Building Cognition: The Construction of External Representations for Discovery

Building Cognition: The Construction of External Representations for Discovery Sanjay Chandrasekharan (sanjayan@cc.gatech.edu) Nancy J. Nersessian (nancyn@cc.gatech.edu) School of Interactive Computing, Georgia Institute of Technology, 85 Fifth Street NW, Atlanta, GA 30308, USA therefore do not transfer well to the ill-structured and open- ended task environment of a scientific laboratory. Further, the DC framework, as it stands now, focuses on the use of existing external representations, not on the processes of generating representations, which play a significant role in scientific practice. Building representations is part of the activity of what Hall et al. (2010) call “distributing” cognition, that is, “how cognition … is produced … out of human activity (p.2).” The DC framework therefore needs to be extended to understand scientific practices that require building novel representations for problem solving. In this paper, we outline some aspects of this extension, using a case study of how computational models are built in a system biology lab. Abstract The analysis of the cognitive role played by external representations – particularly within the distributed cognition (DC) framework – has focused on the use of such representations in cognitive tasks. In this paper, we argue that the processes of building such representations require close attention as well, especially when extending the DC framework to ill-structured domains such as scientific laboratories, where building novel representations is crucial for making discoveries. Based on an ethnographic study of the building of computational models in a systems biology laboratory, we examine the complex cognitive roles played by the external representations built by the lab (pathway diagrams and models), and the building process itself. Keywords: Distributed Cognition, External Representations, Scientific Cognition, Discovery, Creativity, Ethnography Lab G as a Distributed Cognitive System Traveler, there is no path, paths are made by walking. – Antonio Machado In our current project we are studying problem-solving practices in two integrative systems biology labs. We focus here on one lab that does only computational modeling (“Lab G”). The modelers come mainly from engineering fields, but work on building computational models of biochemical pathways, to simulate and understand phenomena as varied as Parkinson’s disease, plant systems for bio-fuels, and atherosclerosis. The problems Lab G modelers work on are provided by outside experimental collaborators, who see modeling primarily as a method for identifying key experiments of scientific or commercial importance. The collaborators provide experimental data for modeling, and also generate data needed for developing or validating the model. In broad terms, the Lab G modeling processes can be understood as occurring within a distributed socio-technical system, which is the primary unit of analysis in DC. This system comprises people working together (modelers, experimentalists) to accomplish a task (discover fruitful changes to biological pathways), and the artifacts they use (models, pathways, diagrams, graphs, papers, databases, search engines) in the process. The task environment of the lab, and the external representations used there, differ drastically from those usually examined in DC, such as the standard example of the cockpit and the speed-bug (Hutchins, 1995a). The main differences can be classified as follows: Actors and Goals: The lab does not have a structured task environment, with synchronous actions connecting individuals or groups. The objective of the lab is to make discoveries, so the lab task environment is one where the specific goal is not known in advance. There are very general goals, such as “discover interesting reactions”, and less general goals, such as “fit model”. These general goals are spread across people who share a resource (experimental The role of external representations in cognitive tasks has received a lot of attention, particularly within the distributed cognition (DC) framework (Hutchins, 1995; 1995a; Kirsh, 2010). Much of the work on external representations within DC focuses on capturing detailed descriptions of the way external representations are used in highly structured task environments, such as ship navigation and landing of aircraft, and the way these representations change the nature of cognitive tasks. Less understood are the processes of building external representations to alter task environments (Kirsh, 1996; Chandrasekharan & Stewart, 2007), and the role played by this building process in cognition and problem-solving. In this paper, we focus on the building of a complex external representation – a computational model – and examine the role this external representation, and its building process, plays in structuring, as well as altering, the discovery task in a systems biology laboratory. Only a handful of studies have examined problem solving in scientific research from a DC perspective (e.g, Nersessian et al, 2003; Alac & Hutchins, 2004; Hall, Wieckert & Wright, 2010; Goodwin, 1997; Giere, 2002). These studies do not consider how external representations are built, largely because the development of a novel external representation, and the changes it makes to the scientific task environment, are complex events that occur over long periods of time, and therefore not easily captured. Even when such data are available (Chandrasekharan, 2009; Nersessian & Chandrasekharan, 2009), it is not easy to understand building using the current DC framework, which is derived using studies of well-structured tasks, and