Learning the Internet and the structure of information behavior

Reviewer: William R. Patterson

Nahl reviews literature and research in the area of human-computer interaction. Data from 1992 through 1997 are examined across varying populations. She presents research conclusions and organizes them in categories concerning surveys, experiments, theory building, and the future. In citing her own previous work as part of this review paper, Nahl argues that “creating user-friendly environments requires more detailed knowledge of the information behaviors of users at both the affective and cognitive levels.” (The affective level supplies the direction for the cognitive.) A strength of this paper is its reporting on studies of both the affective and cognitive aspects of research behavior using the Internet (and on other human-computer interaction areas). Several studies are cited that demonstrate the interrelations of these aspects. Future development directions are suggested by descriptions of new notions of human-computer interfaces, for instance, that “computers can take the user's affect into account in order to adjust content, slow things down, or speed them up, tailoring style and humor to individual users.” New interfaces could use the fact that human motivation to learn can be sensed by the machine, which can then respond appropriately. The author shows where theory is developing around the research, particularly in prototype systems. One example is “affective wearables” (which I interpret to mean “smart clothing”). A weakness of the paper is that there is no attempt to organize or even to identify significant aspects of the various computer-based research tools that are its subject. They are merely placed in the very general category of Internet research tools. Nahl reports on the behavior of people using list servers, gopher, ftp, telnet, Archie, and Web browsers, but there is no discussion of the differences in user aspects among these tools and how they might relate to the various behaviors that are reported. Further, much of the paper, particularly as it relates to theory building, concerns the broader topic of human-computer interfaces rather than just those related to the Internet. Perhaps it should have been titled “A Review of Research on Human-Computer Interfaces.” A tabular display of research summaries would have helped.