Content uploaded by Jaye L Derrick
Author content
All content in this area was uploaded by Jaye L Derrick on Nov 06, 2017
Content may be subject to copyright.
Social surrogacy: How favored television programs provide the experience
of belonging
Jaye L. Derrick
a,*
, Shira Gabriel
b
, Kurt Hugenberg
c
a
Research Institute on Addictions, University at Buffalo, State University of New York, 1021 Main Street, Buffalo, NY 14203-1016, United States
b
Department of Psychology, University at Buffalo, State University of New York, United States
c
Department of Psychology, Miami University, United States
article info
Article history:
Received 16 May 2008
Revised 2 December 2008
Available online 24 December 2008
Keywords:
Need to belong
Rejection
Parasocial relationship
abstract
The current research examines the Social Surrogacy Hypothesis: parasocial relationships in favored televi-
sion programs can provide the experience of belonging. Four studies support the hypothesis. Study 1
demonstrated that people report turning to favored television programs when feeling lonely, and feel less
lonely when viewing those programs. Study 2 demonstrated that experimentally activating belonging-
ness needs leads people to revel longer in descriptions of favored (but not non-favored) television pro-
grams. Study 3 demonstrated that thinking about favored (but not non-favored) television programs
buffers against drops in self-esteem and mood and against increases in feelings of rejection commonly
elicited by threats to close relationships. Finally, Study 4 demonstrated that thinking about favored tele-
vision programs reduces activation of chronically activated rejection-related words. These results yield
provocative preliminary evidence for the Social Surrogacy Hypothesis. Thinking about valued television
programs appears to yield the experience of belongingness.
Ó2008 Elsevier Inc. All rights reserved.
We humans share a great deal with our hominid ancestors. We
require food and drink, shelter and safety, and the experience of
inclusion and connectedness with others (Baumeister & Leary,
1995; Williams, 2007). We also share an affective system evolved
to subserve these needs. When our needs are met, we feel good.
When our needs are not met, or the environment is dangerous or
otherwise problematic, we feel badly (Frijda, 1988). Despite these
sweeping similarities, certain characteristics are uniquely human.
Perhaps the most obvious among them is the robust and continued
use of technology. Importantly, many technological advancements
are designed to help us meet such needs. Indeed, most of our needs
seem to be addressed more easily than ever by using such techno-
logical advancements.
Curiously, not all technology serves to meet human needs. To the
contrary, some technological advancements serve to provide the
experience of having needs met, without actually meeting the need.
Indeed, evolutionary biologist Steven Pinker, in his modern classic,
How the Mind Works (Pinker, 1997), argues that a number of techno-
logical advances are designed not to directly fulfill evolved needs,
but rather to deceive our mind into believing that some needs have
been fulfilled. In Pinker’s own words, such technologies can ‘‘get at
the pleasure circuits of the brain and deliver little jolts of enjoyment
without the inconvenience of wringing bona fide fitness increments
from the harsh world” (p. 524). For example, both humans and our
hominid ancestors evolved to crave food, especially food high in fats,
oils, and sugars, which were historically difficult to obtain, but abso-
lutely vital to ingest. Historically, humans have relied upon naturally
occurring substances (e.g., coca leaves) or meditation to suppress the
appetite. More recently, humans have turned to technologies, such
as diet drugs, or more drastically, gastric bypass surgery, to experi-
ence satiety without eating.
We argue that other commonplace technologies, such as narrative
fiction, television, music, or interactive video games, can also provide
the experience of needfulfillment.We hypothesize that the facsimiles
of social contexts presented in these technologies may beused to sat-
isfy the fulfillment of belongingness needs. Just as Harlow’s (1958) in-
fant monkeys experienced succor from cloth surrogates, satisfying
belongingness needs, so too may beloved books, television programs,
movies,music, or video games potentially serveas ‘‘social surrogates,”
leading to an experience of belongingness even when no real, bona
fide belongingness has been experienced.
Treating the parasocial relationships provided by such technol-
ogies seriously as surrogates for actual belongingness is, we be-
lieve, a novel theoretical treatment of a ubiquitous part of the
human environment. To this end, we first outline the increasing
evidence that the need to belong is, itself, as basic a human need
as food and shelter. We then describe research on parasocial rela-
tionships and means by which those relationships are frequently
achieved. Next we explain why television shows in particular pres-
ent both a ubiquitous and potent vehicle for delivering parasocial
relationships. Finally, we present four studies in which we test
0022-1031/$ - see front matter Ó2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jesp.2008.12.003
*Corresponding author.
E-mail address: jderrick@ria.buffalo.edu (J.L. Derrick).
Journal of Experimental Social Psychology 45 (2009) 352–362
Contents lists available at ScienceDirect
Journal of Experimental Social Psychology
journal homepage: www.elsevier.com/locate/jesp
the hypothesis that calling to mind favored television programs
can yield the experience of belonging.
Human’s sociality and parasociality
The evidence that human beings seek out, value, and maintain
relationships with other humans is myriad, and a full review is cer-
tainly beyond the scope of the current work. Indeed, the evidence
is so strong that Baumeister and Leary (1995) argue the need to be-
long is a fundamental human need (see also Williams, 2007). Yet
belonging is not unconditional, and rejection is an all too common
part of the human experience. Rejection hurts. People who have
been rejected may experience, among a host of other conse-
quences, reduced state self-esteem and negative mood states
(Leary, Haupt, Strausser, & Chokel, 1998; Leary, Tambor, Terdal, &
Downs, 1995; Williams, Cheung, & Choi, 2000; Zadro, Williams, &
Richardson, 2004). Positive social connections or interactions,
however, can buffer against the effects of rejection. For example,
Buckley, Winkel, and Leary (2004) demonstrated that participants
who were initially negatively evaluated but then were evaluated
more positively (increasing acceptance) were protected against
the mood and state self-esteem decrements experienced by partic-
ipants who were only rejected. Similarly, even simple reminders of
positive social interactions may be enough to buffer against the ef-
fects of rejection (Gardner, Pickett, & Knowles, 2005).
Based on this evidence that social connections can buffer
against feelings of rejection, we hypothesize that technologies such
as television, books, music, and video games may afford the expe-
rience of belonging, via one-way parasocial relationships. There is a
well-founded theoretical basis for such a hypothesis. It seems clear
that people can, and frequently do, form parasocial relationships
with sets of favorite media characters or media personalities, both
fictional and real (Horton & Wohl, 1956). These one-sided paraso-
cial relationships also frequently show psychological hallmarks of
real relationships. For example, in the presence of a parasocial rela-
tionship partner or after thinking about a parasocial relationship,
those with strong parasocial attachments react much as they
would when in the presence of ‘‘real” friends. They become more
willing to self-disclose, show more empathy (Knowles, 2007),
and demonstrate social facilitation effects (Gardner & Knowles,
2008). They even become more similar to their ideal selves (Der-
rick, Gabriel, & Tippin, 2008), a phenomenon elicited by only the
closest of relationships (Drigotas, Rusbult, Wieselquist, & Whitton,
1999; Gabriel, Carvallo, Jaremka, & Tippin, 2008).
Some preliminary evidence already exists that technology (and
specifically television) may afford the experience of belonging. First,
Green and Brock (1998) found that people prefer parasocial (or
‘‘ersatz” social) activities when the costs of friendship are salient,
at least among those who are low in trust. Thus, their findings pro-
vide initial evidence suggesting that people may be interested in
seeking parasocial relationships, especially in situations where real
interactions may be perceived as problematic. Similarly, Twenge,
Baumeister, DeWall, Ciarocco, and Bartels (2007, Study 3) found that
the tendency to aggress after social exclusion is reduced among indi-
viduals who think about a well-liked celebrity. Indeed, they provide
provocative preliminary evidence that a faux relationship can ame-
liorate one of the negative sequelae of rejection.
1
Moreover, exclu-
sion by a technological ‘other’ seems to elicit similar effects to that
of a real friend. For example, extensive anecdotal evidence suggests
that people commonly react to the end of a television series as if it were
the end of a relationship (Cohen, 2004). Even being rejected by a com-
puter (Zadro et al., 2004) elicits the same hurt feelings that people
experience after rejection by other real people. Finally, survey evi-
dence also suggests a link between technology and belongingness:
data from the General Social Survey demonstrate that people who
watch more television actually feel as if they have more friends
(Kanazawa, 2002). Thus, despite the one-sided nature of parasocial
relationships, it appears that such relationships are psychologically
real to the people experiencing them.
