Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning

The role of action concepts in physical reasoning: insights from late childhood

A fundamental component of human cognition is the ability to intuitively reason about behaviours of objects and systems in the physical world without resorting to explicit scientific knowledge. This skill was traditionally considered a symbolic process. However, in the last decades, there has been a shift towards ideas of embodiment, suggesting that accessing physical knowledge and predicting physical outcomes is grounded in bodily interactions with the environment. Infants and children, who learn mainly through their embodied experiences, serve as a model to probe the link between reasoning and physical concepts. Here, we tested school-aged children (5- to 15-year-olds) in online reasoning games that involve different physical action concepts such as supporting, launching and clearing. We assessed changes in children's performance and strategies over development and their relationships with the different action concepts. Children reasoned more accurately in problems that involved supporting actions compared to launching or clearing actions. Moreover, when children failed, they were more strategic in subsequent attempts when problems involved support rather than launching or clearing. Children improved with age, but improvements differed across action concepts. Our findings suggest that accessing physical knowledge and predicting physical events are affected by action concepts, and those effects change over development. This article is part of the theme issue 'Minds in movement: embodied cognition in the age of artificial intelligence'.

Transmission Versus Truth, Imitation Versus Innovation: What Children Can Do That Large Language and Language-and-Vision Models Cannot (Yet)

Much discussion about large language models and language-and-vision models has focused on whether these models are intelligent agents. We present an alternative perspective. First, we argue that these artificial intelligence (AI) models are cultural technologies that enhance cultural transmission and are efficient and powerful imitation engines. Second, we explore what AI models can tell us about imitation and innovation by testing whether they can be used to discover new tools and novel causal structures and contrasting their responses with those of human children. Our work serves as a first step in determining which particular representations and competences, as well as which kinds of knowledge or skill, can be derived from particular learning techniques and data. In particular, we explore which kinds of cognitive capacities can be enabled by statistical analysis of large-scale linguistic data. Critically, our findings suggest that machines may need more than large-scale language and image data to allow the kinds of innovation that a small child can produce.

Fundamental processes in sensorimotor learning: Reasoning, refinement, and retrieval

Motor learning is often viewed as a unitary process that operates outside of conscious awareness. This perspective has led to the development of sophisticated models designed to elucidate the mechanisms of implicit sensorimotor learning. In this review, we argue for a broader perspective, emphasizing the contribution of explicit strategies to sensorimotor learning tasks. Furthermore, we propose a theoretical framework for motor learning that consists of three fundamental processes: reasoning, the process of understanding action-outcome relationships; refinement, the process of optimizing sensorimotor and cognitive parameters to achieve motor goals; and retrieval, the process of inferring the context and recalling a control policy. We anticipate that this '3R' framework for understanding how complex movements are learned will open exciting avenues for future research at the intersection between cognition and action.

Repeated rock, paper, scissors play reveals limits in adaptive sequential behavior

How do people adapt to others in adversarial settings? Prior work has shown that people often violate rational models of adversarial decision-making in repeated interactions. In particular, in mixed strategy equilibrium (MSE) games, where optimal action selection entails choosing moves randomly, people often do not play randomly, but instead try to outwit their opponents. However, little is known about the adaptive reasoning that underlies these deviations from random behavior. Here, we examine strategic decision-making across repeated rounds of rock, paper, scissors, a well-known MSE game. In experiment 1, participants were paired with bot opponents that exhibited distinct stable move patterns, allowing us to identify the bounds of the complexity of opponent behavior that people can detect and adapt to. In experiment 2, bot opponents instead exploited stable patterns in the human participants' moves, providing a symmetrical bound on the complexity of patterns people can revise in their own behavior. Across both experiments, people exhibited a robust and flexible attention to transition patterns from one move to the next, exploiting these patterns in opponents and modifying them strategically in their own moves. However, their adaptive reasoning showed strong limitations with respect to more sophisticated patterns. Together, results provide a precise and consistent account of the surprisingly limited scope of people's adaptive decision-making in this setting.

