Socially Distributed Cognition

Socially Distributed Cognition

Any culture contains assumptions that are foundational, unquestioned and, in many cases, unconscious. Perhaps one of the most powerful for Western Eurocentric culture is the duality between mind and body, a Platonist concept implicit in much Christian theology, and perhaps most powerfully developed philosophically by Descartes. This model places "mind," and thus mental processes such as cognition, firmly within each individual (cogito, ergo sum). While such metaphysics may seem innocuous or irrelevant, it can actually be profoundly dysfunctional. Thus, for example, the illusion that cognition occurs only inside individual humans leads most people to seriously mischaracterize the development and nature of distributed human-ICT (information and communication technology) cognitive networks. As regards environmental issues, the model leads to exhortations such as "think globally, act locally," and allocations of blame and responsibility, which assume the individual is the center of cognition for environmental issues. Such approaches may work where the phenomena at issue involve “simple” systems dynamics, but tend to fail badly where complex adaptive systems are involved.

In this regard, it is worth rereading Edwin Hutchins' classic Cognition in the Wild. Rather than begin with hypothesized first principles such as the Cartesian worldview, Hutchins studies the ways in which a navigation team on a naval vessel, composed of four or more individuals, actually navigates or, more broadly, thinks. He explores the way that certain relatively arbitrary models of the world are embedded in the artifacts that have developed over centuries to enable Western-style navigation (as compared to, say, Micronesian navigation, which is founded on a different, but equally competent, worldview). In doing so, he finds a “cognitive ecology” rather than a controlled, centrally directed function understood and structured by a key individual. In this system, cognition is not a function of individuals, but an emergent property of the data, tools, individuals, and relationships that characterize the system: “thinking” is done just as much by the navigation chart as by the quartermaster. The tools do not simply amplify the existing cognitive powers of the individuals involved in the process, but transform the task itself, reallocating cognitive functionality between artifacts and individuals.

The implications of this analysis of socially distributed cognition for industrial ecologists, environmental researchers, and policymakers are obvious. To begin with, many of the issues and systems such communities engage with are clearly complex: non-linear, rapidly changing, reflexive, unpredictable, data intense, and characterized by emergent properties. Moreover, the institutional and governance structures that are frequently encountered, from private firms to NGOs to governments, are also complex. Under such circumstances, it is highly unlikely that single individuals, no matter how well placed, either completely understand, or control, systems response. This is not to argue that powerful individuals, such as CEOs or politicians, cannot have substantial impacts on systems performance (and thus bear ethical responsibility for such behavior), but it does suggest that a more nuanced approach to systems behavior and learning might pay dividends.

For example, suppose one wanted to integrate environmental considerations into a product design process. In a Cartesian world, one might try to teach all engineers how to create environmentally preferable designs. But if the relevant product design system is understood as socially distributed cognition, the goal is not to change the cognition of any individual, but rather to change overall system cognition, focusing, perhaps, on the cognition embedded in design tools (e.g., modifying lists of acceptable components to emphasize environmental preferability, developing environmentally sensitive design heuristics and algorithms), rather than on the designers themselves. The advantage of such an approach is that the human components of the system need not be changed at all: The socially distributed cognitive system becomes “environmentally conscious,” while designers continue in their existing patterns. Similarly, if one wanted to encourage the evolution of environmentally sensitive nanotechnology, one could try to educate each nanotech researcher on relevant environmental and ethical issues, a daunting task. Or, one could try to define the relevant distributed cognitive system, and create artifacts and processes, such as regulatory pathways, that accomplished the same goal from a systems perspective -- not easy, but doable.

Manipulation of distributed cognition systems is not a substitute for decisions about ultimate ends. These are still chosen by humans exercising intentionality (including failure to agree on goals, which is of course still a decision, albeit by default). But understanding, using, and participating in distributed cognition systems is frequently far more effective operationally than reliance on a false construct of solipsistic individual cognition.

Brad Allenby is professor of civil and environmental engineering at Arizona State University, a fellow at the University of Virginia's Darden Graduate School of Business, and previously was AT&T's vice president of environment, health, and safety.