Full Blog: Media Equation

Reference Information
Title: Machines and Mindlessness: Social Responses to Computers
Authors: Clifford Nass, Youngme Moon
Published: Journal of Social Issues, 2000

Title: Computers are Social Actors
Authors: Clifford Nass, Jonathan Steuer, Ellen Tauber
When/Where: CHI 1994

Title: Can Computer Personalities Be Human Personalities?
Authors: Clifford Nass, Youngme Moon, BJ Fogg, Byron Reeves, Chris Dryer
When/Where: CHI 1995

Summary
These papers all had a very similar point that they made. Each one of them discussed similar experiments that related human-human interactions and the psychological theory associated with them to human-computer interactions.


The first paper discussed how people tend to have social responses such as politeness towards computers. The paper suggested that these social responses do not occur on the basis of thinking that computers have human attributed, but are actually due to mindless behavior that results in subconscious social reactions due to similar situations. Experiments were conducted that confirmed psychological theories applied to human-computer interaction in areas of social categories (gender roles, ethnicity, ingroup vs. outgroup, politeness, reciprocity, and self-disclosure), premature cognitive commitment (labeling of specialist vs. generalist), and personality. Finally, the paper listed possible alternative explanations for this behavior, such as ignorance, anthropomorphism, and orientation to the programmer, and provided reasons that these do not hold.

An example of the experiment setups. Source: Computers are Social Actors

 The second paper presented experimental evidence that people's interactions with computers follow psychological social rules that apply to human-human interactions, and are not the result of ignorance or social dysfunctions, much like the first paper. However, the method of reproducing these social interactions was discussed and applied to various situations, including politeness, self vs. other, gender roles, and computer vs. programmer vs. "I" context. Each of the social norms was proven through experimentation.

The third paper presented a study to find the minimum set of clues needed to create a computer-based personality and to show that it can elicit social responses from users. Dominance/submission aspects of personality were tested by assigning particular attributes to computers. The results confirmed all of the hypotheses that were based on psychological theory.


Discussion
These papers were very interesting to read. Most people generally attribute ignorance or anthropomorphism to acting socially towards computers, so it was very intriguing to discover that it is not the case. Reading three papers all discussing the same experiments got a little old, and some of the graphs presented in the papers seemed to be confusing and a bad choice of the type of graph, but the idea presented was good. It seems like knowing that many psychological theories apply to human-computer interaction could be applied to designing better programs. Since simple adjustments to programs, such as adding particular voices or assigning a dominant/submissive trait, can elicit social responses, it seems like all the effort to design mimicry of humans, such as AI and graphical representations, could be eliminated. These studies showed that much simpler programs can still create social responses.