Lifelong learning of cognitive styles for physical problem-solving: The effect of embodied experience

'Embodied cognition' suggests that our bodily experiences broadly shape our cognitive capabilities. We study how embodied experience affects the abstract physical problem-solving styles people use in a virtual task where embodiment does not affect action capabilities. We compare how groups with different embodied experience - 25 children and 35 adults with congenital limb differences versus 45 children and 40 adults born with two hands - perform this task, and find that while there is no difference in overall competence, the groups use different cognitive styles to find solutions. People born with limb differences think more before acting but take fewer attempts to reach solutions. Conversely, development affects the particular actions children use, as well as their persistence with their current strategy. Our findings suggest that while development alters action choices and persistence, differences in embodied experience drive changes in the acquisition of cognitive styles for balancing acting with thinking.

Using games to understand the mind

Board, card or video games have been played by virtually every individual in the world. Games are popular because they are intuitive and fun. These distinctive qualities of games also make them ideal for studying the mind. By being intuitive, games provide a unique vantage point for understanding the inductive biases that support behaviour in more complex, ecological settings than traditional laboratory experiments. By being fun, games allow researchers to study new questions in cognition such as the meaning of 'play' and intrinsic motivation, while also supporting more extensive and diverse data collection by attracting many more participants. We describe the advantages and drawbacks of using games relative to standard laboratory-based experiments and lay out a set of recommendations on how to gain the most from using games to study cognition. We hope this Perspective will lead to a wider use of games as experimental paradigms, elevating the ecological validity, scale and robustness of research on the mind.

A stochastic world model on gravity for stability inference

The fact that objects without proper support will fall to the ground is not only a natural phenomenon, but also common sense in mind. Previous studies suggest that humans may infer objects' stability through a world model that performs mental simulations with a priori knowledge of gravity acting upon the objects. Here we measured participants' sensitivity to gravity to investigate how the world model works. We found that the world model on gravity was not a faithful replica of the physical laws, but instead encoded gravity's vertical direction as a Gaussian distribution. The world model with this stochastic feature fit nicely with participants' subjective sense of objects' stability and explained the illusion that taller objects are perceived as more likely to fall. Furthermore, a computational model with reinforcement learning revealed that the stochastic characteristic likely originated from experience-dependent comparisons between predictions formed by internal simulations and the realities observed in the external world, which illustrated the ecological advantage of stochastic representation in balancing accuracy and speed for efficient stability inference. The stochastic world model on gravity provides an example of how a priori knowledge of the physical world is implemented in mind that helps humans operate flexibly in open-ended environments.

Large-scale citizen science reveals predictors of sensorimotor adaptation

Sensorimotor adaptation is essential for keeping our movements well calibrated in response to changes in the body and environment. For over a century, researchers have studied sensorimotor adaptation in laboratory settings that typically involve small sample sizes. While this approach has proved useful for characterizing different learning processes, laboratory studies are not well suited for exploring the myriad of factors that may modulate human performance. Here, using a citizen science website, we collected over 2,000 sessions of data on a visuomotor rotation task. This unique dataset has allowed us to replicate, reconcile and challenge classic findings in the learning and memory literature, as well as discover unappreciated demographic constraints associated with implicit and explicit processes that support sensorimotor adaptation. More generally, this study exemplifies how a large-scale exploratory approach can complement traditional hypothesis-driven laboratory research in advancing sensorimotor neuroscience.

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.

Object geometry serves humans' intuitive physics of stability

How do humans judge physical stability? A prevalent account emphasizes the mental simulation of physical events implemented by an intuitive physics engine in the mind. Here we test the extent to which the perceptual features of object geometry are sufficient for supporting judgments of falling direction. In all experiments, adults and children judged the falling direction of a tilted object and, across experiments, objects differed in the geometric features (i.e., geometric centroid, object height, base size and/or aspect ratio) relevant to the judgment. Participants' performance was compared to computational models trained on geometric features, as well as a deep convolutional neural network (ResNet-50), none of which incorporated mental simulation. Adult and child participants' performance was well fit by models of object geometry, particularly the geometric centroid. ResNet-50 also provided a good account of human performance. Altogether, our findings suggest that object geometry may be sufficient for judging the falling direction of tilted objects, independent of mental simulation.

