Last thursday I visited the SIGCHI.nl 2004 conference in Amsterdam, the Netherlands. Main theme was: Dutch Directions in CHI. It was an interesting day which provided me with new ways to look at interaction design and user experience. Two different subjects, which I would like to share here now, caught my attention in particular.
1) Could computers (or: adaptive systems) ever get to understand our (emotional) world sufficiently to interact with us in a human way?
1) Keynotespeaker Bas Haring, Phd in Artificial Intelligence and professor at Leiden University in The Netherlands, explained why he thought computers could experience basic emotions and why he questioned whether computers could experience the world like humans do.
To illustrate that computers (or robots) could experience basic emotions, he used a hypothetical story about a small robot dog. He programmed this dog to run away from humans and trucks and to avoid hitting obstacles like walls or street lights. The dog was let to walk freely in town and all went well, until one day he decided to catch the robot dog to see how he was doing: a lot of work, for the dog was programmed to run away from him! At one moment he managed to hunt down the dog into a dead end street (with a wall). He came closer and closer until the dog was only one meter from the wall and one away from him….. the dog was programmed to avoid both obstacles: human and wall. What was he to do??? Slowly all circuits started to smoke and melt as confusion rose and rose in the little dog’s system… what was happening with it?…… it was scared!!
Ok, so possibly they could experience basic emotions. But really, could computers really get to understand the world like we do? Bas believed a person who was born in an isolated situation as in an aquarium, could never experience the world like someone who was able to ‘feel’ and experience things through the human body: we have many different shapes in chairs, nevertheless we know immediately it is a chair with each shape. Why? Because we know how to sit in one, we the movement and experience of sitting down with the shape. A person in an aquarium could only see the shape, without the reference of prior bodily experience with it.
Very interesting concepts! It made me interested in reading more about A.I. in the near future. Especially because I see (as did Norman in his book “Emotional Design”, chapter 7) the need to understand affective computing, as it is called, in order to see new opportunities in emotional design and adaptive systems/ products.
2) I had the chance to speak to Peter Merholz for a moment and tell him about this site, design-emotion.com. Peter is a nice guy and has a lot of experience in the field. His presentation as key-note speaker was interesting. Peter used various models of (user) research as inspiration for his story. Of those models, I liked the one taken from archaeological research the most. He illustrated some key factors, which are of importance for researchers to think of in developing a useful model for user research, with some interesting diagrammatic analyses in archaeology. You can also find this example in one of his earlier posts on peterme.com. Key factors were e.g. simplicity (which this example illustrated very well) and approachability (can someone ‘get it’ with a quick scan of the model?). To me, great inspirational work to get to easy-to-get models in emotional design (user) research.
Reading material recommended by Peter Merholz: “Design Research”, by Brenda Laurel.
Lecture on Emotional Design
For SIGCHI.nl I will give a lecture on Emotional Design (and interaction design) this fall (october, 2004) in the Netherlands. For a date, time and place I will keep you updated.