| III. |
Cognitive
and Behavioral Processes. These approaches "describe" how people actually perform work tasks, so they are generally less formal than the "models" considered in Sections I and II. The methods examine how workers use the tools they currently have available to them to perform tasks in the work domain. Typically, these methods are used when there is already a system (or at least a prototype of a system) in place that is to be evaluated and improved. We also include system evaluation methods here, which are methods used to evaluate the utility, usability, and learnability of a system design. |
| |
|
|
| |
A. |
System
Evaluation Methods. Methods used to evaluate how workers interact with existing or proposed systems. They aim to assess how easy a particular system is to learn and use, and how well the system supports the tasks that worker performs. Although these techniques cannot be used to build a new system from scratch, they should be applied sooner rather than later in the design process when deficiencies with a proposed system are easier to correct. They are typically used in an iterative fashion to test and refine a proposed system design, evolving it from a prototype to a final design. The final design may be reached when certain usability requirements are met (e. g. when workers are able to complete task x in under 5 seconds). Prototyping and testing are integral parts of Cognitive Engineering. New systems change the nature of the cognitive tasks workers perform. It is usually not possible to anticipate the impacts that these changes will have on all aspects of human performance. Thus, it is necessary to iterate and refine designs with actual workers.
|
| |
|
1. |
Heuristic Evaluation: A
technique for assessing the usability of a computer interface that uses
ten rules of thumb, such as "speak the user’s language",
"provide feedback", "be consistent", and
"provide good error messages." In a heuristic evaluation, the
analyst evaluates how well the proposed interface follows the rules of
thumb and provides feedback as to how it could be improved.
References
|
| |
|
2. |
Walk-throughs/Cognitive
Walk-throughs/Talk-throughs: In walk-throughs and talk-throughs, workers who know a system perform a task using an actual system or a realistic mock-up for analysis. When performing a talk-through, the user is removed from realistic surroundings and merely verbalizes the demonstration. Cognitive walk-throughs attempt to evaluate the state of the user's thought processes at each step of task performance, with emphasis on identifying aspects of the interface that are confusing.
References
|
| |
|
3. |
Formal Usability Studies:
Workers are observed or videotaped while they perform tasks using a proposed system in a controlled environment. They may also be instructed to think-aloud when performing the tasks. By observing workers performing the same tasks under such controlled conditions, it is possible to identify aspects of the interface that require improvement.
References
|
| |
|
4. |
Rapid Prototyping:
A prototype of a proposed system is presented to workers for critical comments. Revisions are made to the original prototype, producing a second version that is again presented to users for critical analysis. The process of revising and submitting to users continues until some criterion for acceptability is reached.
References
|
| |
|
5. |
Storyboarding:
Storyboards are nonfunctional pictures or display screens of candidate design concepts and how they would operate in a simulated scenario. Storyboarding enables system developers to receive user input and feedback early in the concept development phase.
References
|
| |
|
6. |
Interface Evaluation Surveys:
These are a group of information collection methods which can be used to identify specific ergonomics problems or deficiencies in interfaces. They address issues such as the labeling and consistency of controls, how well the system works within its environment (e.g. is the environment too noisy for an auditory interface?), and whether operators have modified the system in some way to overcome a deficiency (e.g. are there post-it notes everywhere?). They are applied when a detailed design has been created. There are also a host of surveys, such as the Questionnaire for User Interaction Satisfaction (QUIS), which can be used to assess worker satisfaction with specific aspects of a human-computer interface.
References
|
| |
|
7. |
Ergonomics Checklists:
Checklists that an analyst can use to ascertain whether particular ergonomic criteria are being met by a system. The items within these checklists can range from overall subjective opinions to very specific objective checks. They can be used to evaluate both existing and proposed systems.
References
|
| |
|
8. |
Contextual Inquiry:
Contextual Inquiry is essentially a set of structured field interview methods that are used to collect detailed information about how people work, the context in which the work takes place, which tools are more important and less important, and which design changes to existing tools would help make their work easier.
References
|
| |
|
|
|
| |
|
Theoretical
Frameworks. These are theories about how people perform cognitive work. While there are a host of theoretical frameworks, those discussed here are most relevant to Cognitive Engineering. They can provide focus for conducting knowledge elicitation by positing the important aspects of human-technology interaction that one should take into account.
|
| |
|
1. |
Activity Theory:
Activity theory is a psychological theory of human practices as developmental processes, which has begun to receive increased attention in the HCI community. The fundamental unit of analysis is the human activity, which is directed toward an object, mediated by artifacts (i.e. problem solving tools), and occurs within a social context. Activity theory focuses primarily on the goals workers are trying to satisfy with artifacts. The approach has been used to analyze how workers achieve task goals with an existing system to develop requirements specifications and prototypes for future systems.
References
|
| |
|
2. |
Situated Cognition:
An approach to studying cognition that emphasizes the fact that cognition must be studied in its natural context. The approach is based on the fact that the same set of actions will have different effects at different times. Each situation is a unique context, and, as a result of the situated context, a different set of actions will be required to achieve the same task goal. The approach has primarily been used by researchers studying cognition, but it could be similarly applied to system design as the distributed cognition approach.
References
|
| |
|
3. |
Distributed Cognition:
A hybrid approach to studying cognition developed by Hutchins that attempts to understand how cognitive phenomena are distributed across individuals and artifacts (i.e. tools for solving problems, such as a computer system). It emphasizes the fact that cognition does not lie strictly within the individual, but instead is an emergent, distributed activity, performed by people with tools, within the context of a team or organization, in an evolved cultural context. The approach has been used to study ship navigation, cognition in the cockpit, air traffic control, and engineering processes. In terms of system design, the approach has potential for being able to identify problems with existing work practices and use of technologies, and for highlighting what is important in existing system designs that needs to be retained in future systems.
References
|
| |
|
4. |
Naturalistic Decision Making (NDM): The study of how people make judgements and decisions in environments that have high-stakes, multiple players, ill-defined goals, are uncertain and dynamic, and time pressured. A central tenet of NDM is Klein's Recognition Primed Decision Model (RPD), which asserts that experts assess the current situation and draw on their experience to relate it to previous situations and successful courses of action.
References
|