather than relying on
extensive training to improve chronic self-control, the choice context itself harbors simple cues,
nudges or heuristics that shift consumer cognition, judgment and choice in the direction of more
self-control and so promotes conditions where more impulsive behaviors need not invariantly be
considered indulgent and harmful, but instead can be beneficial and fostering well-being.
2.5. Overview of studies
Five studies examine how friendship reminders affect self-control in both compulsive and
non-compulsive buyers. Our main hypothesis is that the extent to which consumers identify a
goal conflict mediates the impact on self-control of being reminded of friendship on the one hand
and buyer compulsivity on the other. We dissect our main hypothesis across five studies,
focusing on the main constituent parts and relations using a variety of validated measures and
paradigms in Studies 1-3 and ultimately integrating our findings in Studies 4 and 5.
More specifically, the first study is a behavioral experiment designed to investigate the
effect of friendship reminders on self-control by assessing the actual, overt extent to which
compulsive and non-compulsive consumers are able to inhibit the impulse to approach tempting
products. Study 2 shows that the hypothesized effect only occurs with friendship reminders, and
not with acquaintances or positive circumstances in general, thus ruling out mere presence of
others and mood as alternative explanations. It moreover shifts the focus from inhibition to
impulse by zooming-in on acutely experienced product desire as a particularly marketing-
relevant dimension of self-control. Study 3 builds on the previous findings by measuring self-
control exertion in the real physical presence of either a close friend, or a stranger, or when
alone. In Studies 4 and 5, we focus on the role of the proposed underlying mechanism driving the
effects: the extent of experiencing a goal conflict and its implications for exerting inhibitory
control. Specifically, in Study 4 we test our key postulate that friendship reminders are more
likely to induce goal conflict for compulsive than for non-compulsive buyers. Moreover in this
study we directly test the proposed moderated mediation model —that goal conflict perceptions
mediate the impact of the friendship × compulsivity interaction, such that salient conflict
perceptions mediate the impact on self-control for compulsive buyers more than for non-
compulsive buyers. Finally, in Study 5 we study both self-control stages together by addressing
the neural correlates of goal conflict identification and control exertion in an fMRI experiment.
Note that in our studies we will focus on both sides of the “tug-of-war” of self-control.
That is, Studies 2 and 4 will focus on impulse strength by assessing the extent to which
consumers experience a strong impulse or desire to engage in unplanned buying. In contrast,
Studies 1, 3, and 5 will focus on the inhibition strength by assessing the ability to suppress an
impulse and hence the ability to inhibit a prepotent response.
3. Study 1
We started by investigating the substantive phenomenon in a behavioral experiment. This
allowed us to examine the observable behavioral impact of friendship reminders on consumers’
ability to implement self-control. In doing so, we used a buying temptations-specific Go/No-go
task, a pre-validated measure of consumer self-control exertion (Batterink, Yokum, & Stice,
2010; Mishra & Mishra, 2010; Newman, Widom, & Nathan, 1985). The task is based on the
notion that successful self-control exertion manifests itself in the ability to inhibit a prepotent
approach response towards consumption temptations.
If our hypothesis that being reminded of a strong friendship with a close friend may
decrease self-control for non-compulsive buyers while increasing it for compulsive buyers is
correct, then non-compulsive buyers should show more while compulsive buyers should show
fewer inhibition failures towards buying temptations when reminded of friendship compared to
when they are not. In sum, we expect a crossover interaction between friendship reminders and
buyer compulsivity on self-control exertion.
3.1. Participants and design
To warrant a sample size of at least twenty participants per cell (cf. Simmons, Nelson, &
Simonsohn, 2011), we recruited 69 undergraduate students that participated in exchange for
money or partial course credit. Given that around 90 percent of compulsive buyers are female
(Dittmar, Long, & Bond, 2007), only women were invited. Also, in anticipation of the fMRI
experiment performed in Study 5 in which all participants should be right-handed, we chose to
minimize potentially confounding factors by including only right-handed participants. Two
participants were excluded from further analysis. One of them did not comply with experimental
instructions, while the other reported an extremely negative mood. This resulted in an effective
sample of 67 participants (mean age 19.81 years, SD = 1.69; matching the onset age of
compulsive buying; Black, 2007). The experiment employed a design with one between-subjects
factor (friendship reminder present vs. absent), and measured individual differences in
compulsive buying tendency as a continuous moderator.
Upon arrival at the laboratory, participants were randomly assigned to one of two
conditions. After signing an informed-consent form, they were individually seated in cubicles.
Computer instructions told them that the experiment consisted of several unrelated tasks. The
first task involved the reminder of friendship manipulation, using an established mindset
activation procedure (Stillman, Tice, & Fincham, 2009; Zhang & Shrum, 2009). In line with
Sullivan’s (1953) definition of friendship, participants in the friendship reminder condition were
asked to describe a situation in which they felt strong friendship with a close, same-sex friend. In
the control condition, participants were asked to describe the manufacturing process of a wooden
table. All participants were asked to write at least four sentences and to be as accurate as possible
in their description. In the friendship condition, participants were required to use the words: we,
together, friendship and bond. In the control condition, the words trees, wood, sawmill and truck.
Immediately following the friendship reminder procedure, all participants participated in
a Go/No-go task (see next for a detailed description). This task was presented as a
speed/accuracy test and designed to measure participants’ self-control exertion by assessing their
ability to inhibit prepotent approach responses towards buying temptations. Next, participants
completed the compulsive buying scale developed by Faber and O’Guinn (1992), a 7-item scale
with items including: “I bought things even though I couldn't afford them,” and “I felt others
would be horrified if they knew of my spending habits,” ranging from (1) never/strongly
disagree to (5) very often/strongly agree. We used the average score as an indicator of
compulsive buying tendency (M = 1.87, SD = .58; α = 0.7) with higher scores indicating higher
buying compulsivity. Finally, in order to verify that participants in the friendship reminder
condition felt close to the friend, they responded to: “How close is the bond between you and the
friend you wrote about?” and “How close are you to this friend?” Both items ranged from (1) not
close at all to (7) very close and were averaged (r = 0.8) into an indicator of friendship closeness.
3.3. Dependent measure
The main variable of interest was participants’ ability to exert self-control, measured by
the number of failures to inhibit prepotent approach responses towards attractive clothing items
on a buying temptations-specific Go/No-go task. The Go/No-go task is an established measure of
self-control exertion in the consumption domain (Batterink et al., 2010; Mishra & Mishra, 2010)
and requires participants to perform a speeded behavioral response on Go trials (by pushing the /
key), while inhibiting a behavioral response on No-go trials (by refraining from pushing the /
key). While Go trials expose participants to neutral stimuli, No-Go trials typically involve
exposing participants to tempting stimuli, for which inhibiting the approach response is relatively
effortful. To be able to reliably assess inhibitory control, a large number of trials is required, with
a large majority of Go (rather than No-go) trials. This standard procedure warrants that
behavioural responses on the Go trials become maximally incongruent with the inhibition of
those responses on the No-Go trials, which increases the sensitivity of the task to pick-up on
inhibition failures. The much smaller number of No-Go trials prevents the inhibition response
itself to become habitual and hence prepotent. To be comparable to the task used in the fMRI
study (Study 5), we followed Batterink et al. (2010) who, for their fMRI study, used 96 trials
with three times more Go than No-go trials. Hence, in the present study, 75% of our trials were
Go-trials where participants responded to pictures of neutral products of a typically non-hedonic,
non-branded, non-luxurious, utilitarian nature. For these we followed Dittmar and Drury (2000)
who identified basic, non-branded furniture as satisfying these criteria. In contrast, the remaining
25% were No-go trials, consisting of pictures of buying temptations of a more hedonic, luxury
nature (for which inhibiting the approach response is likely more effortful, see Wagner &
Heatherton, 2015). Following Dittmar and Drury (2000) we selected clothing items like dresses,
jewelry and high-heeled shoes as products that typically possess these attributes. By refraining
from responding in the No-go trials, participants thus had to inhibit the impulse to respond to
attractive clothing stimuli, thus exerting self-control. We expected compulsive buyers to have
greater difficulties with this than non-compulsive buyers, given their problems with chronic
impulse inhibition (Ridgway et al., 2008).
Trials were presented for 500 ms. Each trial was followed by a 500 ms inter-trial interval,
showing a fixation cross that appeared directly after the participant’s response. In case of non-
response, the fixation cross appeared automatically 750 ms after stimulus presentation. Trials
were presented in pseudo-randomized order, such that pictures of clothes appeared after one, two
or three pictures of furniture. The total number of failures to inhibit responses to No-go
(clothing) items served as our measure of self-control, with lower numbers indicating a greater
ability to inhibit approach responses to buying temptations. That is, a greater capacity to exert
3.4. Results and discussion
3.4.1. Friendship closeness
An one-sample t-test using the scale's mid-point (i.e., 4.0) as a benchmark confirmed that
participants in the friendship condition indeed wrote about a strong friendship with a close friend
(M = 5.83; SD = 0.85; t(38) = 13.42, p < .001).
3.4.2. Inhibition failures
After standardizing the compulsive buying tendency measure, we performed a regression
analysis (Hayes 2013; Process Model 1) on participants’ total number of failures of inhibition
with the friendship reminder procedure (0 = control, 1 = friendship), compulsive buying
tendency, and their interaction as predictors. This analysis showed a significant main effect for
compulsive buying tendency (B = 2.14, SE = 0.75, t(65) = 2.85, p < .01), such that higher
compulsivity yielded more inhibition failures, thus aligning the present findings with previous
research on buying compulsivity (Ridgway et al., 2008). There was no main effect of the
friendship reminder procedure (B = 0.10, SE = 0.69, t < 1, NS). More importantly, the expected
crossover interaction between the friendship reminder procedure and compulsive buying
tendency was significant (B = -2.55, SE = 0.84, t(65) = -3.05, p < .01). Additional simple slope
analyses to probe the interaction indicated that –in line with predictions– for non-compulsive
buyers (evaluated at -1 SD from the mean), the friendship reminder increased the number of
inhibition failures made compared to the control condition and hence decreased self-control (B =
2.65, SE = 0.98, t(65) = 2.70, p < .01), thus replicating past research. In contrast, and as
predicted, for compulsive buyers (evaluated at +1 SD from the mean), the friendship reminder
decreased the number of inhibition failures made compared to the control condition and hence
increased self-control (B = -2.46, SE = 1.17, t(65) = -2.09, p < .05; Johnson-Neyman values: -
0.55, 0.91, respectively; see Figure 1)
Study 1 showed that compulsive buying tendency is positively related to failures of
inhibition, supporting our suggestion that compulsive buyers show impaired inhibition of
prepotent approach responses and hence are relatively low in self-control when exposed to
buying temptations, i.e., attractive clothing items in the present case. Importantly, however,
when compulsive buyers were asked to remember a situation where they experienced a strong
friendship with a close friend, their ability to inhibit these approach responses to buying
temptations increased. That is, reminding compulsive buyers of friendship enhanced their
capacity to exert self-control. In contrast, and in line with our expectations based on existing
findings, friendship reminders decreased the self-control of ‘regular,’ non-compulsive buyers.
