ArticlePDF Available

Do online social media cut through the constraints that limit the size of offline social networks?


Abstract and Figures

The social brain hypothesis has suggested that natural social network sizes may have a characteristic size in humans. This is determined in part by cognitive constraints and in part by the time costs of servicing relationships. Online social networking offers the potential to break through the glass ceiling imposed by at least the second of these, potentially enabling us to maintain much larger social networks. This is tested using two separate UK surveys, each randomly stratified by age, gender and regional population size. The data show that the size and range of online egocentric social networks, indexed as the number of Facebook friends, is similar to that of offline face-to-face networks. For one sample, respondents also specified the number of individuals in the inner layers of their network (formally identified as support clique and sympathy group), and these were also similar in size to those observed in offline networks. This suggests that, as originally proposed by the social brain hypothesis, there is a cognitive constraint on the size of social networks that even the communication advantages of online media are unable to overcome. In practical terms, it may reflect the fact that real (as opposed to casual) relationships require at least occasional face-to-face interaction to maintain them.
This content is subject to copyright.
Cite this article: Dunbar RIM. 2016 Do online
social media cut through the constraints that
limit the size of oine social networks? R. Soc.
open sci. 3: 150292.
Received: 25 June 2015
Accepted: 9 December 2015
Subject Category:
Psychology and cognitive neuroscience
Subject Areas:
network size, social networking sites (SNSs),
social brain hypothesis, egocentric networks
Author for correspondence:
R. I. M. Dunbar
Electronic supplementary material is available
at or via
Do online social media cut
through the constraints that
limit the size of oine
social networks?
R. I. M. Dunbar
Department of Experimental Psychology, University of Oxford,S outh Parks Road,
Oxford OX1 3UD, UK
The social brain hypothesis has suggested that natural social
network sizes may have a characteristic size in humans. This
is determined in part by cognitive constraints and in part
by the time costs of servicing relationships. Online social
networking offers the potential to break through the glass
ceiling imposed by at least the second of these, potentially
enabling us to maintain much larger social networks. This is
tested using two separate UK surveys, each randomly stratified
by age, gender and regional population size. The data show
that the size and range of online egocentric social networks,
indexed as the number of Facebook friends, is similar to that of
offline face-to-face networks. For one sample, respondents also
specified the number of individuals in the inner layers of their
network (formally identified as support clique and sympathy
group), and these were also similar in size to those observed
in offline networks. This suggests that, as originally proposed
by the social brain hypothesis, there is a cognitive constraint
on the size of social networks that even the communication
advantages of online media are unable to overcome. In practical
terms, it may reflect the fact that real (as opposed to casual)
relationships require at least occasional face-to-face interaction
to maintain them.
1. Introduction
Thanks to the Internet, the past decade has witnessed a dramatic
revolution in our social world. By providing novel channels
that allow us to communicate with individuals that we would
otherwise have difficulty meeting face to face, the Internet has
made it possible to service existing relationships as well as meet
new individuals more efficiently and on a wider geographical
scale. Given the extensive use of social media, the question as
to whether Internet-based social networking sites (SNSs) have a
positive or negative impact on social relationships has been much
debated [1,2]. Cyberpessimists (e.g. [37]) have argued that the
2016 The Authors. Published by the Royal Society under the terms of the Creative Commons
Attribution License, which permits unrestricted
use, provided the original author and source are credited.
2 R. Soc. open sci. 3:150292
Internet has detrimental effects on our social life. In contrast, cyberoptimists (e.g. [8,9]) have insisted that
the effects have been beneficial in many different ways (for reviews, see [10,11]).
There is some evidence to suggest that one of the motivations for using social media among teenagers
is to extend their range of social contacts [12]. Indeed, it has specifically been claimed that those who
are more socially competent use social media to expand their network of friendships, thereby increasing
their social capital [1316]. However, young children, and to some extent teenagers, are relatively poor
at judging relationship quality: very young children, for example, commonly mistake a desire to form
friendships on their part with the assumption that such friendships are reciprocated [17]. Adults tend
to be more attuned to the nuances in different types of relationship and are less prone to signing up to
‘friending’ requests without considering the nature of the relationship involved. They may thus provide
a better test of the hypothesis than the teenagers who have been the focus of most social media research.
One specific respect in which the Internet has been held to change our social world is the size of
our social networks. On the basis of a projection from an equation relating social community size to
neocortex volume in apes, it had previously been suggested that there is a natural group size for humans
[18], and this has been validated against hunter–gatherer community sizes [19,20]aswellasthesizesof
offline personal social networks (egocentric networks) in two European populations [21,22]. This limit
is thought to arise from a combination of a cognitive constraint (the product of the relationship with
neocortex size known as the social brain hypothesis (SBH) [18,23]) and a time constraint associated with
the costs of servicing relationships [24,25]. Implicit evidence for a potential cognitive constraint has been
provided by a number of neuroimaging studies which show that individual variation in adult social
network size correlates with the volume of core areas in the neocortex (notably those regions of the
prefrontal and temporal lobes) that are associated with the ‘theory of mind’ network in humans [2629],
and this also seems to be true of monkeys [30].
An important feature of natural social networks in both humans [19,20,25] and non-human primates
[31] is that they are structured into a distinctive series of hierarchically inclusive layers that have a
natural scaling ratio of approximately 3. These layers reflect both interaction frequencies and, at least
in humans, emotional closeness [22,25,31]. In humans, these layers have values that approximate 5, 15,
50 and 150, and extend beyond this in at least two further layers to 500 and 1500 [32]. The first three layers
have been identified in several online datasets [33] and, at least in humans, appear to be a consequence
of a constraint on available social time [34] combined with a relationship between time invested in a
relationship and its quality (as rated in terms of emotional closeness) [25,35,36]. The two outermost layers
(at 500 and 1500) correspond, respectively, to acquaintances (people we would not consider as personal
friends or family, but know well enough to have a conversation with) and to the number of faces we can
put names to.
It has been suggested that, even if this limit on personal network size exists in the face-to-face world,
the rise of online SNSs has circumvented at least some of these constraints and has thus allowed us to
increase dramatically the number of people we can have as friends [9,3739]. Because there are significant
limits on the number of people we can talk to at any one time in the offline world [4042]aswellason
the amount of time we have available for social interaction [25,43], there is inevitably a limit on the size
of our egocentric social networks when relationships require time investment. In contrast, there are no
limits to the number of people who can read our posts, and SNSs might thus allow us to cut through this
constraint imposed by face-to-face interaction. Being able to interact with many individuals at the same
time could in principle allow us to increase social network size dramatically.
Although many recent studies have undertaken comparisons between online and offline networks,
almost all have focused on the small number of strong ties [44] that individuals have [4550]. The
definitions used in sampling networks in these studies have often been so restrictive that the typical
network size they reveal has been in the order of five individuals (and never more than 20), suggesting
that they are sampling just the two innermost core layers of personal egocentric networks (typically
five and 12–15 individuals, respectively [25,34]). These sampling strategies thus ignore the fact that an
individual’s social network (defined as all meaningful relationships) is considerably larger than this, and
typically in the order of 100–200 individuals.
