ArticlePDF Available

Easier contagion and weaker ties make anger spread faster than joy in social media

Authors:

Abstract and Figures

Similar to face-to-face communication in daily life, more and more evidence suggests that human emotion also spread in online social media through virtual interactions. However, the mechanism underlying the emotion contagion, like whether different feelings diffuse unlikely or how the spread is coupled with the social network, is rarely investigated. Indeed, because of the costly expense and spatio-temporal limitations, it is challenging for conventional questionnaires or controlled experiments. While given the instinct of collecting natural affective responses of massive connected individuals, online social media offer us an ideal proxy to tackle this issue from the perspective of computational social science. In this paper, based on the analysis of millions of tweets in Weibo, a Twitter-like service in China, we surprisingly find that anger is more contagious than joy, indicating that it can sparkle more angry follow-up tweets and anger prefers weaker ties than joy for the dissemination in social network, indicating that it can penetrate different communities and break local traps by more sharings between strangers. We conjecture that easier contagion and weaker ties could together make anger spread faster than joy, which is further justified by the diffusion of bursty events with different dominant emotions. To our best knowledge, for the first time we quantificationally disclose the difference between joy and anger in dissemination mechanism and our findings would shed lights on personal anger management in human communication and collective outrage control in cyber space.
Content may be subject to copyright.
arXiv:1608.03656v1 [cs.SI] 12 Aug 2016
Easier contagion and weaker ties make
anger spread faster than joy in social media
Rui Fan1, Ke Xu1and Jichang Zhao2,*
1State Key Lab of Software Development Environment, Beihang University
2School of Economics and Management, Beihang University
Corresponding author: jichang@buaa.edu.cn
(Dated: August 15, 2016)
Abstract
Similar to face-to-face communication in daily life, more and more evidence suggests that human
emotion also spread in online social media through virtual interactions. However, the mechanism
underlying the emotion contagion, like whether different feelings diffuse unlikely or how the spread
is coupled with the social network, is rarely investigated. Indeed, because of the costly expense
and spatio-temporal limitations, it is challenging for conventional questionnaires or controlled ex-
periments. While given the instinct of collecting natural affective responses of massive connected
individuals, online social media offer us an ideal proxy to tackle this issue from the perspective of
computational social science. In this paper, based on the analysis of millions of tweets in Weibo, a
Twitter-like service in China, we surprisingly find that anger is more contagious than joy, indicat-
ing that it can sparkle more angry follow-up tweets and anger prefers weaker ties than joy for the
dissemination in social network, indicating that it can penetrate different communities and break
local traps by more sharings between strangers. We conjecture that easier contagion and weaker
ties could together make anger spread faster than joy, which is further justified by the diffusion of
bursty events with different dominant emotions. To our best knowledge, for the first time we quan-
tificationally disclose the difference between joy and anger in dissemination mechanism and our
findings would shed lights on personal anger management in human communication and collective
outrage control in cyber space.
1
I. INTRODUCTION
Emotions have a substantial effect on human decision-makings and behaviors [22]. Emo-
tions also transfers between different individuals through their communications and interac-
tions, indicating that emotion contagion, which make others around experience the similar
emotional statuses, could possibly promote social interactions [34, 37] and synchronize col-
lective behavior, especially as the individuals are embedded in social networks [42]. From
this perspective, better understanding of the emotion contagion can essentially help up-
grade emotion management and disclose collective behavior patterns. However, the detailed
mechanism of emotion contagion in context of the social network still remains unclear.
Conventional approaches like laboratory experiments are pervasively employed to testify
the existing of emotion contagion in realistic circumstances [21, 28, 41]. However, to unravel
the detailed mechanism beyond the emotion contagion is much more challenging. It is hard
for the controlled experiments to establish a large social network, stimulate different emo-
tions simultaneously and then track the propagation of emotions in real-time. Meanwhile,
in order to study the coupled dynamics between the social structure and emotion contagion,
properties like relationship strengths should be considered, while which might further import
uncontrolled contextual factors and fundamentally undermine the reliability of the experi-
ment. We argue that it is extremely difficult for conventional approaches to investigate the
spread mechanism in the context of large social network and long-term observation, and
the footprints of natural affective responses from massive individuals in online social media
indeed offer us a new and computational perspective [23, 33, 34].
It is not easy to differentiate online interactions from face-to-face communication in terms
of emotion contagion [17]. More and more evidence from both Facebook and Twitter in
recent years consistently demonstrates the existing of emotion contagion in online social
media [10, 13, 14, 25, 26, 31, 32]. Kramer et al. for the first time provide the experimental
evidence of emotion contagion in Facebook by manipulating the amount of emotional content
in the News Feed [32]. Later Ferrara and Yang disclose the evidence of emotion contagion
in Twitter [19]. Instead of controlling the content, they measure the emotional valence of
the content and show that posting positive and negative tweets both follow significant over-
exposures, indicating the spread of different feelings. Evidence from both studies suggests
that even in the absence of non-verbal cues of in-person interactions [19], emotions like
2
positive and negative feelings can still transfer from one user to others in online social media.
In fact, through posting, reposting and other virtual interactions, users in online social media
express and expose their natural emotional states into the social network in real-time as the
context evolves. Hence being an ubiquitous sensing platform, the online social media collects
emotion expression and dissemination in the most realistic and comprehensive circumstance
and thus provides us an unprecedented proxy to obtain universal insights of the underlying
mechanism of emotion contagion in social network.
However, in the previous study, emotions in online social media are usually simplified into
the positive and the negative [5, 6, 19, 32] and many fine-grained emotional states, especially
the negative ones like anger and disgust are neglected. In fact, negative feelings like anger
on the Internet might greatly dominate online bursts of societal issues or terrorist attack
and play essential roles in driving the collective behavior during the event propagation [11,
44]. Aiming at fill this vital gap, in this paper, we categorize the human emotion into
four categories [18, 44], including joy, anger, disgust and sadness and try to disclose the
mechanism underlying their spread. Fine-grained emotional states enrich the landscape
of emotion contagion and make it possible to systematically understand the underlying
mechanisms between social structures and propagation dynamics.
In this paper, we employ over 11 million tweets posted by around 100 thousand users
of a half year in Weibo, a variant of Twitter in China to perform the computational and
quantificational investigation of emotion contagion. Surprisingly, we find that among the
four emotions we study, joy and anger demonstrate significant evidence of contagion and
anger is the most contagious mood, indicating that under the same volume of exposure,
angry messages will sparkle more follow-up retweets or tweets than joy in the near future.
Besides, from the view of coupled dynamics with the social structure, different measures
of relationship strengths consistently show that anger prefers weaker ties than joy in the
dissemination, indicating that angry tweets can break local traps and arrive the global
coverage with greater odds. We conjecture that easier contagion and weaker ties could make
anger spread faster than joy in online social media, then over 40 million tweets from 616
online bursts in China further solidly testify this argument. In these sudden events, we
find that negative ones with anger as the dominating emotion arrives their diffusion peaks
with shorter intervals and higher velocities than joyful counterparts. Our findings about the
emotion contagion will be insightful for personal anger management and collective behavior
3
understanding in realistic scenarios either online or offine.