Hence, Study 1 provides evidence in favor of our hypothesis that in the face of consumption
temptations, friendship reminders decrease self-control for non-compulsive buyers, but increase
self-control for consumers classified as compulsive buyers.
Figure 1. Failures of inhibition as a function of friendship reminder and compulsive buying tendency
Using average response latencies on the No-go trials as dependent variable yielded similar results. The expected crossover interaction between
the friendship reminder procedure and compulsive buying tendency was significant (B = -0.04, SE = 0.01, t(63) = -3.33, p < .01). Non-
compulsive buyers (-1 SD from the mean) showed higher response latencies in the friendship reminder (M = 0.09 sec) compared to the control
condition (M = 0.04 sec; B = 0.05, SE = 0.01, t(63) = 3.19, p < .01), suggesting lower self-control when reminded of friendship. Conversely,
compulsive buyers (+1 SD from the mean) showed lower response latencies in the friendship reminder (M = 0.07 sec) compared to the control
condition (M = 0.10 sec; B = -0.04, SE = 0.02, t(63) = -2.08, p < .05), suggesting higher self-control when reminded of friendship.
Non-compulsive buyers (-1 SD) Compulsive buyers (+1 SD)
4. Study 2
Study 2 was designed to examine whether the proposed effect is indeed a function of
reminders of friendship, excluding mere acquaintances or generally positive circumstances as
alternate explanations for the effect. Moreover, while the previous study focused on the ability to
inhibit a prepotent approach response towards buying temptations, we extended the previous
findings by focusing on the impulse-strength side of the “tug-of-war” between the opposing
forces of self-control and assessed experienced acute purchase desire.
4.1.1. Participants and design
One hundred thirty-four students (59.7% male; mean age 19.76 years, SD = 1.93)
participated in return for partial course credit. The study employed a single factor (friendship
reminder: close friend vs. acquaintance vs. positive event) between-participants design with
compulsive buying tendency as a continuous moderator.
After signing an informed-consent form, participants were randomly assigned to one of
three conditions. All participants read a hypothetical shopping scenario adapted from Rook and
Fisher (1995). They were asked to imagine a situation in which they went to a department store
to purchase a pair of socks but once in the store saw a great looking pair of jeans on sale.
Participants in the close friend condition were asked to imagine being accompanied by a close
friend while shopping, participants in the acquaintance condition by an acquaintance and
participants in the non-social, positive event condition were asked to imagine that they had just
heard that they passed an exam with a good grade. More specifically, in the close friend and
acquaintance conditions, the scenarios read:
“Imagine you are a college student with a part-time job. You’ve got an
outdoor party this weekend and still need to buy a pair of warm socks.
After work, you go together with a close friend [vague acquaintance] to
a department store to purchase the socks. As you and your close friend
[vague acquaintance] are walking through the store, you see a great
looking jeans on sale. You’ve got already sufficient jeans, but it’s your
favorite brand and it’s a good offer.”
In the positive event condition the scenario read:
“Imagine you are a college student with a part-time job. You’ve got an
outdoor party this weekend and still need to buy a pair of warm socks.
After work, you hear that you passed an important exam with a good
grade. You go to a department store to purchase the socks. As you are
walking through the store, you see a great looking jeans on sale.
You’ve got already sufficient jeans, but it’s your favorite brand and it’s
a good offer.”
We then measured purchase desire by: “I felt a sudden urge to buy” (1 = not at all to 7 = very
much). In this and the next studies, we used a different scale to measure compulsive buying
tendency than in the previous study to increase the generalizability and robustness of our
findings. Hence, participants next completed the 6 item Compulsive Buying Scale (CBS;
Ridgway et al., 2008), using a 7-point scale. In this, and the subsequently reported studies we
used the sum score (M= 20.97, SD = 6.65; α = 0.7) as an indicator of compulsive buying
tendency with higher scores indicating higher buying compulsivity.
4.2. Results and discussion
4.2.1. Purchase desire
After standardizing the compulsive buying measure, we performed a regression analysis
(Hayes & Preacher, 2014; Process Model 1) on purchase desire with the friendship reminder
procedure (dummy coded, positive event condition being the reference category), compulsive
buying tendency, and their interaction as predictors. Although there was no main effect of the
friendship reminder procedure (Bacq = 0.16, SE = 0.24, t(128)acq = 0.65, NS; Bfriend = -0.19, SE =
0.25, t(128) friend = -0.77, NS), the impact of compulsive buying tendency was significant (B =
1.19, SE = 0.17, t(128) = 6.99, p < .0001), indicating that higher levels of compulsive buying
yielded stronger acute purchase desires.
More importantly, while the interaction between the presence of a friend (vs.
acquaintance) and compulsive buying tendency did not reach significance (B = -0.36, SE = 0.26,
t(128) = -1.39, NS), we did find significant interaction effects between the presence of a friend
(vs. positive event) and compulsive buying tendency (B = -0.96, SE = 0.26, t(128) = -3.65, p <
.001) and between the presence of an acquaintance (vs. positive event) and compulsive buying
tendency (B = -0.60, SE = 0.23, t(128) = -2.54, p < .05).
In line with our expectations and previous research, for non-compulsive buyers (-1 SD
from the mean) the presence of a friend increased acute purchase desire relative to the positive
event condition (B = -0.77, SE = 0.38, t(128) = 2.03, p < .05). The difference between the
acquaintance and positive event condition was also significant (B = 0.76, SE = 0.33, t(128) =
2.33, p < .05), while the difference between friend and acquaintance was not (t < 1). Conversely,
for compulsive buyers (+1 SD from the mean) the presence of a close friend reduced acute
purchase desire relative to the positive event (B = -1.16, SE = 0.35, t(128) = -3.34, p < .05) and
acquaintance condition (B = -0.71, SE = 0.35, t(128) = -2.01, p < .05). Moreover, as proposed,
the presence of an acquaintance did not reduce purchase desire relative to the positive event
condition (B = -0.45, SE = 0.35, t(128) = -1.26, NS; see Figure 2).
Figure 2. Purchase desire as function of friendship reminder and compulsive buying tendency
Study 2 shows that for compulsive buyers the presence of a close friend reduces purchase
desire compared to when they are in the presence of an acquaintance or under non-social but
positive circumstances. The presence of an acquaintance did not produce the same effect as a
close friend for compulsive buyers, confirming the importance of friendship reminders rather
than social relationships per se. Also in line with our expectations, the presence of a friend did
not reduce, but enhanced acute purchase desire for non-compulsive buyers compared to the
Non-compulsive buyers (-1 SD) Compulsive buyers (+ 1 SD)
positive event condition, although the present study did not observe a difference between friends
and acquaintances for non-compulsive buyers.
5. Study 3
Study 3 extended on the friendship reminder manipulation by assessing self-control
exertion in the real physical presence of a close friend. To rule out that any effects could be
attributable to the mere presence of any person, rather than a close friend, we included a
condition with the physical presence of a stranger as a control condition. Compared to being
alone, we expected to replicate the previous results that non-compulsive buyers would show
reduced and compulsive buyers enhanced self-control exertion in the company of a close friend.
In addition, the present and following studies extend the previous studies by making the
connection with the presumed underlying process: the extent of conflict identification underlying
the exertion of self-control. To this end, we relied on an established paradigm in the present
study. More specifically, to assess whether the postulated effects of friendship reminders and
buying compulsivity are limited to, or might extend beyond, the product purchase domain, we
used a more generic well-established task assessing intentional self-control exertion as a function
of conflict identification —a Color Word Stroop Task (Fennis, Janssen, & Vohs, 2009; Muraven,
Shmueli, & Burkley, 2006; Webb & Sheeran, 2003; see next for details). If compulsive buying is
indeed typically part of a more general obsessive-compulsive disorder, involving impairment in
inhibiting prepotent responses across multiple domains (DSM-V, 2013; Müller et al., 2015), and
if for these consumers friendship reminders indeed promote perceptions of conflict in self-
control situations, then we should observe facilitated conflict detection in the presence of a close
friend, even on a task that reaches beyond the buying sphere. Please note that such facilitated
conflict detection should translate in faster response latencies (i.e., reduced interference) without
making more errors on tasks requiring conflict identification (Harris, Harris, & Miles, 2017)”
5.1.1. Participants and design
One hundred thirty undergraduate female students (mean age: 19.35 year, SD = 1.71)
participated in the present study in return for money or partial course credit. Only female
participants were recruited for the same reasons as Study 1. The experiment employed a single
factor (friendship reminder: alone vs. stranger vs. close friend) between-participants design with
compulsive buying tendency as a continuous moderator.
Participants were asked to bring a close, same-sex friend to the lab. Pairs of friends
randomly assigned to the alone condition were split and seated alone in a room, whereas pairs of
friends assigned to the close friend-condition were seated together. Pairs of friends randomly
assigned to the stranger condition were split and combined with an unacquainted participant
from another pair of friends. All participants were seated behind a computer, with pairs of
participants seated opposite to each other, not able to watch each other’s responses. Participants
were instructed to not talk to each other. These procedures warranted that any effects found were
attributable to the mere physical presence of a close friend (rather than anyone) and not a
function of the other person acting as an exemplar, inducing behavioral contagion, or a source of
Instructions on the computer screens told participants they would participate in several
unrelated tasks. The first task comprised our dependent variable of interest, participants’ ability
to implement self-control as measured by a Color Word Stroop Task (Fennis et al., 2009;
Muraven et al., 2006; Webb & Sheeran, 2003). Participants received 32 randomized trials. On
each trial a stimulus word was presented in a font color that was either congruent or incongruent
with its semantic meaning, e.g., the word ‘blue’ presented in blue font (congruent), or in red font
(incongruent). Of all trials 8 were congruent and 24 were incongruent. Participants were required
to identify the font color in which the words were printed, by clicking as quickly as possible one
of four buttons on their computer screen. The task presents participants with a response conflict
between font color and task instruction on the incongruent trials only. Responding to the
incongruent trials, one thus has to inhibit one’s impulse to choose the semantic meaning of the
word, rather than its font color, requiring conflict identification and intentional self-control
In line with previous research, we used the average response latencies on the incongruent
trials as our main dependent measure of self-control exertion as a function of conflict
identification, with lower response latencies indicating greater self-control exertion ability
through improved conflict detection (Fennis et al., 2009; Harris et al., 2017; Muraven et al.,
2006; Webb & Sheeran, 2003). Following the Stroop task, participants completed the CBS (M =
24.25, SD = 6.67; α = 0.8; Ridgway et al., 2008). Last, as a manipulation check, participants in
the friend and stranger conditions indicated how long (in months) and how well they knew the
other person in the room (1 = not at all to 7 = very well).