Television as a vehicle for parasocial relationships
Television programming, particularly reliably-followed favorite
programs, allows viewers the opportunity, week after week (or
even day after day), to regularly immerse themselves in a narrative
about a recognizable ‘‘social” world in which familiar people, situ-
ations, landscapes, and events become intimate and comfortable
(Cohen, 2006). The most common themes in these narratives are
social (Hogan, 2003), and strong initial research demonstrates that
narratives engage people in social processing (Mar & Oatley, 2008).
For example, engaging in narratives leads to an increase in
thoughts and emotions congruent with the ones presented in the
narrative (Oatley, 1999), and exposure to narratives is related to
more sophisticated social skills and abilities (Mar, Oatley, Hirsh,
dela Paz, & Peterson, 2006). Indeed, Mar and Oatley (2008) argue
that one core function of narratives is to mentally simulate social
interactions, potentially facilitating subsequent social behavior.
Although there are numerous possible technologies that may
supply narratives (e.g., books, movies, and video games), television
is likely one of the most potent. First, Americans report spending
an average of 3 h per day watching television (US Department of La-
bor, 2006). This is over half the total time spent on leisure activities,
and is substantially more time than is actually spent with friends.
Second, whereas reading narratives requires mental simulation
(Mar & Oatley, 2008), television provides a rich visual and auditory
environment, mirroring almost completely our daily experience,
and requiring few of the cognitive resources necessary to simulate
lexically mediated parasocial relationships. In summary, television
viewing is ubiquitous, copious, and provides multi-sensory stimula-
tion, making it a potent facsimile of social interaction.
Favored television programs as a social surrogate: the current
research
In this research, we sought novel evidence for our Social Surro-
gacy Hypothesis; that is, that parasocial relationships provided by
television programs can yield the experience of belonging. Specif-
ically, we drew three primary predictions from the Social Surro-
gacy Hypothesis. If favorite television programs can yield the
experience of belonging, we hypothesized that (1) events that typ-
ically elicit belongingness needs (e.g., threats to a relationship, a
rejection experience) would elicit a desire to experience a favored
television program, (2) thinking about a favored television pro-
gram could buffer against threats to real-world belongingness,
and (3) thinking about a favored television program should reduce
the accessibility of loneliness related concepts.
Given the novelty of these predictions, we adopted a multi-
pronged approach, involving both correlational and experimental
evidence. We first conducted a preliminary correlational study. In
this study, we examined a large sample of people to determine
whether they report preferring to watch a favorite television pro-
gram relative to enacting other non-social activities when feeling
1
Although supportive of our hypotheses, the (Twenge, Baumeister, et al., 2007;
Twenge, Zhang, et al., 2007, Study 3) study differs from the current research in
important ways. First, it examines a parasocial relationship with a specific individual
rather than with an entire parasocial world. Second, it examines only one effect of a
rejection experience (aggressive behavior) leaving it somewhat more open to
alternate explanations. The current research examines behavior, mood, state self-
esteem, feelings of rejection, and chronically accessible concerns with exclusion.
Nonetheless, the study’s findings are innovative, provocative, and consistent with the
current hypotheses.
J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362 353
lonely, and if so, if they report that this favored television program
reduces the need for social interaction and acceptance. Impor-
tantly, to differentiate between true social surrogacy and mere
escapism, we compared favored television programs to non-fa-
vored television programs, hypothesizing that favored television
programs would be more likely to include parasocial relationships,
and thus, would be more likely to alleviate the belongingness need.
Across three additional studies, we experimentally investigated
whether the parasocial relationships afforded by favored television
programs can address belongingness needs aroused by threats to a
real relationship. In Study 2, we examined the effect of a threat to a
relationship (thinking about a fight with a close other) on the de-
sire to spend time thinking about a favorite television program.
In Study 3, we examined how calling to mind a favorite television
program buffers against the feelings of rejection, self-esteem, and
mood effects often engendered by a threat to a real relationship. Fi-
nally, in Study 4, we investigated how calling to mind a favorite
television program reduces the accessibility of chronically acti-
vated exclusion-related concepts. Overall, we hypothesized that
when individuals’ need for belongingness is aroused, they will turn
to television programs as one means to address those needs; when
belongingness is threatened by experiencing or recalling a rejec-
tion experience, calling to mind a favored television program will
buffer against the negative effects of that belongingness threat.
Study 1: A correlational approach
Can television programs and their attendant parasocial relation-
ships serve as social surrogates? Can these faux ‘‘interactions” pro-
vide the experience of belonging? To provide preliminary evidence
for this Social Surrogacy Hypothesis, we investigated the activities
in which people participate when feeling lonely. We hypothesized
that people would report watching a favored television program as
a preferred activity when belongingness needs were aroused.
Moreover, we hypothesized that participants would report feeling
less lonely when watching this television program, both in com-
parison to less-favored television programs and in comparison to
other non-social activities.
Method
Participants
Seven-hundred and one undergraduate students (233 men, 322
women, and 146 participants who did not indicate their gender;
mean age = 18.86) participated in exchange for course credit. Most
participants were Caucasian (67%). The remainder was predomi-
nantly Asian (15%) and African American (7%). Participants who
did not answer a question were excluded from analyses involving
that question.
Materials and procedure
Participants completed two primary measures, the lonely activ-
ities scale and the likelihood of feeling lonely scale.
2
The order of
these two measures was counter-balanced across participants. Par-
ticipants then completed a demographic measure. Finally, partici-
pants were thanked, debriefed and dismissed.
Lonely activities scale. Participants indicated how likely they would
be to engage in each of 31 activities when feeling lonely (see Table
1). They responded on a scale from 1 (definitely would not do) to 7
(definitely would do).
Likelihood of feeling lonely scale. Participants indicated how likely
they would be to feel lonely when engaging in each of 31 activities
(see Table 1). They responded on a 1 (definitely would feel lonely)
to 7 (definitely would not feel lonely) scale. Participants were also
given the response option ‘‘0” (would not do activity); participants
who responded ‘‘0” on an item were excluded from analyses
involving that item. Responses on this measure were reverse-
scored so that lower scores indicated less likelihood of feeling
lonely.
Results
Overall, this initial correlational investigation was designed to
give us a window into the extent to which people (1) report
engaging in parasocial ‘‘interactions” with television programs
when belongingness needs are activated, and (2) report experi-
encing decreases in belongingness needs when engaging in para-
social ‘‘interactions” with favored television programs.
Importantly, both the lonely activities scale and the likelihood
of feeling lonely scale were designed to compare favored televi-
sion programs to other enjoyable activities and to other less-fa-
vored programs.
Lonely activities scale
Do people report watching favored television programs when
belongingness needs are aroused? As a preliminary test of the
Social Surrogacy Hypothesis, we examined participants’ mean
evaluations of the 31 activities rated by participants. As can be
seen in Table 1, viewing a favored television program was re-
ported as one of the top two most frequent activities in which
participants reported engaging while feeling lonely. Indeed, the
only other activity in which participants engaged with the same
frequency was listening to favored music, t(690) = .09, p= .93,
another technology that can provide parasocial relationships
via connections to celebrities (e.g., Derrick et al., 2008). However,
the mean rated frequency for watching a favored television pro-
gram was significantly higher than the mean for all other activ-
ities, all ps < .001. Importantly, when lonely, viewing a favored
television program is preferred to viewing ‘‘whatever is on tele-
vision,” t(691) = 19.74, p< .001, suggesting that a favored pro-
gram offers more than mere escapism. Thus, people report
viewing more favored television programs than watching what-
ever is on television (and many other activities) when feeling
lonely.
Likelihood of feeling lonely scale
Can favored television programs ease an activated need for
belongingness? As can be seen in Table 1, of all of the investi-
gated activities, participants reported feeling the least lonely
while viewing favored television programs. Although the means
for taking a shower, sleeping, and exercising are not significantly
different from the mean for viewing favored television programs,
ps > .15, participants reported feeling significantly less lonely
while watching a favored television program than after partici-
pating in all other activities, all ps < .001, including watching
‘‘whatever is on television,” t(666) = 19.11, p< .001.
3
In summary,
in line with the Social Surrogacy Hypothesis, people report that
consuming favored television programs buffers against feelings
of loneliness. Importantly, people report that these favored televi-
2
To create the lonely activities scale and the likelihood of feeling lonely scale, we
conducted a preliminary pilot st udy. Twelve undergraduates listed non-social
activities they believed people do when feeling lonely. They each listed as many
items as possible and together came up with 31 viable items (see Table 1).