In addition, these studies have, with very few exceptions, sampled students (and, in many cases,
teenagers), heavy SNS users or members of other specialized communities [2,51], and so cannot be
considered representative of the wider population. The few studies that have undertaken large-scale
randomized samples representative of the population at large [2,48] have focused only on the innermost
layers of the network and have had quite modest sample sizes (N1000 Internet users) by the standards
of sociological sampling. Two attempts to examine the size of extended online social networks with
seriously large samples have considered communities formed among twitterati (i.e. followers of a
3 R. Soc. open sci. 3:150292
particular twitter account) [52] and scientific email communities [53]. Both claimed evidence for a
natural community size between 100 and 200, but neither of these can really be considered conventional
everyday social communities in any meaningful sense.
This study tests the claim that online social environments allow us to significantly increase the size of
our social networks using two large structured random samples of the UK population and the number
of friends listed on Facebook as the test metric. These data constitute the first attempt to determine the
natural limit on network size using unbiased, randomized, stratified sampling of a national population.
As such, this study is the first real attempt to test whether online social media do allow us to increase the
size of our social networks.
2. Material and methods
The data derive from two samples commissioned from the panel provider OnePoll by the WildCard
agency on behalf of the Thomas J. Fudge’s company, sampled from OnePoll’s large in-house panel. The
samples were carried out in the first week of April 2015 and the third week of May 2015, respectively.
Each sample was a nationally structured random sample of adults aged 18–65 years distributed
proportionally to age, sex and regional population across the UK.
Sample 1 included 2000 adults (mean age 39 years; 45.2% male), focused on people’s use of, and
satisfaction with, online social media, with adults who ‘made regular use of social media’ as the sampling
criterion. A total of 85.4% of respondents declared that they checked social media every day; 51%
declared that they had never deleted their social media profile. Only 12.3% of the entire sample declared
that, when they had deleted a profile, they had lasted more than two weeks before signing back in, with
4% declaring that they had lasted more than a year. Sample 2 included 1375 adults (mean age =37.4
years; 39.1% male) and sampled professional adults who worked full time at 9.00 to 17.00 weekday jobs
and had attended business meetings on behalf of their employer. In this sample, respondents were not
necessarily social media users, and Sample 2 thus might be seen as being more representative of the
general population. Each sample included approximately 30 questions related to use of online media
(Sample 1) and social behaviour in relation to management meetings (Sample 2), with network size
questions included as part of these.
In both cases, subjects were asked to state, on a 14-point (Sample 1) or 16-point (Sample 2) categorical
scale ranging from 0 to 1000+, how many friends they had on Facebook (electronic supplementary
material, tables S1 and S8). Categorical answers were used rather than asking for actual numbers of
Facebook friends because, in large-scale sampling, it is important to maintain respondent interest and
focus, and not to distract them by requiring them to break off and open new windows. Respondents in
Sample 1 were also asked to state how many of these individuals they considered to be close friends
(on a 12-point categorical scale ranging from 0 to 100+: electronic supplementary material, table S4) and
how many individuals they would ‘consider going to for advice or sympathy in times of great emotional
or other distress’ (on a 7-point categorical scale ranging between 0 and 16+: electronic supplementary
material, table S5). These correspond to the two innermost circles of the egocentric social network, the
sympathy group (normatively approx. 15 individuals) and the support clique (normatively approx. 5
individuals) [25,54].
3. Results
Figure 1 plots the distribution of total number of friends for the two samples. The mean number of friends
is 155.2 and 182.8 in the two samples, respectively. Neither of these differs significantly from the value
of 150 (95% CI =100–200) predicted by the SBH [18](z=0.20, p=0.842; z=1.29, p=0.198, respectively).
(If we exclude those who responded 0 in Sample 2, the mean increases to 186.5, but the conclusion
is unchanged.) As has been noted in all previous studies reporting network size, both distributions
have long tails to the right, but have marked modes around 150. Although the two distributions differ
significantly from each other (χ2=191.6, d.f. =9, p<0.0001), this is in fact mainly due to the high
frequency of 0 scores in Sample 2. Given the difference in sampling criteria (regular social media users
versus business professionals), this is perhaps hardly surprising. Discounting the 0 category, the two
distributions in fact correlate closely across size categories (r=0.909, N=11, p<0.001; the slope does
not differ significantly from β=1, t9=0.293, p=0.776).
In both cases, women list significantly more friends than men (Sample 1: 165.5 versus 145.0;
Sample 2: 196.2 versus 156.6; χ2=33.3 and χ2=56.1, respectively; d.f. =8, p<0.0001). Network size
4 R. Soc. open sci. 3:150292
25 50 75 100 200 300
400 500 600 700 800 900 1000
0 25 50 75 100 150 200
no. friends
250 300 350 400 450 500 600 700 800
Figure 1. Distribution of network size for (a) Sample 1 (social network users: N=2000) and (b) Sample 2 (business employees:
also varies significantly by decadal age class: in each sample, mean network size is negatively related
to age class, with younger age classes having larger online networks than older age classes (electronic
supplementary material, tables S3 and S9; r=−0.972, p=0.006, and r=−0.973, p=0.005; N=5 in both
cases). Both distributions differ significantly across age classes (χ2=436.7, d.f. =28, and χ2=352.1,
d.f. =24; p<0.0001). Partitioning χ2indicates that, in both samples, each age class differs significantly
from the others (χ254.2 and χ221.5, respectively, p<0.005). Note that the mean values for each age
class are almost identical in the two samples (electronic supplementary material, figure S1: Pearson’s
r=0.993, standardized β=0.993, N=5, p=0.001).
On average, respondents in Sample 1 considered that only 27.6% of their Facebook friends
could be considered ‘genuine’ (i.e. close) friends, with a strong modal value between 0 and 10%
(electronic supplementary material, table S2 and figure S2). These respondents were also asked more
explicit questions about the number of close friends they had. Figure 2 plots the distribution of
the support clique and the number of close friends (sympathy group) for Sample 1. The mean
values are 4.1 and 13.6, respectively, for the support clique (friends on whom you would depend
for emotional/social support in times of crisis) and sympathy group (close friends). Neither mean
is significantly different from the values of 3.8 ±2.29 and 11.3 ±6.19 given for these two grouping
layers by Hill & Dunbar [21] based on the literature (standardized deviates: z=0.13, p=0.897, and
z=0.37, p=0.711, respectively) or from the generic values of 5 and 15 identified by Zhou et al.[19]
(z=0.39, p=0.694, and z=0.23, p=0.818, respectively). It is noteworthy that the mean size of the
support clique and sympathy group hardly vary at all with age (electronic supplementary material,
tables S5 and S7) (r=−0.837, p=0.077, and r=−0.042, p=0.947, respectively; N=5). The distribution of
support clique values does not differ across decadal age classes (χ2=9.79, d.f. =20, p=0.972). Although
the distributions do differ significantly across age classes for the sympathy group data (χ2=45.76,
d.f. =24, p=0.0047), most of the deviations of observed from expected are in fact modest and rather
inconsistently distributed.
