In Human-Compute Interaction: Psychology as a Science of Design, John Carroll argues that “HCI is the visible part of computer science” that is “a science of design” which “seeks to understand and support human beings interacting with and through technology.” (p. 62). In the article, Carroll walks one through the creation of and changes to the HCI field.

HCI started as Software Psychology, which “tended to focus on unrepresentative situations” (p. 63) where researchers often did not understand much about a system’s intended users. (p. 64). Software Psychology presented two problems: to better describe and understand how design could be supported, and to better specify what psychology, social and behavioral science should contribute to HCI. (p. 64). From here, the field transitions to iterative development, with the lesson that “designers should always ‘plan to throw one away.’” (p. 64). Dreyfuss took this idea further, with the following ideas: “(a) early prototyping with (b) the involvement of real users, (c) introduction of new functions through familiar “survival forms,” and (d) many cycles of design iteration.” (p. 65). Next came Card’s GOMS (Goals, Operators, Methods, and Selection rules), which “provided a framework for systematically analyzing the goals, methods, and actions that compromise human-computer interactions.” (p. 66). These models provided the foundation for understanding consistency, as well as the idea of an active user “improvising, hypothesizing, trying to make sense of a very complex environment.” (p. 67)

With the realization that the user is an important part of HCI, the field began to earnestly integrate humans into software development starting around the 1980s. At this time, usability engineering emerged, with three key notions: “iterative development can be managed according to explicit and measurable objectives, called ‘usability specifications,’” “a call to broaden the empirical scope of design” in which users assisted in developing prototypes as opposed to becoming involved after prototypes were already created, and cost effectiveness. (p. 68-70).  The 1990s brought design rationale, with Carroll and Rosson developing an approach “that considers systems to be ‘embodied’ social and behavioral claim about the needs, abilities, and activities of their users.” (p. 72). A prominent design rationale that most process-oriented approaches are based off of is the issue-based information system (IBIS), developed by Rittel. In IBIS, “design deliberations are described according to the issues that arise during the design process” as well as the pros and cons of positions raised in response to these issues. (p. 72).

As the interest in computing grew, HCI changed from being rooted in cognitive psychology to being assisted by sociologists and anthropologists, who urged “a stronger social and contextual orientation.” (p. 74). HCI had been spearheaded by the U.S. and Britain, but now was beginning to involve other countries. Activity theory, influenced by Vygotsky, proposed that computing is more of an activity system, or a tool by which people make themselves. (p. 75). This marked a shift from “characterizing static and individual competencies to characterizing how people can negotiate with the social and technological environment to solve problems and learn.” (p. 75).

Carroll predicts three technical themes for the 2000s: that the engineering scope of HCI will broaden beyond user interface interactions, HCI success in general applications will prompt more domain specific work, and HCI will continue to have an impact on the field of Psychology. (p. 77).