Maurício Martins surveys experimental paradigms designed to test RHE behaviorally and neurally, including discrimination and production tasks. Across domains, studies show that humans can reliably acquire and apply recursive rules, although RHE is more demanding than simpler iterative rules and typically emerges later in development. Learning recursion is often scaffolded by prior mastery of iteration, and recursive performance improves with training and consolidation. Neuroimaging and neuropsychological evidence indicate partial overlap in the brain networks involved across domains during early learning, followed by increasing domain-specific specialization with expertise.

The review also highlights open questions about the computational and biological mechanisms underlying recursion. While behavioral data suggest partially shared cognitive resources, existing models cannot yet fully explain how recursive knowledge is acquired or why its generativity is bounded by memory and attentional constraints. The author calls for tighter integration between cognitive experiments and formal computational modeling.

Overall, the paper proposes a roadmap toward a unified science of fractal cognition, linking the mathematics of self-similarity with experimental approaches that uncover how the human mind builds complex hierarchical representations.

Contact: Maurício Martins (University of Vienna, Vienna, Austria)

Email: mauricio.martins@univie.ac.at

Martins, M. (2025). From fractal geometry to fractal cognition: Experimental tools and future directions for studying recursive hierarchical embedding. Fractal and Fractional 9(10), 654. doi.org/10.3390/fractalfract9100654