5 R. Soc. open sci. 3:150292
0 1–2 3–4 5–7
support clique
8–10 11–15 16+
Figure 2. Distribution of (a) support clique size and (b) sympathy group size for Sample 1 (N=2000).
4. Discussion
These data allow us to draw two conclusions. First, they confirm, using two separate, large, nationally
stratified random samples, that the typical egocentric network size for adult humans is similar to
that predicted by the SBH. The selection criteria for the two samples were quite different and could
have yielded very different outcomes. In fact, both estimates of network size were well within the
95% confidence limits around the predicted value from the SBH (roughly 100–200 [18]). As with
more conventional offline networks, there is considerable variance around the mean values. In offline
networks, this variance is due to personality differences (extroverts have larger networks than introverts
[55]), gender (women have larger networks than men [56]) and, as we show here, age (electronic
supplementary material, tables S3 and S9).
Sample 1 also confirms the existence and size of the two innermost layers of the social network, now
conventionally identified as the support clique and the sympathy group [25]. Numerically, the innermost
layer (five individuals) has been found in almost every online and offline study that has sampled network
sizes, where estimates have varied from 4.0 to 9.0 in online samples [2,46,48,57] and 2.1 to 7.4 in offline
studies [46,48,54,56,5865]. The 15-layer has likewise been widely identified, with online values varying
between 11.2 and 20.9 [2,45,48] and offline between 9.0 and 20.0 [28,48,54,56,6668]. In the present sample,
the mean values for these two layers (4.1 and 13.6) fit comfortably within these ranges.
The second main finding is that the two samples provide a direct test of whether online SNSs allow
individuals to have larger social networks than is possible offline because SNSs allow one or more of the
constraints that limit offline social network size to be circumvented. The results clearly suggest that they
do not. This result concurs with previous findings for a much smaller sample (N=117) which suggested
that heavy users of online social media do not have larger offline social networks than casual users, even
though more of these may appear online for heavy users [69]. Previous studies that have looked at this
question have typically focused exclusively on the innermost (5) network layer, and this study is the first
to test the full social network.
6 R. Soc. open sci. 3:150292
We might argue that the long tails to the right provide evidence that, even if most do not, some people
can maintain larger networks online (i.e. a small number of people really do have large numbers of listed
friends). If so, it would seem that this does not apply universally: only 14.2% of respondents in Sample 1
listed more than 300 friends (i.e. significantly more than the average offline network) and only 13.4% of
those in Sample 2 did so. Since respondents were not asked to specify the quality of their relationships
with individual alters, we cannot say whether the extra members in these larger networks really are
additional high-quality relationships or simply individuals that would normally be included in the circle
of acquaintances that forms a further layer stretching out to 500 individuals in the offline world [19,25]. In
most online platforms, these are all formally subsumed under the single category ‘friends’, whereas in the
offline world we would naturally distinguish between friends and acquaintances of different emotional
quality (and may even make that distinction informally for ourselves online).
In fact, analyses of traffic in online environments such as Facebook and Twitter reproduce rather
faithfully both the nested structure of the inner layers of offline networks and their typical interaction
frequencies [33]. Taken together with the fact that, in this study, the sizes of the two inner friendship
circles did not differ from those previously identified in offline samples (figure 2), this suggests that it is
most likely failure to differentiate relationships of different quality in the outermost layers that leads to
the impression of large numbers of online ‘friends’ (see also [2]). Respondents who had unusually large
networks did not increase the numbers of close friendships they had, but rather added more loosely
defined acquaintances into their friendship circle simply because most social media sites do not allow
one to differentiate between these layers (see also below).
The data in the two samples confirm previous findings of a small but consistent difference in network
size between the sexes (with females generally having larger networks at any given layer than males)
[21,28,54,56]. In both samples in this study, women had significantly larger networks than men, though
the differences remain within the natural range of variation in egocentric social network size. This is in
line with previous studies, which have suggested that women have more friends than men at least in the
innermost layers (though, again, the absolute differences are numerically modest, and sometimes not
significant) [28,56,70]. This may be related to women’s greater social skills, as reflected in their higher
scores on the kinds of cognitive tasks (e.g. mentalizing) that are thought to underpin social relationships
The data also highlight a strong age effect in complete network size: younger respondents (18–24 year
olds) had significantly larger networks than older respondents (55+year olds) (electronic supplementary
material, tables S3 and S9). Rosen et al. [57] report a mean network size of 249 for a sample with a mean
age of 19.5 years, which is close to the mean of 282 found in the 18–24-year-old age group in this study
(electronic supplementary material, table S3). It is noteworthy that this is also close to the ‘optimal’
number of online friends that maximizes social attractiveness: in a study of students (mean age 20.2
years), Tong et al. [72] found that ratings for attractiveness of fictitious profiles peaked at profiles which
listed 302 friends (in a range covering 102–902 in steps of 200). This may reflect well-known differences
in how teenagers (in particular) and older adults use social media, with younger individuals using it in
a more exploratory way to meet new people [12].
Note, however, that there is no equivalent age effect for the inner two layers of the network
(electronic supplementary material, tables S5 and S7), suggesting that the age effect applies only in
the outermost layer(s) of more casual friendships. Even so, this age effect contrasts with findings from
offline networks, where younger respondents tend to have significantly smaller social networks than
older adults [21,24,35]. A likely explanation for this difference probably lies in the fact that SNSs typically
encourage promiscuous ‘friending’ of individuals who often have very tenuous links to ego (X is a friend
[or friend-of-friend-of-a-friend] of Y, so would you like to befriend them?). Given that children are less
discriminating than adults in defining friendships [17], this may cause younger people in general to
respond more positively to these invitations. In addition, teenagers and young adults are in a period of
their lives when searching for new friendship (and especially romantic) opportunities is a particularly
important part of the natural life cycle; this may encourage individuals to establish many weak links
with alters as a means of testing out the opportunities available to them.
It is perhaps worth noting that there has been a notable tendency for teenagers to move away from
using Facebook as a social environment and to make use of media like Snapchat, WeChat, Vine, Flickr and
Instagram instead [73], with Facebook being reserved mainly for managing social arrangements. It is not
yet entirely clear what has driven this, but the fact that Facebook is too open to view by others seems to
have been especially important [74,75]. Teenagers have much smaller offline social networks than adults
[24,76], and forcing them to enlarge their network with large numbers of anonymous ‘friends-of-friends’
may place significant strain on their ability to manage their networks. Thus, this trend towards more
7 R. Soc. open sci. 3:150292
private social media may actually confirm the claim being made here—that open-ended social media do
not in fact allow us to increase the sizes of our social circles beyond that imposed by the SBH and the
constraints of everyday offline interaction.