Consistency and Variation in Reasoning About Physical Assembly

The ability to reason about how things were made is a pervasive aspect of how humans make sense of physical objects. Such reasoning is useful for a range of everyday tasks, from assembling a piece of furniture to making a sandwich and knitting a sweater. What enables people to reason in this way even about novel objects, and how do people draw upon prior experience with an object to continually refine their understanding of how to create it? To explore these questions, we developed a virtual task environment to investigate how people come up with step-by-step procedures for recreating block towers whose composition was not readily apparent, and analyzed how the procedures they used to build them changed across repeated attempts. Specifically, participants (N = 105) viewed 2D silhouettes of eight unique block towers in a virtual environment simulating rigid-body physics, and aimed to reconstruct each one in less than 60 s. We found that people built each tower more accurately and quickly across repeated attempts, and that this improvement reflected both group-level convergence upon a tiny fraction of all possible viable procedures, as well as error-dependent updating across successive attempts by the same individual. Taken together, our study presents a scalable approach to measuring consistency and variation in how people infer solutions to physical assembly problems.

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.

Active inductive inference in children and adults: A constructivist perspective

A defining aspect of being human is an ability to reason about the world by generating and adapting ideas and hypotheses. Here we explore how this ability develops by comparing children's and adults' active search and explicit hypothesis generation patterns in a task that mimics the open-ended process of scientific induction. In our experiment, 54 children (aged 8.97±1.11) and 50 adults performed inductive inferences about a series of causal rules through active testing. Children were more elaborate in their testing behavior and generated substantially more complex guesses about the hidden rules. We take a 'computational constructivist' perspective to explaining these patterns, arguing that these inferences are driven by a combination of thinking (generating and modifying symbolic concepts) and exploring (discovering and investigating patterns in the physical world). We show how this framework and rich new dataset speak to questions about developmental differences in hypothesis generation, active learning and inductive generalization. In particular, we find children's learning is driven by less fine-tuned construction mechanisms than adults', resulting in a greater diversity of ideas but less reliable discovery of simple explanations.

Universality, domain-specificity, and development of psychological responses to music

Humans can find music happy, sad, fearful, or spiritual. They can be soothed by it or urged to dance. Whether these psychological responses reflect cognitive adaptations that evolved expressly for responding to music is an ongoing topic of study. In this Review, we examine three features of music-related psychological responses that help to elucidate whether the underlying cognitive systems are specialized adaptations: universality, domain-specificity, and early expression. Focusing on emotional and behavioural responses, we find evidence that the relevant psychological mechanisms are universal and arise early in development. However, the existing evidence cannot establish that these mechanisms are domain-specific. To the contrary, many findings suggest that universal psychological responses to music reflect more general properties of emotion, auditory perception, and other human cognitive capacities that evolved for non-musical purposes. Cultural evolution, driven by the tinkering of musical performers, evidently crafts music to compellingly appeal to shared psychological mechanisms, resulting in both universal patterns (such as form-function associations) and culturally idiosyncratic styles.

Phy-Q as a measure for physical reasoning intelligence

Humans are well versed in reasoning about the behaviours of physical objects and choosing actions accordingly to accomplish tasks, while this remains a major challenge for artificial intelligence. To facilitate research addressing this problem, we propose a new testbed that requires an agent to reason about physical scenarios and take an action appropriately. Inspired by the physical knowledge acquired in infancy and the capabilities required for robots to operate in real-world environments, we identify 15 essential physical scenarios. We create a wide variety of distinct task templates, and we ensure that all the task templates within the same scenario can be solved by using one specific strategic physical rule. By having such a design, we evaluate two distinct levels of generalization, namely local generalization and broad generalization. We conduct an extensive evaluation with human players, learning agents with various input types and architectures, and heuristic agents with different strategies. Inspired by how the human intelligence quotient is calculated, we define the physical reasoning quotient (Phy-Q score) that reflects the physical reasoning intelligence of an agent using the physical scenarios we considered. Our evaluation shows that (1) all the agents are far below human performance, and (2) learning agents, even with good local generalization ability, struggle to learn the underlying physical reasoning rules and fail to generalize broadly. We encourage the development of intelligent agents that can reach the human-level Phy-Q score.

Bringing cumulative technological culture beyond copying versus reasoning

The dominant view of cumulative technological culture suggests that high-fidelity transmission rests upon a high-fidelity copying ability, which allows individuals to reproduce the tool-use actions performed by others without needing to understand them (i.e., without causal reasoning). The opposition between copying versus reasoning is well accepted but with little supporting evidence. In this article, we investigate this distinction by examining the cognitive science literature on tool use. Evidence indicates that the ability to reproduce others' tool-use actions requires causal understanding, which questions the copying versus reasoning distinction and the cognitive reality of the so-called copying ability. We conclude that new insights might be gained by considering causal understanding as a key driver of cumulative technological culture.

