Social Robots and Emotion: Transcending the Boundary Between Humans and ICTs

Abstract

Since the emergence of modern artificial intelligence research in the 1950s, scholars have striven to model the ways in which humans think, speak, move, and emote. A presumed measure of success is the degree to which computer-driven programs or mechanical constructs can reproduce (that is, simulate) behaviors of their human counterparts. With the development of robotic pets, comparable measures have been considered. The goal of simulation is based upon an assumption regarding the extent to which humans seek authenticity, more generally, in the representations they encounter. By “representation” we mean everything from an audio recording of a musical performance, to an architectural reproduction of an original edifice, to a sculpture depicting a person, to a robotic pet. This essay argues that authenticity in representation is not always a desired goal. As a result, as we think about the extent to which the robots of today (and tomorrow) are capable of expressing emotion and other behaviors that are judged to be life-like, it behooves us to consider the degree to which users of robots are actually seeking verisimilitude. KEY WORDS: artificial intelligence, authenticity, emotion, ICTs, representation, simulation Baron Lessons from Venice 8 OPENING PRELUDE: SOUNDS AND SIGHTS OF VENICE Chiesa di San Vidal stands in Venice, near the Accademia Bridge over the Grand Canal. Though no longer a functioning church, it is home to performances by a talented and vivacious chamber orchestra, Interpreti Veneziani, which specializes in playing music by Baroque composers. The group has recorded a number of CDs, making it possible to enjoy their music anywhere in the world. Yet however good the quality of the recordings, attending a live concert is a radically different aesthetic experience: The dynamics of the interaction between members of the group, their nuanced gestures, and the emotion on their faces render these performances “authentic” in a way that cannot be captured on an audio soundtrack. The issue of authenticity also comes into play with a better-known Venice landmark: the Campanile di San Marco, the 98.6 meter bell tower that stands adjacent to the Basilica di San Marco. The original Campanile, completed by 1514, collapsed in 1902, but was faithfully rebuilt in 1912. Authentic? Not literally, but the very best that could be done, even incorporating pieces from the original structure. Far less faithful is the version of the Campanile that is part of the Venetian Hotel in Las Vegas, which opened in 1999. A bit like the buildings in the national pavilions at Walt Disney World, the Venetian Hotel complex (including the Campanile) is more intended to give the flavor of Venice than to offer verisimilitude. Judging from the commercial success of the hotel, it would appear that guests are not put off by the lack of authenticity. What do issues of authenticity in Venetian music or architecture have to do with information and communication technologies (ICTs)? The answer is personal. A Spring 2011 visit to Venice inspired me to think about issues of authenticity in preparation for the Summer 2011 workshop on Social Robots and Emotions, hosted by Franklin College of Switzerland. The central question of this essay is: How important is authenticity when we are using ICTs, particularly as social robots? We will argue that just as in the Venetian examples – as well as in a host of other auditory, linguistic, or visual contexts – strict authenticity is often less relevant than filling particular aesthetic or interactive functions. THE QUESTION OF AUTHENTICITY To what extent does an artifact such as a musical recording, a building, a piece of art, or an artificial intelligence program constitute a “genuine” rendering of an original? This question contains two parts. The first concerns the extent to which the rendering is faithful. However, the second involves the degree to which the beholder (or the user) takes authenticity as a desirable goal. Bot