Exploring the Power of Cute

By Dr. Susannah Paletz and Dr. Ewa Galonka

Shutterstock photo by Investor


Program managers and researchers at the Office of Naval Research (ONR) have supported and partnered with academic experts to explore the polarization of news. Our research team drew on the latest in the psychology of emotions to create a new methodology to examine the role of emotions in social media sharing.

The internet is a dark place, filled with hate speech, deliberate misinformation, trolls, and bots. After the multipronged Russian influence campaign during the 2016 US election, information warfare has intensified. In 2018, researchers at the University of Washington visualized how trolls at the Russian Internet Research Agency played both sides of the Black Lives Matter debate against each other, increasing polarization.1 With an increase in these types of subtle and not-so-subtle influence campaigns, it becomes even more pressing for researchers to understand the nature of malign influence.

The study of social influence and propaganda is not new. Recently, Lt. Col. David Beskow and Professor Kathleen Carley created the “BEND framework,” to describe communication tactics that might be used on social media (see page 22). Communication tactics support achieving desired, overarching communication objectives. Some typical tactics include distorting the truth and attempting to dismiss a topic. Other tactics include exciting a group, or raising discussions that bring joy and cheer, and distracting attention from a topic, person, or brand, which involves sidelining a conversation to something irrelevant. In the course of this research, it became clear that although the internet is filled with negativity, it also could be a positive experience. Cute messages on social media might serve to excite a group as well as distract—by spreading cute images, social media manipulators, be they genuine humans or bots, can change the conversation to something positive and innocuous.

Cute things have played a role in propaganda before. ISIS used cats and kittens in some of their recruitment materials, including having a soldier posed with a kitten in the 15th issue of Dabiq (ISIS’s official magazine).2 In addition to potentially softening their image, photographs such as these were probably a reference to a companion of the Prophet, Abu Huraira, who was fond of cats.3 Research on cuteness and its emotional reactions has a long history in the scientific literature: as early as 1943, Konrad Lorenz claimed that certain child-like features such as large eyes, a small nose, a round face, and/or playful behavior evoke caretaking behaviors by adults.4 In more recent studies, these features have been linked to cuteness.5 There is research on the characteristics of cuteness and how people are attracted to cute stimuli, as well as on the emotions cuteness elicits.6 The current challenge, however, is to develop ways to measure cuteness and the reactions cuteness evokes. Being able to quantify cute content and reactions to it is especially important when the messenger intends to appeal to specific audiences, such as in marketing and product advertising or in information warfare.

Bad actors using cute things in their messaging is a concern, particularly because social media cross national borders. ONR funded our social media studies to examine specific emotions that might promote social media sharing.7 Our research used a rating scale, which ranged from 0 (not present) to 100 (extremely intense), which included more than 20 emotions. Using a trained research assistant or scientist, each social media post was annotated. These annotators assessed each post independently, but then came together to debate each emotion assessment for each post to achieve consensus and improve validity. The annotators also recorded their reactions to the posts using the same list of emotions. Those reactions were averaged and assessed for breadth across the annotators, but not debated.

By using this process, researchers separated the effects of the content of a post from the reactions elicited by that post, such as when a post does not contain angry content but makes people who read it angry. This methodology was limited in terms of scale—annotation can be time-intensive and depends on human judgment—but it captured the full range of multimedia (e.g., photos, video, memes) on social media, not just text. It also covered a broader array of emotions than is typically examined in social media research. In addition to emotions such as anger, fear, disgust, surprise, and happiness, researchers also looked for gratitude, nostalgia, pride (including ethnonationalism), hate, contempt, love, admiration—and kama muta, a Sanskrit term usually translated as the emotional reaction of feeling heart-warmed when seeing cute, infant-like things.

To continue this research team’s advances in social media research, ONR awarded our team (which includes our colleagues Dr. Anton Rytting and Dr. Cody Buntain and researchers Egle Murauskaite and Devin Ellis) a grant through the Minerva Research Initiative to assess 1,000 Facebook posts and 300 YouTube videos from Polish and Lithuanian sociopolitical influencers. The goal of this research is to examine the role of emotion and narratives in social media sharing by political and social influencers in these two NATO member states.

