Using the Decision-Ladder to add a Formative Element to Naturalistic Decision Making Research
Abstract
Background and reasoning behind the work
This report is submitted under HFI DTC work package 4.6.1, exploring requirements analysis for decision making training. This report is one of a number of reports submitted under the same work package.
The primary reason for the approach developed and described in this report, is in direct response to the complexity of the sociotechnical domains that typify military systems. These complicated, dynamic domains are frequently unpredictable; thus, there is perceived benefit in an approach that models the decision making activity in terms of how actors can proceed through the process rather than how they actually do or should.
It is believed that a more holistic approach is important for the conceptualisation, analysis and study of these domains. Further, it is contended that such an approach is essential in considering the introduction of new combat identification (combat ID) technology and its implications for training and operating procedures.
Specific research question being addressed
The remit of this report is to develop a model of decision making that, like current approaches to Naturalistic Decision Making (NDM), accounts for the relevance of context and expertise. Unlike current NDM approaches, the described approach aims to create a prototypical model that is less constrained to a given time, location and group of actors. This initial report describes such an approach, demonstrating its applicability to military domains. Future work, based upon this approach, is anticipated to investigate the value of the model in investigating decision support systems.
This report considers decision making in-line with theories of macrocognition (Crandall et al, 2006; Klein, 2000; Schraagen et al, 2008). According to Schraagen et al (2008), in contrast to microcognition, which attempts to provide a reductionist, causal-chain account of behaviour, macrocognition seeks to maintain a focus on the phenomena themselves. In short, theories of macrocognition acknowledge that, in complex sociotechnical domains, elements of cognition are inextricably linked; further, the linkage between these elements is neither clear nor static. Thus, to understand decision making in these domains, it should be viewed in its entirety and in the context of a number of other aspects of cognition. This report therefore considers the decision making activity in terms of problem detection, sensemaking, evaluation of competing goals, and the selection of a task.
What was undertaken in the research?
The research involved a detailed examination of contemporary theories on decision making. The report also builds upon work produced for the HFI DTC on Cognitive Work Analysis (CWA). Desk-based research was used to study the implication of decision making for the role of combat ID. A model of decision making for combat ID was then produced in a small table-top study involving a subject matter expert in the field of tank warfare, Lt Col Alan Ellis.
This report presents a prototypical model of decision making that is complimentary to existing NDM techniques. Based upon Rasmussen’s decision-ladder, the approach provides a model of how decision making can proceed within a given domain. The report starts by exploring different types of decision making research, addressing the compatibility of the decision-ladder and the recognition-primed decision making model. Based upon the decision-ladder, an approach for capturing formative descriptions of existing decision making processes is presented. The example of land-based combat ID is used to demonstrate this approach. Two new representations are introduced to aid the understanding of these domains and to support the design of decision support tools. The first maps the links between information elements, system states and options. The second, clusters elements in the decision making process in terms of their location in the world. The approach presented is not intended to replace existing decision making analysis techniques, rather, based on similar data collection procedures, its aim is to compliment them with a more formative integrant.
What was discovered?
This report has illustrated the benefits of supplementing NDM approaches with a more formative element. As stated, the use of the decision-ladder has a number of benefits in considering what can happen within in environment, rather than what does, or what should happen. Thus, it provides a description beyond what actually took place at a prescribed time and location with a given actor group. The model has been specifically developed to remain generic, it is actor-independent and can easily be adapted to consider the introduction of additional technology designed to provide decision makers with additional information. The extensions presented provided a deeper understanding of the domain; furthermore, the model provides a template for exploring the potential impact of changes to technology (such as in-tank interfaces or helmet mounted displays), training support, operating procedures, and allocation of function. Thus, informing design and system optimisation decisions.
Main conclusions and recommendations
Whilst the model has been developed specifically for tank-on-tank warfare, it is broadly applicable to all combat ID activities, regardless of platform. Minor changes are expected throughout the model; however, this model forms a solid basis for adaptation to other domains including maritime and air. The tank-warfare model has subsequently been successfully applied to police operations on the 22nd of July 2005 in Stockwell, with only minor modifications (Jenkins et al, in preparation).
The report has identified significant overlaps in the aspirations and origins of the NDM and CWA approaches. The presented approach is intended as an addition to further inform some of the other aspects of decision making research listed by Rasmussen (1997), namely; the development of decision support tools through the modelling of knowledge-based problem solving, and the relationship between the ecology of work and the design of support systems.
In its current guise, this work originates from a theoretical standpoint supported by a small data collection activity. The approach requires further validation, particularly in support of its claims to aid the design of decision support systems. Extensions to this work, including further validation of the model, are presented in a proceeding HFI DTC report (Jenkins et al, 2008).
Military relevance of the work
Combat ID is always a relevant issue for the Ministry of Deference (MOD); a greater understanding of the way decisions are made is therefore of significant benefit. As combat ID technologies become readily available the design and integration of these systems becomes a significant issue. This work forms the basis for the consideration of these systems, presenting a model to consider their impact on current cognitive processes and the implications for operating procedures and training.
contact us