The fact that social networks remain about the same size despite the communication opportunities
provided by social media suggests that the constraints that limit face-to-face networks are not fully
circumvented by online environments. Instead, it seems that online social networks remain subject to
the same cognitive demands of maintaining relationships that limit offline friendships. These constraints
come in two principal forms: a cognitive constraint derivative of the SBH and a temporal constraint
associated with the time that needs to be invested in a relationship to maintain it at a requisite level
of emotional intensity [25]. We can only interact coherently with a very small number of other people
(about three, in fact) at any one time [40,41]. It seems that even in an online environment, the focus of
our attention is still limited in this way.
This conclusion is reinforced by analyses of the frequencies with which individuals communicate
with members of their network in the inner network layers in online environments (Facebook and
Twitter): these yield interaction rates that are virtually identical to those observed in the offline world
[33]. Data from both face-to-face contacts [25,35] as well as mobile phone databases [34,77] suggest that
there are natural limits to both the amount of time we can devote to social interactions with network
members and how we distribute this time among them. Indeed, it seems that each of us distributes our
social capital, whether this is indexed by frequency of calling or by self-rated emotional closeness, in a
uniquely characteristic way rather like a social signature, and that this signature remains stable over time
despite significant churn in network membership [77]. This appears to be immune to the opportunities
for multiple interactions offered by the Internet.
Respondents were not asked to specify details about the individual alters in their networks (e.g. age,
gender, emotional closeness or spatial proximity). Consequently, we cannot say whether participants’
decisions about whom to include as a Facebook friend reflect ease of access to individuals (e.g. how far
away they lived, and so how easy it might be to meet up with them) or the distinction between family
and friends (an important feature of face-to-face networks [22]). It may well be that the individuals in the
sample with unusually small numbers of Facebook friends were using social media to maintain links only
with distant family and friends. If so, this is only likely to affect the variance in the data by increasing the
size of the left-hand tail (as figure 1 perhaps suggests). However, there is evidence to suggest that people
do not use communication media only to contact geographically distant alters: to the contrary, mobile
phone data show rather clearly that people in fact phone most frequently the people they live closest
to [78].
The fact that people do not seem to use social media to increase the size of their social circles
suggests that social media may function mainly to prevent friendships decaying over time in the
absence of opportunities for face-to-face contact [76,79]. Given that people generally find interactions via
digital media (including the phone as well as instant messaging and other text-based social media) less
satisfying than face-to-face interactions [80], it may be that face-to-face meetings are required from time
to time to prevent friendships, in particular, sliding down through the network layers and eventually
slipping over the edge of the 150 layer into the category of acquaintances (the 500 layer) beyond.
Friendships, in particular, have a natural decay rate in the absence of contact, and social media may well
function to slow down the rate of decay. However, that alone may not be sufficient to prevent friendships
eventually dying naturally if they are not occasionally reinforced by face-to-face interaction.
Ethics. The study was considered to be exempt from the need for ethics approval by the University of Oxford research
ethics board (CUREC).
Competing interests. I declare I have no competing interests.
Funding. R.D.’s research is funded by a European Research Council Advanced grant. The surveys were funded by the
Thomas J. Fudge’s company.
1. Uslaner EM. 2000 Social capital and the Net.
Commun. A CM 43, 60–65. (doi:10.1145/355112.
2. Zhao S. 2006 Do internetusers have more social
ties? A call for dierentiated analyses of internet
use. J. Comput. Mediat. Commun.11, 844–862.
3. Parks MR, Floyd K. 1996 Making friends in
cyberspace. J. Commun. 46, 80–97. (doi:10.1111/j.
4. Kraut R, Patterson M, Lundmark V, Kiesler S,
Mukopadhyay T, Scherlis W. 1998 Internet paradox:
a social technology that reduces social involvement
and psychological well-being? Am. Psychol.53,
1017–1031. (doi:10.1037/0003-066X.53.9.1017)
8 R. Soc. open sci. 3:150292
5. Nie NH. 2001 Sociability, interpersonal relations, and
the internet: reconciling conicting ndings. Am.
Behav.Sci. 45,420435.(doi:10.1177/0002764012
6. Nie NH, Hillygus DS. 2002 The impact of Internet use
on sociability: time-diary ndings. IT Soc. 1,1–20.
7. Mesch GS. 2006 Family relationsand the Internet:
exploring a family boundaries approach. J. Family
Commun. 6, 119–138. (doi:10.1207/s15327698
8. Rheingold H. 1993 The virtual community:
homesteading on the electronic frontier. Reading,
MA: Addison-Wesley.
9. Wellman B, Quan-Haase A, Witte J, Hampton K.
2001 Does the Internet increase, decrease or
supplement social capital? Social networks,
participation and community commitment. Am.
Behav.Sci. 45,436465.(doi:10.1177/0002764
10. Bargh JA, McKenna KYA. 2004 The Internet and
social life. Annu.Rev.Psychol.55, 573–590.
11. ValkenburgP, Peter J. 2009 Social consequences of
the Internet for adolescents: a decade of research.
Curr.Dir.Psychol.Sci.18,1–5. (doi:10.1111/j.1467-
12. BrandtzægPB, Heim J, Kaare B. 2010 Bridging and
bonding in social network sites – investigating
family-based capital. Int. J. WebBa sed Commun.6,
231–353. (doi:10.1504/IJWBC.2010.033750)
13. Quan-HaaseA, Wellman B. 2004 How does the
internet aect social capital? In Social capital and
information technology (eds V Huysma n, M Wulf),
pp. 113–132. Cambridge, MA:MIT Press.
14. Valkenburg PM, Schouten AP, Peter J. 2005
Adolescents’ identity experiments on the Internet.
New Media Soc. 7, 383–402. (doi:10.1177/1461444
15. Freberg K, Adams R, McG aughey K, Freberg L. 201 0
The rich get richer: online and oine social
connectivity predicts subjective loneliness. Media
Psychol. Rev.3.
16. Wang H, Wellman B. 2010 Social connectivity in
America: changes in adult friendship network size
from 2002 to 2007. Am.Behav.Sci.53,1148–1169.
17. HowesC. 1996 The earliest friendships. In The
company they keep: friendship in childhood and
adolescence (eds WMBukowski,AFNewcomb,WW
Hartup), pp. 66–86. Cambridge, UK: Cambridge
University Press.
18. DunbarRIM.1993Coevolutionofneocortexsize,
group size and language in humans. Behav. Brain
Sci. 16, 681–735. (doi:10.1017/S0140525X00032325)
19. ZhouW-X,SornetteD,HillRA,DunbarRIM.2005
Discrete hierarchical organizationof social group
sizes. Proc.R.Soc.B272, 439–444.
