People construct simplified mental representations to plan

Hierarchical Constraints on the Distribution of Attention in Dynamic Displays

Human vision is remarkably good at recovering the latent hierarchical structure of dynamic scenes. Here, we explore how visual attention operates with this hierarchical motion representation. The way in which attention responds to surface physical features has been extensively explored. However, we know little about how the distribution of attention can be distorted by the latent hierarchical structure. To explore this topic, we conducted two experiments to investigate the relationship between minimal graph distance (MGD), one key factor in hierarchical representation, and attentional distribution. In Experiment 1, we constructed three hierarchical structures consisting of two moving objects with different MGDs. In Experiment 2, we generated three moving objects from one hierarchy to eliminate the influence of different structures. Attention was probed by the classic congruent-incongruent cueing paradigm. Our results show that the cueing effect is significantly smaller when the MGD between two objects is shorter, which suggests that attention is not evenly distributed across multiple moving objects but distorted by their latent hierarchical structure. As neither the latent structure nor the graph distance was part of the explicit task, our results also imply that both the construction of hierarchical representation and the attention to that representation are spontaneous and automatic.

From task structures to world models: what do LLMs know?

In what sense does a large language model (LLM) have knowledge? We answer by granting LLMs 'instrumental knowledge': knowledge gained by using next-word generation as an instrument. We then ask how instrumental knowledge is related to the ordinary, 'worldly knowledge' exhibited by humans, and explore this question in terms of the degree to which instrumental knowledge can be said to incorporate the structured world models of cognitive science. We discuss ways LLMs could recover degrees of worldly knowledge and suggest that such recovery will be governed by an implicit, resource-rational tradeoff between world models and tasks. Our answer to this question extends beyond the capabilities of a particular AI system and challenges assumptions about the nature of knowledge and intelligence.

Simplifying social learning

Social learning is complex, but people often seem to navigate social environments with ease. This ability creates a puzzle for traditional accounts of reinforcement learning (RL) that assume people negotiate a tradeoff between easy-but-simple behavior (model-free learning) and complex-but-difficult behavior (e.g., model-based learning). We offer a theoretical framework for resolving this puzzle: although social environments are complex, people have social expertise that helps them behave flexibly with low cognitive cost. Specifically, by using familiar concepts instead of focusing on novel details, people can turn hard learning problems into simpler ones. This ability highlights social learning as a prototype for studying cognitive simplicity in the face of environmental complexity and identifies a role for conceptual knowledge in everyday reward learning.

Bayesian Reinforcement Learning With Limited Cognitive Load

All biological and artificial agents must act given limits on their ability to acquire and process information. As such, a general theory of adaptive behavior should be able to account for the complex interactions between an agent's learning history, decisions, and capacity constraints. Recent work in computer science has begun to clarify the principles that shape these dynamics by bridging ideas from reinforcement learning, Bayesian decision-making, and rate-distortion theory. This body of work provides an account of capacity-limited Bayesian reinforcement learning, a unifying normative framework for modeling the effect of processing constraints on learning and action selection. Here, we provide an accessible review of recent algorithms and theoretical results in this setting, paying special attention to how these ideas can be applied to studying questions in the cognitive and behavioral sciences.

Wagers for work: Decomposing the costs of cognitive effort

Some aspects of cognition are more taxing than others. Accordingly, many people will avoid cognitively demanding tasks in favor of simpler alternatives. Which components of these tasks are costly, and how much, remains unknown. Here, we use a novel task design in which subjects request wages for completing cognitive tasks and a computational modeling procedure that decomposes their wages into the costs driving them. Using working memory as a test case, our approach revealed that gating new information into memory and protecting against interference are costly. Critically, other factors, like memory load, appeared less costly. Other key factors which may drive effort costs, such as error avoidance, had minimal influence on wage requests. Our approach is sensitive to individual differences, and could be used in psychiatric populations to understand the true underlying nature of apparent cognitive deficits.

