Dan Russell and I have been working on a Special Issue of the journal Human Computer Interaction. The special issue emerged from a workshops that were held at the SIGCHI conference over several years, and we had become convinced that a field of research had started to solidify. I thought I would jot done some background on the ideas about sensemaking that we had developed in the early 1990;s

Making sense of the world using information technology has become a ubiquitous activity in the digital era. It involves not only finding information, but also requires learning about new domains, solving ill-structured problems, acquiring situation awareness, and participating in social exchanges of knowledge.  Since the introduction of the notion of sensemaking in the field of human-computer interaction (HCI) in the early 1990s (Russell, Stefik, Pirolli, & Card, 1993) the term has been used to refer to a complex natural kind of phenomena that occurs in everyday life. The general category of sensemaking involves a set of interrelated activities that involve collecting, organizing and creating representations of complex information sets, all centered around some problem that needs to be understood and solved. Examples of sensemaking activities include understanding features, costs, service plans, and trade-offs in consumer decision making (e.g., buying a cell phone), collecting, organizing, and comprehending information about a medical condition, treatment options, and trade-offs in order to choose a treatment, analyzing a subject matter domain in order to develop an efficient and effective training course, and collecting and analyzing open source publications to determine the likelihood that a foreign nation is developing biological weapons. There has been a recent increase of focused interest in this notion of sensemaking activity that involves active, iterative, interaction with massive amounts of amounts of information to distill it into forms that provide insight and support effective action. Such interest is spurred by many forces.  One has been the push of the information explosion from the Web.  Another comes from the library and information sciences as well as HCI communities that have begun to converge on projects trying to help people make sense of the multitude of information resources now available.  Another has come from funding agencies interested in improving homeland security, emergency response, and intelligence analysis.

Sensemaking, as in to make sense, implies an active process as opposed to the achievement of some state of affairs. Sensemaking, in essence, is a process of forming and working with representations, and those representations determine which computations are easy or difficult, and consequently (we will argue) which activities can be performed more or less intelligently. In sensemaking: representation is central; representation shapes computation; computation shapes intelligence.

Russell et al. (1993) developed a conceptual model of sensemaking centered around the formation and manipulation (by people and by computer) of representations. Russell et al. studied a small set of cases involving ill-structured problem solving. According to Russell et al. sensemaking involves interplay between foraging for information and abstracting the information into a representation called a schema that will facilitate a decision or solution. The schema summarizes (abstracts, aggregates) the external information that has been found so far, eliminates irrelevant information, and is structured to efficiently and effectively support the task in which it is embedded.  The Russell et al. analysis focused on re-representational sensemaking processes for a case in which large amounts of information had to be digested to create a curriculum for printer repairmen.  The core of the process is called a “learning loop complex” (Figure  above). First, there is a search for a good representation (the generation loop). Then there is an attempt to encode information in the representation (the data coverage loop). The attempt at encoding information in the representation identifies items that do not fit (“residue”). This gives rise to an attempt to adjust the representation so that it has better coverage (the “representation shift loop”). The result is a more compact representation of the essence of the information relative to the intended task.  Effort is expended up front to create representations that facilitate insight and intelligent action later.

THE DATA/FRAME MODEL

Related, but somewhat different “sensemaking” approaches have developed in fields outside of HCI.   The macrocognitive model of sensemaking (Klein, et al., 2006b) provides a theoretical framework for understanding and predicting how people make sense of experience or events in the world. Klein et al. (2006a) propose that situation awareness can be considered a state-of-knowledge about the world, involving some form of mental model representation of the state-of-affairs in the world. Sensemaking, rather than being that state-of-knowledge, is the process of achieving that outcome. The Data/Frame theory assumes that meaningful representations called frames define what counts as data and how those data are structured for mental processing (Klein, et al., 2006b). Frames can be expressed in a variety of forms including stories, maps, organizational diagrams, or scripts.  Whereas frames define and shape data, data can mandate changes to frames.

In this framework, sensemaking can involve elaboration of a frame (e.g., filling in details), questioning a frame (e.g., due to the detection of anomalies), or reframing (e.g., rejecting a frame and replacing it with another). The Data/Frame theory proposes backward-looking processes are involved in forming mental models that explain past events and forward-looking mental simulations that predict how future events will unfold.

REFERENCES

Russell, D. M., Stefik, M. J., Pirolli, P., & Card, S. K. (1993). The cost structure of sensemaking. Paper presented at the INTERCHI '93 Conference on Human Factors in Computing Systems, Amsterdam.

Klein, G., Moon, B., & Hoffman, R. R. (2006a). Making sense of sensemaking 1: Alternative perspectives. IEEE Intelligent Systems, 21(4), 70-73.

Klein, G., Moon, B., & Hoffman, R. R. (2006b). Making sense of sensemaking 2: A macrocognitive model. IEEE Intelligent Systems, 21(5), 88-92.