The study of Human Computer Interaction (HCI) is all of the rage these days. I was recently talking to some students from the Information School at UW the other day and found out that they even have a program specifically focusing on it now. When I was at the Information School, this program did not exist yet, though I do remember the curriculum having a focus on HCI, and I have even have brought up subject matter from the classes from time to time in casual conversations about technology. Since one could make the argument that the current bottleneck of the efficiency of technology is not necessarily how fast computers are as it has been in the past, but more-so how effectively interfaces can be designed to make users more efficient, a lot of HCI research studies on the nature of how people perceive computer interfaces are of prime importance now.
Of these studies one of the most interesting bodies of research I was introduced to at UW was the research and writings of Clifford Nass. Nass has done some amazing research by doing what one of my professors enthusiastically described as taking a psychology book and a black felt marker, crossing out every mention of “individual”, writing in “interface” instead, and testing whether it holds true. Interestingly, many psychological principles do hold true in the context of human-computer interaction, and there are even instances where taking the psychologist’s perspective can greatly simplify problems that drive computer scientists mad.
Something eluded to in many sci-fi shows, but is nonetheless, a true difficulty with technology is finding a way to make anthropomorphic interfaces have personality. Personality is complicated in the mind of the computer scientist. Just as Lieutenant Commander Data in Star Trek has difficulty understanding human emotion, computer scientists struggle to convey emotion in these types of interfaces. To the psychologist, however, personality is not as difficult; you are introverted or extroverted. Cognition, development, social attitudes, and countless other things are infinitely more complicated, and to convey personality, all one needs to do is to convey the introverted or extroverted nature of the interface in question, which can be done by changing the speed and pitch of voice. You can see this tested out in Nass’ Does Computer-Generated Speech Manifest Personality? An Experimental Test of Similarity-Attraction.
Another interesting psychological principle that also applies to interfaces has to do with human perception of consistency. According to several psychological theories, people generally find consistency more pleasant or trustworthy than inconsistency or asymmetry. Nass mentions Asch’s Gestalt theory and Kelley’s personal attribution model as examples of this in classical psychology. In applying this to the HCI context, Nass found that users preferred a computerized voice over a real recorded voice when the interface was synthetic, showing a preference for consistency. The hypothesis for the reason for this preference is that the same preference for consistency found in other areas of psychology also apply to human-computer interaction.
With the growing focus on HCI in the software industry, Nass’ research is extremely interesting, and I would recommend this reading to those who are interested in HCI. One may wonder what other things we can determine when looking at computer science from the perspective of another discipline. You can see many studies like those mentioned above on Nass’ Com 369 materials list.
References
Nass, C., Lee, K.M. Does Computer-Generated Speech Manifest Personality? An Experimental Test of Similarity-Attraction. Retrieved May 2012 from: http://www.stanford.edu/~nass/p329-nass.pdf
Gong, L., Nass C. (2000). Does Adding a Synthetic Face Always Enhance Speech Interfaces? Retrieved May 2012 from: http://www.stanford.edu/~nass/comm369/pdf/SyntheticFaceandTypeOfVoice.pdf
JeffMW - The Test Ranger Blogs 
Leave a Reply