Distributed Cognition

What goes into making sure a plane takes off in time?

There are many people, systems, and actions required; cooperation and coordination are essential.¹ Pilots and crew members work together to prepare the plane and its passengers, communicating with each other as the necessary steps are completed. Hundreds of sensors and machinery are in operation. Together, this complex system of people and machinery allows a plane to take off in time.

This is one example of what cognitive scientists call “distributed cognition.”² The core idea is that knowledge and cognition not only exist within oneself, but also in one’s social and physical environments. Our memory, decision making, reasoning, and learning are all collectively processed by our minds and by extensions of our minds in the external world.

This is a radical departure from the traditional idea that cognition exists only within the mind or the brain.³ Distributed cognition extends the reach of what is considered cognition (or, at least, what is involved in it) to include interactions between people and things in one’s environment.⁴ It is not a kind of cognition,⁴ but a new way of viewing how cognition works.³

Cognitive anthropologist Edwin Hutchins was part of the U.S. Navy in November 1980, when he spent most of his time on the navigation bridge, the room from where the ship was commanded.⁵ Hutchins studied what the operators of the ship’s steam propulsion plant knew and how they gained this knowledge. The week before, he had spent his time in the bowels of the ship, observing engineering operations and engaging with those who worked in such spaces. Together, these observations sparked Hutchins’ interest in how these two parts of the ship worked together.

In 1984, Hutchins was working for the Navy Personnel Research and Development Center as a research psychologist.⁵ Due to his prior research success, he was granted the freedom to conduct an independent research project. He chose to study what he called at the time “naturally situated cognition.” Based on his love of navigation, Hutchins once again worked in and studied a naval ship’s navigation bridge. In the beginning, Hutchins considered navigation in terms of cognition at an individual level. But at the end of his first study period at sea, he realized the importance of the social distribution of cognition.

Hutchins’ experiences in 1980 and 1984 laid the groundwork for his 1995 book, Cognition in the Wild, where he outlined the principles and applications of distributed cognition.⁵ But much of his work drew from existing research in psychology and sociology.^{2 4} Memory, for example, has a long history of being treated as a socially distributed cognitive function. Other researchers focused on social organization, and proposed a distributional model of culture that emphasized the distribution of beliefs across members of a society.

Hutchins’ book aimed to give a comprehensive, cohesive explanation of these different strands of research. His model of distributed cognition was the result. In distributed cognition, we find a system that can change and redistribute itself for different cognitive processes. A cognitive process is limited by the functional relationships between the elements that participate in it, resulting in three types of distribution of cognitive processes:

1. Cognitive processes may be distributed across the members of a social group;
2. Cognitive processes may involve coordination between internal and external (material or environmental) structures; and,
3. Cognitive processes may be distributed through time such that the products of earlier events can transform the nature of later events.⁴

One of its general assumptions is that cognitive systems consisting of more than one person have cognitive properties that differ from the people who participate in those systems.³ The knowledge possessed by members of the cognitive system is highly variable: people who work together are likely to possess different kinds of knowledge, and will engage in interactions that help them combine what they have and what they know. This makes sharing access, resources, and knowledge key for the broader cognitive system to function.

Distributed cognition provides a framework and method to examine the interactions between people and objects, which is not possible with traditional approaches.³ By doing so, distributed cognition can highlight the complex interdependencies between people and objects.

Distributed cognition has been used to analyse collaborative work environments⁸ and education,^{9 10 11 12 13} due to how dependent their tasks are on their social and physical environment.³ A variety of methods have been used to apply distributed cognition, ranging from detailed analysis of video or audio recordings of real interactions, to neural network simulations and lab experiments. However, most applications have relied on field research:  researchers making rigorous observations to determine the ways cognition is distributed.

Let us focus on education. Distributed cognition has been used for theories of learning, where knowledge acquisition is attributed to people interacting dynamically with cognitive artifacts.⁹ Think about how learning to read involves cognitive processes not only at the level of the reader. Learners also need books, instructors, and peers; these external resources and processes are a key part of the learning too.  Distributed cognition gives us the tools to understand how these components interact together to allow someone to learn how to read.

Distributed cognition has also played a role in discussions of distance learning, focusing on computer supported collaborative learning.¹¹ Distributed cognition holds that humans have limited cognitive resources, making distribution and collaboration necessary. Computer supported collaborative learning can provide spaces for these processes. When learning a new language, for example, students can use computer exercises to practice their writing skills and store their progress.^{12 13}

Some researchers have argued that the distributed cognition framework is hard to apply: it takes time to internalize its concepts and apply it to data captured by more traditional frameworks.³ Understanding a 2 second video clip all of the sudden requires analyzing not just the video, but the website, the viewer, the creator, the internet, etc. Some critics worry that, by making explanations so burdensome, distributed cognition stops us from explaining what we already have easily explained.

