The real world is stochastic, making it difficult to know when planning is worthwhile. Here the authors show that as stochasticity increases, people spend less effort planning and adopt simpler, less ...

Expectations—prior knowledge of environmental statistics—support adaptive behavior by reducing uncertainty during inference. However, the questions of where expectations come from and how that ought t...

Abstract. Prioritizing memories that are motivationally salient is adaptive for goal-directed behavior, ensuring that information being retained is the most relevant to future goals. Past research dem...

Most widely studied option evaluation strategies rely on knowledge accumulated across repeated experiences. But how should options be evaluated in ...

Human inference is often suboptimal in ways that vary across individuals and tasks. We propose that this variability reflects information processing limits and develop a task-general application of th...

Li et al. show that recurrent neural networks optimized for free recall discover diverse, human-like memory strategies beyond classical temporal context models, with top models using an index-based me...

Correctly assigning value to different options and leveraging this information to guide choice is a cornerstone of adaptive decision-making. Reinforcement learning (RL) has provided a computational fr...

Flexible decision making depends on retrieving and recombining memories. Yet because this process unfolds covertly, its governing principles remain unknown. Here we use gaze reinstatement to uncover t...

When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of...

A drawing-like task designed to study compositional generalization identifies a specific neural population in the ventral premotor cortex in primates that encodes action symbols.

Two hallmarks of biological computation are its flexibility and efficiency. These features are often attributed to cognitive control processes that balance external utility against computational cost. However, how the brain could implement such adaptive control remains unknown. Here, we provide one possible answer by combining the computational theory of rational meta-reasoning with a meta-learning algorithm recently proposed as a model of prefrontal cortex. This yields a recurrent neural network model that learns to select computations. In simple choice tasks, the model approximates the algorithms and representations of optimal symbolic models and reproduces neural dynamics observed in macaque orbitofrontal cortex. In multi-step planning tasks, the model replicates key behavioral signatures of human planning strategies and captures human neural dynamics associated with step-by-step mental simulation. Our framework unifies meta-reasoning and meta-learning by showing that learning to reason can be understood as learning to learn from information generated by one’s own cognitive operations, providing a mechanistic account of how adaptive control of thought can be implemented in neural systems. ### Competing Interest Statement The authors have declared no competing interest.

Human curiosity is dynamic, however the principles governing its fluctuations remain debated. Here, we test two competing hypotheses about how past...

The neuroscience of planning has long been analogized to search algorithms in artificial intelligence (AI), which simulate future actions to guide immediate choices. We argue that advances in both neu...

Strategic avoidance of uncertainty emerges under high cognitive demands, enabling faster decisions without impairing learning.

Nature Human Behaviour - Biba et al. show that episodic memory encoding fluctuates at a theta rhythm of 3–10 Hz.

Counterfactual thinking — considering what could have come of choosing the other path — can facilitate inference. Previous studies have demonstrated t…

Nicholas and Mattar found that people use episodic memory to make decisions when it is unclear what will be needed in the future. These findings reveal how the rich representational capacity of episod...

When learning new tasks, both humans and artificial neural networks face a trade-off between reusing prior knowledge to learn faster and avoiding the disruption of earlier learning. This study shows t...

Child & Adolescent Translational Science Lab at Northwestern University

By harnessing a mental model of how the world works, learners can make flexible choices in changing environments. However, while children and adolesce…

When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of machine lear...

The brain seamlessly transforms sensory information into precisely-timed movements, enabling us to type familiar words, play musical instruments, or perform complex motor routines with millisecond pre...

Abstract. Forming memories requires a focus on the external world; retrieving memories requires attention to our internal world. Computational models propose that the hippocampus resolves the tension ...

Modelling biological decision-making with tiny recurrent neural networks enables more accurate predictions of animal choices than classical cognitive models and offers insights into the underlying cog...

Humans can apply solutions used in past problems to new problems. In this study, the authors reveal the neural correlates of this process, known as generalization, and show that humans apply past poli...

Our experiences contain countless details that may be important in the future, yet we rarely know which will matter and which won't. This uncertainty poses a difficult challenge for adaptive decision ...