A Case Study of the Cognitive Task Analysis Approach and Collaborative Information Seeking
To the best of our knowledge, there has not been a CTA focused on specific CIS related tasks. Therefore, for this case study, we will highlight similar work that focuses on the utilization of a cognitive work analysis (CWA) framework to better understand CIR (a context closely related to CIS). Once, we have reviewed this work, we will then present a specific CIS task and explain how the methodology presented by Crandall, Klein, and Hoffman can be applied.
A cognitive work analysis is extremely similar to a CTA in that the goals of each are to understand the cognition associated with adequately performing work. The cognitive work analysis framework is an approach that uses multiple methods to better understand human behaviors relating to work and cognition in context (Vicente 1999; Rasmussen et al. 1994). In general, a CTA is more focused than a CWA, which seeks to typically understand the larger scope of work, not just one specific task. A CWA also does not lend itself as well to specifically understanding the relationship of cognition to performance. A CTA directly seeks to understand how cognition impacts performance, whereas CWAs are typically aligned to understanding cognitive motivations, not examining how cognition actually affects task performance. This is not to discourage researchers from using CWAs to study the cognitive aspects of CIS. We feel that both CTAs and CWAs have value in further developing our understanding of cognition during CIS. The researcher will need to review the research context and the goals of the study to decide which approach to utilize.
CWAs are often employed within the information sciences community, where they are used to develop a better understanding into the design and evaluation of information systems (Fidel and Pejtersen 2004). For this reason, related areas of CIS, specifically CIR, have been studied using this methodological approach. CIS activities are becoming increasingly dependent on information systems and collaborative search platforms. If a CIS researcher is focused on understanding technical development, then a CWA might be helpful to develop new technical insights. Research by Bruce and colleagues (2003) explored CIR activities of two design teams through a CWA oriented field study. The study found that the concept of CIR is ambiguous—not lending well to a single definition that CIR is tied directly to context, and that CIR activities are not exclusive to collaborative activities. In addition, work by Fidel and colleagues (2004) employed a CWA to better understand why design engineers participate in CIR-based activities. Through a field study using observation and interviews, the researchers identified that engineers utilize CIR when they are new to an organization, when information is ambiguous, or when information is not documented (Fidel et al. 2004).
These two studies are beneficial in understanding the cognitive aspects of collaborative information behaviors that occur within the engineering context. Unfortunately, these types of studies are lacking, with very few CWAs taking place with the focus on understanding the cognitive aspects of collaborative information behaviors (with no CTAs taking place). Below we will briefly explain how to use a CTA during the specific task of intelligence analysts collaboratively seeking information. The intelligence context has long been defined by individual work, but collaborative work is becoming more apparent (McNeese et al. 2015).
First, when conducting any CTA, it is extremely important that the researchers understand the specific cognitively oriented task that they are studying. In the proposed setting, the researcher needs to focus their methods on the specific task of collaboratively seeking and sharing information during the analysis of an intelligence problem. This is difficult to achieve, as there are many other activities occurring within the greater scope of work. In order to develop this focus, we suggest the researcher actually observes all aspects of the environment. By doing this (possibly multiple times), the researcher will be able to identify specific instances of where and when CIS is occurring either directly or indirectly. With this knowledge, the researcher can then focus their KE methods to these task specific instances of CIS. This will help determine which of the many KE methods the researcher should employ. At a minimum, for this specific task and context, the researcher needs to observe the task (intelligence analysts collaborating to find and/or share information), conduct both individual and team level interviews aimed at understanding the cognitive activities that occur during the CIS task, and conduct at least one of the process tracing or conceptual techniques (at the individual and team level) previously outlined. The analysis of the data collected by these methods can be done using multiple different analytical techniques, ranging from thematic to statistical analysis. Finally, after the analysis, knowledge representation should occur to report data findings and to improve cognition associated with the task. We encourage researchers to not only view knowledge representation as a means for reporting data, but as a way to develop new recommendations/workflows/technologies to help improve cognition. For example, in this context, a technology that better allows analysts to share or view information in a collaborative manner may be developed based on the overall findings of the CTA.
