Pathfinding: a neurodynamical account of intuition
We examine the neurobiology of intuition, a term often inconsistently defined in scientific literature. While researchers generally agree that intuition represents “an experienced-based process resulting in a spontaneous tendency toward a hunch or hypothesis,” we establish a firmer neurobiological foundation by framing intuition evolutionarily as a pathfinding mechanism emerging from the brain’s optimization of its relationship with the environment. Our review synthesizes empirical findings on intuition’s neurobiological basis, including relevant brain networks and their relationship to cognitive states like insight.
We propose that unsolved problems dynamically alter attractor landscapes, guiding future intuitions. We investigate “opportunistic assimilation” through nonlinear neurodynamics and identify hippocampal sharp wave ripples as potential neural correlates of intuition, citing their role in creativity, choice, action planning, and abstract thinking.
Finally, we explore intuition through two complementary perspectives: the free energy principle, which models brains as minimizing uncertainty through predictive hierarchical coding, and metastable coordination dynamics, describing the brain’s simultaneous tendencies toward regional cooperation and functional autonomy.
Together, these principles provide a comprehensive neurodynamical account of intuition’s neurophenomenology.
The figure illustrates the distributed neural architecture underlying intuitive cognition, characterized by metastable coordinated activity across multiple brain regions. Key structures include:
orbitofrontal cortex (red) integrating sensory-emotional inputs for coherence assessment;
anterior cingulate cortex (blue) regulating cognitive transitions;
anterior insula (green) processing interoceptive signals;
basal ganglia (purple) facilitating implicit pattern recognition;
hippocampal-entorhinal complex (orange) mediating experience replay via sharp-wave ripples;
amygdala (yellow) incorporating emotional valence in risk evaluation; and
precuneus (teal) supporting self-referential processing.
Together, these regions form a dynamic network that oscillates between integration and segregation states, with intuition emerging as a product of neurodynamical coordination that minimizes uncertainty through rapid action-path prediction and selection.
The free energy principle and metastable coordination dynamics could be conceptually viewed as two sides of the same coin, although further empirical validation is needed to support this alignment. This follows from the fact that the free energy principle casts internal (e.g., neuronal) dynamics as a gradient flow on variational free energy, thereby furnishing paths of least action. Variational free energy provides an upper bound on surprise. Therefore, minimizing variational free energy minimizes surprise and uncertainty. This can also be expressed as maximizing the evidence for generative models of the lived world, or self-evidencing. Crucially, minimizing variational free energy maximizes the entropy of posterior beliefs, in the spirit of the maximum entropy principle, to which the free energy principle is dual. This may help explain features of metastability and related neuronal dynamics, such as criticality. Heuristically, this means that to self-evidence is to keep an open mind (cf., Occam’s principle), through coordinated flows on variational free energy landscapes that—by construction—feature metastability.
The schematic depicts how intuitive cognition emerges through the dynamic integration of sensory and motor hierarchies. The central dynamic attractor landscape represents the internal model that integrates sensory input and motor output to facilitate intuitive judgments and decision-making. The sharp wave ripple inlet symbolizes hippocampal contributions, combining past experiences with present stimuli to shape predictions and pathfinding. The framework is grounded in the interaction with the external environment, highlighting the embodied and predictive nature of intuition.
The role of interoception in intuition is also underpinned by embodied cognition and the free energy principle. It is well known that interoceptive signals, such as heart rate and respiratory rate, and visceral sensations like gut movements, impact cognitive processes. By providing valuable information about internal states, these signals also contribute to the brain’s predictive imperative, helping to optimize the body-brain-environment system by reducing uncertainty.
While our primary focus has been on the neurodynamical mechanisms underpinning intuition, it is important to acknowledge that intuitive judgments are not always accurate. The process of intuition, like all predictive processes, is inherently probabilistic and therefore susceptible to error. In certain contexts, especially under conditions of ambiguity or novelty, intuitions may yield maladaptive or suboptimal inferences. This is evident in reversal learning paradigms, where prior associations must be rapidly unlearned, and intuitive responses can persist despite environmental change. Nonetheless, such outcomes do not disqualify the process as intuitive. Rather, they reflect the broader adaptive function of intuition as a fast, resource-efficient mode of navigating uncertainty. This process trades precision for speed and operates below the threshold of conscious deliberation.
As such, even erroneous intuitions remain part of the same underlying pathfinding mechanism. Overall, we hope our proposal contributes to the ongoing effort to develop a neurobiological account of intuition and offers testable hypotheses for future investigation.
Surtil : 