Systems, Man & Cybernetics - April 2015 - 15

that speak louder tend to be perThe experiments presented
ceived as more extraverted [35]).
in [28] address this question, and
SSP fills a gap
the results suggest that pictures
left open by other
play a role in impression manageSocial Physics
ment among the users of Flickr,
Are there signals generated by
machine intelligence
one of the most popular photogroups of people as well as indidomains dealing
sharing platforms on the Web. In
viduals? During the last decade,
[28], a pool of raters were asked
the SSP research community has
with people involved
to assess the big-five personaldeveloped the ability to analyze
in colocated social
ity traits (five major dimensions
the signals of entire social organinteractions.
known to capture most individual
isms, groups, companies, and
differences) of 300 Flickr users
whole communities on a millisecbased on the images the users
ond-by-millisecond basis for up to
tagged as favorites, i.e., as particuyears at a time, an implementation
larly appealing from an aesthetic and/or affective point
of what is known as a living lab [36], [37]. The method
of view. The setting reflects the situation depicted in Figis simple. Measurements are made by collecting digital
ureĀ 2, where Flickr users interact with others through the
breadcrumbs, such as from the sensors in cell phones,
photo-sharing platform, and, therefore, the pictures are
postings on social media, and purchases with credit
the only social signals that they have at their disposal.
cards. By combining these fine-grain, objective measureThe results show that the personality scores attributed
ments with traditional surveys and other social science
to a given user can be inferred with significant accuracy,
tools, it is possible to provide both richer context and
with a correlation up to 0.6 between the actual and pregreater detail to the traditional measurement techniques.
dicted scores, from low-level, content-independent feaIn many cases, this allows for the accurate prediction of
tures extracted automatically from the favorite images
effect sizes even in complex, natural circumstances. It is
(e.g., statistics about the distribution of colors or textural
also possible to estimate subjective biases and compare
properties). Therefore, users' favorite pictures convey an
socially constructed reality to objective reality. We can
impression, and the statistical relationship between lowtrack how even the most apparently insignificant signals
level, visual characteristics and attributed personality
help shape social phenomena at the scale of communities
traits is solid enough to allow one to perform automatic
as large as entire cities.
prediction. The visual characteristics of users' favorite
Using this methodology, it was possible to find, for
pictures covariate with the users' personality traits (e.g.,
instance, that there are behavioral markers that signal
people that tend to use brighter colors tend to be perthe impending onset of flu [38], a fact that is now being
ceived as more extraverted [28]) in the same way that
used commercially by Ginger.io (http://ginger.io). It was
nonverbal cues do in face-to-face interactions (e.g., people
also found that certain aggregate changes in behavior, e.g.,

Pictures

Functions:

Psychological Level

Impression
Management
e nt
Emotion
Expression
Sending Relational
Messages
M
Deceiving and
Deception
Detection
D
ece ti D
Social Ve
V rticality
ityy
Expression
E

Figure 2. Pictures as social signals: in the case of social media, pictures might act as social signals, i.e., they

might allow people to perform the functions necessary for social interaction (at least impression management
according to the experiments presented in [28]). The functions that have been investigated only in face-to-face
settings are shaded.

Ap ri l 2015

IEEE Systems, Man, & Cybernetics Magazine

15


http://www.Ginger.io http://www.ginger.io

Table of Contents for the Digital Edition of Systems, Man & Cybernetics - April 2015