II. RESULTS
In this study, 11,753,609 tweets posted by 92,176 users from Sept., 2014 to Mar., 2015
are collected, including the following network of these users. Through a Bayesian classifier
trained in [44], each emotional tweet in our data set can be automatically labeled to joy,
anger, disgust or sadness. Then the diffusive probability and structure preferences are
accordingly investigated in the social network of around 100 thousand subjects. Finally the
realistic bursts in Weibo will be employed to testify the conjecture that anger spreads faster
than joy, in which 40,005,242 tweets of 616 different events are employed to perform the
computational analysis [36].
A. Anger is more contagious than joy
Rather than manipulating the content users receive [31], we measure the emotion con-
tagion through the extent to which a user will be influenced emotionally by the tweets she
received in Weibo during a certain period, i.e., an observation window. Similar to [19], it
is assumed that the emotion that users are exposed to in the recent past will stimulate
similar emotional status and which can be reflected by users’ newly posted tweets. Each
individual in Weibo can be embedded into a online social network with nodes being users
and directed links being following relationships, for example, if a user u(follower) follows
a user w(followee), a tie will be established from uto w, indicating that tweets posted by
wwill be flooded to uin real-time. Supposing the observation window is ∆ hours, a vector
v(tu) = (panger , pdisgust, pj oy, psadness) represents the emotion distribution of tweets exposed
to uwithin ∆ before posting the tweet tu. Note that we omit the tweets with total exposure
less than 20 tweets to ensure the reliability of the measurement. By averaging over each
dimension of all tweets’ exposure vectors, a vector vall can be obtained to reflect the baseline
of emotion distribution before any tweet is posted in Weibo. While for a certain emotion
i(i= 1,2,3 or 4), by averaging over each dimension of all tweets with emotion i, a vector
viis similarly obtained to stand for the average emotion distribution before any tweet with
emotion iis posted in Weibo. Then for emotion i, the difference (denoted as di) on dimen-
4
0.5
1.0
1.5
2.0
1 2 3 4 5 6 7 8 9 10 11 12
Difference (%)
a
47
48
49
50
51
52
1 2 3 4 5 6 7 8 9 10 11 12
Percentage (%)
b
Anger
Disgust
Joy
Sadness
Figure 1. Contagion of different emotions. (a) The contagion tendency of different
emotions declines with the length of observation window (in hours). (b) The percentage of
influenced tweets for both anger and joy rises with the growth of observation window and
the fraction of influenced tweets in anger even exceeds 50% when ∆ >4 hours.
sion ibetween vall and vican intuitively reflect the contagion tendency, i.e., the significance
of contagion, of iin Weibo. Higher diimplies that emotion iwill stimulate more tweets with
the same emotion in the near future than the average level, suggesting it is more contagious
from the perspective of emotional influence.
As can be seen in Fig. 1(a), dang er and djoy are significantly higher than that of other
emotions, which indicates that both emotions can spread in the social network, especially
for anger which possesses the highest significance as compared to the baseline. For example,
danger is about 2.1% when ∆ = 1, which means that users received 2.1% more angry tweets
than ordinary time before they post an angry message. However, the significance of contagion
is relatively low for disgust and sadness, especially for disgust, suggesting the possibility of
contagion is trivial. Because of this, in the following content we only focus on anger and joy.
Note that dialso decrease as ∆ grows, because the influence of the emotion decays with time
and thus users are easier to be influenced by messages they received in more recent hours,
which is consistent with our intuition. While even ∆ grows, anger persistently shows the
most significant difference, indicating that it is more contagious than joy and is a stronger
stimuli in emotion expression of human in the social network.
Emotion influence is closely entangled with emotion contagion. When emotion spread in
5
the social network, connected individuals could influence each other emotionally and later
might lead to the clustering of emotion, i.e., networked individuals demonstrate homophily
in sentimental statuses. For example, Bollen et. al find that happiness of users is assorta-
tive across Twitter [7] and Bliss et. al reveal that average happiness scores are positively
correlated between friends within three hops [5]. Particularly, Fan et. al even show that in
Weibo anger possesses stronger correlation than joy for friends within three hops, suggesting
that anger is more influential than joy [18]. Inspired by this, similar to the metric presented
in [19], here we also evaluate the contagion tendency of anger and joy from the perspective
of emotion influence. For each tweet tuwith emotion iposted by u, we calculate the Eu-
clidean distance ej(j= 1,2,3 or 4) between v(tu) and vj(j= 1,2,3 or 4). We define tuis
an emotionally influenced tweet, i.e., tuis stimulated by emotion icarried by tweets that u
received within ∆ if the emotion with the smallest distance is i. Then for anger and joy,
the one with higher percentage of influenced tweets will be more influential and contagious.
As shown in Fig. 1(b), anger possesses the higher percentage of influenced tweets than joy,
indicating that as compared to joy, it is easier to be stimulated by tweets that users are
exposed to. Meanwhile, as ∆ grows both emotions’ influenced percentages increase because
users are exposed to more tweets when ∆ is large, while anger persistently possesses the
higher percentage than joy and the gap is even enlarged for greater ∆.
We further investigate the emotion contagion from the view of emontionally influenced
users. Specifically, all users are sorted in the descending order of the percentage of their
influenced tweets, i.e., the faction of influenced tweets in their tweeting timelines. The
top and bottom 15% users are then defined as the high susceptible and the low susceptible,
respectively. As can be seen in Fig. 2, low susceptible users are easier to be affected by anger
while high susceptible users are more inclined to be stimulated by joy. Different from [19],
in which it is claimed that both high and low susceptible users are more susceptible to
positive feelings than the negative, however, our results demonstrate that in fact anger, one
of the negative emotions, is more susceptible than joy for low susceptible users. It means
that a fine-grained emotion category can lead to new insights in discussing the emotion
contagion and anger, out of expectation, is more contagious than joy. Meanwhile, relatively
lower number of followers for the low susceptible user indicates its higher occupation than
the high susceptible in the soical network and given the power-law like distribuiton of the
number of followers, it is further suggested that anger is more contagious for the majority
6
0
20
40
60
80
Low susceptible High susceptible
Percentage (%)
Anger
Joy
Figure 2. Influenced percentage of anger and joy in both high and low susceptible users,
respectively.
of the social network that joy.
Being more contagious and more susecptible to the majority indicate that anger could be
transfered faster than joy, because it will inspire more follow-up tweets under the same in-
tensity of stimuli. However, the mechanism underlying emotion contagion also relies heavily
on structure of the social network and thus how structure functions on contagion of joy and
anger indeed deserves further explorations.