3
Participants also reported feeling significantly less lonely while watching a
favored television program than when listening to favored music, t(675) = 4.88,
p< .001, watching a favored movie, t(662) = 11.98, p< .001, or reading a favored book,
t(559) = 8.66, p< .001, providing confirmation for the decision to focus on favored
television programs as the vehicle for social surrogacy in this population.
354 J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362
sion programs have benefits over-and-above non-favored televi-
sion programs.
Discussion
The Social Surrogacy Hypothesis predicts that people are
more likely to watch favored television programs over other
such stimuli when feeling lonely because favored television pro-
grams can buffer against loneliness. While this initial investiga-
tion yielded data congenial with this hypothesis, these data are
self-reported and descriptive, leaving numerous possible alterna-
tive explanations. For example, participants’ tendency to report
watching favored television programs when lonely may simply
be due to base rates. That is, the tendency to watch a favored
program when lonely was not compared to the tendency to
watch such programs when in other need states (or when needs
have been addressed). Instead, it may be that watching favored
programs simply occurs more frequently than does non-favored
programs.
To examine this question, an additional 73 participants indi-
cated how many hours per week they spent on each of the 31
activities employed in the previous study (see Table 1). Partici-
pants reported spending 5.31 h per week (SD = 4.52) on average
watching favored television programs, while spending 8.18 h per
week (SD = 8.91) on average ‘‘watching whatever was on televi-
sion.” Therefore, it appears unlikely that participants watch more
of their favorite television programs when feeling lonely simply
because they spend more time watching their favorite television
programs overall. The base rates appear to go in the opposite
direction. Instead, the Social Surrogacy Hypothesis predicts that
belongingness needs draw individuals to favored television
programs.
Study 2: Belongingness needs enhance thinking about favorite
television programs
The results of our correlational study suggest that belonging-
ness needs elicit high frequencies of self-reported consumption
of favorite television programs, and that self-reported feelings of
loneliness are minimized when watching favorite television pro-
grams. Due to the provocative, yet correlational nature of this evi-
dence, we examined each of these associations experimentally. In
Study 2, we examined how activating belongingness needs may in-
crease the desire to think about favored television programs.
Do people spend more time thinking about favored television
programs after a threat to a relationship? In Study 2, we manipu-
lated belongingness needs by having participants write about a
fight with a close other. Previous research has used similar reliving
tasks to successfully induce concerns about belongingness (see
Williams, 2007; see also Murray, Derrick, Leder, & Holmes, 2008;
Pickett, Gardner, & Knowles, 2004). Participants in the belonging-
ness needs control condition simply listed objects in their dwell-
ing. All participants then spent time writing about a television
program, but we manipulated the program about which partici-
pants wrote. Participants in the television control condition de-
scribed a time when they watched ‘‘whatever was on television.”
Participants in the favored television condition wrote about a time
when they watched their favorite television program.
We predicted that participants who had aroused belongingness
needs (those who wrote about fights with close others) would
write for longer about a favored television program than would
participants who did not have aroused belongingness needs. How-
ever, we predicted this difference between belongingness needs
conditions would not be observed when participants were writing
about ‘‘whatever was on television,” thereby distinguishing be-
Table 1
Mean ratings of participants’ likelihood of enacting each activity when feeling lonely, the likelihood of feeling lonely when enacting each activity, and time spent enacting each
activity.
Activity When feeling lonely Likely to feel lonely Time spent (in h)
M SD M SD M SD
Listen to music – a particular CD/tape 5.82 1.37 2.97 1.70 10.82 15.49
Watch television – a favorite TV program 5.81 1.34 2.64 1.55 5.31 4.52
Sleep 5.43 1.53 2.70 1.86 43.58 15.49
Surf the web 5.22 1.54 3.26 1.65 14.42 15.48
Eat 5.20 1.51 3.39 1.77 8.99 4.90
Exercise 5.02 1.66 2.73 1.52 5.66 4.38
Rent/watch a movie – an old favorite 5.00 1.67 3.43 1.68 3.38 3.88
Rent a movie – one you haven’t seen before 4.88 1.70 3.47 1.74 1.82 2.37
Take a shower 4.87 1.61 2.67 1.66 3.49 2.39
Watch television – anything that’s on 4.86 1.65 3.69 1.70 8.18 8.91
Focus on work or schoolwork 4.79 1.59 3.18 1.66 15.96 12.61
Clean your house/apartment/room 4.62 1.74 3.34 1.73 2.39 2.37
Go for a drive 4.57 1.86 3.87 1.76 1.15 1.85
Go for a walk 4.49 1.75 4.03 1.71 2.27 3.64
Look at old photographs 4.44 1.91 4.67 1.78 .80 1.15
Listen to music – anything that’s on the radio 4.41 1.71 3.57 1.73 7.30 10.45
Go to the mall; go shopping 4.36 1.91 3.54 1.77 1.53 1.81
Spend time with a pet 4.15 2.07 3.08 1.80 2.51 11.70
Read a book – a new one 3.89 1.94 3.16 1.77 1.57 3.23
Watch television – a sports game 3.87 2.17 3.45 1.81 4.43 4.92
Read a book – an old favorite 3.75 1.95 3.29 1.87 1.11 2.81
Play video games 3.63 2.15 3.65 1.79 4.21 5.98
Go to a movie 3.56 1.97 4.15 2.05 1.27 1.69
Engage in a favorite hobby 3.43 1.72 2.90 1.57 9.32 9.73
Look through old stuff or wear old clothes 3.20 1.76 4.27 1.87 1.19 2.19
Write in a journal or a diary 3.09 2.01 4.19 1.89 .37 1.04
Go to a bar or club 2.94 2.01 3.54 2.04 2.52 3.61
Go to church 2.85 1.96 3.16 1.81 .42 1.21
Drink alcohol (at home) 2.44 1.80 4.75 1.94 2.27 5.78
Smoke 1.85 1.72 4.23 2.03 1.29 5.22
Do drugs 1.66 1.45 4.29 2.11 .59 2.18
J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362 355
tween watching television for social surrogacy and watching tele-
vision merely for escapism. In line with the Social Surrogacy
Hypothesis, if belongingness needs elicit a desire for favored para-
social interactions, then exacerbating belongingness needs should
increase the amount of time participants spent writing about
(and therefore, thinking about) favored television programs, but
not non-favored television programs.
Method
Participants and design
One hundred and two undergraduates (49 women; mean
age = 20.23) participated in exchange for course credit. Most par-
ticipants (53%) were Caucasian; the remainder was predominantly
Asian (20%) and African American (19%). The experiment employed
a 2 (Belongingness Needs: Aroused Needs vs. Control) 2 (Paraso-
cial Essay: Favored vs. Control) between-subjects design. Partici-
pants were randomly assigned to condition.
Materials and procedure
Participation occurred in private cubicles on computers. All par-
ticipants wrote two essays. The first manipulated belongingness
needs. The second manipulated the importance of the television
program. The length of time spent writing about the television pro-
gram (measured via computer) served as the primary dependent
measure. Finally, participants were debriefed, thanked and
dismissed.
Belongingness needs (essay 1). As a manipulation of belongingness
needs, all participants wrote an initial essay. Participants assigned
to the Aroused Needs condition wrote about a time they fought
with a close other. Participants in the Control condition listed as
many items in their residence as they could remember. All partic-
ipants had 6 min to write their belongingness needs essay.
Parasocial essay (essay 2). All participants were asked to write
about a time in which they viewed a television program. Partici-
pants in the Favored condition wrote about a time they watched
their favorite television program, describing it in as much detail
as possible. Participants in the Control condition wrote about a
time when they had watched ‘‘whatever was on” television,
describing it in as much detail as possible. Participants were asked
to describe as much as they could about the content of the program
and their experience watching it.
4
Length of time writing this Para-
social essay served as the primary dependent measure.
Results and discussion
Of primary interest was the extent to which aroused belonging-
ness needs elicited an increase in the amount of time participants
spent elaborating upon (and thus, thinking about) favored, but not
non-favored television programs. To test this aspect of the Social
Surrogacy Hypothesis, we submitted time spent writing the para-
social essay to a 2 (Belongingness Needs: Aroused Needs vs. Con-
trol) 2 (Parasocial Essay: Favored vs. Control) ANOVA, which
yielded the predicted Belongingness Needs Parasocial Essay
interaction, F(1,98) = 7.91, p< .01 (Fig. 1).
