Home » Uncategorized » Prof. James R. Hamilton, ‘Spectating Animated Objects: The “Uncanny Valley” Puzzle’.

Prof. James R. Hamilton, ‘Spectating Animated Objects: The “Uncanny Valley” Puzzle’.

posted in: Uncategorized

Monday 12 May 2014, 6-8pm. Jarman 1, University of Kent

There are a number of puzzles about spectating animated objects. One puzzle is that it seems spectators should not, at least when talking “seriously,” attribute mental states to them. That is, it appears spectators simply should not ask of them, seriously, if they have a theory of mind. For example, when we see a Punch doll, say, we should not ask what Punch knows or believes about Judy’s mental states. Call this the ToM question. It seems we should not ask the ToM question. And yet we do, frequently and with little concern about what might be entailed by those attributions.

A second puzzle concerns how spectators ascribe agency in performances featuring animated objects, specifically to puppets and puppeteers, and whether they perceive the relationship between puppet and puppeteer differently from that between an animation and an animator (or team of animators), and a robot and its engineer(s) and programmer(s).

A third puzzle concerns how the information delivered in a performance involving animated objects, such as a standard Punch & Judy show, is actually processed in time by the performance’s spectators. Once attended to, spectators figure the behavior of animated objects into the stream of experiences that they are learning from and that they rely upon in order to determine what is happening in the performance. This is a bit vague, of course. How this happens is a deep puzzle that connects spectatorship of animated objects with our understanding of how any agents learn to adjust to changing environmental cues.

A fourth puzzle, and the one I address in this presentation, concerns the apparent fact (which, if true, should be troubling not only to puppeteers and scholars of puppetry but also to those working on human-robot interactions and video animations) that the closer to human features an animated object is—in movement, sound, and appearance—the more responsive the human participant in the exchange is until, that is, the so-called “uncanny valley” is reached, at which point human participants become frightened, distrustful, and even disgusted by the animated object. This puzzling phenomenon, if it were real, would be both a practical concern to makers of animated objects but also a theoretical concern for those who wish to understand how animated objects are experienced and understood by spectators.