B. Anger prefers weaker ties than joy
The dynamics of emotion contagion is essentially coupled with the underlying social
network that providing channels for disseminating the sentiment from one individual to
others. Except the willingness of posting an angry or joyful tweet after exposed to anger
or joy from followees, another key factor that determines the user’s action is that on which
relationship the contagion will happen with higher likelihood. Specifically, the prediction of
which friend will retweet the emotional post in the future will be insightful for the contagion
modeling and control.
The social relationship, or the tie in social network, can be measured by its strength, which
7
is instrumental for spreading both online and real-world human behaviors [8]. The strength
of a tie in online social media can be intuitively measured through online interactions, i.e.,
retweets, between its two ends. Here three metrics are presented to depict tie strengths
quantificationally. The first one is the proportion of common friends [38, 46], which is
defined as cij /(ki1 + kj1cij ) for the tie between users iand j, where cij is the
number of common friends, kiand kjrepresent degrees of iand j, respectively. Note that
for the metric of common friends, the social network of Weibo is converted to undirected
with each link representing the possible interaction between both ends. The second metric
is inspired by the reciprocity in Twitter-like services and higher ratio of reciprocity indicates
more trust and more significant homophily [47]. Hence for a pair of users, the proportion
of reciprocal retweets in the total flux between them is defined as the tie strength. The
third metric is the number of reweets between two ends of a tie in Weibo and greater values
stand for more frequent interactions. Note that different from the previous two metrics,
the retweet strength is time-evolving and we only count the retweets happening before the
corresponding emotional retweet. And in order to smooth the comparison between anger
and joy, the retweet strength, denoted as S, is also normalized by (SSmin)/(Smax Smin),
in which Smin and Smax separately represent the minimum and maximum values in all
obervations.
By investigating each emotional retweet, i.e., reposting an angry or joyful tweet posted
by the followee, we correlate emotion contagion with tie strengths. As can be seen in
Fig. 3, all the metrics consistently demonstrate that anger prefers weaker ties than joy in
contagion, suggesting that angry tweets spread through weak ties with greater odds than
the joyful ones. It is well known that weak ties play essential roles in diffusion of social
networks [24], especially in break the local trap caused by communities in terms of bridging
different clusters together [16, 38, 45]. For example, a typical snapshot of emotion contagion
with four communities are demonstrated in Fig. 4, in which anger disseminates through
more weak ties of inter-communities than joy. Because of this, as compared to joy, anger
has more chances to penetrate different communities through its preference on weak ties in
emotion contagion. And more communities lead to more global coverage, indicating that
anger could reach broader dissemination than joy over the same time period.
Both evidence from contagion tendency and relationship strength suggests that anger
could spread faster than joy, because it will stimulate more follow-up tweets and penetrate
8
0.00
0.01
0.02
0.03
Anger Joy
Common Friends
a
0.00
0.05
0.10
0.15
Anger Joy
Reciprocity (%)
b
0.000
0.005
0.010
0.015
Anger Joy
Retweets
c
Figure 3. Anger prefers more weak ties than joy. Three different metrics are averaged
over all emotional retweets in the dataset. Lower metrics for anger suggests that in
contagion anger disseminates through weaker ties than joy. Error bars represent standard
errors in (a) and (c), while in (b) there is no stand error beause the reciprocity is just a
ratio obtained from all emotional retweets.
more communities in the social network of Weibo. Next direct evidence from online bursts
in Weibo will be demonstrated to testify this conjecture further.
C. Anger spreads faster than joy
No moods are created equally online [15], the difference in contagion of positive and
negative feelings is a trending but controversial topic for years. Berger and Milkman reveal
that more positive content is more viral than negative content in study of NY Times [4]. Tadi
and ˇ
Suvakov find that for human-like bots in online social network, positive emotion bots are
more effective than the negative ones [42]. Wu et al. even claim that bad news containing
more negative words fades more rapidly in Twitter [43]. While Chmiel et. al point out that
negative sentiments boost user activity in BBC forum [11]. Pfitzner et. al state that users
tend to retweet tweets with high emotional diversity [40]. Ferrara and Yang demonstrate
that though people are more likely to share positive content in Twitter, negative messages
9
Figure 4. Emotion contagion in a sampled snapshot of the Weibo network with four
communities and links with more angry retweets will be orange and otherwise will be
green. It can be seen that anger prefers more weak ties bridging different communities
than joy. Besides, the left two communities dominated by anger and thus a large percent
of messages disseminate inter communities. While for the two right joy-dominated
communities, emotion tends to diffuse inside the community.
spread faster than positive ones at the level of contents [20]. Thus Hansen et al. conclude
that the relation between emotion and virality is much complicated [27]. Here we argue that
fine-grained division of human emotion, especially the negative, will enrich the background
of investigating the contagion difference. And in the meantime, explicit definitions of fast
spread will further facilitate the elimination of debates on this issue.
Over 600 bursty events are extracted from Weibo and for each event, the emotion occu-
pying more than 60% emotional tweets will be defined as the dominant emotion of the event.
In total we get 37 anger-dominated events and 200 joy-dominated events, as can be seen
Fig. 5(a), in which the volume of emotional tweets fluctuates with time and anger takes over
the majority. For an online burst, the fast spread means the volume of emotional tweets
grow quickly in the period from the awakening instant to the diffusion peak. Through a
parameterless approach presented in [30], we can precisely locate the instant of awakening
10
0
200
400
600
800
0 10 20 30 40 50
τ
Volume
Anger
Disgusting
Joy
Sadness
315 party
0
500
1000
1500
0 10 20 30 40
τ
Volume
0
2000
4000
6000
0 10 20 30 40
τ
Volume
a
b c
Figure 5. Illustration of the online bursts. (a) An example of an anger-dominated event
and the newly posted emotional tweets varies as time τevolves. (b) Examples of locating
the awakening and the peak for both anger and joy dominated events. xAdenotes the
instant of awakening, xPdenotes the peaking time and yAand yPstand for the volume of
emotional tweets at the instants of awakening and peak, respectively. The slope of the line
(blue) between points (xA, yA) and (xP, yP) can reflect the averaged velocity.
and peak. Then how fast the spread is can be reflected by the interval between the awak-
ening and the peak (shorter intervals suggest faster spread) or the averaged velocity of the
growth in diffusion from awakening to the peak (higher velocities indicate faster spread).
Specifically, as demonstrated in Fig. 5(b), regarding to interval, 1/(xPxA) can reflect how
fast the spread is, while with respect to averaged velocity, it can be defined as the slope,
i.e., (yPyA)/(xPxA) and for both measures higher values suggest faster spread. The
averaged results over all anger and joy dominated events are listed in Table I and it can be
seen that both metrics testify significantly that anger-dominated events incline to arrive the
peak from the awakening more quickly than joy. We can conclude that anger indeed spreads
faster than the joy in social media like Weibo.