In line with the Social Surrogacy Hypothesis, participants spent
significantly longer writing about favored television programs after
having belongingness needs aroused than after recalling objects in
their residence, t(98) = 2.07, p= .04. Importantly, this tendency for
belongingness needs to increase the time spent writing about a fa-
vored televised program does not appear to be mere escapism. If
this were true, activating belongingness needs should increase
the length of time spent elaborating upon and writing about any
recalled televised material. To the contrary, participants with
aroused belongingness needs spent significantly more time writing
about their favorite television program than writing about what-
ever was on television, t(98) = 2.82, p< .01. This effect was not sig-
nificant for participants who listed items in their residence,
t(98) = 1.18, p= .24. Moreover, participants spent marginally less
time writing about ‘‘whatever was on television” when belonging-
ness needs were aroused, as compared to when those needs were
not aroused, t(98) = 1.91, p= .06.
The Social Surrogacy Hypothesis predicts that television pro-
grams can serve as a surrogate for real interactions, that these faux
relationships can provide the experience of belonging. If true,
when belongingness needs are aroused favored television pro-
grams should become particularly attractive. In line with this
hypothesis, participants preferred to spend more time thinking
about (and thus writing about) a favored television program when
belongingness needs were aroused. Moreover, this effect does not
appear to be simple escapism; participants spent less time writing
about non-favored television programs when those same belong-
ingness needs were aroused.
5
Study 3: Favored television programs buffer against
relationship threats
Study 2 demonstrated that people spend more time thinking
and writing about favored television programs, with all of their
attendant parasocial relationships, when belongingness needs are
aroused. We also predict that such parasocial relationships can
actually buffer against the potential pain elicited by threats to val-
ued relationships. Drawing on extensive research on belongingness
needs, we hypothesized that recalling a relationship threat (fights
with close others) would decrease state self-esteem and mood and
increase feelings of rejection (e.g., Baumeister & Leary, 1995; Leary
et al., 1995, 1998).
6
However, in line with the Social Surrogacy
4
We were able to code 83 of the essays for the social vs. non-social nature of the
program described (e.g., sitcoms or dramas vs. news or cartoons). The vast majority of
participants described a social program: 78.7% in the Favored condition and 88.9% in
the Control condition. Analyses controlling for the type of program described yielded
consistent results in all studies.
0
60
120
180
240
300
360
420
480
Control Aroused
Belongingness Needs Condition
Time on Parasocial Essay (in sec)
Television Control Television Favorite
Fig. 1. Length of time spent writing television essay as a function of social needs
condition and type of television program.
5
Additional follow-up analyses were also conducted on the actual word length of
the essays. These analyses yielded a nearly identical interaction, F(1, 98) = 5.71,
p= .02.
6
We chose this manipulation (recalling a negative social event) because it has been
linked to the outcome variables in which we were interested, whereas more
immediate and intense rejection experiences tend to affect different dependent
variables (e.g., DeWall & Baumeister, 2006; Twenge, Baumeister, Tice, & Stucke, 2001;
Twenge, Catanese, & Baumeister, 2003; Twenge, Baumeister, et al., 2007; Twenge,
Zhang, et al., 2007; Zadro et al., 2004). It would be interesting for future research to
examine whether favored programming can buffer against those outcomes as well.
356 J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362
Hypothesis, recalling a favored television program (but not a non-fa-
vored television program) should buffer against these deleterious
effects.
Method
Participants and design
One hundred and sixteen undergraduates (59 women; mean
age = 19.81) participated in exchange for course credit. Most par-
ticipants (71%) were Caucasian; the remainder was predominantly
Asian (16%) and African American (8%). Participants who did not
answer a particular question were excluded from analyses involv-
ing that question. The same experimental design was used as in
Study 2; specifically, the experiment employed a 2 (Belongingness
Needs: Aroused Needs vs. Control) 2 (Parasocial Essay: Favored
vs. Control) between-subjects design. We were able to code 90 es-
says for the social vs. non-social nature of the program described.
The vast majority of participants again described a social program
in both the Favored (84.6%) and Control (94.7%) conditions. Partic-
ipants were randomly assigned to condition.
Materials and procedure
First, all participants completed a measure of global self-esteem
at a large preliminary testing session. This measure was collected
to serve as a covariate, to control for participants’ overall levels
of self-esteem. Participants were subsequently called to an osten-
sibly unrelated laboratory experiment. Participation in this second
phase of the experiment occurred in private cubicles on computers.
The manipulations in this study were nearly identical to those
utilized in Study 2. Participants wrote for 6 min describing a fight
with a close other or listing the contents of their residence. Partic-
ipants then wrote about either a time when they watched their
favorite television program or a time when they watched whatever
was on television. In Study 2, participants were not limited in the
amount of time they could spend writing the Parasocial Essay.
However, in the current experiment, participants in both condi-
tions were limited to writing for 6 min. After completing both es-
says, participants then completed two measures of state self-
esteem, a measure of mood, and a measure of feelings of rejection,
which served as the primary dependent measures. Participants
then completed some demographic questionnaires, after which
they were debriefed, thanked, and dismissed.
Global self-esteem measure. Participants completed the 10-item
Rosenberg Self-Esteem Scale (1965;
a
= .90) in an unrelated session
early in the semester. Participants responded to statements related
to global self-evaluations on 1 (not at all true) to 5 (completely
true) scales. Higher scores indicate higher global self-esteem.
State self-esteem measures. As in Leary et al. (1998), two measures of
state self-esteem were used. First, participants indicated how they
felt on ten self-relevant emotions (Leary et al., 1998; see also McFar-
land & Ross, 1982). Participants rated each emotion (
a
= .73) on a 1
(not at all) to 5 (extremely) scale. Second, participants completed
the Heatherton and Polivy (1991) measure of state self-esteem
(
a
= .88). Participants indicated their agreement with each of 20
self-related statements on a 1 (not at all) to 5 (extremely) scale. An
overall state self-esteem composite (see Leary et al., 1998) was com-
puted by standardizing the total scores on each scale and averaging
them (
a
= .74). Higher scores indicate higher state self-esteem.
Mood measure. The 20-item Positive and Negative Affect Schedule
(PANAS; Watson, Clark, & Tellegen, 1988) assessed participants’
positive (
a
= .76) and negative (
a
= .89) mood. Participants indi-
cated on a 1 (not at all) to 9 (extremely) scale how much they felt
each emotion (e.g., distressed, enthusiastic). The scores for positive
mood and negative mood were computed separately. Higher scores
indicate more positive and more negative mood, respectively.
Feelings of rejection. An eight item measure (
a
= .87) assessed par-
ticipants feelings of rejection. For three items, participants indi-
cated on a 1 (strongly disagree) to 7 (strongly agree) scale how
much they agreed with each statement (e.g., ‘‘Right now, I feel re-
jected”). For five items, participants indicated on a 1 (not at all
true) to 5 (extremely true) scale how much they agreed with rejec-
tion-related statements (e.g., ‘‘I feel alone right now”). The items
were standardized and averaged to compute the overall score.
Higher scores indicate stronger feelings of rejection.
Results
Of primary interest was the extent to which thinking about fa-
vored television programs buffers against relationship threat.
When belongingness needs are aroused by an insult to a relation-
ship (e.g., fighting with close others), this typically elicits reduc-
tions in state self-esteem, increases in negative mood, and
increases in feelings of rejection. If thinking about favored televi-
sion programs and the parasocial relationships they provide can
alleviate those increases in belongingness needs, then favored tele-
vision programs should buffer against the relationship threat elic-
ited by recalling a fight with a close other. Moreover, insofar as this
occurs uniquely for favored, but not for non-favored television pro-
grams, we can distinguish the benefits of these parasocial relation-
ships from mere escapism.
State self-esteem
Did the drop in state self-esteem due to relationship threat dis-
appear when participants wrote (and thus thought) about their
favorite television program? To test this hypothesis, we submitted
participants’ composite state self-esteem scores to a 2 (Belonging-
ness Needs: Aroused Needs vs. Control) 2 (Parasocial Essay: Fa-
vored vs. Control) Analysis of Covariance (ANCOVA), with global
self-esteem entered as a covariate. This was done to control for
participants’ overall levels of self-esteem. Naturally, global self-es-
teem predicted state self-esteem, F(1,100) = 79.81, p< .001. As pre-
dicted by the Social Surrogacy Hypothesis, a significant
Belongingness Needs Parasocial Essay interaction also emerged,
F(1,100) = 4.51, p= .04. As Fig. 2 demonstrates, when participants
wrote about non-favored television programs, they experienced
lower state self-esteem after belongingness needs were aroused,
as compared to when they simply recalled objects in their resi-
dence, t(100) = 2.45, p= .02. Thus, the manipulation of relation-
ship threat seems successful.