Subjective selection and the evolution of complex culture

Why is culture the way it is? Here I argue that a major force shaping culture is subjective (cultural) selection, or the selective retention of cultural variants that people subjectively perceive as satisfying their goals. I show that people evaluate behaviors and beliefs according to how useful they are, especially for achieving goals. As they adopt and pass on those variants that seem best, they iteratively craft culture into increasingly effective-seeming forms. I argue that this process drives the development of many cumulatively complex cultural products, including effective technology, magic and ritual, aesthetic traditions, and institutions. I show that it can explain cultural dependencies, such as how certain beliefs create corresponding new practices, and I outline how it interacts with other cultural evolutionary processes. Cultural practices everywhere, from spears to shamanism, develop because people subjectively evaluate them to be effective means of satisfying regular goals.

IntPhys 2019: A Benchmark for Visual Intuitive Physics Understanding

In order to reach human performance on complex visual tasks, artificial systems need to incorporate a significant amount of understanding of the world in terms of macroscopic objects, movements, forces, etc. Inspired by work on intuitive physics in infants, we propose an evaluation benchmark which diagnoses how much a given system understands about physics by testing whether it can tell apart well matched videos of possible versus impossible events constructed with a game engine. The test requires systems to compute a physical plausibility score over an entire video. To prevent perceptual biases, the dataset is made of pixel matched quadruplets of videos, enforcing systems to focus on high level temporal dependencies between frames rather than pixel-level details. We then describe two Deep Neural Networks systems aimed at learning intuitive physics in an unsupervised way, using only physically possible videos. The systems are trained with a future semantic mask prediction objective and tested on the possible versus impossible discrimination task. The analysis of their results compared to human data gives novel insights in the potentials and limitations of next frame prediction architectures.

Invariant representation of physical stability in the human brain

Successful engagement with the world requires the ability to predict what will happen next. Here, we investigate how the brain makes a fundamental prediction about the physical world: whether the situation in front of us is stable, and hence likely to stay the same, or unstable, and hence likely to change in the immediate future. Specifically, we ask if judgments of stability can be supported by the kinds of representations that have proven to be highly effective at visual object recognition in both machines and brains, or instead if the ability to determine the physical stability of natural scenes may require generative algorithms that simulate the physics of the world. To find out, we measured responses in both convolutional neural networks (CNNs) and the brain (using fMRI) to natural images of physically stable versus unstable scenarios. We find no evidence for generalizable representations of physical stability in either standard CNNs trained on visual object and scene classification (ImageNet), or in the human ventral visual pathway, which has long been implicated in the same process. However, in frontoparietal regions previously implicated in intuitive physical reasoning we find both scenario-invariant representations of physical stability, and higher univariate responses to unstable than stable scenes. These results demonstrate abstract representations of physical stability in the dorsal but not ventral pathway, consistent with the hypothesis that the computations underlying stability entail not just pattern classification but forward physical simulation.

Physical understanding in neurodegenerative diseases

This quantitative review gives an overview of physical understanding (i.e., the ability to represent and use the laws of physics to interact with the physical world) impairments in Alzheimer's disease (AD), semantic dementia (SD), and corticobasal syndrome (CBS), as assessed mainly with mechanical problem-solving and tool use tests. This review shows that: (1) SD patients have apraxia of tool use because of semantic tool knowledge deficits, but normal performance in tests of physical understanding; (2) AD and CBS patients show impaired performance in mechanical problem-solving tests, probably not because of intrinsic deficits of physical understanding, but rather because of additional cognitive (AD) or motor impairments (CBS); (3) As a result, the performance in mechanical problem-solving tests is not a good predictor of familiar tool use in dementia; (4) Actual deficits of physical understanding are probably observed only in late stages of neurodegenerative diseases, and associated with functional loss.

Naturalistic embodied interactions elicit intuitive physical behaviour in accordance with Newtonian physics

The success of visuomotor interactions in everyday activities such as grasping or sliding a cup is inescapably governed by the laws of physics. Research on intuitive physics has predominantly investigated reasoning about objects' behaviour involving binary forced choice responses. We investigated how the type of visuomotor response influences participants' beliefs about physical quantities and their lawful relationship implicit in their active behaviour. Participants propelled pucks towards targets positioned at different distances. Analysis with a probabilistic model of interactions showed that subjects adopted the non-linear control prescribed by Newtonian physics when sliding real pucks in a virtual environment even in the absence of visual feedback. However, they used a linear heuristic when viewing the scene on a monitor and interactions were implemented through key presses. These results support the notion of probabilistic internal physics models but additionally suggest that humans can take advantage of embodied, sensorimotor, multimodal representations in physical scenarios.