11
Measures Anger Joy
Time-difference 0.179 0.102
Slope 0.329 0.238
TABLE I. Averaged values over all anger and joy dominated events.
III. DISCUSSION
Bad is always stronger than good [3]. From the previous studies, it is profoundly disclosed
that anger is more influential than joy in Weibo [18] and the negative contents in Twitter
spread faster than positive ones [20]. However, in this study, by finely splitting the negative
feelings into anger, disgust and sadness, we for the first time offer the empirical evidence
that anger spreads faster than joy in social media, implying that anger on the Internet
should be considered primarily in either personal emotion management or collective mood
understanding.
The way how anger is expressed and experienced on the Internet is arousing attention.
It has been found from self-reports of rant-site visitors that posting anger produces relaxed
feelings immediately [35], which makes posting anger an effective way of self-regulation
in mood. However, considering the easy contagion of anger, more angry expressions on the
Internet might arouse negative shifts in mood of the crowd that connected to posters [35, 39].
Moreover, anger’s preference of weak ties will make it likely to spread to “strangers” in the
Internet. The users want to ease anger by posting on the Internet should be suggested to
underhand the possible impact to their online social network. Even in offline scenario, like
the “road rage” [1] at rush hours of China, in which anger spreads quickly between strangers
and might cause aggressive driving or even accidents. We suggest that in personal anger
management, the unexpected contagion to strangers should be seriously considered.
The online social media has already been the most ubiquitous platform for collective
intelligence, in which various signals generated from massive connected individuals provide
the foundation of collective behavior understanding. However, how emotion contagion affects
the aggregation of the individual behavior, particularly the individual intelligence, is rarely
considered. In fact from our findings, emotion spread, especially the anger’s fast contagion,
might imply a lot to the collective behavior in the cyber space, especially the crowdsourcing
12
results. It is even stated that emotion, such as anger, can be a threat to reasoning and
arguing [34]. For example, the outrages of massive emotional individuals, which would badly
bias the public opinion, might just come from the fast spread of anger, but not because the
event itself. Meanwhile, fast contagion of anger also offers a new perspective to picture
the emotional behavior of the crowd on the Internet. We suggest that for scenarios like
crowdsourcing or collective behavior understanding, angry users should be carefully treated
to omit their possible impact on the fair judgment of the observations in the epoch of big
data. Besides, diminishing weak ties will function effectively in controlling the diffusion of
the Internet outrages and make the crowd rational and smart.
Contrarily, happiness is believed to unify and clustering the community [12] and our
finding about the joy’s preference on less weak ties also supports this (see Fig.4), implying
that strong ties inside the community disseminate more joyful content in online social media.
Meanwhile, self-reports from Facebook users also testify that communication with strong ties
is associated with improvements in well-being [9], which further testifies that our findings
from the computational view is solid.
IV. CONCLUSION
Instead of self-reports in controlled experiments, the natural and emotional postings
in Weibo social network are collected to investigate the detailed mechanism of emotion
contagion from a new view of computational social science. For the first time, we offer the
solid evidence for the fact that anger spreads faster than joy in social media because of
being more contagious and more preferential on weak ties. Our findings shed lights on both
personal anger management and collective behavior understanding.
This study has inevitable limitations. It is generally accepted that emotion expression is
culture dependent and demographics like gender also matter [2, 29], suggesting that exploring
how anger spreads in Twitter and how the male and the female response differently to
emotion contagion will be of great significance in the future work.
13
ACKNOWLEDGMENT
This work was supported by the National Natural Science Foundation of China (Grant
Nos. 71501005, 71531001 and 61421003) and the fund of the State Key Lab of Software
Development Environment (Grant Nos. SKLSDE-2015ZX-05 and SKLSDE-2015ZX-28).
[1] Road rage. https://en.wikipedia.org/wiki/Road_rage, 2016.
[2] Richard P. Bagozzi, Nancy Wong, and Youjae Yi. The role of culture and gender in the
relationship between positive and negative affect. COGNITION AND EMOTION, 13(6):641–
672, 1983.
[3] Roy F. Baumeister, Ellen Bratslavsky, Catrin Finkenauer, and Kathleen D. Vohs. Bad is
stronger than good. Review of General Psychology, 5(4):323–370, 2001.
[4] Jonah Berger and Katherine L. Milkman. Social transmission, emotion, and the virality of
online content. Wharton Research Paper, 2010.
[5] Catherine A Bliss, Isabel M Kloumann, Kameron Decker Harris, Christopher M Danforth, and
Peter Sheridan Dodds. Twitter reciprocal reply networks exhibit assortativity with respect to
happiness. Journal of Computational Science, 3(5):388–397, 2012.
[6] Johan Bollen, Bruno Gon¸calves, Guangchen Ruan, and Huina Mao. Happiness is assortative
in online social networks. Artificial life, 17(3):237–251, 2011.
[7] Johan Bollen, Huina Mao, and Xiaojun Zeng. Twitter mood predicts the stock market. Journal
of Computational Science, 2(1):1–8, 2011.
[8] Robert M. Bond, Christopher J. Fariss, Jason J. Jones, Adam D. I. Kramer, Cameron Marlow,
Jaime E. Settle, and James H. Fowler. A 61-million-person experiment in social influence and
political mobilization. Nature, 489(7415):295–298, 2012.
[9] Moira Burke and Robert E. Kraut. The relationship between facebook use and well-being
depends on communication type and tie strength. Journal of Computer-Mediated Communi-
cation, 21:265281, 2016.
[10] Anna Chmiel, Julian Sienkiewicz, Mike Thelwall, Georgios Paltoglou, Kevan Buckley, Arvid
Kappas, and Janusz A Ho lyst. Collective emotions online and their influence on community
life. PloS ONE, 6(7):e22207, 2011.
14
[11] Anna Chmiel, Pawel Sobkowicz, Julian Sienkiewicz, Georgios Paltoglou, Kevan Buckley, Mike
Thelwall, and Janusz A Ho lyst. Negative emotions boost user activity at BBC forum. Physica
A: statistical mechanics and its applications, 390(16):2936–2944, 2011.
[12] Nicholas A. Christakis and James H. Fowler. Connected: The Surprising Power of Our Social
Networks and How They Shape Our Lives. Back Bay Books, 2011.
[13] Lorenzo Coviello, Yunkyu Sohn, Adam DI Kramer, Cameron Marlow, Massimo Franceschetti,
Nicholas A Christakis, and James H Fowler. Detecting emotional contagion in massive social
networks. PloS one, 9(3):e90315, 2014.
[14] Linh Dang-Xuan and Stefan Stieglitz. Impact and diffusion of sentiment in political
communication-an empirical analysis of political weblogs. In ICWSM, 2012.
[15] Munmun De Choudhury, Scott Counts, and Michael Gamon. Not all moods are created equal!
exploring human emotional states in social media. In Proceedings of the Sixth International
Conference on Weblogs and Social Media, ICWSM ’12, 2012.
