Thinking with a tool involves a close coupling between user and tool
When people use a tool to think (and I’m specifically interested in computational tools here) the locus of thought lies neither in the heads of users, nor in the software tools that they use, but instead in the system created by the tight coupling of user and tool. Cognition is thus distributed through the tool-person system. We see evidence of this, for instance, in how people use epistemic actions within a tool (or game) to understand a particular problem space. The tool itself becomes critical to the thinking process, and much of the “state” of cognition has been offloaded to the tool. This is one of the central tenets of distributed cognition. (Though, it’s worth noting Using an external external referrent in cognition is about more than offloading memory!)
As people develop more familiarity with a tool, they tend to take even greater advantage of this structural coupling by more expertly leveraging the tool for decision-making. For instance, Kirsh and Maglio have shown that Experts manipulate and perform more epistemic actions than non-experts. All of this is precisely why The fundamental unit of design analysis should be the tool-user system.
In Distributed Cognition: a methodological note, Kirsh cites the example of writing on a piece of paper (p. 251):
The person causes paper changes, paper changes cause person changes.
We can say the same for software; the person causes changes in the state of the computer system, and the changes in that state cause person changes. I believe this principle of close coupling holds true for all tools for thought and creative tools.
However, and this is critical, the inverse may also be true, where the computer system serves as the first actor. Unlike paper, computers can have autonomy and some agency (especially AI systems). In Toolsforthoughts Shaffer and Clinton echo exactly Kirsh’s claim, but for software:
The computer causes person changes, the person causes computer (state) changes.
And Gordon Brander writes something similar in Notes are conversations across time:
A conversation can happen between yourself and a pencil and paper. “There is the person who draws, and the other, yet the same, person who looks.” My pencil loops once, my mind loops three times. Perhaps this causes me to fruitfully trip over my ideas, to see them from a different angle, creating a subject-object shift.
All of it reminiscent of McLuhan’s earliest observation about media:
We become what we behold. We shape our tools and then our tools shape us.
References
Kirsh, D. (2006) Distributed Cognition: A Methodological Note.
Shaffer, D. and K. Clinton. (2006) Toolforthoughts: Reexamining Thinking in the Digital Age.
Links to this note
-
The fundamental unit of design analysis should be the tool-user system
…This is because Thinking with a tool involves a close coupling between user and tool. This implies thinking less about screens and manipulation of objects, and more about the information and state changes in tool and mind that occur when using a tool.