Work

What Do You Think I Think? Theory of Mind and Schizophrenia

Year: 2005
Type: article
Abstract: Multi-tasking and Cost Structure: Implications for Design David Kirsh ([email protected]) Department of Cognitive Science 0515 La Jolla, CA 92093-0515 state. This focus on task relevance is necessary if there is to be a principled basis for a metric evaluating the costs and benefits of actions. Environments are conceptualized as spaces where an agent tries to perform a task, and all that really matters are the features or properties of that space that can support or hinder task performance (Newell & Simon 1972). The role of technology and design can then be understood as shaping the cost benefit structures of activity in that space. One design will be better than another if it enables completing a task faster or with reduced error. Better designs have better cost structures. It may help to give an example of costing an environment before I critique the approach. My own favorite case of how technology has changed the cost structure of activity is the way remote control changed the activity of watching TV. The obvious activities to cost in TV environments are the efforts associated with turning the TV on and off, changing channels and changing volume. Other costs, such as those associated with discovering the time and channel of programs, or with controlling aspects such as muting, titling or color intensity we can treat as of secondary importance. Prior to the introduction of remote control the cost to change channels depended on the distance from seat to TV, the design of the channel switch on the television, and the speed a viewer can move. The greater the physical distance, slower the viewer, or the more channels, the greater the cost. A designer intent on improving the cost structure of watching TV would think about changing one or more parameters: moving the couch closer to the TV, adding random access to the channel dial. An even better solution would be to provide a hand sized controller that allows a viewer to control the TV from anywhere in the room without moving. In each case the objective of design is to lower the costs of some aspect of TV watching activity. Competing designs are evaluated by how much they lower costs. Since adaptation to a new cost structure is an equilibrium process it takes time for agent environment interactions to stabilize. But once a new equilibrium is reached the design process can start over again. In TV watching the next cost to be reduced was the previously non existent routine of channel surfing. Remotes soon began offering ‘last’ buttons to allow for quick toggling between channels to allow ‘timing’ the advertisement. More radically, TV’s themselves were changed by adding ‘picture in picture’ to support watching two channels at once. Again the force driving innovation was the desire to lower the cost of TV watching activities. Abstract I argue that it is not possible to accurately represent our task settings as closed environments with a single well defined cost structure. Natural environments are places where many things are done, often at the same time, and often by many people. To appreciate the way such invariants of everyday life affect design I present a case study, a micro-analysis of espresso making at Starbucks to show the challenges facing a cost structure approach. Introduction Almost all natural environments are places where many things are done, often simultaneously. How shall we conceptualize the structure of such environments? Designers interested in analyzing and re-engineering artifacts, layouts and settings, expect cognitive science to deliver representations and theories about agent- environment interaction that will help them understand real environments. But interruption, multi-tasking, and workspace sharing, pose special problems for most theories of action. The approach I shall evaluate is an extension of Herbert Simon’s seminal ideas about a science of design (1997). Simon argued that if we assume that agents’ make rational choices given their preferences and information, as well as cognitive and physical limitations, we can predict, at least in principle, how changes in an environment will lead to new equilibrium states of agent environment interaction. Although this is a radical extension of economic methods to non-economic environments – to activity spaces where agents respond to a universe of non-economic signals – it is one I personally find compelling and one which has driven a healthy research program in HCI. Card et al (1994), Russell (1994) and others since then have used this principle successfully to explain many of the regularities we find in the behavior of information seekers foraging for high value content in large information spaces, and to offer plausible accounts of why one application interface or one visualization of complex information is better than another (Kirsh 2005). It has also been used with success by ethologists and ecologists to explain regularities in the way animals behave given detailed descriptions of the niche they inhabit. The heart of the idea is that the resources in a setting define a particular cost structure for an activity. When applied to distinctively human settings resources need to include technologies and artifacts, as well as layouts, surfaces and other humanly modified physical features. To compute a cost structure for an activity a cost function is applied to each task relevant action at each task relevant
Language: English
Field: Medicine
Open Access status: green
OpenAlex ID: W2765412282
FWCI:
Cites: 18
Cited by: 2
Related to: 10