Naturalistic Decision-Making (NDM) models are used to study how people use previous experience to make decisions in real-life situations. NDM models can be used in a wide variety of research contexts, such as teamwork studies, simulations, error analysis, and more! Modern experts define NDM models as consisting of 4 main characteristics: process orientation, situation-action matching feature decision rules, context-bound informal modeling, and empirical-based prescription. Each of these 4 characteristics is described in the sections below.
NDM research is interested in the cognitive processes of experts and how those affect their decisions. This can include what information the experts are seeking, how this information is being interpreted, and which decision rules they are using when making their decisions [1].
Research has shown that experts make decisions based on matching ("Do x because it is appropriate for situation y") rather than concurrent choice ("Do x because its outcome is better than the alternatives") [1]. These experts will often compare their choices against a standard to narrow down their options, and then make their decision based on the situation.
In cognitive research, formal models don't always take things like context-specificity and semantics into account. Because the real world is based purely around semantics and context, NDM models tend to be more informal and specific to the research question, rather than trying to look at all human decision. When developing an NDM model, researchers look at the information their experts are paying attention to and which arguments they are using to make their decisions.
Finally, the NDM model must be usable in the real world. After all, why study real-life decisions if they aren't applicable to real life? While previous models of decision-making tried to generalize their findings to all types of decision-making, NDM models are what's known as prescriptive models. Prescriptive models are used by researchers to help describe actions that lead to a specific outcome [2]. From Lipshitz et al. (2001), "The goal of empirical-based prescription [...] is to improve feasible decision makers' characteristic modes of making decisions [...], rather than replacing them all together, by basing prescription on demonstrations of feasible expert performance" (p. 335). In other words, empirical-based prescriptive models (and therefore NDM models) look to augment decision-making strategies that experts are already using, rather than completely replace them with new strategies.
The NDM framework has been used to inform many of the other methods described in this site. Some of the most popular methods that use NDM are:
Lipshitz, R., Klein, G., Orasanu, J., & Salas, E. (2001). Taking stock of naturalistic decision making. Journal of Behavioral Decision Making 14(5), 331-352. doi: 10.1002/bdm.381
Ullrich, C. (2008). Descriptive and prescriptive learning theories. In Pedagogically Founded Courseware Generation for Web-Based Learning (pp. 37-42_. Springer, Berlin, Heidelberg. doi: 10.1007/978-3-540-88215-2_3