Omega Point

Emotional cues in virtual spaces

While fre­quently used to great effect in prose, text is a noto­ri­ously poor medium for con­vey­ing the emo­tional meta­data humans rely on for face-to-face con­ver­sa­tion. How do we know exactly how to inter­pret some­one else’s words, stripped of their emo­tional con­text? What was intended as a sim­ple request for infor­ma­tion may be taken by one reader as a joke, while another may see it as a per­sonal attack.

The sys­tem used by many mod­ern inter­net users was pro­posed in the early 1980s by Scott Fahlman, a com­puter sci­en­tist with Carnegie Mel­lon Uni­ver­sity. He sug­gested that users employ a short series of char­ac­ters, evok­ing the iconic smi­ley face, to demon­strate that their words were to be taken lightheartedly: :-)

While at the time, the pro­posal was viewed by many as some­what tongue-in-cheek, the smi­ley quickly caught on, and is as rec­og­niz­able as the let­ters “www” today, demon­strat­ing its effec­tive­ness in clar­i­fy­ing human-to-human inter­ac­tion in text-based com­mu­ni­ca­tion. In the decades since, inter­net users have extended the orig­i­nal sys­tem by adopt­ing many other emoti­cons, con­vey­ing dis­plea­sure, sad­ness, dis­gust, exhaus­tion, and many oth­ers, insert­ing much-needed emo­tional con­text to their chat and email conversations.

Just as vir­tual envi­ron­ments like Sec­ond Life are fre­quently described as updated MUDs or cha­t­rooms, user inter­ac­tions within them can be sim­i­larly enhanced by the use of body lan­guage and ges­tures based on that of real-world humans. Con­sider the image of an avatar fac­ing another and smil­ing, look­ing away dis­in­ter­est­edly, or stand­ing with arms crossed; each con­veys a rad­i­cally dif­fer­ent mes­sage even when asso­ci­ated with the same text.

But what about cases in which we see avatars’ body lan­guage injected into our com­mu­ni­ca­tions with­out our explicit per­mis­sion? There have been count­less posts to the Sec­ond Life forums by newer users, angry and hurt by the dis­dain­ful, supe­rior man­ner of an estab­lished res­i­dent, and how they were delib­er­ately ignored.

These new users describe an inci­dent that usu­ally fol­lows a set pat­tern. They approached a Lin­den employee or an older res­i­dent, usu­ally a fairly high-profile con­tent cre­ator, and greeted them. The estab­lished res­i­dent turned to face them, looked down their nose, and turned back to what they were doing. In actual fact, this is a client-side avatar animation–when chat is “heard” on the client, avatars appear to turn their heads to face it with­out any input from the user con­trol­ling that avatar.

From the per­spec­tive of the Lin­den or long­time res­i­dent, they are unlikely to have even known any­one approached them, as they were busy doing some­thing else: pro­gram­ming, brows­ing the web, or work­ing on tex­tures, leav­ing behind a pup­pet with its strings cut.

This is an exam­ple of a “sub­con­scious” mes­sage injected to the com­mu­ni­ca­tions chan­nel. While no infor­ma­tion has delib­er­ately been con­veyed, to a human observer, a clear mes­sage has been sent. The body lan­guage of the avatar has effec­tively spo­ken for its user. Yet, to the recip­i­ent of this mes­sage, the avatar is the human. From their per­spec­tive, they’ve just been snubbed by some stand­off­ish per­son who clearly can’t be both­ered to even give them the time of day.

Next: Delib­er­ate sub­con­scious fil­ters and their implications.