Are recurrent, multiscale circuit dynamics—reverberation, operating regime (near‑criticality), and neuromodulated plasticity—the common mechanistic substrate implementing attention, working memory, and basal‑ganglia mediated action selection?

Neural mechanisms

Neural mechanisms are typically investigated as organized processes linking neural structure and activity to perception, learning, decision-making, and behavior. Researchers use multiple levels of analysis, from synapses and circuits to cognition and behavior. Established research programs emphasize mechanisms such as hierarchical sensory processing, reciprocal feedback, synaptic plasticity, distributed synchrony, and interactions among cortical, medial-temporal, and basal-ganglia systems. Some models provide specific mechanistic accounts: for example, neuromodulated plasticity can support relational learning in artificial neural networks, while coordinated activity and plasticity may help link attention and working memory. The central disagreement is whether current circuit- and computation-based frameworks adequately explain cognition, or whether they overinterpret correlational signatures and require broader conceptual foundations or alternative organizing principles.

9 sources

Brain criticality debate

The criticality hypothesis proposes that neural networks operate near a transition between activity that dies out and activity that amplifies. Advocates link this regime to information processing, homeostasis, and possibly health. Evidence often cited includes neuronal avalanches and scale-invariant or long-range temporal correlations, but these signatures can have alternative explanations and may arise from measurement choices or simple noncritical models. The main disagreement is whether criticality is a necessary, unified operating principle of the brain, rather than one useful description of some neural dynamics that can also be produced by memory, reverberation, or self-regulation.

10 sources

Basal ganglia loops

Basal ganglia loops are recurrent circuits linking cerebral cortex, basal ganglia nuclei, thalamus, and cortex. They include motor, associative or cognitive, limbic, and oculomotor domains with partly distinct functions. The canonical model describes direct and indirect striatal pathways as having opposing effects on movement, broadly facilitating or suppressing motor output, and has helped explain movement disorders. More recent evidence suggests the pathways are not independent on–off levers: they can be coactivated, interact through intrastriatal and other connections, and contribute to action selection, uncertainty, reinforcement learning, and habitual behavior. The main disagreement is whether the classical segregated, opponent-pathway model is a sufficient explanation or should be replaced by an integrated account emphasizing interacting pathways and loops.

12 sources

Attention working-memory

Research broadly agrees that attention and working memory interact closely: attention can select, prioritize, and maintain information, while working memory can guide attention toward goal-relevant representations. One mainstream account treats attention and working memory as relying on overlapping control mechanisms, supported by behavioral and neural evidence that attentional prioritization influences working-memory contents and performance. A significant alternative view argues that attention is not simply a gatekeeper or limited-capacity resource for working memory. Findings such as attribute amnesia and evidence for dissociable processes suggest that attended information need not automatically enter or remain in working memory. The central disagreement is whether attention primarily supplies or controls a limited working-memory resource, or instead selects and prioritizes information through mechanisms that can operate partly independently of working memory.

9 sources