Experience in online networks- some massive

For one week in 2012 the emotional content of Newsfeeds from 689,003 Facebook users was manipulated; by reducing the number of positive expressions, or the number negative expressions. The findings were reported in PNAS http://www.pnas.org/content/111/24/8788.short?rss=1&ssource=mfr

The publication resulted in a huge outcry about the ethics of Facebook manipulating Newsfeeds in this way - see  here  and here  and comments posted directly to PNAS, as to whether Facebook users actually  give  'informed' consent for this kind of experiment and what the consequences for individuals could be. It is a debate worth having. However it seems that it was at the expense of a critical look at the authors ( A.D. Kramer, J.E. Guillory, and J.T. Hancock) claim about 'emotional contagion on a massive scale'.  From the perspective of  someone interested in social learning at a distance using computer mediated communication the article missed some key points about the emotional experience of  'written' interpersonal interaction .  So I wrote a letter to PNAS but it will not be published (yes it was rejected).  Neverthess here it is.

The outcome of the manipulation was as follows;  a reduction of positive expressions produced fewer positive posts and more negative posts with the opposite pattern occurring when negative expressions were manipulated. The authors claim that these findings show that emotional contagion can be elicited in the absence of nonverbal cues and that given the link between emotion and well being the finding could be important for public health. There are implications in another domain, education, specifically for social learning online.  In higher education an increasing number of universities and MOOC platforms are providing facilities for students to communicate with each other remotely using social media, e.g. text based forums.  The opportunity for students to  ‘learn by discussion’, to collaborate with each other on a shared task, ‘learning by collaboration’, are extremely effective ways of achieving depth of understanding (3) and also equip students for the teamwork skills required by modern organizations. However, the emotional valence of the experience is an important factor.  Reports that social interaction online is less satisfactory and experientially ‘uncanny’ (1,4) is why the findings of (2) are relevant.  Interpersonal interaction defines social learning however (2) claim that for their study contagion and social interaction are disentangled.

Two constructs are challenged: the conceptualization of emotional contagion and mimicry, assumptions about ‘direct interaction’. A taxonomy, on which to base the definitional nuance of the emotional experience that results when others share one’s emotional experience: mimicry, contagion, personal distress, affective empathy, cognition empathy, sympathy, and the criteria that distinguish each, has been formulated (5). Mimicry and emotional contagion are both characterized by affective behavior while emotional contagion involves affective experience and isomorphism. Although affective behavior was elicited the authors (based on the cross-emotional encouragement effect) concluded that mimicry, as an explanation, was not sufficient. However claiming contagion is not substantiated; they do not have any data to confirm that affective experience and affective isomorphy occurred.

That ‘emotional contagion occurs without direct interaction’ is the significance claim. However, conceptually ‘direct interaction’ is underdeveloped.  ‘Direct interaction’, as when individuals meet face-to-face, involves two shared aspects, time frame and place. Asynchronous interaction means conversational turns may be disrupted and response is usually delayed, while interacting in an online space means that multisensory information is not available and real time context is not shared. The specification of all these factors is crucial when evaluating emotion communicated online through writing.  For Facebook, a written communication from ‘friends’ will sometimes results in  ‘friends’ moderating their responses (the data that the authors collected).  That is the essence of human communication: the use of written language and metacognition. The usefulness of the study lies in the foregrounding the power of language. However, the value of this approach, theoretical and practical, will depend on substantial integration with research in two fields, emotion and computer mediated communication.

(1) BAYNE, S. (2008) Uncanny spaces for higher education: Teaching and learning in virtual worlds. ALT-J  Research in Learning Technology, 16, 197-205.

(2) KRAMER, A. D. I., GUILLORY, J. E. & HANCOCK, J. T. (2014) Experimental evidence of massive-scale emotional contagion through social networks. Proceedings of the National Academy of Sciences (PNAS), 111, 8788-8790.

(3) LAURILLARD, D. (2012) Teaching as a design science. Building pedagogical patterns for learning and technology., New York, London, Routledge.

(4) ROBINSON, K. (2013) The interrelationship of emotion and cognition when students undertake collaborative group work online: An interdiscplinary approach. Computers & Education, 62, 298-307.

(5) WALTER, H. (2011) Social Cognitive Neuroscience of Empathy: Concepts, Circuits and Genes. Emotion Review, 4, 9-23.