5.2. Results and discussion
5.2.1. Manipulation check
Confirming the effectiveness of our manipulation, participants in the friend condition
knew the other person in the room longer (M = 47.9 months, SD = 42.72) than participants in the
stranger condition (M = 2.6 months, SD = 11.49; t(83) = -6.71, p < .001). They also indicated that
they knew the other person better (Mfriend = 5.48, SD = 1.35 versus Mstranger = 1.47, SD = 1.28;
t(83) = -14.08, p < .001).
5.2.2. Self-control exertion
After log transforming the response latencies to account for non-normality, we performed
a regression analysis on the average response latencies with the friendship reminder procedure
(dummy coded, alone being the reference category), compulsive buying tendency (standardized),
and their interaction as predictors
. There were no main effects of the friendship reminder
procedure (Bstranger = 0.05, SEstranger = 0.03, t(124)stranger = 1.33, NS; Bfriend = -0.01, SEfriend = 0.03,
tfriend < 1, NS) and compulsive buying tendency (B = 0.04, SE = 0.03, t(124) = 1.39, NS). More
importantly, replicating our previous findings, we found the predicted crossover interaction
between the presence of a friend and compulsive buying tendency (B = -0.09, SE = 0.04, t(124) =
-2.31, p < .05). For non-compulsive buyers (-1 SD from the mean), the pattern of results was in
the predicted direction, although the simple slopes analyses did not reach significance (see
When using the total number of inhibition failures on the incongruent trials as dependent variable, there were no main effects of the friendship
reminder procedure (all t’s < -1.01) nor of compulsive buying tendency ( t(124) = -1.63, NS). There were also no interaction effects (all t’s <
1.59) Please note that compulsive buyers’ faster responding without them making more errors (i.e., inhibition failures on incongruent trials) in the
presence of a friend indicates a net increase in self-control performance (see Harris et al., 2017).
Figure 3). That is, the presence of a close friend tended to increase response latencies relative to
being alone (B = 0.07, SE = 0.05, t(124) = 1.42, NS), as did the presence of a stranger (B = 0.07,
SE = 0.05, t(124) = 1.48, NS). In contrast, for compulsive buyers (+1 SD from the mean) the
presence of a close friend significantly reduced response latencies relative to being alone (B = -
0.10, SE = 0.05, t(124) = -2.00, p < .05) and with a stranger (B = -0.12, SE = 0.05, t(124) = -2.54,
p < .05). The presence of a stranger did not reduce response latencies relative to being alone (B =
0.02, SE = 0.05, t < 1, NS).
Figure 3. Average response latencies on Stroop task as function of friendship reminder and compulsive buying tendency
Study 3 shows that for compulsive buyers the mere, physical presence of a close friend
reduces response latencies on a Stroop Task, a well-established indicator of self-control exertion,
relative to being with a stranger or alone. The presence of a stranger did not produce the same
effect as a close friend for compulsive buyers, confirming the importance of friendship reminders
rather than any social presence per se in the current theorizing. These results support our
reasoning that friendship reminders engender increased perceptions of conflict in self-control
situations, enabling compulsive buyers to display restraint and control in the real or imagined
presence of a close friend, even in domains unrelated to buying and shopping. In addition, our
results dovetail nicely with previous research that has established that Stroop performance is
boosted when participants are reminded of important personal goals and values (see Harris et al.,
2017). The present findings imply that the improved self-control ability of compulsive buyers
reminded of friendship may extend to other spheres of life beyond the purchasing domain. This
Non-compulsive buyers (-1 SD) Compulsive buyers (+1 SD)
is in line with the observation that buying compulsivity is frequently part of a more general
obsessive-compulsive disorder, involving a chronic, more generic, domain unspecific
impairment in inhibiting prepotent responses across multiple domains (DSM-V, 2013; Müller et
al., 2015). As such, the present findings also attest to the potential of research in the marketing
and consumer behavior domains to inform neighboring fields and disciplines.
6. Study 4
The previous study demonstrated improved self-control for compulsive buyers reminded
of friendship using a Stroop Task. This task bridges the previous three studies that focused on
self-control exertion with the next two that zoom-in on the role of the main concept proposed to
drive our effects —conflict identification. That is, in addition to being a measure of self-control
exertion, performance on a Stroop Task is typically measured on the incongruent trials, which
represent a response conflict. Indeed, research in neuroscience has shown that better (i.e., faster)
Stroop performance is associated with higher ACC activation, a brain area associated with
conflict identification (Inzlicht & Gutsell, 2007; Kerns et al., 2004). Hence, in the following two
experiments we will directly address the role of (goal) conflict identification in the relationship
between friendship reminders, compulsive buying tendency and self-control. In Study 4 we test
our postulate that friendship reminders will induce a higher (lower) sense of conflict between
indulging in buying temptations on the one hand and the goal of maintaining satisfactory
friendship relations on the other for compulsive (non-compulsive) buyers. Moreover, in this
study we directly test the proposed mediating role of goal conflict identification in the
relationship between friendship reminders, buyer compulsivity and self-control exertion
identified in Studies 1-3. Hence, not only do we aim to replicate the crossover interaction we
systematically observed in the previous studies, we also test the proposed moderated mediation
model that may account for our effects. In this study we also oscillate back to the impulse
strength side of the self-control tug-of-war. In so doing, we included another, more elaborate,
measure of purchase desire further corroborating the converging support for, as well as the
marketing relevance of, our notions.
6.1.1. Participants and design
Sixty-four undergraduate students (51.6% male; mean age 19.70 years, SD = 3.18)
participated for partial course credit. The experiment employed a single factor (friendship
reminder: close friend vs. acquaintance) between-participants design with compulsive buying
tendency as a continuous individual difference variable.
Participants were randomly assigned to one of two conditions. All participants read the
same hypothetical shopping scenario as used in Study 2. Participants in the close friend condition
were with a close friend, whereas participants in the acquaintance condition were with an
acquaintance. We then asked participants to list the thoughts they had while imagining the
shopping situation. Two coders blind to experimental hypotheses categorized the thought listings
as either or not containing references to the identification of a goal conflict (inter-rater
agreement: 94%; all disagreement between coders was resolved by consensus), thus creating a
dichotomous measure of goal conflict identification (0 = absent vs. 1 = present) per participant.
Examples of references to a sense of conflict were: “I’d feel guilty” and “I’m aware that Iʼm
giving up something else in return”. We added another, more direct self-report measure of the
extent of goal conflict participants perceived between giving in to the consumption temptation
(i.e., buying the jeans) and maintaining the relationship with their friend/acquaintance, by asking
participants: “What do you think that your friend [acquaintance] would think of it if you would
buy the jeans?”, ranging from very positive (1) to very negative (7). We measured participants
desire to purchase by: “I experienced a sudden urge to buy”, “I had a strong urge to buy
impulsively”, and “I felt a sudden urge to buy” (1 = not at all to 7 = very much; α = 0.9). Before
participants finished the experiment by completing the CBS (M = 22.50, SD = 6.66; α = 0.7;
Ridgway et al., 2008), they indicated how well they were able to imagine the described scenario
(1 = not at all to 7 = very much).
6.2. Results and discussion
6.2.1. Manipulation check
An independent-samples t-test showed that participants in the close friend (M = 5.77, SD
= 0.96) and acquaintance (M = 5.45, SD = 0.97) condition imagined the described scenario
equally well (t(62) = -1.33, NS).
6.2.2. Goal conflict identification
A logistic regression analysis on the dichotomous goal conflict identification measure
with the friendship reminder procedure (0 = acquaintance, 1 = friend), compulsive buying
tendency (standardized), and their interaction as predictors revealed no significant main effect of
compulsive buying tendency (Wald < 1), but did reveal a significant main effect of friendship
reminder procedure (B = 2.47, SE = 0.69, Wald = 12.94, Exp(B) = 11.79, p < .0001), indicating
higher odds of perceiving a goal conflict with a friend than with an acquaintance. Furthermore,
the interaction effect was also significant (B = 1.80, SE = 0.77, Wald = 9.36, Exp(B) = 6.06, p <
.05), indicating that with a friend (relative to an acquaintance), a decrease in compulsive buying
tendency decreased the odds of identifying a goal conflict. Also in line with our expectations, the
opposite holds as well: with a friend relative to an acquaintance, an increase in compulsive
buying tendency increased the odds of identifying a goal conflict.
We furthermore performed a regression analysis (Hayes, 2013; Process Model 1) on the
self-report measure of the extent of perceived goal conflict with the friendship reminder
procedure, the standardized compulsive buying tendency, and their interaction as predictors,
showing no main effects of the friendship reminder procedure (B = 0.37, SE = 0.36, t(60) = 1.03,
NS) and compulsive buying tendency (B = -0.002, SE = 0.26, t(60) = -0.07, NS). In line with our
predictions, there was a significant crossover interaction between the friendship reminder
procedure and compulsive buying tendency (B = 0.90, SE = 0.37, t(60) = 2.46, p < .05). For non-
compulsive buyers (-1 SD from the mean), although in the expected direction, the difference
between the presence of a friend and acquaintance did not reach significance (B = -0.53, SE =
0.52, t(60) = -1.02, NS). In contrast, and as predicted, compulsive buyers (+1 SD from the mean)
who were in the presence of a friend showed a higher extent of goal conflict identification than
those in the presence of an acquaintance (B = 1.27, SE = 0.51, t(60) = 2.48, p < .05; Johnson-
Neyman value 0.48).