[16] Pasquale De Meo, Emilio Ferrara, Giacomo Fiumara, and Alessandro Provetti. On facebook,
most ties are weak. Communications of the ACM, 57:78–84, 2014.
[17] Daantje Derks, Agneta H. Fischer, and Arjan E.R. Bos. The role of emotion in computer-
mediated communication: A review. Computers in Human Behavior, 24:766–785, 2008.
[18] Rui Fan, Jichang Zhao, Yan Chen, and Ke Xu. Anger is more influential than joy: sentiment
correlation in weibo. PLoS ONE, 9:e110184, 2014.
[19] Emilio Ferrara and Zeyao Yang. Measuring emotional contagion in social media. PloS ONE,
10(11):e0142390, 2015.
[20] Emilio Ferrara and Zeyao Yang. Quantifying the effect of sentiment on information diffusion
in social media. PeerJ Computer Science, 1:e26, 2015.
[21] James H Fowler and Nicholas A Christakis. Dynamic spread of happiness in a large social
network: longitudinal analysis over 20 years in the framingham heart study. BMJ, 337:a2338,
2008.
[22] Uri Gneezy and Alex Imas. Materazzi effect and the strategic use of anger in competitive
interactions. Proceedings of the National Academy of Sciences, 111(4):1334–1337, 2014.
[23] Scott A. Golder and Michael W. Macy. Diurnal and seasonal mood vary with work, sleep,
and daylength across diverse cultures. Science, 333(6051):1878–1881, 2011.
15
[24] Mark S Granovetter. The strength of weak ties. American journal of sociology, pages 1360–
1380, 1973.
[25] Anatoliy Gruzd, Sophie Doiron, and Philip Mai. Is happiness contagious online? a case
of twitter and the 2010 Winter Olympics. In System Sciences (HICSS), 2011 44th Hawaii
International Conference on, pages 1–9. IEEE, 2011.
[26] Jamie Guillory, Jason Spiegel, Molly Drislane, Benjamin Weiss, Walter Donner, and Jeffrey
Hancock. Upset now?: Emotion contagion in distributed groups. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems, CHI ’11, pages 745–748, New York,
NY, USA, 2011. ACM.
[27] Lars Kai Hansen, Adam Arvidsson, Finn Aarup Nielsen, Elanor Colleoni, and Michael Et-
ter. Good Friends, Bad News - Affect and Virality in Twitter, pages 34–43. Springer Berlin
Heidelberg, Berlin, Heidelberg, 2011.
[28] Elaine Hatfield, John T Cacioppo, and Richard L Rapson. Emotional contagion. Curr Dir
Psych Sci, 2(3):96–100, 1993.
[29] Yue Hu, Jichang Zhao, and Junjie Wu. Emoticon-based ambivalent expression: A hidden
indicator for unusual behaviors in weibo. PLoS ONE, 11:e0147079, 2016.
[30] Qing Ke, Emilio Ferrara, Filippo Radicchi, and Alessandro Flammini. Defining and identifying
sleeping beauties in science. Proceedings of the National Academy of Sciences, 112(24):7426–
7431, 2015.
[31] Adam D.I. Kramer. The spread of emotion via facebook. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems, CHI ’12, pages 767–770, New York,
NY, USA, 2012. ACM.
[32] Adam DI Kramer, Jamie E Guillory, and Jeffrey T Hancock. Experimental evidence of
massive-scale emotional contagion through social networks. Proceedings of the National
Academy of Sciences, 111(24):8788–8790, 2014.
[33] David Lazer, Alex Pentland, Lada Adamic, Sinan Aral, Albert-L´aszl´o Barab´asi, Devon Brewer,
Nicholas Christakis, Noshir Contractor, James Fowler, Myron Gutmann, Tony Jebara, Gary
King, Michael Macy, Deb Roy, and Marshall Van Alstyne. Computational social science.
Science, 323(5915):721–723, 2009.
[34] Stacy Marsella and Jonathan Gratch. Computationally modeling human emotion. Commu-
nications of The ACM, 57(12):56–67, 2014.
16
[35] Ryan C Martin, Kelsey Ryan Coyier, Leah M Vansistine, and Kelly Schroeder. Anger on the
internet: The perceived value of rant-sites. Cyberpsychology, Behavior, and Social Networking,
16(2):119, 2013.
[36] All the data sets will be publicly available soon. Any information please contact
jichang@buaa.edu.cn.
[37] Lauri Nummenmaa, Enrico Glerean, Mikko Viinikainen, Iiro P. Jskelinen, Riitta Hari, and
Mikko Sams. Emotions promote social interaction by synchronizing brain activity across
individuals. Proceedings of the National Academy of Sciences, 109(24):9599–9604, 2012.
[38] J-P Onnela, Jari Saram¨aki, Jorkki Hyv¨onen, Gy¨orgy Szao, David Lazer, Kimmo Kaski, J´anos
Kert´esz, and A-L Barab´asi. Structure and tie strengths in mobile communication networks.
Proceedings of the National Academy of Sciences, 104(18):7332–7336, 2007.
[39] Minsu Park, Chiyoung Cha, and Meeyoung Cha. Depressive moods of users portrayed in
twitter. In Proceedings of the ACM SIGKDD Workshop on Healthcare Informatics (HI-KDD),
2012.
[40] Ren´e Pfitzner, Antonios Garas, and Frank Schweitzer. Emotional divergence influences infor-
mation spreading in twitter. In Proceedings of the Sixth International AAAI Conference on
Weblogs and Social Media, pages 2–5, 2012.
[41] J Niels Rosenquist, James H Fowler, and Nicholas A Christakis. Social network determinants
of depression. Molecular psychiatry, 16(3):273–281, 2011.
[42] Bosiljka Tadi and Milovan uvakov. Can human-like bots control collective mood: agent-
based simulations of online chats. Journal of Statistical Mechanics: Theory and Experiment,
2013(10):P10014, 2013.
[43] Shaowei Wu, Chenhao Tan, Jon Kleinberg, and Michael Macy. Does bad news go away faster?
In Proceedings of the Fifth International Conference on Weblogs and Social Media, ICWSM
’11, 2011.
[44] Jichang Zhao, Li Dong, Junjie Wu, and Ke Xu. Moodlens: An emoticon-based sentiment
analysis system for Chinese tweets. In Proceedings of the 18th ACM SIGKDD International
Conference on Knowledge Discovery and Data Mining, KDD ’12, pages 1528–1531, New York,
NY, USA, 2012. ACM.
[45] Jichang Zhao, Junjie Wu, Xu Feng, Hui Xiong, and Ke Xu. Information propagation in online
social networks: A tie strength perspective. Knowledge And Information System(KAIS),
17
32:589–608, 2012.
[46] Jichang Zhao, Junjie Wu, and Ke Xu. Weak ties: Subtle role of information diffusion in online
social networks. Phys. Rev. E, 82:016105, Jul 2010.