However, the insult to self-esteem typically engendered by
threats to close relationships was not observed when participants
wrote about favored programs, t(100) = .81, p= .42. Among partic-
ipants for whom belongingness needs had been aroused, those
who wrote about a favorite television program had higher state
self-esteem than did those who wrote about a non-favored televi-
sion program (i.e., whatever was on television), t(100) = 2.06,
p= .04. Self-esteem among participants who simply recalled items
from their residence, however, appeared to be unaffected by think-
ing about a favored or non-favored television program,
t(100) = .81, p= .42. As predicted by the Social Surrogacy Hypoth-
esis, therefore, it appears that calling to mind the parasocial rela-
tionships embedded in a favored television program can buffer
state self-esteem against the negative effects of a threat to a close
relationship.
7
7
The degrees of freedom for state self-esteem analyses are lower than those for
mood and feelings of rejection because several participants did not answer the self-
esteem questions in mass testing.
J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362 357
Negative mood
Negative mood is also a common result of a threat to a relation-
ship. The Social Surrogacy Hypothesis predicts that thinking about
favored television can also buffer against such negative moods. To
test this hypothesis, we submitted participants’ average negative
mood scores from the PANAS to a 2 (Belongingness Needs: Aroused
Needs vs. Control) 2 (Parasocial Essay: Favored vs. Control) ANO-
VA. As seen in Fig. 3, the Social Needs Parasocial Essay interac-
tion predicted by the Social Surrogacy Hypothesis again emerged,
F(1,112) = 4.61, p= .03.
8
For participants who wrote about a non-fa-
vored television program (i.e., whatever was on television), recalling
a fight with a close other led to stronger negative moods than did
simply recalling objects in their residence, t(112) = 2.76, p< .001.
As predicted by the Social Surrogacy Hypothesis, the effect of arous-
ing belongingness needs on negative moods was eliminated by
thinking about a favored television program, t(112) = .30, p= .77.
Additionally, among participants for whom belongingness needs
had been aroused, writing about a favored television program elic-
ited marginally lessened negative moods than did writing about a
non-favored television program, t(112) = 1.69, p= .09. Among partic-
ipants who simply recalled objects in their residence, however, writ-
ing about a favored or a non-favored television program did not have
a significant influence on negative mood states, t(112) = 1.33,
p= .19. Therefore, it appears that writing about a favorite television
program also buffered participants’ mood against the negative effect
of writing about a fight. In other words, participants who were given
the opportunity to call to mind parasocial relationships that com-
monly appear in favorite television programs were protected against
the negative mood effects of a relationship threat.
Feelings of rejection
The Social Surrogacy Hypothesis predicts that thinking about a
favored television program can also buffer against feelings of rejec-
tion. To test this hypothesis, we submitted participants’ scores on
the Feelings of Rejection measure to a 2 (Belongingness Needs:
Aroused Needs vs. Control) 2 (Parasocial Essay: Favored vs. Con-
trol) ANOVA. As can be seen in Fig. 4, the Social Needs Parasocial
Essay interaction predicted by the Social Surrogacy Hypothesis was
marginally significant, F(1,112) = 3.01, p= .09. For participants
who wrote about a non-favored television program (i.e., whatever
was on television), recalling a fight with a close other led to mar-
ginally greater feelings of rejection than did simply recalling ob-
jects in their residence, t(112) = 1.85, p= .07. As predicted by
the Social Surrogacy Hypothesis, the effect of arousing belonging-
ness needs on feelings of rejection was eliminated by thinking
about a favored television program, t(112) = .61, p= .54. Addi-
tionally, among participants for whom belongingness needs had
been aroused, writing about a favored television program elicited
marginally lessened feelings of rejection than did writing about a
non-favored television program, t(112) = 1.78, p= .08. Among par-
ticipants who simply recalled objects in their residence, however,
writing about a favored or a non-favored television program had
no influence on feelings of rejection, t(112) = .65, p= .51. There-
fore, it appears that writing about a favorite television show buf-
fered participants’ feelings of rejection against the negative effect
of writing about a fight.
Discussion
The results of Study 3 support the Social Surrogacy Hypothesis
across all three dependent measures. Whereas recalling a fight
with a close other serves to depress self-esteem, increase negative
mood, and increase feelings of rejection, thinking about a favored
television program buffers against such negative effects. Moreover,
these effects appear unique to favorite television programs, not any
television program, suggesting the results are due to more than
just mere distraction. Thus, it seems that only well-established
parasocial relationships found in beloved television programs offer
any protective benefit; mere escapism via the consumption of
other televised media does not seem sufficient.
Study 4: Television programs assuage chronic belongingness
needs
The results of the previous studies provide support for the So-
cial Surrogacy Hypothesis. Study 2 suggests that people revel long-
1
1.5
2
2.5
3
Control Aroused
Belongingness Needs Condition
Negative Mood
Television Control Television Favorite
Fig. 3. Negative mood as a function of social needs condition and type of television
program.
-0.4
-0.2
0
0.2
0.4
Control Aroused
Belongingness Needs Condition
State Self-Esteem
Te
levision Control
Television Favorite
Fig. 2. State self-esteem as a function of social needs condition and type of
television program.
8
Analyses of positive mood data yielded no significant effects.
-0.4
-0.2
0
0.2
0.4
Control Aroused
Belongingness Needs Condition
Feelings of Rejection
Television Control Television Favorite
Fig. 4. Feelings of rejection as a function of social needs condition and type of
television program.
358 J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362
er in the recall of television programs when their belongingness
needs have been aroused. Study 3 suggests that thinking about fa-
vored (but not non-favored) television programs can actually buf-
fer against the drops in self-esteem, increases in negative mood,
and feelings of rejection typically experienced when relationships
suffer a threat. Yet in both experiments, the possibility remains
that the results were obtained because of the generally positive
experience of thinking about a favored television program, rather
than because of the increased experience of belongingness that
the Social Surrogacy Hypothesis suggests. Perhaps participants
turn to favorite television programs, not because they make them
feel less lonely, but because they make them generally happier. To
examine this hypothesis, Study 4 directly examined the effects of
favorite television programs on the reduction of belongingness
needs and on mood.
As psychologists have long known, unalleviated goal states are
accompanied by the increased accessibility of goal-related con-
cepts (Zeigarnik, 1927). Indeed, one of the signature effects of hav-
ing an unmet need or goal is sustained activation of need- or goal-
relevant concepts (Förster, Liberman, & Friedman, 2007). More-
over, a certain chronic level of anxiety about social rejection or lack
of acceptance is common among most people (Leary & Kowalski,
1995). Drawing on this logic, we hypothesized that if thinking of
favored television programs alleviates chronically activated
belongingness needs, the accessibility of exclusion-related con-
cepts would be reduced.
To examine this, we had participants think about either watch-
ing a favored television program, watching whatever is on televi-
sion (television control), or experiencing an academic success (a
non-social, positive control). We then employed a word comple-
tion task (e.g., Bassili & Smith, 1986; Sinclair & Kunda, 1999) de-
signed to measure the accessibility of words related to loneliness
or exclusion, and words related to positive and negative mood. Pre-
vious researchers have found that people who are depressed or in
negative moods show priming effects for negative words stems
(e.g., Ruiz-Caballero & Gonzalez, 1994, 1997; Watkins, Vache, Ver-
ney, & Mathews, 1996). We predicted that relative to the two con-
trol conditions, participants primed with favored television
programs would show weaker accessibility of exclusion-related
concepts because their belongingness needs should be reduced.
However, we did not expect that the accessibility of mood related
words would vary by condition. In other words, we did not expect
that the experience of writing about a favored television program
would be any more generally positive or negative of an experience
than writing about non-favored programs or about a non-social
control. After all, academic successes and leisure time ‘‘zoning out”
in front of the television are both positive experiences that many
people seek out and enjoy. Thus, we predicted that favored televi-
sion programs exert their power by providing the experience of
belonging, not by altering mood. At first glance, this may seem at
odds with the results of Study 3 in which we found that thinking
about a favored television program reduced negative mood; how-
ever, it should be noted that the reduction was only evident when
participants’ social needs were first activated and thus mood threa-
tened. For participants in the control condition, writing about a fa-
vored or a non-favored television program had no influence on
mood states.
