Successful attempts to explain expertise in human beings, or to capture its properties in expert systems, will have to contend with issues of rationality and generalization. Rationality and generalization pose enough difficulties on a purely synchronic basis. But an account of expertise must be diachronic—it must account for the development of rationality and generalization, even in those who are already experts. We describe the obstacles in the path of standard approaches to rationality and generalization, and present an alternative, interactivist treatment of rationality and its development (space forbids us to do likewise for generalization). In the interactivist account, rationality cannot be defined in general as adherence to the rules of a system of formal logic; we propose instead that rationality be understood in terms of the development of negative knowledge—knowing what kinds of errors to avoid. We examine the development of negative knowledge using examples from the history of science, and consider the consequences of an orientation towards negative knowledge for classroom instruction as well as the development of expert systems.

The empirical study of human expertise has grown up largely within the information-processing framework. We contend that assumptions and practices inherited from that framework have significantly hampered the study of expertise. Most studies have been conducted within a binary paradigm that sets novices apart from experts while ignoring any qualitative distinctions in between. End-state description has been preferred over accounts of learning; accounts of learning as a smoothly continuous process have been preferred over accounts of qualitative change. To overcome the binary paradigm, psychologists need to adopt an explicitly developmental approach, one that identifies multiple way stations toward expertise and seeks to explain qualitative changes in knowledge and motivation. We consider two insufficiently developmental conceptions (‘theory-change’ accounts and global stage models) and two others that, in our view, harbour greater promise (the Piagetian framework as elaborated by Feldman and the Vygotskian framework as extended by Scribner). Progress in the study of human expertise doesn't just depend on the more widespread adoption of developmental research programmes. There also needs to be more open and frank discussion of the strengths and weaknesses of different frameworks than has been customary.

In order to understand long-term learning and the acquisition of expertise, human-computer interaction needs to take a developmental turn. Adopting a developmental approach means using longitudinal research methods, building developmental sequence models of the acquisition of expertise, and analyzing tasks as scenarios specific to developmental levels. The psychology of programming seems particularly amenable to a developmental approach because of the length of time that it takes to become an expert. We propose a model of seven developmental levels for Smalltalk/V programming, and provide sample scenarios for each level. We conclude that developmentally ordered scenarios convey valuable design information that would be lost in the standard “average user” approach to scenarios.

This panel session examines major issues in the design and implementation of online help systems.

A source of intellectual overhead periodically encountered by scientists is the call to be "hard," to ensure good science by imposing severe methodological strictures. Newel1 and Card (1985) undertook to impose such strictures on the psychology of human-computer interaction. Although their discussion contributes to theoretical debate in human-computer interaction by setting a reference point, their specific argument fails. Their program is unmotivated, is severely limited, and suffers from these limitations in principle. A top priority for the psychology of human-computer interaction should be the articulation of an alternative explanatory program, one that takes as its starting point the need to understand the real problems involved in providing better computer tools for people to use.