20. Hamilton MJ, Milne BT, WalkerRS, Burger O, Brown
JH. 2007 The complex structure of hunter–gatherer
social networks. Proc.R.Soc.B274,2195–2202.
21. HillRA, D unbar RIM. 2003 Social network sizein
humans. Hum. Nat. 14,53–72.
22. Roberts SBG, Dunbar RIM, Pollet TV, Kuppens T.
2009 Exploring variations in active network size:
constraints and ego characteristics. Soc. Netw.31,
138–146. (doi:10.1016/j.socnet.2008.12.002)
23. Dunbar RIM. 1998 The social brain hypothesis. Evol.
Anthropol. 6, 178–190. (doi:10.1002/(SICI)1520-
24. Roberts SBG, Dunbar RIM. 2011 The costs of family
and friends: an 18-month longitudinal study of
relationship maintenance and decay.Evol. Hum .
Behav.32, 186–197. (doi:10.1016/j.evolhumbehav.
25. Sutclie AJ, Dunbar RIM, Binder J, Arrow H. 2012
Relationships and the social brain: integrating
psychological and evolutionary perspectives. Br.J.
Psychol. 103, 149–168. (doi:10.1111/j.2044-8295.
26. LewisPA,RezaieR,BrowneR,RobertsN,Dunbar
RIM. 2011 Ventromedialprefrontal volume predicts
understanding of others and social network size.
NeuroImage 57,16241629.(doi:10.1016/j.neuro
27. Bickar t KC,Wright CI, Dauto RJ, Dickerson BC,
Barrett LF. 2011 Amygdala volume and social
network size in humans. Nat. Neurosci. 14, 163–164.
28. Powell J, Lewis PA, Roberts N, García-Fiñana M,
Dunbar RIM. 2012 Orbital prefrontal cortex volume
predicts social network size: an imaging study of
individual dierences in humans. Proc.R.Soc.B279,
2157–2162. (doi:10.1098/rspb.2011.2574)
29. Kanai R, Bahrami B, Roylance R, Rees G. 2011 Online
social network size is reected in human brain
structure. Proc.R.Soc.B279, 1327–1334.
30. Sallet J et al. 2011 Social network size aects neural
circuits in macaques. Science 334, 697–700.
31. HillRA, B entleyA, D unbar RIM. 2008 Network
scaling reveals consistent fractal pattern in
hierarchical mammalian societies. Biol.Lett. 4,
748– 751. (doi:10.1098/rsbl.2008.0393)
32. Dunbar RIM. 2014 The social brain: psychological
underpinnings and implications for the structure of
organizations. Curr.Dir.Psychol.Sci.24, 109–114.
33. Dunbar RIM, Arnaboldi V, Conti M, Passarella A.
2015 The structure of online social networks mirrors
those in the oine world. Soc. Netw.43,3947.
34. Miritello G, Moro E, Lara R, Martínez-López R,
Belchamber J, Roberts SBG, Dunbar RIM. 2013 Time
as a limited resource: communicationstrategy in
mobile phone networks. Soc. Netw.35, 89–95.
35. Roberts SBG, Dunbar RIM. 2011 Communication in
social networks: eects of kinship, network size and
emotional closeness. Pers. Relationships 18,
439–452. (doi:10.1111/j.1475-6811.2010.01310.x)
36. Roberts SBG, Arrow H, Lehmann J, Dunbar RIM.
2014 Close social relationships: an evolutionary
perspective. In Lucy to language: the benchmark
papers (eds RIM Dunbar,C G amble,JA J Gowlett),
pp. 151–180. Oxford,UK: Oxford University
37. Boase J, Horrigan JB, Wellman B, Rainie L. 2006 The
strength of internet ties. Washington,DC: Pew
Internet and American Life Project.
38. Hansen M. 2009 Collaboration. Cambridge, MA:
Harvard Business School Press.
39. Wellman B. 2012 Is Dunbar’s number up? Br.J.
Psychol. 103,174176.(doi:10.1111/j.2044-8295.
40. DunbarRIM,DuncanNDC,NettleD.1995Sizeand
structure of freely forming conversational groups.
Hum. Nat. 6, 67–78. (doi:10.1007/BF02734136)
41. Dezecache G, Dunbar RIM. 2012 Sharing the joke:
thesizeofnaturallaughtergroups.Evol. H um.
42. Krems J, Dunbar RIM. 2013 Clique size and network
characteristics in hyperlink cinema: constraints of
evolved psychology.Hum. Nat. 24,414429.
43. Dunbar RIM. 1998 Theory of mind and the evolution
of language. In Approaches to the evolutionof
language (eds J Hurford, M Studdart-Kennedy,
C Knight), pp. 92–110. Cambridge,UK: Cambridge
University Press.
44. Granovetter M. 1973 The strength of weak ties. Am.
J. Sociol. 78, 1360–1380. (doi:10.1086/225469)
45. Haythornthwaite C. 2000 Online personal networks:
size, composition and media use among distance
learners. New Media Soc. 2,195226.(doi:10.1177/
46. Kim K, Kim GJ, Park HW, Rice RE. 2007
Congurations of relationships in dierentmedia:
FtF,email, instant messenger, mobile phone, and
SMS. J. Comput. Mediat. Commun.12, 1183–1207.
47. Sooryamoorthy R, Miller BP, Shrum W. 2008
Untangling the technology cluster: mobile
telephony,internet use and the location of social
ties. New Media Soc. 10,729–749.(doi:10.1177/1461
48. Vergeer M, Pelzer B. 2009 Consequences of media
and internet use for oine and online network
capital and well-being. A causal model approach.
J. Comput. Mediat. Commun.15, 189–210. (doi:10.
49. Palackal A, Mbatia PN, Dzorgbo D-B, Duque RB,
Ynalvez MA, Shrum WM. 2011 Are mobile phones
changing social networks? A longitudinal study of
core networks in Kerala. New Media Soc. 13,
391–410. (doi:10.1177/1461444810393900)
50. Stefanone MA, KwonKH, Lacka D. 2012 Exploring
the relationship between perceptions of social
capital and enacted support online. J. Comput.
Mediat. Commun. 17, 451–466. (doi:10.1111/j.1083-
51. BrandtzægPB. 2012 Social networking sites: their
users and social implications: a longitudinal study.
J. Comput. Mediat. Commun.17, 467–488.
52. Gonçalves B, Perra N, Vespignani A. 2011 Validation
of Dunbar’s number in Twitter conversations.PLoS
ONE 6, e22656. (doi:10.1371/journal.pone.0022
53. Haerter JO, Jamtveit B, Mathiesen J. 2012
Communication dynamics in nite capacity social
networks. Phys.Rev.Lett.109, 168701.
54. Dunbar RIM, Spoors M. 1995 Social networks,
support cliques and kinship. Hum. Nat. 6, 273–290.