World model learning and inference

Understanding information processing in the brain-and creating general-purpose artificial intelligence-are long-standing aspirations of scientists and engineers worldwide. The distinctive features of human intelligence are high-level cognition and control in various interactions with the world including the self, which are not defined in advance and are vary over time. The challenge of building human-like intelligent machines, as well as progress in brain science and behavioural analyses, robotics, and their associated theoretical formalisations, speaks to the importance of the world-model learning and inference. In this article, after briefly surveying the history and challenges of internal model learning and probabilistic learning, we introduce the free energy principle, which provides a useful framework within which to consider neuronal computation and probabilistic world models. Next, we showcase examples of human behaviour and cognition explained under that principle. We then describe symbol emergence in the context of probabilistic modelling, as a topic at the frontiers of cognitive robotics. Lastly, we review recent progress in creating human-like intelligence by using novel probabilistic programming languages. The striking consensus that emerges from these studies is that probabilistic descriptions of learning and inference are powerful and effective ways to create human-like artificial intelligent machines and to understand intelligence in the context of how humans interact with their world.

Capturing the objects of vision with neural networks

Human visual perception carves a scene at its physical joints, decomposing the world into objects, which are selectively attended, tracked and predicted as we engage our surroundings. Object representations emancipate perception from the sensory input, enabling us to keep in mind that which is out of sight and to use perceptual content as a basis for action and symbolic cognition. Human behavioural studies have documented how object representations emerge through grouping, amodal completion, proto-objects and object files. By contrast, deep neural network models of visual object recognition remain largely tethered to sensory input, despite achieving human-level performance at labelling objects. Here, we review related work in both fields and examine how these fields can help each other. The cognitive literature provides a starting point for the development of new experimental tasks that reveal mechanisms of human object perception and serve as benchmarks driving the development of deep neural network models that will put the object into object recognition.

Characterising factors underlying praxis deficits in chronic left hemisphere stroke patients

Limb apraxia, a disorder of skilled action not consequent on primary motor or sensory deficits, has traditionally been defined according to errors patients make on neuropsychological tasks. Previous models of the disorder have failed to provide a unified account of patients' deficits, due to heterogeneity in the patients and tasks used. In this study we hypothesised that we may be able to map apraxic deficits onto principal components, some of which may be specific, whilst others may align with other cognitive disorders. We implemented principal component analysis (PCA) to elucidate core factors of the disorder in a preliminary cohort of 41 unselected left hemisphere chronic stroke patients who were tested on a comprehensive and validated apraxia screen. Three principal components were identified: posture selection, semantic control and multi-demand sequencing. These were submitted to a lesion symptom mapping (VBCM) analysis in a subset of 24 patients, controlled for lesion volume, age and time post-stroke. The first component revealed no significant structural correlates. The second component was related to regions in inferior frontal gyrus, primary motor area, and adjacent parietal opercular (including inferior parietal and supramarginal gyrus) areas. The third component was associated with lesions within the white matter underlying the left sensorimotor cortex, likely involving the 2nd branch of the left superior longitudinal fasciculus as well as the posterior orbitofrontal cortex (pOFC). These results highlight a significant role of common cognitive functions in apraxia, which include action selection, and sequencing, whilst more specific deficits may relate to semantic control. Moreover, they suggest that previously described 'ideomotor' and 'ideational' deficits may have a common neural basis within semantic control. Further research using this technique would help elucidate the cognitive processes underlying limb apraxia, its neural correlates and their relationship with other cognitive disorders.

The Ouroboros Model, Proposal for Self-Organizing General Cognition Substantiated

The Ouroboros Model has been proposed as a biologically-inspired comprehensive cognitive architecture for general intelligence, comprising natural and artificial manifestations. The approach addresses very diverse fundamental desiderata of research in natural cognition and also artificial intelligence, AI. Here, it is described how the postulated structures have met with supportive evidence over recent years. The associated hypothesized processes could remedy pressing problems plaguing many, and even the most powerful current implementations of AI, including in particular deep neural networks. Some selected recent findings from very different fields are summoned, which illustrate the status and substantiate the proposal.