6.2.3. Purchase desire
We performed a regression analysis (Hayes, 2013; Process Model 1) on purchase desire
with the friendship reminder procedure, the standardized compulsive buying tendency, and their
interaction as predictors, showing no main effect of the friendship reminder procedure (B = -
0.003, SE = 0.27, t(60) = -0.01, NS), while the impact of compulsive buying tendency was
significant (B = 1.13, SE = 0.19, t(60) = 5.91, p < .001), indicating that higher levels of
compulsive buying tendency yielded higher purchase desire. In line with our hypotheses, we
replicated our basic finding as the crossover interaction between the friendship reminder
procedure and compulsive buying tendency was also significant (B = -0.79, SE = 0.27, t(60) = -
2.89), p < .01). Confirming our hypotheses, for non-compulsive buyers (-1 SD from the mean),
the presence of a friend increased purchase desire relative to the presence of an acquaintance (B
= 0.78, SE = 0.38, t(60) = 2.04, p < .05). Conversely, for compulsive buyers (+1 SD from the
mean) the presence of a friend decreased purchase desire relative to the presence of an
acquaintance (B = -0.79, SE = 0.38, t(60) = -2.07, p < .05; Johnson-Neyman moderator values -
0.96, 0.94 respectively; see Figure 5).
Figures 4-5. Perceived goal conflict and purchase desire as function of friendship reminder and compulsive buying
6.2.4. Moderated mediation
To assess whether the impact of the interaction between friendship reminders and
compulsive buying tendency on purchase desire is indeed mediated by the extent of identified
goal conflict for more compulsive consumers as we hypothesize, we performed a moderated
mediation analysis (Hayes, 2013; Muller, Judd, & Yzerbyt 2005; Process Model 8) using the
continuous measure of perceived goal conflict as proposed mediator
. A bootstrapping procedure
with 5000 re-samples showed a significant overall index of moderated mediation, as the 95%
confidence interval of the indirect effect did not include zero: [0.005, 0.51]. Further analysis
showed that, in line with the non-significant findings on (self-reported) perceived goal conflict
for non-compulsive buyers (at -1 SD from the mean), the mediation analysis did not reach
significance for non-compulsive buyers [-0.47, 0.08]. In contrast, for compulsive buyers (at + 1
SD of the mean), the impact of the friendship reminder on purchase desire was indeed mediated
by the extent of goal conflict identification as the confidence interval did not include zero [0.004,
These findings provided evidence for our key postulates that reminders of friendship in
the context of buying temptations are more likely to induce the identification of a goal conflict
for compulsive than for non-compulsive buyers. We again found the predicted crossover
interaction, not only on our measure of self-control exertion (experienced acute purchase desire),
but also on the measures of goal conflict identification. Moreover, for compulsive buyers
perceived goal conflict mediated the impact of the friendship reminder on purchase desire.
Hence, the results of Study 4 support our hypothesis that the extent of goal conflict identification
between giving in to buying temptations on the one hand and the perceived consequences for
We selected the continuous measure of perceived goal conflict since the dichotomous goal conflict index cannot function in a mediating role in
Process’ (Hayes, 2013) multiple regression-based approach.
maintaining satisfactory relationships with close friends on the other, drives the impact of
friendship reminders and buying compulsiveness on self-control.
7. Study 5
The goal of Study 5 was twofold. First, to study the neural correlates of friendship
reminders on self-control exertion, thereby conceptually replicating the pattern of findings from
the previous four studies. And second, to build upon Study 4 to find converging evidence for the
role of goal conflict identification as the main driver underlying the effect of friendship
reminders for (non-)compulsive consumers on self-control exertion. As such, this and the
previous study replicate and extend the results of the previous three studies by showing how the
extent of conflict identification underlies the crossover interaction between friendship reminders
and compulsive buying tendency on self-control exertion. Moreover, the present study not only
complements Study 4 in its relatively unique fMRI approach, but also in revisiting the other end
of the self-control “tug-of-war” —the motivation and ability to restrain an acute impulse, i.e., the
ability to inhibit a prepotent approach response toward buying temptations.
Using fMRI, conflict identification and control exertion can reliably be dissociated from
each other as they rely on two separate neural systems (MacDonald et al., 2000). While
consumers’ conflict identification is closely associated with activation in the anterior cingulate
cortex (ACC; BA’s 24/32), their ability to inhibit prepotent responses and thus exert self-control
is associated with the dorsolateral prefrontal cortex (DLPFC; BA 9; Batterink et al., 2010;
MacDonald et al., 2000; Simmonds et al., 2008). More specifically, the ACC or ‘conflict
detection center’ serves a general role in identifying conflicting situations. It is known to detect
incompatibilities in both the cognitive and social domain (Lieberman & Eisenberger, 2005;
Robinson, Schmeichel, & Inzlicht, 2010). Moreover, several authors have suggested that the
ACC not only plays a role in cognitive control but in self-control as well, affecting people’s
ability to identify possible goal conflicts between short term gratifications on the one hand, and
the successful pursuit of valued long-term goals on the other (Goldberg, 2001; Hedgcock et al.,
2012; Robinson et al., 2010). In contrast, the DLPFC is recognized as the brain area engaged
with the actual implementation of control, both cognitive control as well as self-control
(Goldberg 2001). It has been linked to the successful inhibition of approach responses to No-go
trials in Go/No-go tasks (Batterink et al., 2010; Simmonds et al., 2008) and to choices that favor
long-term goals over short-term gratifications (McClure, Botvinick, Yeung, Greene, & Cohen,
In sum, conflict identification is associated with increased ACC activation, while failures
of conflict identification have been associated with reduced activation in the ACC (Inzlicht &
Gutsell, 2007; MacDonald et al., 2000). Furthermore, successful inhibition of prepotent
responses is associated with DLPFC activation (Batterink et al., 2010; Knoch & Fehr, 2007;
Liddle, Kiehl, & Smith, 2001; Simmonds et al., 2008), while failures of response inhibition have
been associated with reduced activation in the DLPFC (Hare, Camerer, & Rangel, 2009;
Hedgcock et al., 2012; Liddle et al., 2001; MacDonald et al., 2000; Menon, Adleman, White,
Glover, & Reiss, 2001). Hence, translated to the issue under consideration, we hypothesized that
non-compulsive buyers when faced with buying temptations would show decreased ACC and
DLPFC activation, indicating reduced goal conflict identification and prepotent response
inhibition. In contrast, compulsive buyers should show the opposite and hence would show
increased ACC and DLPFC activation under these conditions, indicating increased goal conflict
identification and successful approach response inhibition towards buying temptations.
7.1. Participants and design
After approval by the medical ethical committee, forty female undergraduate students
(mean age 20.35 years, SD = 1.82) participated in the fMRI experiment, receiving 20 Euros for
their participation. We screened all participants to ensure they could safely undergo the
procedure in the MRI scanner, were right-handed, had normal or corrected-to-normal vision, and
were using the contraceptive pill, to control for idiosyncratic differences in neural responses
throughout the menstrual cycle (Dreher et al., 2007). All participants were native in the language
the study was conducted in and indicated a preference for buying clothing items while shopping.
Using a Go/No-go task procedure in line with Study 1, the experiment employed a
repeated measures design, with friendship (not reminded vs. reminded of friendship) and product
type (neutral vs. tempting) as within-subjects factors and compulsive buying tendency as a
continuous moderator. Friendship reminder and product type were manipulated whereas
compulsive buying was measured. Similar to Study 1, we used basic furniture as neutral products
on Go trials and clothing items as our tempting products on No-go trials.
Before the fMRI session, all participants provided written informed consent. Participants
were placed inside the MRI scanner with their head securely placed in a head coil to prevent
excessive movement. Stimuli were projected on a screen outside the scanner, visible to
participants via a mirror located above their eyes. Participants could respond with their right
hand, pushing a button on a response box.
During the fMRI scan, which lasted for approximately one hour, participants took part in
four sessions. They started with two subsequent sessions in which they performed two Go/No-go
tasks (a target and a control version). Afterwards, they proceeded with a friendship reminder
procedure, before taking part in another two sessions performing two Go/No-go tasks (target and
control). After participation in all four Go/No-go tasks, participants were helped out of the
scanner and indicated their willingness to pay for a random sample of the clothing and furniture
items, functioning as manipulation checks for the desirability of both product categories. As a
manipulation check for the friendship reminder procedure, they indicated their agreement with:
“During my description of the situation in which I felt strong friendship, I re-experienced the
bond with this friend,” ranging from (1) totally disagree to (7) totally agree. We lack data of one
participant on this variable. Participants then completed the CBS (M = 21.05, SD = 6.83; α =
0.8; Ridgway et al., 2008).
7.3. Friendship reminder procedure
The friendship reminder procedure was similar to Study 1. Because of the (physical)
limitations of the scanner, participants now orally described a situation in which they felt strong
friendship. They were instructed to use the following words: we, together, friendship and bond
and to keep telling about the friendship experience for five minutes. To encourage participants to
describe a real and personal friendship, it was stressed that they would not be recorded and could
only be heard by the experimenters. If participants finished their description before five minutes
had passed, they were asked to keep thinking about the friendship.
7.4. Go/No-go tasks
Participants performed Go/No-go tasks that were similar to the Go/No-go task of Study 1
in terms of instructions and nature of stimuli, but were adapted to the needs of fMRI data
acquisition and analysis in terms of timing and number of stimuli (Wager, Sylvester, Ching-
Yune, Lacey, Nee, Franklin, & John, 2005). More specifically, we adopted a mixed block/event-
related design following Petersen et al. (2012) and Visscher et al. (2003). This type of design has
been demonstrated to be optimal for Go/No-go tasks in fMRI (Vissher et al., 2003) as it allows
for the simultaneous modeling of the transient, trial-related activity and the sustained, block-
related BOLD activity (Peterson et al., 2012).
All participants participated twice in both a target and a control Go/No-go task, requiring
speeded responses on Go trials and response inhibition to No-go trials. In both the target and
control Go/No-go task, 80 pictures (trials) were presented in randomized order. As in Study 1, in
the target version of the Go/No-go task, the Go trials (70% of all trials) consisted of pictures of
neutral products (i.e., basic, non-luxurious furniture), whereas the No-go trials (30% of all trials)
consisted of pictures of tempting products (attractive women’s clothing, like dresses and high-
heeled shoes). In the control version of the Go/No-go task the Go and No-go stimuli were
mirrored compared to the target version (i.e., the Go trials (70%) consisted of clothes and the
No-go trials of basic furniture). To control for potential confounds, half of the participants
(randomly assigned) started the fMRI session with the target version and continued with the
control version of the Go/No-go task (path A, see Figure 6). The other half started with the
control version and continued with the target version (path B, see Figure 6). After the friendship
reminder, participants again participated in a target and control version of the Go/No-go tasks,
similar to the two Go/No-go tasks before.
Figure 6. Sequence of the four Go/No-go tasks and friendship reminder procedure dependent on path A or B
All pictures were presented on a light grey background, using E-prime 2.0. Each Go/No-
go task consisted of 10 trial-blocks and 11 fixation-blocks. Each trial-block consisted of eight
(randomly presented Go or No-go) trials (500 ms) alternated by eight inter-trial intervals
consisting of a fixation cross (2080 ms). The fixation blocks showing a fixation cross (20 sec),
served to start and finish the Go/No-go tasks and to separate the 10 trial-blocks from each other.
