IEEE Technology and Society Magazine - March 2020 - 64

Among the various types of
Internet addiction, addiction to social
media has particularly insidious
aspects that make it stand out.
views. This amplification of existing preferences leads to
polarized viewership and to filter bubbles. In response
to criticism, YouTube is attempting to curate sources to
provide reliable news, not only sensational and attention-grabbing content, following a similar trend Google
embarked on in 2017. As described in a Wall Street
Journal article, "if it's outrageous, it's contagious" -
extreme content leads to viral spreading, to longer
views, and to more advertising revenues [19]. Similar prediction software is used to curate and display the news
feed for Facebook users, with similar addictive results.
To maximize revenue for online companies, users
are programmed akin to B. F. Skinner's pigeons, who
were incessantly pressing levers to get rewards [20]. A
Google employee concerned with the ethics of designing such addictive behavior said that major tech companies, "have 100 of the smartest statisticians and
computer scientists, who went to top schools, whose
job it is to break your willpower." Just like the Skinner
pigeons, online users receive small rewards that are
infrequent and difficult to predict, a schedule that creates addiction and dependence.
Already in the early days of SM a study found that
among fifteen strongest human desires, the desire to use
online media is one of the hardest to resist. Although the
desire to use online media is among the weaker ones in
strength (not as strong as hunger, the need for sleep or the
desire for sex) access to online media is so easily available
and has such a low apparent opportunity cost that 42% of
users were unable to resist the desire to go online whenever they felt it. This inability to resist the desire is only comparable to that for the desire to work (43%). In
comparison, survey respondents were more likely to resist
other desires, even including the most primal ones [21].

Special Concerns about SM Addiction
Although the most troubling statistics above are those
about Internet use or Internet gaming, SM addiction is
also concerning. Roger McNamee, an early investor in
Google and Facebook, argues that a third of U.S. population is not willing to consider new ideas, including
demonstrable facts, and that people can be manipulated
via SM akin to "brain hacking" - as evidenced by the
2016 U.S. election and Brexit vote [22]. When addicted to

gambling or pornography, users are "only" manipulated
into wasting time and spending more money than they
intended. With SM addiction, users can be influenced
not only to purchase products, but to engage in more
complex activities as directed by the SM manipulators.
In 2014 Facebook researchers experimented with the
users' emotions by slightly biasing the users' news feed
to be either more positive or more negative, and finding
out, not surprisingly, that moods are contagious: users
exposed to more positive news tended to be in a more
positive mood and to post more positive items, while the
reverse was true for users exposed to more negative
news [23]. Aside from arousing public outrage, the experiment probably inspired the widespread use of Facebook
ads by Russian operatives during the 2016 electoral campaign [24]. The same idea fueled the infamous Pizza
Gate, when a 28-year-old man traveled from Salisbury,
NC, to Washington, DC, to confront what he was led to
believe was a child sex trafficking ring run by the Hilary
Clinton campaign out of the Comet Ping Pong pizzeria in
Washington, DC. Spurred by weeks of postings on SM,
the man entered the pizza restaurant armed, threatened
staff and patrons, and even shot his weapon inside the
establishment [25]. The main concern with SM addiction
above and beyond that for other types of online excesses
is that the extreme content fuels the addiction, and then
the addiction drives the extreme behavior in real life, be
it in the voting booth or in other spheres of life.
While these were experiments run either by Facebook staff or by manipulators outside, driven by political or economic agendas, things could get much worse
in the future with the increased use of artificial intelligence to curate and enhance online content. Already,
artificial intelligence algorithms are able to accurately
predict private traits and attributes (including sexual orientation, parental separation, political and religious
views, or substance abuse) from SM records [26], with
accuracies a bit higher than those of humans [27]. As
algorithms are designed to increase engagement and
stickiness of the site, if stirring up strong emotions in
viewers is the way to do so, artificial intelligence algorithms will likely curate and even create news that is
most likely to engage viewers. As users and their behavior coevolve with machine learning algorithms, it is likely that closely coupled ecosystems can arise, involving
complex behaviors and interactions.
Several such examples in nature are of one species
(parasite) taking over another species' brain and causing a series of complex actions intended only to benefit
the parasite, sometimes at the expense of the victim's
life [28]. For example, Toxoplasma gondii is a parasite
that has evolved to reproduce in cats, but that can infect
a range of other mammals. When the parasite infects
mice, the mice lose their fear of predators and may

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

∕

MARCH 2020

IEEE Technology and Society Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - March 2020