Peter T. Doohan

The prefrontal cortex (PFC) transforms what we know about the world into intelligent, adaptive behaviour. Existing theories have traditionally explained this function through cross-circuit interactions, where PFC neurons control sensory-action mappings elsewhere in the brain. However, recent findings reveal a remarkable organisation of neuronal activity that enables flexible computation within the PFC itself. These findings motivate a framework of duplicated and structured representations (DSRs), where the abstract structure of the world is embedded locally in connections between prefrontal neurons encoding conjunctions of action and context. We review an extensive body of empirical findings showing that DSRs explain canonical PFCdependent behaviours as diverse as sequence working memory, planning, and value-guided decisions. In each case, PFC neurons represent multiple actions simultaneously. Structured connectivity between these action representations then guides local network dynamics to infer how behaviour should be organised - simplifying both computation and readout. Taken further, the same logic offers a mechanistic foothold on some of the PFC's most elusive functions, including hierarchical reasoning and social computations. Overall, the DSR framework provides a unified understanding of prefrontal computations and a tractable basis for a new generation of mechanistic research.

Planning is critical for adaptive behaviour in a changing world, because it lets us anticipate the future and adjust our actions accordingly. While prefrontal cortex is crucial for this process, it remains unknown how planning is implemented in neural circuits. Prefrontal representations were recently discovered in simpler sequence memory tasks, where different populations of neurons represent different future time points. We demonstrate that combining such representations with the ubiquitous principle of neural attractor dynamics allows circuits to solve much richer problems, including planning. This is achieved by embedding the environment structure directly in synaptic connections to implement an attractor network that infers desirable futures. The resulting ‘spacetime attractor’ excels at planning in challenging tasks known to depend on the prefrontal cortex. Recurrent neural networks trained by gradient descent on such tasks learn a solution that precisely recapitulates the spacetime attractor – in representation, in dynamics, and in connectivity. Analyses of networks trained across different environment structures reveal a generalisation mechanism that rapidly reconfigures the world model used for planning, without the need for synaptic plasticity. The spacetime attractor is a testable mechanistic theory of planning. If true, it would provide a path towards detailed mechanistic understanding of how prefrontal cortex structures adaptive behaviour.

Humans and animals plan actions to achieve goals in worlds that are complex and continually changing. While planning is critically dependent on the prefrontal cortex in humans, little is known about its cellular underpinnings. Mechanistic understanding has been limited by a scarcity of controlled animal experiments in which subjects must flexibly plan novel behaviours on every trial. Here we characterise the neural representations and dynamics of mouse medial frontal cortex (mFC) during flexible navigation in structured environments. We trained mice to navigate complex mazes, to goals that changed location on every trial. Optogenetic silencing established that mFC was necessary for efficient navigation. mFC activity was dominated by two factorised components: (i) a structured representation of subjects' position within the maze that formed an efficient code for behavioural trajectories, and (ii) a flexible representation of the shortest path-distance to the current goal. Both representations oscillated within local field potential (LFP) theta cycles, processing from further to closer to the goal at a systematic offset. These data suggest a computation in which mFC evaluates possible futures by their distance-to-goal to update a structured behavioural policy.

Medicinal cannabis is being used worldwide and there is increasing use of novel cannabis products in the community. Cannabis contains the major cannabinoids, Δ⁹-tetrahydrocannabinol (Δ⁹-THC) and cannabidiol (CBD), but also an array of minor cannabinoids that have undergone much less pharmacological characterization. Cannabinol (CBN) is a minor cannabinoid used in the community in “isolate’ products and is claimed to have pro-sleep effects comparable to conventional sleep medications. However, no study has yet examined whether it impacts sleep architecture using objective sleep measures. The effects of CBN on sleep in rats using polysomnography were therefore examined. CBN increased total sleep time, although there was evidence of biphasic effects with initial sleep suppression before a dramatic increase in sleep. CBN increased both non-rapid eye movement (NREM) and rapid eye movement (REM) sleep. The magnitude of the effect of CBN on NREM was comparable to the sleep aid zolpidem, although, unlike CBN, zolpidem did not influence REM sleep. Following CBN dosing, 11-hydroxy-CBN, a primary metabolite of CBN surprisingly attained equivalently high brain concentrations to CBN. 11-hydroxy-CBN was active at cannabinoid CB₁ receptors with comparable potency and efficacy to Δ⁹-THC, however, CBN had much lower activity. We then discovered that the metabolite 11-hydroxy-CBN also influenced sleep architecture, albeit with some subtle differences from CBN itself. This study shows CBN affects sleep using objective sleep measures and suggests an active metabolite may contribute to its hypnotic action.