Others outright reject the legitimacy of distributed cognition.¹⁴ These critics believe that distributed cognition describes social phenomena with redundant cognitive science jargon just because they’re involved in cognitive tasks. When applying distributed cognition to the tools necessary for navigation, like maps, Hutchins categorized all such tools as cognitive artifacts. But these critics argue that categorizing maps in this way gets us nowhere closer to understanding what a map is or how we use it. All distributed cognition tells us, is that they are also “cognitive artifacts.”

Emergency medical dispatches

Emergency medical dispatches require coordination and communication between team members:¹⁵ This type of work has two main aspects:

1. Call-taking, which consists of answering calls for emergency medical assistance; and,
2. Controlling, which consists of dispatching the most appropriate ambulance to the emergency and optimizing ambulance resources.

A team of researchers developed a model fitting for emergency medical dispatch based on case studies.¹⁵ The model incorporated principles from distributed cognition such as:

• The perceptual principle, which holds that spatial representations provide more support for cognition than non-spatial ones, if there is a clear mapping between the spatial layout and what it represents;
• The naturalness principle, which holds that cognition is easier when the form of the representation matches the properties of what it represents; and,
• Situation awareness, which holds that people are informed of what is going on in their external surroundings and what is planned by their peers.

The model also highlighted the importance of physical and communication structures to support the work of ambulance controllers, and emphasized why emergency medical dispatchers work well.¹⁵ For example, it showed the effectiveness of positioning radio operators next to the allocators who decide where and when ambulances go. By being next to each other, radio operators and allocators can share cognitive artifacts and communicate with each other, making dispatches more efficient.

Ultimately, the researchers showed how distributed cognition can be a useful model for work as important as emergency medical dispatches, and they suggest that distributed cognition should be considered as a design tool for similar work environments.¹⁵ This way, planners could alleviate some difficulties that come with designing large and complex systems that require an abundance of cognitive resources.

Group decision making: How to be effective and efficient

Distributed cognition highlights the benefits of collaborative work, as people can pool their cognitive resources together to solve problems and manipulate cognitive artifacts to reach a common goal. However, behavioral science cautions against the consequences of groupthink, causing group members to conform to the norm, hindering problem solving. Take a look at this article for some tips to combat groupthink and be efficient under distributed cognition.

1. Hutchins, E., & Klausen, T. (1996). Distributed cognition in an airline cockpit. Cognition and Communication at Work, 15-34.
2. Hutchins, E. (2000). Distributed cognition. International Encyclopedia of the Social and Behavioral Sciences. Elsevier Science, 138.
3. Descartes, R., (2013). In A. Bailey (Ed.) & I Johnston (Trans), Meditations on first philosophy: In which the existence of God and the difference between the human soul and body are demonstrated.
4. Rogers, Y. (1997). A brief introduction to distributed cognition.
5. Hollan, J., Hutchins, E., & Kirsh, D. (2000). Distributed cognition: Toward a new foundation for human-computer interaction research. ACM Transactions on Computer-Human Interaction, 7(2), 174-196.
6. Hutchins, E. (1995). Cognition in the Wild. MIT Press.
7. Edwin Hutchins. (2021). University of California, San Diego. https://pages.ucsd.edu/~ehutchins/
8. D’Andrade, R. (1995). The Development of Cognitive Anthropology. Cambridge University Press.
9. Rogers, Y., & Ellis, J. (1994). Distributed cognition: An alternative framework for analysing and explaining collaborative working. Journal of Information Technology, 9(2), 119-128.
10. Karasavvidis, I. (2002). Distributed cognition and educational practice. Journal of Interactive Learning Research, 13(1/2), 11-29.
11. Jaramillo, J. A. (1996). Vygotsky’s sociocultural theory and contributions to the development of constructivist curricula. Education, 117(1), 133-141.
12. Lehtinen, E., Hakkarainen, K., Lipponen, L., Rahikainen, M., & Muukkonen, H. (1999). Computer supported collaborative learning: A review. The JHGI Giesbers Reports on Education, 10.
13. Lehtonen, T., & Tuomainen, S. (2003). CSCL – A tool to motivate foreign language learners: The Finnish application. Recall, 15(1), 51-67.
14. Goda, Y., & Yamada, M. (2013). Application of Col to design CSCL for EFL online asynchronous discussion. In Educational Communities of Inquiry: Theoretical Framework, Research and Practice (pp. 295-216). IGI Global.
15. Button, G. (2008). Against “distributed cognition”. Theory, Culture & Society, 25(2), 87-104.
16. Furniss, D., & Blandford, A. (2006). Understanding emergency medical dispatch in terms of distributed cognition: A case study. Ergonomics, 49(12-13), 1174-1203.