Method
Participants and design
Two hundred and twenty two undergraduates (129 women;
mean age = 19.27) participated in exchange for course credit. Most
participants were Caucasian (68%); the remainder was predomi-
nantly Asian (17%) and African American (8%). Twenty-four partic-
ipants (11%) did not speak English as their primary language and
were excluded from analyses. An additional five participants did
not answer the word completions and were excluded from analy-
ses. The final sample consisted of 193 participants (110 women).
Participants were randomly assigned to one of three conditions: fa-
vored television program, whatever was on television, or academic
success.
Procedure and materials
Participation occurred in private cubicles on computers. Partic-
ipants first completed the parasocial manipulation: they wrote an
essay about their favorite television program, about programs they
watch on television ‘‘when nothing else is on” (television control),
or about experiencing a positive academic achievement (non-so-
cial control). Participants were asked to write about the topic for
10 min, and they were asked to describe it in as much detail as pos-
sible. Participants then attempted 48 word completions. All word
stems were presented at the same time on the computer screen,
with letters missing from each word. They were presented to-
gether so that participants could answer them in any order they
chose. Participants were tasked with entering the missing letters
in provided blank spaces. Each item could be completed in more
than one way. The target words to be completed included three
exclusion words, eight negative non-exclusion words, and five po-
sitive non-exclusion words, embedded among an additional 32 fil-
ler words, for a total of 48 word stem completions. Finally,
participants completed some demographic information and were
debriefed, thanked, and dismissed.
Exclusion words. The exclusion-related words were exclude, reject
and hate. They were presented to participants as exc___, rej_ _ _,
and ha__. Examples of other ways the items could be completed
are excite, rejoin, and hard. Items completed as exclusion words
were coded as 1, and items completed in other ways were coded
as 0. The three items were averaged to create an exclusion compos-
ite, or a percentage of the words that were completed using exclu-
sion words. Higher scores indicate greater accessibility of
exclusion-related concepts.
Positive mood related words. The positive words were happy, calm,
cheer, joy, and glad. They were presented to participants as ha_ _ _,
ca__, che_ _, jo_, and gl_ _. Examples of other ways the items could
be completed are hands, cake, check, job, and glow. Items com-
pleted as positive mood related words were coded as 1, and items
completed in other ways were coded as 0. The five items were
averaged to create a positive composite. Higher scores indicate
greater accessibility of positive mood related words.
Negative mood related words. The negative words were sad,
afraid, fear, anger, upset, bad, scared, and mad. They were pre-
sented to participants as sa_, af__ __, fe_ _, ang__, up_ _ _, ba_,
sca__ _, and ma_. Examples of other ways the items could be com-
pleted are say, afford, feet, angel, upper, bat, scales, and man. Items
completed as negative mood related words were coded as 1, and
items completed in other ways were coded as 0. The eight items
were averaged to create a negative composite. Higher scores indi-
cate greater accessibility of negative mood related words.
Filler words. Thirty-two additional word stems were included that
could not be completed with valid English words in ways that de-
noted or connoted exclusion, positive mood, or negative mood. For
example the__ could be completed as there, their, theme, etc., none
of which is exclusion relevant.
Results and discussion
Are chronic belongingness needs uniquely reduced among indi-
viduals who have thought about a favored television program, as
J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362 359
compared to those who have thought about either a non-favored
television program or about a positive, but non-social life event?
To answer this question, we submitted the exclusion composite
to a one-way ANOVA with three levels. As predicted, the effect of
condition on the completion of exclusion words emerged,
F(2,190) = 4.34, p= .01. Participants in the favored television con-
dition (M= .35, SD = .16) completed significantly fewer items as
exclusion words than participants in either the television control,
M= .43, SD = .16, t(190) = 2.56, p= .01, or the positive non-social
control, M= .43, SD = .18, t(190) = 2.56, p= .01.
The manipulation, however, had no effect on the completion
of positive, F(2,190) = .92, p= .40, or negative, F(2,190) = .43,
p= .65, mood related words. Additionally, the manipulation had
no effect on participants’ overall capacity to complete word
stems, F(2,190) = 1.56, p= .21. We also conducted a follow-up
ANCOVA on the index of exclusion accessibility that included
the number of accurately completed filler word stems as a
covariate. This analysis yielded a nearly identical pattern of data
to the initial analysis, F(2, 189) = 3.45, p= .03, with participants
in the positive parasocial condition completing fewer exclu-
sion-related words than the participants in the other two condi-
tions. Therefore, it appears that thinking about a favorite
television program decreased the accessibility of exclusion-re-
lated words, even controlling for participants’ ability to accu-
rately complete word stems. Just as completing a goal or
reducing a need can inhibit the accessibility of goal- or need-re-
lated words, so too does thinking about valued parasocial rela-
tionships reduce the accessibility of belongingness related
concepts. This effect appears to be a result of the reduction of
belongingness needs, not improved mood; there was no differ-
ence across conditions in the completion of positive or negative
mood words.
General discussion
Across four studies, one correlational and three experimental,
and seven different dependent measures, we have provided novel
evidence for the Social Surrogacy Hypothesis: humans can use tech-
nologies, such as television, to provide the experience of belonging.
In line with our initial correlational evidence, Study 2 found that
belongingness needs lead people to revel for longer in a description
of their favored (but not non-favored) television program. Also in
line with our correlational evidence, Study 3 found that thinking
about a favored (but not non-favored) television program can buf-
fer against the effects on self-esteem, mood, and feelings of rejec-
tion that are commonly elicited by a threat to a close relationship.
Extending this logic, just as addressing a goal reduces the activa-
tion of goal-related concepts, Study 4 found that thinking of a val-
ued television program reduced activation of exclusion-related
concepts, providing evidence that these televised parasocial rela-
tionships can ease latent belongingness needs.
Alternate explanations and limitations
Although we believe our data demonstrate the ability of favorite
television programs to provide the experience of belonging, there
are alternative explanations for the findings that need to be ad-
dressed. First, it is possible that people are more likely to watch
a favored television program with another actual person, as com-
pared to other televised content. Therefore, the effects of Studies
2 and 3 may have been driven by the social atmosphere of watch-
ing favorite programs rather than by the parasocial ‘‘interaction.”
Indeed, participants did report watching favored television pro-
grams with others more frequently than non-favored programs,
t(100) = 4.52, p< .01 in Study 2, and t(114) = 7.06, p< .01 in Study
3. However, follow-up analyses of each dependent variable, con-
trolling for whether participants reported having watched the pro-
gram with others, still yielded the predicted Social
Needs Parasocial Essay interactions: F(1,97) = 8.38, p< .01 for
time spent writing in Study 2; F(1, 99) = 4.19, p= .04 for state
self-esteem; F(1,111) = 5.45, p= .02 for negative mood; and
F(1,111) = 3.12, p= .09 for the marginally significant interaction
for feelings of rejection in Study 3. Therefore, the presence of oth-
ers during the initial viewing of the televised programming cannot
clearly account for the effects.
Second, we acknowledge that the use of the same control con-
dition (watching whatever is on television) in all of the experi-
ments is a potential limitation. Because only one control
condition was used, it is possible that some other difference be-
tween watching whatever is on television vs. watching one’s favor-
ite program (other than the propensity to provide a social
surrogate) accounts for some of the differences. For example, per-
haps watching a favorite program is more enjoyable, easier, or
more rewarding. Our concerns about this limitation are tempered
by the findings in the control (belongingness needs not activated)
conditions of Study 2 and Study 3. Across these studies, control
participants who wrote about favored television programs did
not differ from those who wrote about non-favored television pro-
grams in the time they spent writing, mood, state self-esteem, or
feelings of rejection, suggesting that the two conditions only differ
in their ability to address deficits on belonging. Nonetheless, we
thought it prudent to run a replication of one of the studies (Study
4) using a different control condition. In the replication, partici-
pants in the control condition also wrote about their favorite tele-
vision program. However, they were specifically told not to write
about the characters or plots but to focus on the non-social aspects
of their favorite program (e.g., they wrote about what time it was
on, where it was filmed, what the costumes and sets were like,
etc.). Participants then completed the PANAS mood measure (Wat-
son et al., 1988) and the differential loneliness scale (Schmidt &
Sermat, 1983). Thus, the follow-up study was a conceptual replica-
tion of Study 4 with a different control condition, a different mea-
sure of mood, and a different measure of belonging. The results of
Study 4 were replicated exactly. Although both positive (p=.72)
and negative (p= .95) mood were not affected by the manipulation,
loneliness was t(45) = 2.41, p= .02. Participants who wrote about
their favorite program felt significantly less isolated from social
groups than those in the control condition. Thus, our ability to rep-
licate Study 4 using a different essay in the control condition re-
duces the likelihood that the social surrogacy effects are driven
by confounds of writing about whatever is on television.
