Intelligence is dynamic rather than static. The primary sense of distributed intelligence arises from thinking of people in action. A dominant notion in cognitive psychology, educational psychology,and related fields was that learning was essentially the development of the ability to form symbolic representations by exercising the mind. Intelligence was present in one’s mind. In contrast, when we say that intelligence is distributed, we mean that the resources that enable and mediate activity are distributed in configuration across people, environments, situations,and time. Intelligence is assembled and accomplished rather than possessed. Therefore, the boundary unit of analysis for learning is different with this orientation, since intelligence ‘comes to life’ in human activities.

There are two dimensions of distributed intelligence: the material and the social. First, the material dimension originates from the leveraging of affordances of designed artifacts for supporting the purpose of an activity. For example, Papert and colleagues’ work cites how students, on their own, through their own discovery with no teacher intervention, built machines. However, a distributed intelligence lens would recognize the intelligence of the physical lab that Papert and colleagues built as constraints in the Lego Mindstorms kits, with the constraints serving to direct actions toward the activity’s purpose. They did not think of this as instructional because the features that constrained students’ activities were in the objects and in the use of the objects. Second, the social dimension is co-constructed in activities, such as the joint action and guided participation common in parent-child interaction or apprenticeship, or through people’s collaborative efforts to achieve shared aims. For example, in early mother-infant interactions, full sentences are often jointly created by a parent and a child, marking a distributed act of sentence completion.