A Developmental Model of Intelligence Offloading

Cognitive offloading is a process of engaging with environmental affordances, expanding cognitive capacity to acquire information. But this definition restricts our understanding of how capacity-related constraints develop, particularly the process of externalization. Vast differences in brain size and offloading capacity exist across the tree of life. Taking an evo-devo perspective, we utilize computational agents to understand perceptual/behavioral and innate contributions to this process. As developing nervous systems consist of regions that scale with functional importance, offloading capacity is thus driven by growth of specific brain regions, which are in turn driven by ecological selection pressures.We hypothesize that as the nervous system comes online during biological development, it sets into motion a sorting process that determines the information capacity of a particular region, functional dependencies between regions, and the degree to which information must be externalized. Different nervous system regions come online in a certain order and grow at a certain rate. This provides a means to investigate changes in the rate and duration of growth. As growth is accelerated or decelerated over developmental time, it changes the nature of acquisition.In serving the purpose of clarifying the origins of cognitive offloading, our model also extends traditional models of sequence heterochrony in two ways: demonstrate temporal and mechanistic distinctions between ecological plasticity and natural selection, while also examining the interplay between brain region growth and the ability for these regions to encode and represent information. This capacity limitation determines externalization in our agents, and is driven by factors such as genetic control of growth, environmental sampling, and goal-driven behaviors. These can only be understood in the common framework of a constraint-oriented developmental model.