[47] Yu-Xiao Zhu, Xiao-Guang Zhang, Gui-Quan Sun, Ming Tang, Tao Zhou, and Zi-Ke Zhang.
Influence of reciprocal links in social networks. PloS ONE, 9(7):e103007, 2014.
18
... Thus, it was reasonable to say that different emotions would contribute to different behaviors. A very vivid example used by Fan, Xu and Zhao [15] in their article was the scenario of road rage, in which they depicted that the negative emotions of aggression, anger, etc. spread swiftly among drivers during rush hour, producing a collective tendency of uncivil language and rude behaviors. ...
... Though scholars were on the same page, admitting that emotions indeed influenced people's behaviors, they held divergent opinions concerning the salience and potency of positive and negative emotions that are operative. Some said negative emotions were more potent in motivating people than their positive counterparts [7,15]. They claimed that negative emotions could spread faster than positive ones and incite people to take more rash actions. ...
... Scholars continue to debate whether positive or negative content is more widespread and influential. Many have found that negative emotions contribute to more shares and are more powerful than positive emotions [5,15,19]. For example, according to Chmiel et al. [19], negative emotions drove Internet discussions; and users with more negative emotions tended to write more posts in online BBC fora. ...
Article
Social media are becoming new venues for more and more people to get information, express opinions, and connect with others. Various online communication behaviors carry and, meanwhile, are influenced by certain emotions. This study adopted a mixed method of survey and in-depth interviews to examine how the emotionally loaded content influenced people's online engagement. The targeted population focused on those active users on Chinese social media. The results revealed a unique phenomenon on social media in post-Covid China that both joy and anger attracted greater attention and more interaction. Also, the emotion of shock turned out to be unexpectedly influential on people's sharing behavior. It ranked ahead of anger and was second only to joy in getting people's attention. Furthermore, this study found that the content virality depended on the intended audience groupthat people tended to forward the angry messages to friends in private and retweet the joyful content publicly. These findings added nuances to previous literature concerning the role of emotions on social media. It reflected a relatively new reality of the post-Covid online phenomena in China.
... Based on epidemiological mechanisms, spreading is seen as a continuous process, with emotional contagion occurring in individuals who are carriers of emotional infection, while individuals in susceptible and recovery states cannot transmit emotions to others. Additionally, the emotional spread is influenced by factors such as emotional valence, the contagion intensity of different emotions, paths of emotional contagion, and individuals' susceptibility (Fan et al. 2016;Wei-dong et al. 2015;Xu et al. 2022;Ferrara and Yang, 2015a). Emotional propagation on social media following natural disaster events can have dramatic individual and societal consequences. ...
Article
Full-text available
With the advent of climate change and the 5 G era, online communities are increasingly becoming the main medium for information dissemination after emergencies such as natural disasters. The widespread dissemination of negative online information may generate cyber violence or lead to serious adverse psychological outcomes. This study considered a natural disaster event involving avoidable deaths and child casualties as an example to identify emotional contagion and conduct simulation interventions. Data about the aftermath of the 8·13 flash flood in the Longcaogou Scenic Area, Sichuan Province, China, were derived from the Chinese Sina microblog. We analyzed key parameters and modeled them in a dynamic model. We further evaluated the effects of implementing intervention measures (such as transmission path interruption and changing the number of different emotions) on emotional spread. The overall sentiment of posters after this flood was negative, with three epidemic peaks. Negative emotions were more persistent and contagious than positive emotions. Reducing the number of negative blog posts by half could have led to a 14.97% reduction in negative comments and a 7.17% reduction in positive comments. Simultaneously, reducing the number of negative blog posts and increasing the number of positive posts would have helped reduce the relative ratio of negative to positive comments. The findings have theoretical and practical implications for developing an emotional contagion model and formulating intervention strategies to guide public opinion after an emergency that involves extensive online debate.
... Another observational study of Twitter users and their followees' posts arrived at a similar conclusion, thereby giving evidence to the presence of emotional contagion [33]. Some works suggested also that negative emotions, such as anger, expressed in the initial social media content, are more contagious than positive ones, such as joy [34,35]. Yet another study demonstrated that, although the emotional tone of users' messages is usually adapted to that of the chat, chat's emotion tone can be characterized as persistent, rather than fluctuating towards particular negative or positive emotions [36]. ...
Article
Full-text available
In this work, we explore the role of immigrant-critical alternative media in shaping collective emotions and users’ evaluations of the immigration issue, using a conversational approach and an empirical case of Flashback, a prominent Swedish online platform where many immigration-related discussions take place. Our text and network-based analysis of more than 9,000 conversations during the last election period reveals that the platform users consume and distribute diverging types of media content across a wide ideological spectrum which, however, has a limited influence on the evolution of conversations and users’ stances in the immigration debate. Nevertheless, we find that the conversation networks with alternative media content tend to include more negative evaluations of the immigration issue, attracting fewer participants and lasting less than other conversations. We contextualise our findings using Collins’ Interaction Ritual Chains (IRC) theory and discuss the conditions under which such online conversations can produce high user involvement and, potentially, participants’ radicalisation.
... As new forms of communication develop, new practices of violence spread in different fields and at different levels. This is particularly true of the dissemination of violent language through computer-mediated communication (CMC) and social media, as numerous studies have described, in the fields of politics (Breeze, 2020;Ekman, 2019), racism (van Dijk, 2000Fairclough, 2003), or misogyny (Blake et al., 2021;Centola, 2010;Fan et al., 2016), for example. Due to its immediacy and global reach, violence in CMC spreads like wildfire and contaminates other environments, even when it originates in offline environments. ...
Article
Full-text available
Violence against women is a gruesome problem in our society. In 2021 alone, there were around 170,000 complaints of gender-based violence in Spain, and this number has been steadily increasing since 2009, when the first data was collected. Complaints of gender-based violence in Spain are handled and decided by exclusive, specialised or compatible courts; if convicted, the perpetrators can then appeal to higher courts. Given the performative nature of language (Austin, 1962; Searle, 1969), it is of the utmost importance to analyse the language used in court by judges and lawyers to refer to violence against women, as this influences the mainstream social perception by facilitating a construal of gender violence. The ways in which language and violence are intertwined have long been studied from various perspectives (Derrida, 2001; Fairclough, 1989; Foucault, 1980). Other authors have explored how language can inflict (and therefore provoke) violence in various forms (van Dijk, 1995; Silva, 2017a, 2017b). In this paper, we examine the representation of gender-based violence in legal language in Spain. We address the language used to refer to violence in the texts analysed; the portrayal in court judgements of participants in acts of gender-based violence; and the resulting social representation of violence, victims and aggressors. We also examine the representation of women from a constructionist perspective (Butler, 1992; 1997; 2004), using corpus pragmatics (Romero-Trillo, 2008) to analyse lexical items and expressions in 20 court judgments in the context of gender-based violence. The results indicate some bias in the representation of violence, aggressors and victims in court judgments.