Finally, we acknowledge that although all of the stages in our
model are supported by at least one of the studies presented, none
of our studies simultaneously support all stages of our model. Spe-
cifically, our model argues that people turn to favored television
programs when they feel lonely (Studies 1 and 2) because those
programs reduce the negative effects of loneliness (Studies 1 and
3) by providing the experience of belonging (Study 4). Because
none of the studies measures all of the stages together, we cannot
be certain of the mediating role of providing the experience of
belonging; it is possible that the effects of Studies 1–3 are driven
by something other than addressing these belongingness needs.
For example, perhaps watching a favored television program is
enjoyable and thus increases mood, which leads to other positive
outcomes.
9
Although intriguing, we think this alternate explanation
is unlikely for a number of reasons. First, Study 3 demonstrated that
writing about a favored television program only affected mood and
self-esteem when belongingness needs are activated, indicating a
9
We thank a helpful reviewer for this suggestion.
360 J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362
link specific to social situations. Second, both Study 4 and its replica-
tion, each using different measures, found direct effects of television
on loneliness but no effects on mood. Thus, we believe that our
explanation of the data is the most parsimonious. Nonetheless, we
acknowledge that these data present only a first examination of
the Social Surrogacy Hypothesis and that additional research is
needed.
Trick or treat: Does social surrogacy suppress or fulfill belongingness
needs?
Throughout, we have discussed social surrogacy as providing
the experience of belonging, even without having a ‘‘true” social
interaction. As yet, it remains an open question as to whether such
social surrogacy merely suppresses belongingness needs, or
whether such surrogacy actually fulfills the need. While beyond
the scope of this initial investigation, investigating the extent to
which social surrogacy suppresses or fulfills belongingness needs
is certainly a necessary next step.
Similarly, seemingly inherent in the argument that social surro-
gacy can serve as an alternative to ‘‘real” social interaction is the
argument that social surrogacy is maladaptive, or at least an impov-
erished experience as compared to ‘‘real” interaction. We do not,
however, necessarily endorse such an evaluative position. We be-
lieve that extracting the experience of belonging from technology
is neither an inherently adaptive nor maladaptive behavior. For
those with very high belongingness needs, such need management
may reduce chronic feelings of rejection or isolation. For others
who might have difficulty experiencing social interaction due to
physical or environmental constraints, such technologically induced
belongingness may provide otherwise difficult to achieve succor. Of
course, using such technologies to the exclusion of true social inter-
action is certainly maladaptive, and likely self-defeating. Turning
one’s back on friends and family for the solace of televised media
for extended periods of time will likely leave an individual with few-
er rather than greater social resources over time.
Parasocial relationships and relationship processes
This research also raises the question of the boundary of what
might be considered a relationship. Specifically, in the current re-
search, we allowed participants the complete freedom to think
about favored (or non-favored) television programs, and did not
specify that they consider a favorite character or media persona.
Analyses of participants’ essays suggest that individuals frequently
do isolate a favorite character or characters within a program,
especially when writing about their favorite television program
(87%). However, the Social Surrogacy Hypothesis does not specify
what about the favorite television program allows social surrogacy
to occur. It is possible that people identify with a particular charac-
ter (e.g., Cohen, 2006), that they interact parasocially with a char-
acter (e.g., Horton & Wohl, 1956), or that they experience
transportation into the narrative world, allowing them to transport
themselves into the program and enjoy the full story and the full
gamut of relationships (e.g., Green, 2005; Green, Brock, & Kaufman,
2004). Although the current model does not specify at what level
(e.g., individual, collective) or through what framework (e.g.,
attachment, identification) the parasocial relationship and social
surrogacy itself occurs, it is certainly possible that different types
of parasocial relationships may yield the experience of belonging
differently. For example, given that men seem to prefer social
interactions within groups or hierarchies and women seem to pre-
fer dyads (Baumeister & Sommer, 1997; Gabriel & Gardner, 1999),
it is possible that men may prefer parasocial interactions with
groups or teams, and have their needs best met by such collective
parasocial relationships.
Moreover, we have discussed social surrogacy as though it were
something that all people are inclined to do. However, it is possible
that the tendency to use television in this manner is limited to a
particular portion of the population. For example, past research
has demonstrated that people with low self-esteem, people with
an anxious-ambivalent or preoccupied attachment style, people
low in trust, and people high in the need to belong are more likely
to experience parasocial relationships, experience stronger paraso-
cial relationships, or are more likely to turn to parasocial activities
than their more securely-attached or ‘‘better-adjusted” counter-
parts (Babb, 1995; Cohen, 2004; Cole & Leets, 1999; Green & Brock,
1998; Knowles, 2007). To investigate this possibility in the current
research, we measured several individual difference variables in a
large testing session at the beginning of the semester that each
study was run. We assessed self-esteem (Rosenberg, 1965), attach-
ment style (Bartholomew & Horowitz, 1991), need to belong
(Leary, Kelly, Cottrell, & Schreindorfer, 2005; not assessed in Study
2), and gender. None of the individual difference variables moder-
ated our findings.
10
Although this list of individual difference vari-
ables is certainly not exhaustive, the results of these analyses
suggest that social surrogacy does not appear limited to just one
gender or to ‘‘insecure” or ‘‘maladjusted” individuals.
Conclusion
In our current research, we have endeavored to take seriously
the hypothesis that seemingly asocial human technologies, such
as television, can actually serve a social function. Indeed, we be-
lieve that the Social Surrogacy Hypothesis is a theoretically-driven
treatment of a truly ubiquitous part of the modern human environ-
ment, one that moves beyond mere descriptive evidence to test no-
vel, provocative, social psychological hypotheses about basic
human needs. We believe that investigating the underlying moti-
vations for such human technologies offers a new way for social
psychologists to consider the functional underpinnings of human
behavior. Perhaps more importantly, we hope that the current
work serves to elucidate a new mechanism by which humans
can address an important need: the need to belong.
Acknowledgments
We thank a team of research assistants for their assistance in
conducting this research and the social psychology area at the Uni-
versity at Buffalo for their helpful comments during early stages of
this research.
References
Babb, V. D. (1995). The influence of psychological, sociological, and communication
predictors on adolescents’ development of parasocial relationships and their
television uses and gratifications. Dissertation Abstracts International Section A:
Humanities and Social Sciences, 56(6A), 2027. US: Univ Microfilms International.
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults:
A test of a four-category model. Journal of Personality and Social Psychology, 61,
226–244.
Bassili, J. N., & Smith, M. C. (1986). On the spontaneity of trait attribution:
Converging evidence for the role of cognitive strategy. Journal of Personality and
Social Psychology, 50, 239–245.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117,
497–529.
Baumeister, R. F., & Sommer, K. L. (1997). What do men want? Gender differences
across two spheres of belongingness: Comment on Cross and Madson (1997).
Psychological Bulletin, 122, 38–44.
Buckley, K. E., Winkel, R. E., & Leary, M. R. (2004). Reactions to acceptance and
10
A marginal relationship between attachment anxiety and the likelihood of feeling
lonely while watching a favorite television program trended toward significance in
Study 1, r(617) = .08, p= .06, all other ps > .68. Given the large sample size in this
study (all ns > 617), the absence of any significant correlations seems noteworthy.
J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362 361
rejection: Effects of level and sequence of relational evaluation. Journal of
Experimental Social Psychology, 40, 14–28.
Cohen, J. (2004). Parasocial breakup from favorite television characters: The role of
attachment styles and relationship intensity. Journal of Social and Personal
Relationships, 21, 187–202.
Cohen, J. (2006). Audience identification with media characters. In J. Bryant & P.
Vorderer (Eds.), Psychology of entertainment (pp. 183–197). Mahwah, NJ:
Lawrence Erlbaum.