Naturalistic reinforcement learning

Humans possess a remarkable ability to make decisions within real-world environments that are expansive, complex, and multidimensional. Human cognitive computational neuroscience has sought to exploit reinforcement learning (RL) as a framework within which to explain human decision-making, often focusing on constrained, artificial experimental tasks. In this article, we review recent efforts that use naturalistic approaches to determine how humans make decisions in complex environments that better approximate the real world, providing a clearer picture of how humans navigate the challenges posed by real-world decisions. These studies purposely embed elements of naturalistic complexity within experimental paradigms, rather than focusing on simplification, generating insights into the processes that likely underpin humans' ability to navigate complex, multidimensional real-world environments so successfully.

Computational Social Psychology

Social psychologists attempt to explain how we interact by appealing to basic principles of how we think. To make good on this ambition, they are increasingly relying on an interconnected set of formal tools that model inference, attribution, value-guided decision making, and multi-agent interactions. By reviewing progress in each of these areas and highlighting the connections between them, we can better appreciate the structure of social thought and behavior, while also coming to understand when, why, and how formal tools can be useful for social psychologists.

Interpretability Is in the Mind of the Beholder: A Causal Framework for Human-Interpretable Representation Learning

Research on Explainable Artificial Intelligence has recently started exploring the idea of producing explanations that, rather than being expressed in terms of low-level features, are encoded in terms of interpretable concepts learned from data. How to reliably acquire such concepts is, however, still fundamentally unclear. An agreed-upon notion of concept interpretability is missing, with the result that concepts used by both post hoc explainers and concept-based neural networks are acquired through a variety of mutually incompatible strategies. Critically, most of these neglect the human side of the problem: a representation is understandable only insofar as it can be understood by the human at the receiving end. The key challenge in human-interpretable representation learning (hrl) is how to model and operationalize this human element. In this work, we propose a mathematical framework for acquiring interpretable representations suitable for both post hoc explainers and concept-based neural networks. Our formalization of hrl builds on recent advances in causal representation learning and explicitly models a human stakeholder as an external observer. This allows us derive a principled notion of alignment between the machine's representation and the vocabulary of concepts understood by the human. In doing so, we link alignment and interpretability through a simple and intuitive name transfer game, and clarify the relationship between alignment and a well-known property of representations, namely disentanglement. We also show that alignment is linked to the issue of undesirable correlations among concepts, also known as concept leakage, and to content-style separation, all through a general information-theoretic reformulation of these properties. Our conceptualization aims to bridge the gap between the human and algorithmic sides of interpretability and establish a stepping stone for new research on human-interpretable representations.

Using deep neural networks as a guide for modeling human planning

When developing models in cognitive science, researchers typically start with their own intuitions about human behavior in a given task and then build in mechanisms that explain additional aspects of the data. This refinement step is often hindered by how difficult it is to distinguish the unpredictable randomness of people's decisions from meaningful deviations between those decisions and the model. One solution for this problem is to compare the model against deep neural networks trained on behavioral data, which can detect almost any pattern given sufficient data. Here, we apply this method to the domain of planning with a heuristic search model for human play in 4-in-a-row, a combinatorial game where participants think multiple steps into the future. Using a data set consisting of 10,874,547 games, we train deep neural networks to predict human moves and find that they accurately do so while capturing meaningful patterns in the data. Thus, deviations between the model and the best network allow us to identify opportunities for model improvement despite starting with a model that has undergone substantial testing in previous work. Based on this analysis, we add three extensions to the model that range from a simple opening bias to specific adjustments regarding endgame planning. Overall, our work demonstrates the advantages of model comparison with a high-performance deep neural network as well as the feasibility of scaling cognitive models to massive data sets for systematically investigating the processes underlying human sequential decision-making.