55. PolletTV,RobertsS,DunbarR.2011Extravertshave
larger social network layers but do not feel
emotionally closer to individuals at any layer. J.
Individual Dier.32,161169.(doi:10.1027/1614-
56. Stiller J, Dunbar RIM. 2007 Perspective-taking and
memory capacity predict social network size. Soc.
9 R. Soc. open sci. 3:150292
Netw. 29,93–104.(doi:10.1016/j.socnet.2006.04.
57. Rosen D, Stefanone MA, Lacka D. 2010 Online and
oine social networks: investigating
culturally-specic behavior and satisfaction. In Proc.
of the 43rd Hawaii Int.Conf. on System Sciences,
pp. 1–10. New Brunswick, NJ: Institute of Electrical
and Electronics Engineers (IEEE).
58. Fischer CS. 1982 What do we mean by friend? An
inductive study. Soc. Netw.3, 287–306.
59. Johnson MP, Milardo RM. 1984 Network
interference in pair relationships: a social
psychological recasting of Slater’s theory of social
regression. J. Marriage Family 46, 893–899.
60. Bernard HR, Johnsen EC, Killworth PD, McCarty C,
Shelley GA, Robinson S. 1990 Comparing four
dierent methods for measuring personal social
networks. Soc. Netw.12,179215.
61. vanSonderenE,OrmelJ,BrilmanE,vandenHeuvell
C. 1990 Personal network delineation. In Social
network research (eds CPM Knipscheer,
TC Antonucci), pp.101–120. Amsterdam,
The Netherlands: Swets and Zeitlinger.
62. Hartup WW, Stevens N. 1997 Friendship and
adaptation in the life course. Psychol.Bull.
12, 355–370. (doi:10.1037/0033-2909.121.
63. Lang FR. 2000 Endings and continuity of social
relationships: maximizing intrinsic benets within
personal networks when feeling near to death. J.
Soc. Person. Relationships 17, 155–182. (doi:10.1177/
64. Marin A. 2004 Are respondents more likely to list
alters with certain characteristics? Implications for
name generator data. Soc. Netw.26, 289–307.
65. McPherson M, Smith-Lovin L, Brashears ME. 2006
Social isolation in America: changes in core
discussion networks over two decades. Am. Sociol.
Rev. 71,353–375. (doi:10.117 7/00031224060710
66. Buys CJ, Larson KL. 1979 Human sympathy groups.
Psychol. Rep.45, 547–553. (doi:10.2466/pr0.1979.45.
67. McCannell K. 1988 Social networks and the
transition to motherhood. In Familiesand social
networks (ed. RM Milardo), pp.83–106. Newbur y
Park, C A: Sage.
68. Rands M. 1988 Changes in social networks following
marital separation and divorce. In Familiesand
social networks (ed. RM Milardo), pp.127–146.
Newbury Park, CA: Sage Publications.
69. Pollet TV, Roberts SBG, Dunbar RIM. 2011 Use of
social network sites and instant messaging
does not lead to increased oine social network
size, or to emotionally closer relationships with
oine network members. Cyberpsychol.Behav.
Soc. Netw.14, 253–258. (doi:10.1089/cyber.
70. Cairns RB, Leung M, Buchanan L, Cairns BD. 1995
Friendships and social networks in childhood and
adolescence: uidity, reliability,and interrelations.
Child Dev. 66, 1330–1345. (doi:10.2307/1131650)
71. PowellJ, Lewis PA, Dunbar RIM, García-Fiñana M,
Roberts N. 2010 Orbital prefrontal cortex volume
correlates with social cognitive competence.
Neuropsychologia 48, 3554–3562.
72. Tong ST, Van Der Heide B, Langwell L. 2008 Too
much of a good thing? The relationship between
number of friends and interpersonal impressions on
Facebook. J. Comput.Mediat. Commun. 13, 531–549.
73. Petersen B. 2013 Where are all the Facebook teens
going? See
blog/facebook-teens (accessed 5 December
74. Brandtzæg PB, Lüders M, Skjetne JH. 2010 Too many
Facebook ‘friends’? Contentsharing and sociability
versus the need for privacy in social network sites.
Int. J. Hum. Comput. Interact.26, 1006–1030.
75. Madden M, Lenhart A, Cortesi S, Gasser U,
Duggan M, Smith A, Beaton M. 2013 Teens,social
media, and privacy. Washington, DC: Pew Research
76. Roberts SBG, Dunbar RIM. 2015 Managing
relationship decay: network, gender and contextual
eects. Hum. Nat. 26, 426–450.
77. SaramäkiJ, Leicht E, López E, Roberts SBG,
Reed-TsochasF, Dunbar RIM. 2014 The persistence
of social signatures in human communication. Proc.
NatlAcad.Sci.USA111, 942–947. (doi:10.1073/pnas.
78. Jo H-H, Saramäki J, Dunbar R, Kaski K. 2014
Spatial patterns of close relationships across
the lifespan. Sci. Rep. 4, 6988. (doi:10.1038/
79. Dunbar RIM. 2012 Social cognition on the internet:
testing constraints on social network size. Phil.
Trans . R. Soc . B 367,2192–2201.(doi:10.1098/
80. VlahovicT,RobertsSBG,DunbarRIM.2012Eectsof
duration and laughter on subjective happiness
within dierent modes of communication.
J. Comput. Mediat. Commun.17, 436–450.

Supplementary resource (1)

... The classification of social media platforms as small (vs. medium, and large) was determined through Dunbar and co.'s research on the average group size for humans (i.e., Dunbar's number; Dunbar, 2016;Konnikova, 2014;Zhou et al., 2005; see also Dunbar, 1993). It has been consistently reported that the average person has a social group/network of about 150 (what would be considered intimate friends and family, even extending to casual friends). ...
... It has been consistently reported that the average person has a social group/network of about 150 (what would be considered intimate friends and family, even extending to casual friends). More recent research has expanded on this (especially through a social media lens) and demonstrated that the size of one's social network can expand incrementally up to 1500, with around 200-300 still being considered within the scope of "casual friends", while around 500 included acquaintances, and over 900 up to 1500 is considered the maximum social network capacity that people can remember (Dunbar, 2016;Zhou et al., 2005). ...