No additional jittering was applied.
7.5. fMRI data acquisition and preprocessing
Functional data were collected using a Philips 3T Intera MRI scanner with a 32 channel
head coil. We first recorded a T1-weighted anatomical image (170 axial slices, voxel size 1 x 0.6
x 1 mm) for each participant, providing a high-resolution image of the anatomy of the brain over
which the functional data would be overlaid (Yoon et al., 2006). Following the anatomical
image, functional data were recorded based on Echo Planar Imaging (39 axial slices, TE = 30
ms, TR = 2000 ms, FA = 70 degrees, FOV = 224 x 136.5 x 224 mm with in-plane resolution of
3.5 x 3.5 and slice thickness of 3.5 mm). For each participant four runs of functional data were
recorded (one for each Go/No-go task), each run comprising 215 volumes and lasting circa 7
minutes. Data preprocessing was performed using FSL version 5.0 (Jenkinson, Beckmann,
Behrens, Woolrich, & Smith, 2012), correcting for the following types of task-unrelated
variability: motion correction (using MCFLIRT; Jenkinson et al., 2012), non-brain removal
(Brain Extraction Tool), spatial smoothing (kernel size 10 mm FWHM), and high-pass filtering
(set at 100 sec, to filter out scanner drifts). Spatial normalization (i.e., correcting for variability in
brain size and shape by mapping the data on a ‘standard’ brain template; Yoon et al., 2006) was
performed by registering the anatomical image of participants to a MNI–152 stereotactic
template, using an affine procedure with a 12 parameter fit. The data of two participants required
further normalization using nonlinear registration (FNIRT).
7.6. fMRI data analysis
Statistical analyses were performed in three stages. During the first stage, general linear
models were built and convolved with a single gamma function to compensate for the
haemodynamic delay. In line with our mixed block/event-related design, four Fixed Effects
General Linear Models (FFX GLM) were built per participant: before and after the friendship
reminder for both product types. Each of these FFX GLMs comprised two regressors: The first
regressor represented the block condition for all stimuli presented during a trial-block, to capture
the sustained effects of the task at hand (Petersen et al., 2012). The second regressor was built
specifically using the onsets and durations of the No-go trials (lasting 2580 ms each), measuring
the inhibition (transient) effect on the No-go trials. Contrasting the second regressor versus the
fixation baseline allowed us to study the inhibition effects that are central to this research,
revealing which brain areas were activated during No-go trials above and beyond what was
activated during fixation baseline. Following Batterink et al. (2010) we choose to compare No-
go trials versus baseline, rather than No-go versus Go trials, to prevent the incorrect conclusion
that certain brain areas do not play a role in response inhibition because they are involved in
response selection as well (Simmonds, Pekar, & Mostofsky, 2008). Since we adopted a block
design we did not model response correctness.
In the second stage, we took the beta values of the second regressor of all four FFX
GLMs to a “within-subject between-sessions” FFX GLM (following Visscher et al., 2003 and
Petersen et al., 2012). This model thus only contained information about the transient effect from
each of the four sessions. Within this model, we calculated, per participant, main effects of
friendship reminder procedure and product type, and the interaction effect between friendship
reminder and product type using the following formula: [inhibition during clothing pictures after
the friendship reminder procedure – inhibition during clothing pictures before the friendship
reminder procedure] – [inhibition during furniture pictures after the friendship reminder
procedure – inhibition during furniture pictures before the friendship reminder procedure].
Since our predictions were limited to specific anatomical regions, we adopted a region-
of-interest (ROI) approach, testing our hypotheses only in those specific regions rather than
across the entire brain (Batterink et al., 2010; Dietvorst et al., 2009; Poldrack, 2007). We
constructed the ROIs using fslmaths, employing a sphere with a 5 mm radius in line with the
coordinates identified in previous research. Specifically, we used the coordinates defined by
Kerns et al. (2004) and MacDonald et al. (2000) for the ACC (transformed from Talairach to the
following MNI coordinates: x = 1, y = 8, z = 44, and x = 4, y = –1, z = 47 respectively) and the
coordinates defined by Batterink et al. (2010) and Simmonds et al. (2008) for the DLPFC (MNI
coordinates: x = -6, y = 30, z = 57; x = -3, y = 39, z = 51; x = 21, y = 48, z = 39 and transformed
from Talairach to the following MNI coordinates: x = 24, y = 50, z = 41 respectively; see Figure
Figure 7. Locations of the a priori ROIs
In stage three, the correlational analysis (Dietvorst et al., 2009), we calculated four beta
estimates per participant (one for each FFX GLM model), representing the average percentage
signal change within each ROI. Using these four betas we calculated the friendship reminder ×
product type interaction per participant using a mixed effects model in Matlab, and subsequently
examined their correlations with participants’ compulsive buying score. Significance was
determined at p < 0.05 (corrected for multiple comparison).
7.7. fMRI results and discussion
7.7.1. Manipulation checks
A paired-samples t-test showed that participants were on average willing to pay more for
the clothing items (M = 21.86 Euros; SD = 11.17) than for the furniture items (M = 16.79; SD =
14.45; t(40) = 2.68, p = .01). Using willingness to pay as an indicator of desirability, this result
suggests that we successfully manipulated a tempting versus neutral product category. A one-
sample t-test using the scale’s mid-point as a benchmark confirmed that participants had re-
experienced the bond with their friend during their description of the friendship experience (M =
5.56; SD = 1.02; t(38) = 9.57, p < .001). The friendship reminder procedure can thus be
7.7.2. Behavioral results
We performed a regression analysis (Judd, Kenny, & McClelland, 2001; Judd, Kenny,
& McClelland 2001; Montoya and Hayes, 2017; MEMORE, Model 2) on inhibition failures
during clothing trials with the friendship reminder procedure as within-subjects factor and the
standardized compulsive buying tendency as predictor, showing the predicted crossover
interaction between the friendship reminder procedure and compulsive buying tendency (B =
1.84, SE = 0.59, t(38) = 3.12, p < .01).
In line with our results from Study 1, non-compulsive buyers (-1 SD from the mean)
made more inhibition failures on clothing trials after (M = 16.31) compared to before the
friendship reminder (M = 14.45; B = -1.87, SE = 0.83, t(38) = -2.25, p < .05). In contrast,
compulsive buyers (+1 SD from the mean) showed the opposite pattern and made less inhibition
failures on clothing trials after (M = 13.59) than before being reminded of friendship (M = 15.40;
B = 1.82, SE = 0.82, t(38) = 2.19, p < .05; Johnson-Neyman moderator values -0.82, 0.87
When using response latencies as dependent variable the interaction between the friendship reminder procedure and compulsive buying
tendency did not reach significance (t < 1, NS).
7.7.3. Goal conflict identification
Correlation analysis showed that compulsive buying tendency was significantly
positively correlated with activation during inhibition (No-go) trials as a function of the
friendship reminder × product type interaction in the regions of interest previously implicated in
goal conflict identification (ACC, see Table 2)
. That is, the lower (higher) the compulsive
buying tendency, the lower (higher) the ACC activation when asked to inhibit responses towards
pictures of tempting clothing items (versus neutral furniture) after being reminded of friendship
(versus not being reminded of friendship). Hence non-compulsive buyers showed reduced ACC
activation compared to the average participant, while compulsive buyers showed increased ACC
activation compared to the average participant under these conditions.
For the main effects of the friendship reminder procedure and product type per ROI we kindly refer to Table 1.
(x, y, z)
1, 8, 44
Kerns et al.
4, -1, 47
-6, 30, 57
Batterink et al.
-3, 39, 51
Batterink et al.
Table 1. Results of Regions of Interest analyses: main effects of friendship reminder procedure and product type on
and neural activity during inhibition trials
(x, y, z)
Friendship reminder ×
friendship reminder ×
product type with
1, 8, 44
Kerns et al.
4, -1, 47
MacDonald et al.
-6, 30, 57
Batterink et al.
-3, 39, 51
Batterink et al.
21, 48, 39
Batterink et al.
24, 50, 41
Simmonds et al.
Table 2. Results of Regions of Interest analyses: significant positive correlations between compulsive buying
tendency and neural activity during inhibition trials as a function of friendship reminder and product type
Following previous research that identified the ACC as the conflict detection center in the
brain (Botvinick, Cohen, & Carter, 2004; Botvinick, Nystrom, Fissell, Carter, & Cohen, 1999;
Hedgcock & Rao, 2009; Kerns et al., 2004; MacDonald et al., 2000), the current findings
identify the underlying mechanism of the interaction effect of friendship reminders and
compulsive buying tendency identified in the previous studies. That is, the results confirm our
hypothesis that being reminded of friendship reduces goal conflict identification for non-
compulsive buyers, whereas it increases goal conflict identification in consumers classified as
7.7.4. Self-control exertion
Correlation analysis showed that compulsive buying tendency was significantly
positively correlated with activation during inhibition (No-go) trials as a function of the
21, 48, 39
Batterink et al.
24, 50, 41
friendship reminder × product type interaction in the region of interest previously implicated in
self-control exertion (DLPFC, see Table 2)5. Similar to the ACC, the results showed that the
lower (higher) the compulsive buying score, the lower (higher) the DLPFC activation when
asked to inhibit responses towards tempting clothes (versus neutral furniture) after being
reminded of friendship (versus not being reminded of friendship). Hence, non-compulsive buyers
showed reduced DLPFC activation compared to the average participant, while compulsive
buyers showed increased DLPFC activation under these conditions.
As previous research has identified the DLPFC as the brain area involved with the actual
inhibition of (consumption) impulses (Batterink et al., 2010; Hare et al., 2009; Liddle et al.,
2001; MacDonald et al., 2000; Menon et al., 2001; Simmonds et al., 2008), the current findings
are fully in line with our findings of Study 1 in which we identified a crossover interaction
consisting of an adverse effect of friendship reminders on behavioral self-control exertion (lower
inhibition of prepotent responses toward attractive clothing stimuli) for non-compulsive buyers,
and a beneficial effect of friendship reminders on self-control exertion for compulsive buyers.