By determining ways to measure cuteness and emotional reactions to cute stimuli, it is possible to then conduct statistical analyses on the impact of cute content and reactions on social media sharing. To prove this hypothesis, researchers conducted a pilot study in 2018 using Twitter to test the role of different emotions, including kama muta, on social media sharing. Our methodology was based on previous work on cuteness that focused on the specific characteristics of infants, or baby schema (such as appearance features such as small size, chubby cheeks) as well as child-like behavior (clumsy walking, playfulness, etc.). Researchers also adapted the concept of kama muta as a particular emotional reaction to cuteness.

As expected, researchers found that tweets containing cute images or cute behavior and tweets that evoked heartwarming responses were more likely to be shared.8 However, after controlling for relevant confounds (extraneous factors that are potentially associated with both cuteness and sharing, such as the amount of time between when a tweet was posted and when researchers downloaded the retweet), a simultaneous positive effect for the heartwarming feeling and negative effect of the cute content on sharing was found. This complex finding suggested that the heartwarming feeling is essential to sharing on social media. Specifically, when a tweet depicting or discussing something cute did not evoke kama muta in the reader (not all people perceive cuteness the same way), that tweet was less likely to lead to sharing.

Although intuitive, research that directly measures and tests the effect of cute content, or, more importantly, the heartwarming emotion that arises from viewing cute content, is relatively new. Recent research proves that while online content tends to favor negative emotions, it is possible for positive emotions, such as kama muta, to stifle content designed to divide.


1 A. Arif, L. G. Stewart, and K. Starbird, “Acting the part: examining information operations within #BlackLivesMatter discourse,” Proceedings of the ACM on Human-Computer Interaction 2 (2018): Article 20.

2 Lizzie Dearden, “Isis using kittens and honey bees in bid to soften image in Dabiq propaganda magazine,” Independent, August 2, 2016, https://www. independent.co.uk/news/world/middle-east/isis-kittens-honey-bees-dabiq-propaganda-recruits-photo-soften-image-terror-a7168586.html.

3 J. P. Farwell, “The media strategy of ISIS,” Survival 45 (2014): 49-55.

4 K. Lorenz, K. (1943). “Die angeborenen Formen möglicher Erfahrung,” Z. Tierpsychol 5 (1943): 233–519; cited in M. Borgi, I. Cogliati-Dezza,, V. Brelsford, K. Meints, and F. Cirulli, “Baby schema in human and animal faces induces cuteness perception and gaze allocation in children,” Frontiers in Psychology 5 (2014): 411.

5 M. L. Glocker, D. D. Langleben, K. Ruparel, J. W. Loughead, R. C. Gur, and N. Sachser, “Baby schema in infant faces induces cuteness perception and motivation for caretaking in adults,” Ethology 115 (2009): 257-63.

6 D. Jones, “Sexual selection, physical attractiveness, and facial neoteny: Cross-cultural evidence and implications,” Current Anthropology 36, no. 5 (1995): 723-48; K. K. Steinnes, J. K. Blomster, B. Seibt, J. H. Zickfeld, and A. P. Fiske, “Too cute for words: Cuteness evokes the heartwarming emotion of kama muta,” Frontiers in Psychology (2019).

7 S. B. F. Paletz, ed., Measuring emotions in social media: Examining the relationship between emotional content and propagation (College Park, MD: University of Maryland Center for Advanced Study of Language, 2018).

8 “Kama Muta Lab,” accessed November 5, 2019, http://kamamutalab.org/.

About the authors:

Dr. Paletz is a research professor at the University of Maryland’s College of Information Studies and an affiliate at the University of Maryland’s Applied Research Laboratory for Intelligence and Security (ARLIS).

Dr. Golonka is an associate research scientist at ARLIS.

About Future Force Staff