Diminished State Space Theory of Human Aging

Many new technologies, such as smartphones, computers, or public-access systems (like ticket-vending machines), are a challenge for older adults. One feature that these technologies have in common is that they involve underlying, partially observable, structures (state spaces) that determine the actions that are necessary to reach a certain goal (e.g., to move from one menu to another, to change a function, or to activate a new service). In this work we provide a theoretical, neurocomputational account to explain these behavioral difficulties in older adults. Based on recent findings from age-comparative computational- and cognitive-neuroscience studies, we propose that age-related impairments in complex goal-directed behavior result from an underlying deficit in the representation of state spaces of cognitive tasks. Furthermore, we suggest that these age-related deficits in adaptive decision-making are due to impoverished neural representations in the orbitofrontal cortex and hippocampus.

Rational Simplification and Rigidity in Human Planning

Planning underpins the impressive flexibility of goal-directed behavior. However, even when planning, people can display surprising rigidity in how they think about problems (e.g., "functional fixedness") that lead them astray. How can our capacity for behavioral flexibility be reconciled with our susceptibility to conceptual inflexibility? We propose that these tendencies reflect avoidance of two cognitive costs: the cost of representing task details and the cost of switching between representations. To test this hypothesis, we developed a novel paradigm that affords participants opportunities to choose different families of simplified representations to plan. In two preregistered, online studies (Ns = 377 and 294 adults), we found that participants' optimal behavior, suboptimal behavior, and reaction time were explained by a computational model that formalized people's avoidance of representational complexity and switching. These results demonstrate how the selection of simplified, rigid representations leads to the otherwise puzzling combination of flexibility and inflexibility observed in problem solving.

Intention beyond desire: Spontaneous intentional commitment regulates conflicting desires

The human mind is a mosaic composed of multiple selves with conflicting desires. How can coherent actions emerge from such conflicts? Classical desire theory argues that rational action depends on maximizing the expected utilities evaluated by all desires. In contrast, intention theory suggests that humans regulate conflicting desires with an intentional commitment that constrains action planning towards a fixed goal. Here, we designed a series of 2D navigation games in which participants were instructed to navigate to two equally desirable destinations. We focused on the critical moments in navigation to test whether humans spontaneously commit to an intention and take actions that would be qualitatively different from those of a purely desire-driven agent. Across four experiments, we found three distinctive signatures of intentional commitment that only exist in human actions: "goal perseverance" as the persistent pursuit of an original intention despite unexpected drift making the intention suboptimal; "self-binding" as the proactive binding of oneself to a committed future by avoiding a path that could lead to many futures; and "temporal leap" as the commitment to a distant future even before reaching the proximal one. These results suggest that humans spontaneously form an intention with a committed plan to quarantine conflicting desires from actions, supporting intention as a distinctive mental state beyond desire. Additionally, our findings shed light on the possible functions of intention, such as reducing computational load and making one's actions more predictable in the eyes of a third-party observer.

Humans decompose tasks by trading off utility and computational cost

Human behavior emerges from planning over elaborate decompositions of tasks into goals, subgoals, and low-level actions. How are these decompositions created and used? Here, we propose and evaluate a normative framework for task decomposition based on the simple idea that people decompose tasks to reduce the overall cost of planning while maintaining task performance. Analyzing 11,117 distinct graph-structured planning tasks, we find that our framework justifies several existing heuristics for task decomposition and makes predictions that can be distinguished from two alternative normative accounts. We report a behavioral study of task decomposition (N = 806) that uses 30 randomly sampled graphs, a larger and more diverse set than that of any previous behavioral study on this topic. We find that human responses are more consistent with our framework for task decomposition than alternative normative accounts and are most consistent with a heuristic-betweenness centrality-that is justified by our approach. Taken together, our results suggest the computational cost of planning is a key principle guiding the intelligent structuring of goal-directed behavior.