When asked to inhibit their impulses towards buying temptations, non-compulsive buyers
showed lower levels of activation in the ACC and DLPFC after friendship reminders, whereas
compulsive buyers showed higher levels of activation in these brain areas associated with goal
conflict identification and overt self-control exertion respectively. These findings provide neural
support for our key postulates that in the face of temptations, friendship reminders decrease the
self-control of non-compulsive buyers by decreasing goal conflict identification, whereas
friendship reminders enhance the self-control of compulsive buyers by increasing goal conflict
identification. As such, the present results converge with the previous findings of Study 4 in
highlighting the critical role of the extent of goal conflict identification in underlying the impact
of friendship reminders on self-control exertion for non-compulsive and compulsive buyers.
Moreover, the present results replicate the basic crossover pattern postulated as our key
hypothesis and observed systematically in this and the previous studies.
8. General Discussion
Drawing from the literature on social influences on self-control, we aimed to reconcile
past conflicting findings and proposed that friendship reminders may help or hinder self-control,
depending on whether or not they represent a goal conflict. That is, the conflict between giving
in to (buying) temptations on the one hand and the perceived consequences for maintaining
satisfactory relationships with close friends on the other. We proposed that such reminders of
close friendships may reduce the extent of an experienced goal conflict for non-compulsive,
‘regular’, buyers, while they increase the salience of such a conflict for their compulsive
counterparts. Consequently, to the extent that the identification of such a conflict is a necessary
and sufficient antecedent of the actual exertion of overt and intentional self-control, we should
observe less self-control among non-compulsive but more self-control for compulsive buyers.
The findings of a series of five studies were in line with our predictions and were robust across
various methodological and conceptual variations. Indeed, we found the predicted crossover
interaction between friendship reminder procedure and buying compulsivity on self-control
exertion and the underlying role of goal conflict identification among a total of 437 respondents,
for self-report, overt behavior indices using prevalidated inhibition tasks and functional
neuroimaging, zooming in on both the impulse and inhibition sides of the self-control “tug-of-
war”, both within and beyond the buying domain, and for different reminders of friendship,
including, but not limited to, the real physical presence of a close friend.
More specifically, in line with our predictions, behavioral and fMRI studies provided
converging evidence that for non-compulsive buyers, friendship reminders decrease self-control,
whereas they increase self-control for non-compulsive buyers. That is, for ‘regular,’ non-
compulsive buyers, friendship reminders reduce goal conflict identification and hence, friendship
reminders will produce less controlled (more disinhibited) behaviors. In contrast, as their
excessive buying tendencies are in conflict with maintaining the relationships they have with
close friends, friendship reminders do increase goal conflict identification for compulsive buyers
and as such self-control exertion among this group.
8.1. Scientific, societal and managerial relevance
Our findings both replicate, reconcile, and extend previous research and theory. With the
finding that friendship reminders decrease self-control for non-compulsive buyers, we
conceptually replicate past findings that suggest that consumers tend to eat, drink, spend, and
generally indulge more in the safety and comfort that the (psychological) presence of friends
seem to offer (DeCastro, 1994; Hofmann et al., 2012; Kurt et al., 2011; Luo, 2005; Redd &
DeCastro, 1992; Zhang & Shrum, 2009). By finding that compulsive buyers display higher levels
of self-control when they are reminded of friendship, we show the flipside of the coin and show
when and why such reminders may do the opposite and actually aid in self-control. A perhaps
ironic observation is that our findings show the ‘dark side’ of one of the key assets of
friendships: the experience of safety and comfort that they offer. On the other hand we also
demonstrate the ‘bright side’ of a downside of friendships: their inherently optional nature (at
least when compared to family members, see Chopik, 2017), may exacerbate the sense of threat
for compulsive buyers, but this also boosts motivation and actual display of self-control among
these consumers. Our work thus extends the literature on compulsive buying, that is heavily
focused on the motives and consequences of compulsive buying (Dittmar, 2005; Dittmar &
Drury, 2000; Faber & OʼGuinn, 2008), but has neglected the role of social factors in the self-
regulatory behaviors of these consumers. Moreover, by focusing on how and when friendship
reminders affect consumer self-control, we extend and replicate research on social influences on
consumer behavior more in general. Specifically, the current studies respond to a call by Finkel
and Fitzsimons (2011), arguing for research addressing the interpersonal aspects of the self-
control process to add to our knowledge on the social basis of self-control. Also, the current
studies add by combining a behavioral and neuroscientific approach to study consumer self-
control, demonstrating convergent validity (Yoon, Gonzalez, & Bettman, 2009). Whereas
previous research on the effectiveness of interventions aiming to improve goal conflict
identification and/or self-control exertion is largely based on inferences from self-report
measures (Hedgcock et al., 2012), the current research employs a more direct and comprehensive
approach by directly measuring the impact of such an intervention on behavioral measures
indicative of self-control in real-life situations (Batterink et al., 2010) as well as on the associated
brain areas. This is especially relevant as self-report measures provide a more retrospective
account and at best give estimations of what may happen in the consumer’ brain, whereas fMRI
provides real-time observations of the cognitive processes that underlie the (self-reported or
observed) behaviors (Craig, Loureiro, Wood, & Vendemia, 2012; Huettel et al., 2009). Indeed,
Heatherton and Wagner (2011) highlight the importance of neuroscientific methods in
uncovering the workings and failures of (consumer) self-control.
In addition, the current research is also of value to marketing practitioners. In addition to
enhancing sales among non-compulsive buyers, the use of friendship reminders will allow
marketing managers to adapt their strategies for compulsive buyers towards a “loyalty borne of
genuine need, rather than habitual dependence” (Dodd, Linaker, & Grigg 2005, p. 382). A
structural focus on long-term customer relationships has been shown to be more profitable than
maximizing current sales at the expense of the consumer (Reinartz, Krafft, & Hoyer, 2004).
Moreover, another financial motivation for firms to limit compulsive buying tendencies stems
from observations that compulsive buyers often regret their purchases and return a substantial
part of them (Friese, 2000; Hassay & Smith, 1996). As such, reducing compulsive buying
behavior will substantially reduce handling costs for organizations, the significance of which is
illustrated by research estimating that a 1% decrease in return rate among U.S. consumers will
reduce annual costs of logistics with an average of $17 million for large retailers (Accenture,
2011). Our research highlights an efficient approach that marketing practitioners could consider
in order to change compulsive buying into less disturbing, more fulfilling buying behavior
among consumers chronically challenged in self-control. Possibly, such an approach that
involves subtly reminding these consumers of the value of their long-term friendships may be
rewarded by them with increased and more “authentic” brand and product loyalty, something
that future research might address.
8.2. Limitations and future research
A frequently encountered limitation of neuroimaging research is its sample size. Because
data collection is extraordinarily expensive (around $500 per hour) and time consuming, sample
sizes in neuroscientific research tend to be small. Our sample size of 40 participants is large
compared to other fMRI studies (e.g., Dietvorst et al., 2009; Hedgcock et al., 2012; Yoon et al.,
2006), including similar paradigms (Batterink et al., 2010). Even though our sample size is large
enough to render reliable insights (Thirion, Pinel, Roche, Dehaene, & Poline, 2006), our sample
is limited to female undergraduate students and results may thus not be generalizable to other
demographic groups. Note however that around 90 per cent of compulsive buyers are female
(Dittmar et al., 2007) and that the age of our sample matches the onset age of compulsive buying
(late teens or early twenties; Black, 2007), supporting the external validity of our sample.
Our theoretical reasoning implies that ACC activity impacts DLPFC activity, which
subsequently affects inhibitory behavior, as supported by existing literature. Although we did
examine activation in both the ACC and DLPFC, it should be noted that we did not examine the
causal relationship between both regions, or their impact on overt behavior. That is, one would
ideally then like to test for a pattern of sequential moderated mediation with friendship (not
reminded vs. reminded of friendship) as within-subjects factor, compulsive buying tendency as
continuous (between-subjects) moderator, neural activation in the ACC as a first mediator, neural
activation in the DLPFC as second mediator and number of inhibition failures on clothing trials
as dependent variable. Unfortunately, the current ‘state of the art’(i.e., PROCESS, or the more
recent MEMORE macro, Montoya & Hayes, 2017) does presently not allow for moderated
mediation analyses in within-subjects/repeated measures designs with a continuous between-
subjects moderator (please see Montoya & Hayes, 2017, p. 20/22) and so such testing should
await future research.
Furthermore, care should be taken when inferring cognitive processes based on changes
in brain activation. Since activity in a single brain area is associated with several processes,
activation in a particular area is not uniquely associated with a particular psychological process
(Hedgcock et al., 2012; Yoon et al., 2009). We consider this problem of reverse inference
partially mitigated as our hypotheses center on well-defined brain areas that have been reliably
implicated in goal conflict identification and self-control exertion (Batterink et al., 2010; Kerns
et al., 2004; MacDonald et al., 2000; Simmonds et al. 2008). Moreover, we used neuroimaging
data (Study 5) to test predictions derived from behavioral and self-reported data (Studies 1-4),
further mitigating the problem of reversed interference (Huettel et al., 2009).
It is interesting to note that for compulsive buyers we consistently found that friendship
reminders increase self-control. Moreover, we demonstrate that this effect occurs exclusively
with reminders of friendship; the (imagined or real) presence of acquaintances and strangers, or
non-social but otherwise positive circumstances did not yield the same effects on compulsive
buyers’ self-control. For non-compulsive buyers, while the results also consistently show that
friendship reminders produce lower self-control, the reference point varies slightly over studies.
We observe less self-control for these participants when compared to a neutral control condition
(Study 1), when compared to a positive event condition (Study 2), and sometimes when
compared to an acquaintance/stranger condition (Study 4). However, as to the latter observation,
in Studies 2 and 3 we observe similar levels of self-control between the presence of a friend and
acquaintance/stranger conditions. One of the reasons for the mixed results when comparing the
friendship with the acquaintance/stranger conditions might be the type of self-control we
examined. While the few previous studies that distinguished between both social groups focused
on eating behavior (Clendenen et al., 1994; Salvy et al., 2007), we selected (over) spending as
our target behavior. It may well be that while overeating (or drinking) may invariantly trigger a
more negative impression among strangers than among friends and so may spur impression
management concerns aimed at controlling these behaviors, this may not be the case when it
comes to excessive spending. Indeed, if anything, the connotation of excessive spending may
also have a more positive aspect as it may signal wealth and status. As such, impression
management concerns among non-compulsive buyers may still play a role in their spending
behavior, but possibly these concerns may act to further disinhibit, rather than control their
spending in the real or imagined presence of mere acquaintances and strangers, something that
future research may profitably explore.
The current research focused on the mental activation of the concept of friendship. If a
friend was present in a shopping situation, it was not because (s)he had an active support role. As
a meaningful extension, studying the dynamics of self-control in situations where consumers
who are vulnerable to consumption temptations are directly supported by friends, may be a
fruitful direction for future research. For example, some interventions focusing on compulsive
consumers involve support groups. Future research may investigate under which particular
circumstances compulsive consumers’ self-control benefits from the active support provided by
friends in real life situations.