... Another dimension where digital technology's effects are deemed to have been destructive is the domain of spirituality. Digital media are regularly described as robbing users of their sense of presence, bringing them into states of distraction, restlessness, fear of missing out (FOMO), anger, and envy that runs counter to spiritual ideals of presence, firmness and peace (Carr, 2010;Dossey, 2014;Fan et al., 2016;Faiola et al., 2018;Rosen et al., 2018;Rozgonjuk et al., 2020;Serrano, 2020;Hartogsohn, 2022). ...
Article
Full-text available
The concept of cyberdelics emerged in the 1980s and 1990s as an umbrella term denoting the nexus connecting cybernetic (digital) technologies and psychedelic (mind manifesting) drugs. Cyberdelic technologies, in particular the then newly emerging field of virtual reality, were touted by psychedelic cultural icons including Timothy Leary and Terence McKenna as auguring a new era of digital mind-expansion where psychedelic experiences will be recreated online inside virtual worlds. Cyberdelic culture waned in the 2000s. However, recent years have seen the return of the cyberdelic imaginary, following on the heels of a psychedelic resurgence and a renewed interest in virtual reality technologies and their use in therapy. Cyberdelic advocates speak of the necessity of creating transformative technologies that steer humanity away from mindless consumerism and distractedness, and towards expanded states of awe, presence, and transcendence. Nevertheless, much like psychedelics, cyberdelic technologies are seen as running against the grain of current sociocultural arrangements and economic models which threaten to quell their transformative potential. Research on psychedelics within the humanities over the past decade has emphasized the role of cultural set and setting: the significance of the cultural embeddedness of these psychoactive agents and the dependence of their effects on surrounding sociocultural conditions. Building on the notion of information technologies as mind-manifesting technologies, this paper sets out to consider what psychedelics can teach us about cyberdelics: how the principles of set and setting and current discussions within the psychedelic humanities can inform our understanding of the resurgence of interest in cyberdelic media, its prospects, and challenges.
Conference Paper
Understanding how we respond to information-manipulation cyber-attacks is critical to devise disinformation countermeasures that increase resilience against them. We focused on the emotions that promoted secondary social sharing of disinformation, that is sharing others' episodes with third parties. This study aimed to clarify the possibility that emotion had a stronger influence on the secondary social sharing of disinformation compared to authenticity judgments and other factors. We created stimuli that imitated disinformation or true information posted on social media and manipulated their emotional characteristics. Using these stimuli, we asked participants regarding the intensity and type of emotions aroused, authenticity judgments, and sharing intentions. Results showed that strong emotions had a greater influence on sharing intentions than on authenticity judgments. This tendency was seen when anger was evoked. Disinformation often evokes strong anger, which suggests that it promotes secondary social sharing, regardless of authenticity judgments.
Article
Full-text available
Social media users tend to produce content that contains more positive than negative emotional language. However, negative emotional language is more likely to be shared. To understand why, research has thus far focused on psychological processes associated with tweets’ content. In the current study, we investigate if the content producer influences the extent to which their negative content is shared. More specifically, we focus on a group of users that are central to the diffusion of content on social media – public figures. We found that an increase in negativity was associated with a stronger increase in sharing for public figures compared to ordinary users. This effect was explained by two user characteristics, the number of followers and thus the strength of ties, and proportion of political tweets. The results shed light on whose negativity is most viral, allowing future research to develop interventions aimed to mitigate overexposure to negative content.
Article
Instagram has been perceived as the least appropriate platform to express negativity online, yet the number of posts tagged under #complain, #complaint, #complaints, and #complaining are increasing. We conducted a controlled web-based experiment to examine the extent to which exposure to other's quotes of complaints contributed to increased similarity in emotions for the audience (i.e., digital emotion contagion). Participants (n = 591 Instagram users in Indonesia; 82.23% females; Mage = 28.06, SD = 6.39) were randomly assigned to exposure of complaint quotes containing seven basic emotions. We found that exposure to three of the five complaint quotes (i.e., anger, disgust, and sadness) induced similar emotions in the participants while the two other complaint quotes (i.e., fear and anxiety) induced overlapping emotions, but a non-complaint quote (i.e., desire and satisfaction) induced alternative emotions. Taken together, digital emotion contagion was likely produced by exposure to complaint quotes, while exposure to non-complaint quotes induced different, possibly complementary-like, emotions. Although these findings should be considered as a snapshot of the complex emotional dynamics online, it highlights that exposure to simple Instagram quotes has the potential to go beyond mere contagion.
Article
Full-text available
Social media are used as main discussion channels by millions of individuals every day. The content individuals produce in daily social-media-based micro-communications, and the emotions therein expressed, may impact the emotional states of others. A recent experiment performed on Facebook hypothesized that emotions spread online, even in absence of non-verbal cues typical of in-person interactions, and that individuals are more likely to adopt positive or negative emotions if these are over-expressed in their social network. Experiments of this type, however, raise ethical concerns, as they require massive-scale content manipulation with unknown consequences for the individuals therein involved. Here, we study the dynamics of emotional contagion using Twitter. Rather than manipulating content, we devise a null model that discounts some confounding factors (including the effect of emotional contagion). We measure the emotional valence of content the users are exposed to before posting their own tweets. We determine that on average a negative post follows an over-exposure to 4.34% more negative content than baseline, while positive posts occur after an average over-exposure to 4.50% more positive contents. We highlight the presence of a linear relationship between the average emotional valence of the stimuli users are exposed to, and that of the responses they produce. We also identify two different classes of individuals: highly and scarcely susceptible to emotional contagion. Highly susceptible users are significantly less inclined to adopt negative emotions than the scarcely susceptible ones, but equally likely to adopt positive emotions. In general, the likelihood of adopting positive emotions is much greater than that of negative emotions.
Article
Full-text available
Social media has become the main vehicle of information production and consumption online. Millions of users every day log on their Facebook or Twitter accounts to get updates and news, read about their topics of interest, and become exposed to new opportunities and interactions. Although recent studies suggest that the contents users produce will affect the emotions of their readers, we still lack a rigorous understanding of the role and effects of contents sentiment on the dynamics of information diffusion. This work aims at quantifying the effect of sentiment on information diffusion, to understand: (i) whether positive conversations spread faster and/or broader than negative ones (or vice-versa); (ii) what kind of emotions are more typical of popular conversations on social media; and, (iii) what type of sentiment is expressed in conversations characterized by different temporal dynamics. Our findings show that, at the level of contents, negative messages spread faster than positive ones, but positive ones reach larger audiences, suggesting that people are more inclined to share and favorite positive contents, the so-called positive bias . As for the entire conversations, we highlight how different temporal dynamics exhibit different sentiment patterns: for example, positive sentiment builds up for highly-anticipated events, while unexpected events are mainly characterized by negative sentiment. Our contribution represents a step forward to understand how the emotions expressed in short texts correlate with their spreading in online social ecosystems, and may help to craft effective policies and strategies for content generation and diffusion.