Does the social status of victims in emergencies play a role in bystanders’ compassionate orientations towards them? In this thesis, I examine the hitherto unexplored proposition that bystanders may be more inclined toward expressing compassion in favor of victims who signal high (rather than low) social status. I tested this novel thesis in six experiments that systematically varied the social status of victims of fabricated emergencies and afterward measured their compassion to investigate whether the expression of this emotion was stronger for higher (relative to lower) status victims. In doing so, I considered a variety of situational and individual difference factors that could enable (or constrain) a compassion bias favoring victims from high-status backgrounds. In the first empirical chapter (3), I showed that participants (N = 436) reported higher compassionate intentions toward victims of a terror attack that were described as coming from a high-status (vs. low-status) background, while also providing indirect evidence that cost calculations may play a role in this process. In the second empirical chapter (4), I directly investigated the cost-calculus caveat and explored the role of ideological persuasions. The initial experiment in Chapter 4 (N = 273) showed that even participants with an egalitarian ideology do sometimes succumb to the high-status compassion bias, but this occurs when the cost of doing so is trivial for them: a trend that was largely replicated in a subsequent experiment in that chapter (N = 288). The final empirical chapter (5) explored the role that threat appraisals might play in the process, testing the idea of whether high-status victims will continue to benefit from a compassion bias even when they seem threatening to bystanders. In the three experiments (N = 1,373) reported in Chapter 5, I showed that threat appraisals undermine a compassion bias favoring victims from both high and low-status backgrounds. Hence, overall, the preponderance of the evidence across Chapters 3-5 affirms the existence of compassion bias favoring victims from high-status backgrounds, although they also do outline important situational and individual difference factors that can sometimes eliminate or even reverse this trend. This is an important contribution in terms of not only theoretical advancements (i.e., helping to show that status plays a role in compassion during emergencies) but also practice (e.g., it could be useful in the training of frontline emergency responders).
... This allows one to follow groups with specific expertise, such as the Surgical Infection Society (SIS @SurgInfxSoc), IDSA (@IDSAInfo), SIDP (@SIDPharm), and SCCM (@SCCM), and use the targeted Twitter strategy we describe to influence practice and policy around antibiotic use among surgeons. Healthcare professionals, like all humans, are subject to the inherent limitations of their primate brain when trying to exchange information during social interactions [21]. There are barriers to efficient communication, such as time, geographical distance, resources, and language, that can be partially overcome with specific tools. ...
... There are barriers to efficient communication, such as time, geographical distance, resources, and language, that can be partially overcome with specific tools. Twitter has the advantage that it does not restrict social network size in the same way in which it may be limited by a surgeon's cognition [21]. In fact, Twitter can be used to create and coordinate global surgical communities and ecosystems, avoiding time, distance, and language restrictions, a clear advantage considering all the restrictions imposed by the COVID-19 pandemic. ...
Many infectious diseases (ID) clinicians join Twitter to follow other ID colleagues or "like" people. While there is great value in engaging with people who have similar interests, there is equal value in engaging with "unlike" or non-ID people. Here, we describe how Twitter connected an ID pharmacist with a pediatric surgeon, a vice chair of surgery, a surgeon chief medical officer from Spain, and a surgical intensive care unit pharmacist. This Twitter collaboration resulted in several scholarly activities related to antibiotic resistance and antibiotic stewardship and served as a conduit for global collaboration.
... In stable social relationships, people know other people who have a relationship with them and know their relations with other people (Wikipedia, Dunbar number). Although it is believed that online social media develop the size of social networks and the set of related people, Goncalves et al. (2011) and Dunbar (2016) showed that the number of users' friends in online social media are close to the real social networks (Goncalves et al., 2011 andDunbar 2016). However, users rarely share their attention. ...
... In stable social relationships, people know other people who have a relationship with them and know their relations with other people (Wikipedia, Dunbar number). Although it is believed that online social media develop the size of social networks and the set of related people, Goncalves et al. (2011) and Dunbar (2016) showed that the number of users' friends in online social media are close to the real social networks (Goncalves et al., 2011 andDunbar 2016). However, users rarely share their attention. ...
... These parameters were explicitly chosen to achieve clustering coefficients and in-degrees that were realistic for real-life community interactions, namely, a consistent clustering coefficient of approximately 0.3 as well as an average in-degree between 7 and 9 (for populations that begin with a strong consensus belief, we expect marginally higher average degree). Though a person might interact with a larger number of individuals through their online social networks, or a smaller number of individuals through in-person interactions, surveys have shown that people report feeling genuinely close to between 5 and 10 individuals in their social circle, broadly construed [27]. The clustering coefficient was chosen to be consistent with values for average clustering coefficients on directed graphs using random walks on social networks [28]. ...
... Fig 5 depicts examples of the initialized networks for each of the four scenarios we analyzed. For each of the scenarios the average-in degree was between 6 and 9, which corresponds to realistic social networks [27,28]. Our simulation results are presented in Section 5. ...
Full-text available
Covid-19 vaccines are widely available in the United States, yet our Covid-19 vaccination rates have remained far below 100%. Not only that, but CDC data shows that even in places where vaccine acceptance was proportionally high at the outset of the Covid-19 vaccination effort, that willingness has not necessarily translated into high rates of vaccination over the subsequent months. We model how such a shift could have arisen, using parameters in agreement with data from the state of Alabama. The simulations suggest that in Alabama, local interactions would have favored the emergence of tight consensus around the initial majority view, which was to accept the Covid-19 vaccine. Yet this is not what happened. We therefore add to our model the impact of mega-influencers such as mass media, the governor of the state, etc. Our simulations show that a single vaccine-hesitant mega-influencer, reaching a large fraction of the population, can indeed cause the consensus to shift radically, from acceptance to hesitancy. Surprisingly this is true even when the mega-influencer only reaches individuals who are already somewhat inclined to agree with them, and under the conservative assumption that individuals give no more weight to the mega-influencer than they would give to a single one of their friends or neighbors. Our simulations also suggest that a competing mega-influencer with the opposite view can shift the mean population opinion back, but cannot restore the tightness of consensus around that view. Our code and data are distributed in the ODyN (Opinion Dynamic Networks) library available at
... We define the following dimensions of the online image: possible threats [50], privacy [51], resource security [52], user security and simplicity [53,54,55]. Earlier empirical results show that customer satisfaction increases when they perceive the company and its image as appropriate [56,57], indicating that image in social media is associated with customer trust and ultimately with the company's reputation [58]. User safety is perceived as the ability to navigate on social media in accordance with the law, with negligible risk, while ensuring that the activities carried out by enterprises are reliable, clear and professional [59,60,61,62,63]. ...
Full-text available
COVID-19, mobility, socio-social changes have transferred to the world of social media communication, purchasing activities, the use of services. Corporate social media has been created to support clients in using various services, give them the possibility of easy communication without time and local barriers. Unfortunately, they still very rarely take into account the security and privacy of customers. Considering that the purpose of this article is to investigate the impact of social media on the company’s image, it should be remembered that this image also works for the security and privacy of customer data. Data leaks or their sale are not welcomed by customers. The results of empirical research show that the safety, simplicity and variety of services offered on social media have a significant impact on the perceived quality, which in turn positively affects the reputation. The authors proposed a methodology based on the Kano model and customer satisfaction in order to examine the declared needs and undefined desires and divide them into different groups with different impacts on consumer satisfaction. The interview participants were employees of 10 randomly selected companies using social media to conduct sales or service activities. 5,000 people from Poland, Portugal and Germany participated in the study. 4,894 correctly completed questionnaires were received.