Further research may test as well whether friendship reminders also improve self-control
in other domains where problems with inhibiting impulses underlie the phenomenon, such as
obesity and substance addiction.
In sum, the current research employed an interdisciplinary approach by using behavioral
and neuroscientific methods to study when and why friendship reminders increase or decrease
consumer self-control. The studies provide converging evidence that friendship reminders may
lead to more controlled behavior for compulsive buyers, as their excessive buying behavior may
be in conflict with the relationships they have with friends. For non-compulsive buyers, there is
no special conflict between their friendships and encountered consumption temptations. Sporadic
disinhibited consumption may even be good for the relationship (Luo, 2005). For these
consumers we observe a tendency consistent with previous findings: non-compulsive buyers
behave in a less controlled manner in the (psychological) presence of friends. Thus, for both type
of consumers there seems truth in the belief: a faithful friend is the medicine of life.
Accenture. (2011). Reducing the Quantity and Cost of Product Returns in Consumer Electronics.
Argo, J. J., Dahl, D. W., & Manchanda, R. V. (2005). The Influence Of a Mere Social Presence In a Retail Context.
Journal of Consumer Research, 32(2), 207–212.
Ariely, D., & Levav, J. (2000). Sequential Choice in Group Settings: Taking the Road Less Traveled and Less
Enjoyed. Journal of Consumer Research, 27(3), 279–290.
Batterink, L., Yokum, S., & Stice, E. (2010). Body Mass Correlates Inversely With InhibitoryCcontrol in Response
to Food Among Adolescent Girls: An fMRI study. NeuroImage, 52(4), 1696–1703.
Baumeister, R. F., & Leary, M. R. (1995). The Need to Belong: Desire for Interpersonal Attachments as a
Fundamental Human Motivation. Psychological Bulletin, 117(3), 497–529.
Berns, G. S., & Moore, S. E. (2012). A Neural Predictor of Cultural Popularity. Journal of Consumer Psychology,
Black, D. W. (2007). A Review of Compulsive Buying Disorder. World Psychiatry, 6(1), 14–18.
Black, D. W. (2011). Epidemiology and Phenomenology of Compulsive Buying Disorder.
Oxford Handbook of Impulse Control Disorders.
Borda, T., Feinstein, B. A., Neziroglu, F., & Veccia, T. (2013). Are Children with Obsessive–Compulsive Disorder
at Risk for Problematic Peer Relationships? Journal of Obsessive-Compulsive and Related Disorders, 2(4),
Botvinick, M. M., Cohen, J. D., & Carter, C. S. (2004). Conflict Monitoring and Anterior Cingulate Cortex: An
Update. Trends in Cognitive Sciences, 8(12), 539–546.
Botvinick, M.M., Nystrom, L. E., Fissell, K., Carter, C. S., & Cohen, J. D. C. (1999). Conflict Monitoring Versus
Selection-for-Action in Anterior Cingulate Cortex Nature. Nature, 402(6758), 179–181.
Carver, C. S., & Scheier, M. F. (1990). Origins and Functions of Positive and Negative Affect: A Control-Process
View. Psychological Review, 97(1), 19–35.
Childers, T. L., & Rao, A. R. (1992). The Influence of Familial and Peer-Based Reference Groups on Consumer
Decisions. Journal of Consumer Research, 19(2), 198–211.
Chopik, W. J. (2017). Associations Among Relational Values, Support, Health, and Well‐Being Across the Adult
Lifespan. Personal Relationships, 24(2), 408–422.
Clendenen, V. I., Herman, C. P., & Polivy, J. (1994). Social Facilitation of Eating Among Friends and Strangers.
Appetite, 23(1), 1–13.
Couwenberg, L. E., Boksem, M. A. S., Dietvorst, R. C., Worm, L., Verbeken, W. J. M. I., & Smidts, A. (2017).
Neural Responses to Functional and Experiential Ad Appeals: Explaining Ad Effectiveness. International
Journal of Research in Marketing, 34(2), 355–366.
Craig, A. W., Loureiro, Y. K., Wood, S., & Vendemia, J. M. C. (2012). Suspicious Minds: Exploring Neural
Processes During Exposure to Deceptive Advertising. Journal of Marketing Research, 49(3), 361–372.
DeCastro, J. M. (1994). Family and Friends Produce Greater Social Facilitation of Food Intake than Other
Companions. Physiology and Behavior, 56(3), 445–455.
DeCastro, J. M. (1997). Socio-Cultural Determinants of Meal Size and Frequency. British Journal of Nutrition,
Dietvorst, R. C., Verbeke, W. J. M. I., Bagozzi, R. P., Yoon, C., Smits, M., & Van der Lught, A. (2009). A Sales
Force–Specific Theory-of-Mind Scale: Tests of its Validity by Classical Methods and Functional Magnetic
Resonance Imaging. Journal of Marketing Research, 46(5), 653–668.
Dittmar, H. (2005). A New Look at “Compulsive Buying”: Self-Discrepancies and Materialistic Values as
Predictors of Compulsive Buying Tendency. Journal of Social and Clinical Psychology, 24(6), 832–859.
Dittmar, H., & Drury, J. (2000). Self-Image – Is it in the Bbag? A Qualitative Comparison Between “Ordinary” and
“Excessive” Consumers. Journal of Economic Psychology, 21(2), 109–142.
Dittmar, H., Long, K., & Bond, R. (2007). When a Better Self Is Only a Button Click Away: Associations Between
Materialistic Values, Emotional and Identity–Related Buying Motives, and Compulsive Buying Tendency
Online. Journal of Social and Clinical Psychology, 26(3), 334–361.
Dodd, C. A., Linaker, A., & Grigg, N. P. (2005). He’s Gotta Have It: Shopping Dependence and the Homosexual
Male Clothing Consumer. Journal of Consumer Behavior, 4(5), 374–389.
Dreher, J.-C., Schmidt, P. J., Kohn, P., Furman, D., Rubinow, D., & Berman, K. F. (2007). Menstrual Cycle Phase
Modulates Reward-Related Neural Function in Women. PNAS Proceedings of the National Academy of
Sciences of the United States of America, 104(7), 2465–2470.
DSM-V. (2013). Diagnostic and Statistical Manual of Mental Disorders, 5th Edition. American Psychiatric
Dzhogleva, H., & Lamberton, C. P. (2014). Should Birds of a Feather Flock Together? Understanding Self-Control
Decisions in Dyads. Journal of Consumer Research, 41(2), 361–380.
Faber, R. J., & O’guinn, T. C. (1992). A Clinical Screener for Compulsive Buying. Journal of Consumer Research,
Faber, R. J., & O’Guinn, T. C. (2008). Compulsive Buying: Review and Reflection. In Handbook of Consumer
Psychology (pp. 1039–1056).
Fennis, B. M., Janssen, L., & Vohs, K. D. (2009). Acts of Benevolence: A Limited-Resource Account of
Compliance with Charitable Requests. Journal of Consumer Research, 35(6), 906–924.
Fennis, B.M. (2017). How to Foster Health and Well-Being When Self-control is Low. In D.T.D.
De Ridder, M.A. Adriaanse & K. Fujita (Eds.), The Routledge International Handbook of
Self-control in Health and Well-being. New York: Routledge.
Finkel, E. J., & Fitzsimons, G. M. (2011). The Effects of Social Relationships on Self-regulation.
In K. D. Vohs & R. F. Baumeister (Eds.), Handbook of Self-regulation: Research, theory
and applications (pp. 390–406). New York: Guilford Press.
Fitzsimons, G. M., Finkel, E. J., & VanDellen, M. R. (2015). Transactive goal dynamics. Psychological Review,
Friese, S. (2000). Self-concept and Identity in a Consumer Society: Aspects of Symbolic Product Meaning. Marburg,
Goldberg, E. (2001). The Executive Brain: Frontal Lobes and the Civilized Mind. New York: Oxford University
Hare, T. A., Camerer, C. F., & Rangel, A. (2009). Self-Control in Decision-Making Involves Modulation of the
vmPFC Valuation System. Science, 646–648.
Harris, P. S., Harris, P. R., & Miles, E. (2017). Self-Affirmation Improves Performance on Tasks Related to
Executive Functioning. Journal of Experimental Social Psychology, 70, 281–285.
Hassay, D. N., & Smith, M. C. (1996). Compulsive Buying: An Examination of the Consumption Motive.
Psychology & Marketing, 13(18), 741.
Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process Aanalysis: a Regression-
Based Approach. Guilford Press.
Hayes, A. F., & Preacher, K. J. (2014). Statistical Mediation Analysis with a Multicategorical Independent Variable.
British Journal of Mathematical and Statistical Psychology, 76(3), 451–470.
Hedgcock, W. M., Vohs, K. D., & Rao, A. R. (2012). Reducing Self-Control Depletion Effects Through Enhanced
Sensitivity to Implementation: Evidence from fMRI and Behavioral Studies. Journal of Consumer
Psychology, 22(4), 486–495.
Hedgcock, W.M., & Rao, A. R. (2009). Trade-off Aversion As An Explanation for the Attraction Effect: A
Functional Magnetic Resonance Imaging Study. Journal of Marketing Research, 46(1), 1–13.
Herman, C. P. (2015). The Social Facilitation of Eating. A Review. Appetite, 86, 61–71.
Hetherington, M. M., Anderson, A. S., & Norton, G. (2006). Situational Effects on Meal Intake: A Comparison of
Eating Alone and Eating with Others. Physiology & Behavior, 88(4), 498–505.
Hofmann, W., Baumeister, R. F., Förster, G., & Vohs, K. D. (2012). Everyday Temptations: An Experience
Sampling Study of Desire, Conflict, and Self-control. Journal of Personality and Social Psychology, 102(6),
Hofmann, W., Friese, M., & Strack, F. (2009). Impulse and Self-control From a Dual-Systems Perspective.
Perspectives on Psychological Science, 4(2), 162–176.
Howland, M., Hunger, J. M., & Mann, T. (2012). Friends Don’t Let Friends Eat Cookies: Effects of Restrictive
Eating Norms on Consumption Among Friends. Appetite, 59(2), 505–509.
Huettel, S. A., Payne, J. W., Yoon, C., Gonzales, R., Bettman, J. R., Hedgcock, W., & Rao, A. R. (2009).
Commentaries and Rejoinder to “Trade-OffTrade-Off Aversion as an Explanation for the Attraction Effect: A
Functional Magnetic Resonance Imaging Study.” Journal of Marketing Research, 46(1), 14–24.
