upa - home page JUS - Journal of usability studies
An international peer-reviewed journal

Switching Between Tools in Complex Applications

Will Schroeder

Journal of Usability Studies, Volume 3, Issue 4, August 2008, pp. 173-188

Article Contents


Planning and switching

According to activity theory, an expert beginning an operation initiates a preset chain of actions. The operation's internalized chain of action is a form of planning. Contradictions result when these plans fail. In activity theory terms a contradiction signaling a breakdown or focus shift should occur at any point where the tool is not "doing its best." The contradiction alerts the user that the current tool or its usage is (at the moment, at least) suboptimal. The user must then decide whether to change tools or persevere.

These plans (chains of actions) may develop from experience (trial and error) or they may result from training, or simply arise from users striving to implement best practices. The result is the same. Actions proceed without evaluation until a contradiction intrudes.

The importance of tool switching

As soon as task success includes an aesthetic evaluation or quality goal, Kuutti's definition of success as "error-free" must be replaced by "optimal" or at least "satisfactory." Error-free is not enough. Steps must be of adequate quality, performed at an acceptable rate.

Having assembled actions into an operation that may be carried out without considering what to do next, the expert must remain aware of the quality of the work and the rate of progress in order to detect problems and act on them. These problems include not only contradictions, but also inefficiency from using the wrong tool or missing information. Any or all of these problems may require modification of the sequence of actions or a search for a more effective tool. It is precisely in this area that theories clash. Suchman warns that the planned selection of actions of an expert's operation do not account for choices and decisions made in specific situations that advance the work, but that they result from the user's engagement with the task(s) and could not be planned in detail in advance. Activity theory holds that operations disintegrate into actions separated by evaluation and choice when the operation encounters a contradiction, problem, or difficulty.

Unfortunately, "expert users, when asked, cannot report reliably about such cases of breakdown of their expectations, because they happen in the unconscious middle phase of an action" (Raeithel & Velichovsky, 1996, p. 201). In other words, the expert is unaware of any monitoring process until the operation breaks down, so the expert feedback reports only on the problem solving activity that it collapses into. We have no cognitive description of "checking for breakdown." As a practical matter we are left with the following:

Tool switching is the only contradiction-generated events that can be reliably detected. We know that a contradiction has been considered and a choice made.

Contradictions and interruptions

Activity theory is a theory of learning. Contradictions "are not the same as problems or conflicts. Contradictions are historically accumulating structural tensions within and between activity systems" (Engestrom, 1999, p. 3). For our purposes it is not important, at least initially, to distinguish between contradictions whose resolution generates a new form of operation and interruptions that cause the user to notice that the workflow has bogged down. Our interest begins when any event initiates consideration of a tool change tactic.

Interruptions have received more concrete study than contradictions. Mark, Gonzales, and Harris (2005) call interruptions from within the working sphere "interactions," those from outside the working sphere are "disruptions" that reflects a negative impact on task progress. Adamczyk and Bailey (2004) show that the cognitive impact of interruption varies according to which stage of the task (beginning, middle, end) it occurs. Cutrell, Czerwinski, and Horvitz (2000) attribute more impact to interruption of chunking behaviors (highly focused subtasks) than to interruptions happening between them. Speier, Valacich, and Vessey (1997) demonstrate that the disruptive impact of an interruption increases with the complexity of the task. These definitions and characterizations are useful in describing what is observed in the tool switching process.

Previous | Next