Systems, Man & Cybernetics - April 2015 - Cover1
Systems, Man & Cybernetics - April 2015 - Cover2
Systems, Man & Cybernetics - April 2015 - 1
Systems, Man & Cybernetics - April 2015 - 2
Systems, Man & Cybernetics - April 2015 - 3
Systems, Man & Cybernetics - April 2015 - 4
Systems, Man & Cybernetics - April 2015 - 5
Systems, Man & Cybernetics - April 2015 - 6
Systems, Man & Cybernetics - April 2015 - 7
Systems, Man & Cybernetics - April 2015 - 8
Systems, Man & Cybernetics - April 2015 - 9
Systems, Man & Cybernetics - April 2015 - 10
Systems, Man & Cybernetics - April 2015 - 11
Systems, Man & Cybernetics - April 2015 - 12
Systems, Man & Cybernetics - April 2015 - 13
Systems, Man & Cybernetics - April 2015 - 14
Systems, Man & Cybernetics - April 2015 - 15
Systems, Man & Cybernetics - April 2015 - 16
Systems, Man & Cybernetics - April 2015 - 17
Systems, Man & Cybernetics - April 2015 - 18
Systems, Man & Cybernetics - April 2015 - 19
Systems, Man & Cybernetics - April 2015 - 20
Systems, Man & Cybernetics - April 2015 - 21
Systems, Man & Cybernetics - April 2015 - 22
Systems, Man & Cybernetics - April 2015 - 23
Systems, Man & Cybernetics - April 2015 - 24
Systems, Man & Cybernetics - April 2015 - 25
Systems, Man & Cybernetics - April 2015 - 26
Systems, Man & Cybernetics - April 2015 - 27
Systems, Man & Cybernetics - April 2015 - 28
Systems, Man & Cybernetics - April 2015 - 29
Systems, Man & Cybernetics - April 2015 - 30
Systems, Man & Cybernetics - April 2015 - 31
Systems, Man & Cybernetics - April 2015 - 32
Systems, Man & Cybernetics - April 2015 - 33
Systems, Man & Cybernetics - April 2015 - 34
Systems, Man & Cybernetics - April 2015 - 35
Systems, Man & Cybernetics - April 2015 - 36
Systems, Man & Cybernetics - April 2015 - 37
Systems, Man & Cybernetics - April 2015 - 38
Systems, Man & Cybernetics - April 2015 - 39
Systems, Man & Cybernetics - April 2015 - 40
Systems, Man & Cybernetics - April 2015 - 41
Systems, Man & Cybernetics - April 2015 - 42
Systems, Man & Cybernetics - April 2015 - 43
Systems, Man & Cybernetics - April 2015 - 44
Systems, Man & Cybernetics - April 2015 - 45
Systems, Man & Cybernetics - April 2015 - 46
Systems, Man & Cybernetics - April 2015 - 47
Systems, Man & Cybernetics - April 2015 - 48
Systems, Man & Cybernetics - April 2015 - Cover3
Systems, Man & Cybernetics - April 2015 - Cover4
https://www.nxtbook.com/nxtbooks/ieee/smc_202110
https://www.nxtbook.com/nxtbooks/ieee/smc_202107
https://www.nxtbook.com/nxtbooks/ieee/smc_202104
https://www.nxtbook.com/nxtbooks/ieee/smc_202101
https://www.nxtbook.com/nxtbooks/ieee/smc_202010
https://www.nxtbook.com/nxtbooks/ieee/smc_202007
https://www.nxtbook.com/nxtbooks/ieee/smc_202004
https://www.nxtbook.com/nxtbooks/ieee/smc_202001
https://www.nxtbook.com/nxtbooks/ieee/smc_201910
https://www.nxtbook.com/nxtbooks/ieee/smc_201907
https://www.nxtbook.com/nxtbooks/ieee/smc_201904
https://www.nxtbook.com/nxtbooks/ieee/smc_201901
https://www.nxtbook.com/nxtbooks/ieee/smc_201810
https://www.nxtbook.com/nxtbooks/ieee/smc_201807
https://www.nxtbook.com/nxtbooks/ieee/smc_201804
https://www.nxtbook.com/nxtbooks/ieee/smc_201801
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1017
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0717
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0417
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0117
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1016
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0716
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0416
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0116
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_1015
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0715
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0415
https://www.nxtbook.com/nxtbooks/ieee/systems_man_cybernetics_0115
https://www.nxtbookmedia.com