Pushing the Bounds of Bounded Optimality and Rationality

All forms of cognition, whether natural or artificial, are subject to constraints of their computing architecture. This assumption forms the tenet of virtually all general theories of cognition, including those deriving from bounded optimality and bounded rationality. In this letter, we highlight an unresolved puzzle related to this premise: what are these constraints, and why are cognitive architectures subject to cognitive constraints in the first place? First, we lay out some pieces along the puzzle edge, such as computational tradeoffs inherent to neural architectures that give rise to rational bounds of cognition. We then outline critical next steps for characterizing cognitive bounds, proposing that some of these bounds can be subject to modification by cognition and, as such, are part of what is being optimized when cognitive agents decide how to allocate cognitive resources. We conclude that these emerging views may contribute to a more holistic perspective on the nature of cognitive bounds, as well as their alteration subject to cognition.

Goals, usefulness and abstraction in value-based choice

Colombian drug lord Pablo Escobar, while on the run, purportedly burned two million dollars in banknotes to keep his daughter warm. A stark reminder that, in life, circumstances and goals can quickly change, forcing us to reassess and modify our values on-the-fly. Studies in decision-making and neuroeconomics have often implicitly equated value to reward, emphasising the hedonic and automatic aspect of the value computation, while overlooking its functional (concept-like) nature. Here we outline the computational and biological principles that enable the brain to compute the usefulness of an option or action by creating abstractions that flexibly adapt to changing goals. We present different algorithmic architectures, comparing ideas from artificial intelligence (AI) and cognitive neuroscience with psychological theories and, when possible, drawing parallels.

Compositional Sequence Generation in the Entorhinal-Hippocampal System

Neurons in the medial entorhinal cortex exhibit multiple, periodically organized, firing fields which collectively appear to form an internal representation of space. Neuroimaging data suggest that this grid coding is also present in other cortical areas such as the prefrontal cortex, indicating that it may be a general principle of neural functionality in the brain. In a recent analysis through the lens of dynamical systems theory, we showed how grid coding can lead to the generation of a diversity of empirically observed sequential reactivations of hippocampal place cells corresponding to traversals of cognitive maps. Here, we extend this sequence generation model by describing how the synthesis of multiple dynamical systems can support compositional cognitive computations. To empirically validate the model, we simulate two experiments demonstrating compositionality in space or in time during sequence generation. Finally, we describe several neural network architectures supporting various types of compositionality based on grid coding and highlight connections to recent work in machine learning leveraging analogous techniques.

A space of goals: the cognitive geometry of informationally bounded agents

Traditionally, Euclidean geometry is treated by scientists as a priori and objective. However, when we take the position of an agent, the problem of selecting a best route should also factor in the abilities of the agent, its embodiment and particularly its cognitive effort. In this paper, we consider geometry in terms of travel between states within a world by incorporating information processing costs with the appropriate spatial distances. This induces a geometry that increasingly differs from the original geometry of the given world as information costs become increasingly important. We visualize this 'cognitive geometry' by projecting it onto two- and three-dimensional spaces showing distinct distortions reflecting the emergence of epistemic and information-saving strategies as well as pivot states. The analogies between traditional cost-based geometries and those induced by additional informational costs invite a generalization of the notion of geodesics as cheapest routes towards the notion of infodesics. In this perspective, the concept of infodesics is inspired by the property of geodesics that, travelling from a given start location to a given goal location along a geodesic, not only the goal, but all points along the way are visited at optimal cost from the start.

Planning with Theory of Mind

Understanding Theory of Mind should begin with an analysis of the problems it solves. The traditional answer is that Theory of Mind is used for predicting others' thoughts and actions. However, the same Theory of Mind is also used for planning to change others' thoughts and actions. Planning requires that Theory of Mind consists of abstract structured causal representations and supports efficient search and selection from innumerable possible actions. Theory of Mind contrasts with less cognitively demanding alternatives: statistical predictive models of other people's actions, or model-free reinforcement of actions by their effects on other people. Theory of Mind is likely used to plan novel interventions and predict their effects, for example, in pedagogy, emotion regulation, and impression management.