Inzlicht, M., & Gutsell, J. N. (2007). Running on Empty: Neural Signals for Self-Control Failure. Psychological
Science, 18(11), 933–937.
Jenkinson, M., Beckmann, Christian F. Behrens, T. E., Woolrich, M. W., & Smith, S. M. (2012). FSL. NeuroImage,
Judd, C. M., Kenny, D.A., & McClelland, G.H. (2001). Estimating and Testing Mediation and
Moderation in Within-Subject Designs. Psychological Methods, 6(2), 115-134.
Judd, C. M., McClelland, G.H., & Smith, E.R. (1996). Testing Treatment by Covariate
Interactions when Treatment Varies Within Subjects. Psychological Methods, 1(4), 366-
Kerns, J. G., Cohen, J. D., MacDonald, A. W., Cho, R. Y., Stenger, V. A., & Carter, C. S. (2004). Anterior
Cingulate Conflict Monitoring and Adjustments in Control. Science, 303(5660), 782–790.
Kirkova, D. (2013). Why Shopping With the Girls is Bad for Your Bank Balance: Women Spend £900 More a Year
When They Hit the Stores with Friends. Retrieved from http://www.dailymail.co.uk/femail/article-
Knoch, D., & Fehr, E. (2007). Resisting the Power of Temptations: The Right Prefrontal Cortex and Self-Control.
Annals of the New York Academy of Sciences, 1104, 123–134.
Kurt, D., Inman, J. J., & Argo, J. (2011). The influence of friends on consumer spending: The role of agency-
communion orientation and self-monitoring. Journal of Marketing Research, 48(4), 741–754.
Liddle, P. F., Kiehl, K. A., & Smith, A. M. (2001). Event-Related fMRI Study of Response Inhibition. Human Brain
Mapping, 12(2), 100–109.
Lieberman, M. D., & Eisenberger, N. I. (2005). Conflict and Habit: A Social Cognitive Neuroscience Approach to
the Self. On Building, Defending and Regulating the Self: A Psychological Perspective, 77–102.
Luo, X. (2005). How Does Shopping With Others Influence Impulsive Purchasing? Journal of Consumer
Psychology, 15(4), 288–294.
MacDonald, A. W., Cohen, J. D., Stenger, V. A., & Carter, C. S. (2000). Dissociating the Role of the Dorsolateral
Prefrontal and Anterior Cingulate Cortex in Cognitive Control. Science, 288(5472), 1835–1838.
Maslow, A. H. (1954). The Instinctoid Nature of Basic Needs. Journal of Personality, 22(3), 326–347.
McClure, S. M., Botvinick, M. M., Yeung, N., Greene, J. D., & Cohen, J. (2007). Conflict Monitoring in Cognition-
Emotion Competition. In J. J. Gross (Ed.), Handbook of Emotion Regulation (pp. 204–226). New York:
Menon, V., Adleman, N. E., White, C. D., Glover, G. H., & Reiss, A. L. (2001). Error-Related Brain Activation
During a Go/NoGo Response Inhibition Task. Human Brain Mapping, 12(April), 131–143.
Mishra, A., & Mishra, H. (2010). We Are What We Consume: The Influence of Food Consumption on Impulsive
Choice. Journal of Marketing Research, 47(6), 1129–1137.
Montoya, A. K., & Hayes, A. F. (2017). Two-Condition Within-Participant Statistical Mediation
Analysis: A path-Analytic Framework. Psychological Methods, 22(1), 6-27.
Müller, A., Mitchell, J. E., & De Zwaan, M. (2015). Compulsive Buying. The American Journal on Addictions,
Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When Moderation is Mediated and Mediation is Moderated.
Journal of Personality and Social Pschology, 89(6), 852–863.
Muraven, M., & Baumeister, R. F. (2000). Self-Regulation and Depletion of Limited Resources: Does Self-Control
Resemble a Muscle? Psychological Bulletin, 126(2), 247–259.
Muraven, M., Shmueli, D., & Burkley, E. (2006). Conserving Self-Control Strength. Journal of Personality and
Social Psychology, 91(3), 524–537.
Myrseth, K. O. R., & Fishbach, A. (2009). Self-Control: A Function of Knowing When and How to Exercise
Restraint. Current Directions inPsychological Science, 18(4), 247–252.
Newman, J. P., Widom, C. S., & Nathan, S. (1985). Passive Avoidance in Syndromes of Disinhibition: Psychopathy
and Extraversion. Journal of Personality and Social Psychology, 48(5), 1316–1327.
O’Guinn, T. C., & Faber, R. J. (1989). Compulsive buying: A Phenomenological Exploration. Journal of Consumer
Research, 16(2), 147–157.
O'Guinn, T. C., & Faber, R. J. (1989). Compulsive Buying: A Phenomenological Exploration.
Journal of Consumer Research, 16(2), 147-157.
Petersen, S. E., & Dubis, J. W. (2012). The Mixed Block/Event-related Design. Neuroimage, 62(2), 1177–1184.
Poldrack, R. A. (2007). Region of Interest Analysis for fMRI. Social Cognitive and Affective Neuroscience, 2(1),
Redd, M., & DeCastro, J. (1992). Social Facilitation of Eating: Effects of Social Instruction on Food Intake.
Physiology & Behavior, 52(4), 749–754.
Reinartz, W., Krafft, M., & Hoyer, W. D. (2004). The Customer Relationship Management Process: Its
Measurement and Impact on Performance. Journal of Marketing Research, 41(3), 293–305.
Ridgway, N. M., Kukar-Kinney, M., & Monroe, K. B. (2008). An Expanded Conceptualization and a New Measure
of Compulsive Buying. Journal of Consumer Research , 35(4), 622–639.
Righetti, F., & Finkenauer, C. (2011). If You Are Able to Control Yourself, I Will Trust You: The Tole of Perceived
Self-Control in Interpersonal Trust. Journal of Personality and Social Psychology, 100(5), 874–887.
Robinson, M. D., Schmeichel, B. J., & Inzlicht, M. (2010). A Cognitive Control Perspective of Self‐Control
Strength and its Depletion. Social and Personality Psychology Compass, 4(3), 189–200.
Rook, D. W., & Fisher, R. J. (1995). Normative Influences on Impulsive Buying Behavior. Journal of Consumer
Research, 22(3), 305–313.
Salvy, S.-J., Jarrin, D., Paluch, R., Irfan, N., & Pliner, P. (2007). Effects of Social Influence on Eating in Couples,
Friends and Strangers. Appetite, 49(1), 92–99.
Schwarz, N. (2006). Feelings, fit, and funny effects: A situated cognition perspective. Journal of Marketing
Research, 43(1), 20-2
Simmonds, D. J., Pekar, J. J., & Mostofsky, S. H. (2008). Meta-Analysis of Go/No-go tasks Demonstrating that
fMRI Activation Associated with Response Inhibition is Task-Dependent. Neuropsychologia, 46(1), 224–232.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-Positive Psychology: Undisclosed Flexibility in Data
Collection and Analysis Allows Presenting Anything as Significant. Psychological Science, 22(11), 1359–
Stillman, T. F., Tice, D. M., & Fincham, F. D. (2009). The Psychological Presence of Family Improves Self-
Control. Journal of Social and Clinical Psychology, 28(4), 498.
Strack, F., & Deutsch, R. (2004). Reflective and Impulsive Determinants of Social Behavior. Personality and Social
Psychology, 8(3), 220–247.
Sullivan, H. S. (1953). The Interpersonal Theory of Psychiatry. New York: Norton.
Thirion, B., Pinel, P., Roche, A., Dehaene, S., & Poline, J. (2006). Analysis of fMRI Data Sampled From Large
Populations: Statistical and Methodological Issues. NeuroImage, 35(1), 1–28.
Torres, A. R., Cruz, B. L., Vicentini, H. C., Lima, M. C. P., & Ramos-Cerqueria, A. T. A. (2016). Obsessive-
Compulsive Symptoms in Medical Students: Prevalence, Severity, and Correlates. Academic Psychiatry,
Valence, G., d’Astous, A., & Fortier, L. (1988). Compulsive Buying: Concept and Measurement. Journal of
Consumer Policy, 11(4), 419–433.
Visscher, K. M., Miezin, F. M., Kelly, J. E., & Buckner, R. L. (2003). Mixed Blocked/Event-Related Designs
Separate Transient and Sustained Activity in fMRI. Neuroimage, 19(4), 1694–1708.
Wager, T. D., Sylvester, Ching-Yune C., Lacey, S. C., Nee, D. E., Franklin, M., & John, J. (2005). Common and
Unique Components of Response Inhibition Revealed by fMRI. NeuroImage, 27(2), 323–340.
Wagner, D. D., & Heatherton, T. F. (2011). Giving in to Temptation: The Emerging Cognitive Neuroscience of Self-
Regulatory Failure. In K. D. Vohs & R. F. Baumeister (Eds.), Handbook of Self-Regulation: Research,
Theory, and Applications (pp. 41–63). New York: Guilford Press.
Wagner, D. D., & Heatherton, T. F. (2015). Self-Regulation and its Failure: The Seven Deadly Threats to Self-
Regulation. APA Handbook of Personality and Social Psychology, 1, 805–842.
Webb, T. L., & Sheeran, P. (2003). Can Implementation Intentions Help to Overcome Ego-Depletion? Journal of
Experimental Social Psychology, 39(3), 279–286.
Wilcox, K., & Stephen, A. T. (2013). Are Close Friends the Enemy? Online Social Networks, Self-Esteem, and
Self-Control. Journal of Consumer Research, 40(1), 90–103.
Wilkinson, R. G. (1999). Health, Hierarchy, and Social Anxiety. Annals of the New York Academy of Sciences,
Yoon, C., Gonzalez, R., Bechara, A., Berns, G. S., Dagher, A. A., Dubé, L., & Huettel, S. A. (2012). Decision
Neuroscience and Consumer Decision Making. Marketing Letters, 23(2), 473–485.
Yoon, C., Gonzalez, R., & Bettman, J. R. (2009). Using fMRI to Inform Marketing Research: Challenges and
Opportunities. Journal of Marketing Research, 46(1), 17–19.
Yoon, C., Gutchess, A. H., Feinberg, F., & Polk, T. A. (2006). A Functional Magnetic Resonance Imaging Study of
Neural Dissociations Between Brand and Person Judgments. Journal of Consumer Research, 33(1), 31–40.
Zhang, Y., & Shrum, L. J. (2009). The Influence of Self-Construal on Impulsive Consumption. Journal of Consumer
Research, 35(5), 838–850.