Cole, T., & Leets, L. (1999). Attachment styles and intimate television viewing:
Insecurely forming relationships in a parasocial way. Journal of Social and
Personal Relationships, 16, 495–511.
Derrick, J. L., Gabriel, S., & Tippin, B. (2008). Parasocial relationships and self-
discrepancies: Faux relationships have benefits for low self-esteem individuals.
Personal Relationships, 15, 261–280.
DeWall, C. N., & Baumeister, R. F. (2006). Alone but feeling no pain: Effects of social
exclusion on physical pain tolerance and pain threshold, affective forecasting,
and interpersonal empathy. Journal of Personality and Social Psychology, 91,
1–15.
Drigotas, S. M., Rusbult, C. E., Wieselquist, J., & Whitton, S. W. (1999). Close partner
as sculptor of the ideal self: Behavioral affirmation and the Michelangelo
phenomenon. Journal of Personality and Social Psychology, 77, 293–323.
Förster, J., Liberman, N., & Friedman, J. S. (2007). Seven principles of goal activation:
A systematic approach to distinguishing goal priming from priming of non-goal
constructs. Personality and Social Psychology Review, 11, 211–233.
Frijda, N. H. (1988). The laws of emotion. American Psychologist, 43, 349–358.
Gabriel, S., Carvallo, M., Jaremka, L., & Tippin, B. D. (2008). A friend is a present you
give to your ‘‘self”: Avoidance of intimacy moderates the effects of friends on
self-liking. Journal of Experimental Social Psychology, 44, 330–343.
Gabriel, S., & Gardner, W. L. (1999). Are there ‘‘his” and ‘‘hers” types of
interdependence? The implications of gender differences in collective versus
relational interdependence for affect, behavior, and cognition. Journal of
Personality and Social Psychology, 77, 642–655.
Gardner, W. L., & Knowles, M. L. (2008). Love makes you real: Favorite television
characters are perceived as ‘‘real” in a social facilitation paradigm. Social
Cognition, 26, 156–168.
Gardner, W. L., Pickett, C. L., & Knowles, M. (2005). Social snacking and shielding:
Using social symbols, selves, and surrogates in the service of belong needs. In K.
D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism,
social exclusion, rejection, and bullying (pp. 243–262). New York: Psychology
Press.
Green, M. C. (2005). Transportation into narrative worlds: Implications for the self.
In A. Tesser, J. V. Wood, & D. A. Stapel (Eds.), On building, defending and regulating
the self: A psychological perspective (pp. 53–75). New York: Psychology Press.
Green, M. C., & Brock, T. C. (1998). Trust, mood, and outcomes of friendship
determine preferences for real versus ersatz social capital. Political Psychology,
19, 527–544.
Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment:
The role of transportation into narrative worlds. Communication Theory, 14,
311–327.
Harlow, H. (1958). The nature of love. American Psychologist, 13, 573–685.
Heatherton, T. F., & Polivy, J. (1991). Development and measurement of a scale for
measuring state self-esteem. Journal of Personality and Social Psychology, 60,
895–910.
Hogan, P. C. (2003). The mind and its stories: Narrative universals and human emotion.
New York: Cambridge University Press.
Horton, D., & Wohl, R. R. (1956). Mass communication and para-social interaction:
Observations on intimacy at a distance. Psychiatry, 19, 215–229.
Kanazawa, S. (2002). Bowling with our imaginary friends. Evolution and Human
Behavior, 23, 167–171.
Knowles, M. L. (2007). The nature of parasocial relationships. Unpublished doctoral
dissertation. Northwestern University, Evanston, IL.
Leary, M. R., Kelly, K. M., Cottrell, C. A., & Schreindorfer, L. S. (2005). Individual
differences in the need to belong: Mapping the nomological network. Unpublished
manuscript. Wake Forest University.
Leary, M. R., Haupt, A. L., Strausser, K. S., & Chokel, J. T. (1998). Calibrating the
sociometer: The relationship between interpersonal appraisals and state self-
esteem. Journal of Personality and Social Psychology, 74, 1290–1299.
Leary & Kowalski (1995). Social anxiety. New York: Guilford Press.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an
interpersonal monitor: The sociometer hypothesis. Journal of Personality and
Social Psychology, 68, 518–530.
Mar, R. A., & Oatley, K. (2008). The function of fiction is the abstraction and
simulation of social experience. Perspectives on Psychological Science, 3.
Mar, R. A., Oatley, K., Hirsh, J., dela Paz, J., & Peterson, J. B. (2006). Bookworms versus
nerds: Exposure to fiction versus non-fiction, divergent associations with social
ability, and the simulation of fictional social worlds. Journal of Research in
Personality, 40, 694–712.
McFarland, C., & Ross, M. (1982). Impact of causal attributions on affective reactions
to success and failure. Journal of Personality and Social Psychology, 43, 937–946.
Murray, S. L., Derrick, J. L., Leder, S., & Holmes, J. G. (2008). Balancing connectedness
and self-protection goals in close relationships: A levels of processing perspective
on risk regulation. Journal of Personality and Social Psychology, 94, 429–459.
Oatley, K. (1999). Meetings of minds: Dialogue, sympathy, and identification, in
reading fiction. Poetics, 26, 439–454.
Pickett, C., Gardner, W. L., & Knowles, M. L. (2004). Getting a cue: The need to belong
influences attention to subtle social cues. Personality and Social Psychology
Bulletin, 30, 1095–1107.
Pinker, S. (1997). How the mind works. New York: Norton & Company.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.
Ruiz-Caballero, J. A., & Gonzalez, P. (1994). Implicit and explicit memory bias in
depressed and nondepressed subjects. Cognition and Emotion, 8, 555–569.
Ruiz-Caballero, J. A., & Gonzalez, P. (1997). Effects of level of processing on implicit
and explicit memory in depressed mood. Motivation and Emotion, 21, 195–209.
Schmidt, N., & Sermat, V. (1983). Measuring loneliness in different relationships.
Journal of Personality and Social Psychology, 44, 1038–1047.
Sinclair, L., & Kunda, Z. (1999). Reactions to a black professional: Motivated
inhibition and activation of conflicting stereotypes. Journal of Personality and
Social Psychology, 77, 885–904.
Twenge, J. M., Baumeister, R. F., DeWall, C. N., Ciarocco, N. J., & Bartels, J. M. (2007).
Social exclusion decreases prosocial behavior. Journal of Personality and Social
Psychology, 92, 56–66.
Twenge, J. M., Baumeister, R. F., Tice, D. M., & Stucke, T. S. (2001). If you can’t join
them, beat them: Effects of social exclusion on aggressive behavior. Journal of
Personality and Social Psychology, 81, 1058–1069.
Twenge, J. M., Catanese, K. R., & Baumeister, R. F. (2003). Social exclusion and the
deconstructed state: Time perception, meaninglessness, lack of emotion, and
self-awareness. Journal of Personality and Social Psychology, 85, 409–423.
Twenge, J. M., Zhang, L., Catanese, K. R., Dolan-Pascoe, B., Lyche, L. F., & Baumeister,
R. F. (2007). Replenishing connectedness: Reminders of social activity reduce
aggression after social exclusion. British Journal of Social Psychology, 46,
205–224.
US Department of Labor (2006). American time use survey. <http://www.bls.gov/
tus/>. Retrieved 28.07.07.
Watkins, P. C., Vache, K., Verney, S. P., & Mathews, A. (1996). Unconscious mood-
congruent memory bias in depression. Journal ofAbnormal Psychology, 105, 34–41.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief
measures of positive and negative affect: The PANAS scales. Journal of
Personality and Social Psychology, 54, 1063–1070.
Williams, K. D. (2007). Ostracism. Annual Review of Psychology, 58, 425–452.
Williams, K. D., Cheung, K. T., & Choi, W. (2000). Cyberostracism: Effects of being
ignored over the internet. Journal of Personality and Social Psychology, 79,
748–762.
Zadro, L., Williams, K. D., & Richardson, R. (2004). How low can you go? Ostracism
by a computer lowers belonging, control, self-esteem, and meaningful
existence. Journal of Experimental Social Psychology, 40, 560–567.
Zeigarnik, B. (1927). Das Behalten erledigter und unerledigter Handlungen (On the
recall of completed and incompleted tasks). Psychologische Forschung, 9, 1–85.
362 J.L. Derrick et al. / Journal of Experimental Social Psychology 45 (2009) 352–362