Article
Full-text available
There has been rapid growth of cross-disciplinary research employing computational methods to understand human behavior, as well as facilitate interaction between people and machines, with work on computational models of emotion becoming an important component. Modeling appraisal theory in an agent provides an interesting perspective on the relation between emotion and rationality. Appraisal theory argues that emotion serves to generalize stimulus response by providing more general ways to characterize types of stimuli in terms of classes of viable organism responses. Assessments such as desirability, coping potential, unexpectedness, and causal attribution are clearly relevant to any social agent, whether deemed emotional or not. Having been characterized in a uniform fashion, the appraisal results coordinate system-wide coping responses that serve to guide the agent's specific responses to the eliciting event, essentially helping the agent find its ecological niche. Models designed to inform intelligent systems might avoid some of the seemingly irrational application of coping humans adopt.
Article
Full-text available
Recent decades have witnessed online social media being a big-data window for quantificationally testifying conventional social theories and exploring much detailed human behavioral patterns. In this paper, by tracing the emoticon use in Weibo, a group of hidden "ambivalent users" are disclosed for frequently posting ambivalent tweets containing both positive and negative emotions. Further investigation reveals that this ambivalent expression could be a novel indicator of many unusual social behaviors. For instance, ambivalent users with the female as the majority like to make a sound in midnights or at weekends. They mention their close friends frequently in ambivalent tweets, which attract more replies and thus serve as a more private communication way. Ambivalent users also respond differently to public affairs from others and demonstrate more interests in entertainment and sports events. Moreover, the sentiment shift of words adopted in ambivalent tweets is more evident than usual and exhibits a clear "negative to positive" pattern. The above observations, though being promiscuous seemingly, actually point to the self regulation of negative mood in Weibo, which could find its base from the emotion management theories in sociology but makes an interesting extension to the online environment. Finally, as an interesting corollary, ambivalent users are found connected with compulsive buyers and turn out to be perfect targets for online marketing.
Article
Full-text available
Significance We show, via a massive ( N = 689,003) experiment on Facebook, that emotional states can be transferred to others via emotional contagion, leading people to experience the same emotions without their awareness. We provide experimental evidence that emotional contagion occurs without direct interaction between people (exposure to a friend expressing an emotion is sufficient), and in the complete absence of nonverbal cues.
Article
PERVASIVE SOCIO-TECHNICAL NETWORKS bring new conceptual and technological challenges to developers and users alike. A central research theme is evaluation of the intensity of relations linking users and how they facilitate communication and the spread of information. These aspects of human relationships have been studied extensively in the social sciences under the framework of the “strength of weak ties” theory proposed by Mark Granovetter. Some research has considered whether that theory can be extended to online social networks like Facebook, suggesting interaction data can be used to predict the strength of ties. The approaches being used require handling user-generated data that is often not publicly available due to privacy concerns. Here, we propose an alternative defnition of weak and strong ties that requires knowledge of only the topology of the social network (such as who is a friend of whom on Facebook), relying on the fact that online key insights.
Article
An extensive literature shows that social relationships influence psychological well-being, but the underlying mechanisms remain unclear. We test predictions about online interactions and well-being made by theories of belongingness, relationship maintenance, relational investment, social support, and social comparison. An opt-in panel study of 1,910 Facebook users linked self-reported measures of well-being to counts of respondents' Facebook activities from server logs. Specific uses of the site were associated with improvements in well-being: Receiving targeted, composed communication from strong ties was associated with improvements in well-being while viewing friends' wide-audience broadcasts and receiving one-click feedback were not. These results suggest that people derive benefits from online communication, as long it comes from people they care about and has been tailored for them.
Article
A Sleeping Beauty (SB) in science refers to a paper whose importance is not recognized for several years after publication. Its citation history exhibits a long hibernation period followed by a sudden spike of popularity. Previous studies suggest a relative scarcity of SBs. The reliability of this conclusion is, however, heavily dependent on identification methods based on arbitrary threshold parameters for sleeping time and number of citations, applied to small or monodisciplinary bibliographic datasets. Here we present a systematic, large-scale, and multidisciplinary analysis of the SB phenomenon in science. We introduce a parameter-free measure that quantifies the extent to which a specific paper can be considered an SB. We apply our method to 22 million scientific papers published in all disciplines of natural and social sciences over a time span longer than a century. Our results reveal that the SB phenomenon is not exceptional. There is a continuous spectrum of delayed recognition where both the hibernation period and the awakening intensity are taken into account. Although many cases of SBs can be identified by looking at monodisciplinary bibliographic data, the SB phenomenon becomes much more apparent with the analysis of multidisciplinary datasets, where we can observe many examples of papers achieving delayed yet exceptional importance in disciplines different from those where they were originally published. Our analysis emphasizes a complex feature of citation dynamics that so far has received little attention, and also provides empirical evidence against the use of short-term citation metrics in the quantification of scientific impact.
Article
Emotions have ubiquitous effects in human affairs. Vivian Gornick, in Fierce Attachments,^ recounts a typ-ical exchange with her mother. Gor-nick always begins these encounters with high hopes. "Somehow," de-spite her best intentions, the conver-sations always spiral downward: Today is promising, tremendously prom-ising I go to meet my mother. I'm flying. Flying! I want to give her some of this shiningness bursting in me, siphon into her my immense happiness at being alive. Just because she is my oldest inti-mate and at this moment I love every-body, even her. "Oh, Ma! What a day I've had," I say. "Tell me," she says. "Do you have the rent this month?" "Ma, listen ..." I say. "That review you wrote for the Times," she says. "Ifs for sure they'll pay you?" "Ma, stop it. Let me tell you what I've been feeling," I say. "Why aren't you wearing something warmer?" she cries. "It's nearly winter." The space inside begins to shimmer. The walls collapse inward. I feel breath-less. Swallow slowly, I say to myself, slowly. To my mother I say, "You do know how to say the right thing at the right time. Ifs remarkable, this gift of yours. It quite takes my breath away." But she doesn't get it. She doesn't loiow I'm being ironic. Nor does she Elaine Hatfield is a Professor of Psychology and Richard L. Rapson is a Professor of History at the Uni-versity of Hawaii. John T. Ca-cioppo is a Professor of Psychology at the Ohio State University. Ad-dress correspondence to Elaine Hatfield, 2430 Campus Rd., Honolulu, HI 96822; BITNET: psych@uhunix; INTERNET: psych@uhunix.uhcc.hawaii.edu. know she's wiping me out. She doesn't know I take her anxiety personally, feel annihilated by her depression. How can she know this? She doesn't even know I'm there. Were I to tell her that it's death to me, her not knowing I'm there, she would stare at me out of her eyes crowd-ing up with puzzled desolation, this young girl of seventy-seven, and she would cry angrily, "You don't under-stand! You have never understood!" (pp. 103-104)