... Empirical studies have validated this and put the number of people with whom we can meaningfully interact at approximately 150. Recent research showed that this is true of both off-line and on-line networks (Dunbar 2016). If that is the case, why then are our online social networks on average five to ten times larger? ...
In this paper I take the view that social media, and Web 2.0 in general, constitute an environment that structures our communicative behavior. I argue that this environment lacks mechanisms to fulfill some of the classical social functions which traditionally were provided by professional journalism and mass media in general. Specifically, in their current form, social media favor visibility at the expense accuracy and social relevance. Second, I argue that their reliance on artificial intelligence, bypassing both the rational and moral dimensions of human communication, favors strategic communication at the expense of rational discourse. I start by describing certain salient features of journalism, such as sensationalism, mediatization, and criteria of newsworthiness. In the second part of the chapter, I discuss the implications for social communication of some phenomena that are characteristic of social media and Web 2.0, such as: homophily, echo chambers, polarization, and irrationality—all of which are characteristic of the phenomenon called “fake news.” I argue that social media are not neutral or natural environments. Rather, they are created and maintained with specific interests of their owners in mind. Second, the rules that underlie their functioning are biased against meaningful conversation, discourse and collective action. Finally, they trigger and reinforce specific social–psychological mechanisms in their users which favor polarization and identification with the extreme positions, on the one hand, but also political disengagement, on the other hand. Moderate positions, capacity for discourse and dialogue across ideological lines do not thrive in this environment. I conclude with a vision of ethical communication for peace in this new com- munication environment. Christians are called to make their contribution: indi- vidually, corporatively, and institutionally.
We use a paradox approach (Mick and Fournier, 1998) to explore how consumers use and experience their smartphones. To do so, we use a mixed method approach where we interviewed 28 participants across seven focus groups to learn more about when and how they used their smartphones. Participants reported many tensions with regard to their smartphone use, from which we derived one overarching paradox and five specific paradoxes, including two new paradoxes. To support and extend our qualitative findings, we also administered a questionnaire examining the proposed paradoxes and their possible connections to important consumer consequences such as ambivalence, attachment, and well‐being. Overall, we found evidence of a push and pull (or ambivalent) relationship between participants and their smartphones. Specifically, its great functionality and reliability make the connection cherished, but this ongoing reliance takes away the very same things it was meant to help build.
The multidisciplinary field of personal relationships has focused primarily on strong ties (romantic relationships, friendships, family relationships). However, acquaintances (weak ties) are pervasive in people's lives, contribute to well‐being, influence strong ties, and can become strong ties over time. This review article synthesizes several areas of literature about the role of acquaintances (weak ties) in the web of relationships and about the formation of acquaintanceships. The terms acquaintances and weak ties are used interchangeably in this article to refer to the type of relationship that exists in the peripheral layers of social networks. In the first section, I discuss the literature on factors associated with the size of people's acquaintance network, needs met by acquaintances (compared to those of closer ties), health and happiness benefits of interaction with acquaintances, and the dark side of acquaintances including having unwanted acquaintances. In the second section, I discuss how acquaintanceships are formed, and particularly the type that can develop into a closer tie. This section summarizes research from the literatures on friendship formation, relationship initiation, attraction, and first interactions of dyads at zero‐acquaintance. I end the article by identifying several research topics on acquaintances that could be studied by the next generation of scholars.
Full-text available
There are rising concerns that social network sites (SNS) facilitate the creation of echo chambers, in which attitude-consistent information becomes the norm while attitude-challenging information is avoided. This study aims to investigate theoretically derived predictors of attitude-consistent and attitude-challenging exposure on SNS. We theorize that three key sets of predictors may influence attitude-consistent and attitude-challenging exposure: ideology, cognitive, and behavioral indicators of political involvement, and network characteristics. In a two-wave panel study, we predict the frequency of attitude-consistent and attitude-challenging exposure as well as relative attitude-consistent exposure, measured as attitude-consistent exposure as a share of overall opinion exposure. The results demonstrate that extreme ideological positions, higher political knowledge, and low-effort political participation predicted an increase in (relative) attitude-consistent exposure. Cross-social class exposure predicted a decrease in (relative) attitude-consistent exposure. The findings challenge existing arguments that SNS may per se facilitate attitude-consistent exposure.
Full-text available
Relationships are central to human life strategies and have crucial fitness consequences. Yet, at the same time, they incur significant maintenance costs that are rarely considered in either social psychological or evolutionary studies. Although many social psychological studies have explored their dynamics, these studies have typically focused on a small number of emotionally intense ties, whereas social networks in fact consist of a large number of ties that serve a variety of different functions. In this study, we examined how entire active personal networks changed over 18 months across a major life transition. Family relationships and friendships differed strikingly in this respect. The decline in friendship quality was mitigated by increased effort invested in the relationship, but with a striking gender difference: relationship decline was prevented most by increased contact frequency (talking together) for females but by doing more activities together in the case of males.
Full-text available
This paper examines social network size in contemporary Western society based on the exchange of Christmas cards. Maximum network size averaged 153.5 individuals, with a mean network size of 124.9 for those individuals explicitly contacted; these values are remarkably close to the group size of 150 predicted for humans on the basis of the size of their neocortex. Age, household type, and the relationship to the individual influence network structure, although the proportion of kin remained relatively constant at around 21%. Frequency of contact between network members was primarily determined by two classes of variable: passive factors (distance, work colleague, overseas) and active factors (emotional closeness, genetic relatedness). Controlling for the influence of passive factors on contact rates allowed the hierarchical structure of human social groups to be delimited. These findings suggest that there may be cognitive constraints on network size.
Full-text available
The rapid adoption of social networking sites (SNSs) raises important questions about the social implications of such usage. Drawing on unique longitudinal data, surveying a representative sample of Norwegian online users (N=2,000, age 15-75 years) in 3 annual waves (2008, 2009, and 2010), this study found a significantly higher score among SNS users in comparison to nonusers in 3 out of 4 social capital dimensions: face-to-face interactions, number of acquaintances, and bridging capital. However, SNS-users, and in particular males, reported more loneliness than nonusers. Furthermore, cluster analyses identified 5 distinct types of SNS users: Sporadics, Lurkers, Socializers, Debaters, and Advanced. Results indicate that Socializers report higher levels of social capital compared to other user types.
Conventional wisdom over the past 160 years in the cognitive and neurosciences has assumed that brains evolved to process factual information about the world. Most attention has therefore been focused on such features as pattern recognition, color vision, and speech perception. By extension, it was assumed that brains evolved to deal with essentially ecological problem-solving tasks. 1.
Human quantitative boundaries for sympathizing are perceived as having been roughly established during human evolution into social beings. Data obtained on 125 American subjects are viewed as consistent with this highly speculative notion.