%0 Journal Article %J psyArXiv %D 2022 %T An approximate representation of objects underlies physical reasoning %A Yichen Li %A YingQiao Wang %A Tal Boger %A Kevin A Smith %A Samuel J Gershman %A Tomer Ullman %X

People make fast and reasonable predictions about the physical behavior of everyday objects. To do so, people may be using principled approximations, similar to models developed by engineers for the purposes of real-time physical simulations. We hypothesize that people use simplified object approximations for tracking and action (the "body" representation), as opposed to fine-grained forms for recognition (the "shape" representation). We used three classic psychophysical tasks (causality perception, collision detection, and change detection) in novel settings that dissociate body and shape. People's behavior across tasks indicates that they rely on approximate bodies for physical reasoning, and that this approximation lies between convex hulls and fine-grained shapes.

%B psyArXiv %8 03/2022 %G eng %U https://psyarxiv.com/vebu5/ %0 Journal Article %J Attention, Perception, & Psychophysics %D 2021 %T Confidence and central tendency in perceptual judgment %A Xiang, Yang %A Graeber, Thomas %A Enke, Benjamin %A Samuel J Gershman %X

This paper theoretically and empirically investigates the role of noisy cognition in perceptual judgment, focusing on the central tendency effect: the well-known empirical regularity that perceptual judgments are biased towards the center of the stimulus distribution. Based on a formal Bayesian framework, we generate predictions about the relationships between subjective confidence, central tendency, and response variability. Specifically, our model clarifies that lower subjective confidence as a measure of posterior uncertainty about a judgment should predict (i) a lower sensitivity of magnitude estimates to objective stimuli; (ii) a higher sensitivity to the mean of the stimulus distribution; (iii) a stronger central tendency effect at higher stimulus magnitudes; and (iv) higher response variability. To test these predictions, we collect a large-scale experimental data set and additionally re-analyze perceptual judgment data from several previous experiments. Across data sets, subjective confidence is strongly predictive of the central tendency effect and response variability, both correlationally and when we exogenously manipulate the magnitude of sensory noise. Our results are consistent with (but not necessarily uniquely explained by) Bayesian models of confidence and the central tendency.

%B Attention, Perception, & Psychophysics %V 83 %P 3024 - 3034 %8 J10/2021 %G eng %U https://link.springer.com/10.3758/s13414-021-02300-6 %N 7 %! Atten Percept Psychophys %R 10.3758/s13414-021-02300-6 %0 Journal Article %J Nature Neuroscience %D 2021 %T Flexible modulation of sequence generation in the entorhinal-hippocampal system %A McNamee, D. %A Stachenfeld, K. %A Botvinick, M.M. %A Samuel J Gershman %X

Exploration, consolidation and planning depend on the generation of sequential state representations. However, these algorithms require disparate forms of sampling dynamics for optimal performance. We theorize how the brain should adapt internally generated sequences for particular cognitive functions and propose a neural mechanism by which this may be accomplished within the entorhinal–hippocampal circuit. Specifically, we demonstrate that the systematic modulation along the medial entorhinal cortex dorsoventral axis of grid population input into the hippocampus facilitates a flexible generative process that can interpolate between qualitatively distinct regimes of sequential hippocampal reactivations. By relating the emergent hippocampal activity patterns drawn from our model to empirical data, we explain and reconcile a diversity of recently observed, but apparently unrelated, phenomena such as generative cycling, diffusive hippocampal reactivations and jumping trajectory events.

%B Nature Neuroscience %8 04/2021 %G eng %U https://www.nature.com/articles/s41593-021-00831-7 %R 10.1038/s41593-021-00831-7 %0 Journal Article %J Scientific Reports %D 2021 %T Human visual motion perception shows hallmarks of Bayesian structural inference %A Yang, Sichao %A Bill, Johannes %A Drugowitsch, Jan %A Samuel J Gershman %X

Motion relations in visual scenes carry an abundance of behaviorally relevant information, but little is known about how humans identify the structure underlying a scene’s motion in the first place. We studied the computations governing human motion structure identification in two psychophysics experiments and found that perception of motion relations showed hallmarks of Bayesian structural inference. At the heart of our research lies a tractable task design that enabled us to reveal the signatures of probabilistic reasoning about latent structure. We found that a choice model based on the task’s Bayesian ideal observer accurately matched many facets of human structural inference, including task performance, perceptual error patterns, single-trial responses, participant-specific differences, and subjective decision confidence—especially, when motion scenes were ambiguous and when object motion was hierarchically nested within other moving reference frames. Our work can guide future neuroscience experiments to reveal the neural mechanisms underlying higher-level visual motion perception.

%B Scientific Reports %V 11 %8 02/2021 %G eng %U http://www.nature.com/articles/s41598-021-82175-7 %N 1 %! Sci Rep %R 10.1038/s41598-021-82175-7 %0 Journal Article %J Trends in Cognitive Sciences %D 2021 %T Memory as a Computational Resource %A Ishita Dasgupta %A Samuel J Gershman %K amortization %K inference %K memory %K mental arithmetic %K mental imagery %K planning %X

Most computations that people do in everyday life are very expensive. Recent research highlights that humans make efficient use of their limited computational resources to tackle these problems. Memory is a crucial aspect of algorithmic efficiency and permits the reuse of past computation through memoization. We review neural and behavioral evidence of humans reusing past computations across several domains, including mental imagery, arithmetic, planning, and probabilistic inference. Recent developments in neural networks expand the scope of computational reuse with a distributed form of memoization called amortization. This opens many new avenues of research. Computer scientists have long recognized that naive implementations of algorithms often result in a paralyzing degree of redundant computation. More sophisticated implementations harness the power of memory by storing computational results and reusing them later. We review the application of these ideas to cognitive science, in four case studies (mental arithmetic, mental imagery, planning, and probabilistic inference). Despite their superficial differences, these cognitive processes share a common reliance on memory that enables efficient computation.

%B Trends in Cognitive Sciences %V 25 %P 240 - 251 %8 03/2021 %G eng %U https://linkinghub.elsevier.com/retrieve/pii/S1364661320303053 %N 3 %! Trends in Cognitive Sciences %R 10.1016/j.tics.2020.12.008 %0 Journal Article %J Cognition %D 2021 %T Moral dynamics: Grounding moral judgment in intuitive physics and intuitive psychology %A Sosa, Felix A. %A Ullman, Tomer %A Joshua B. Tenenbaum %A Samuel J Gershman %A Gerstenberg, Tobias %X

When holding others morally responsible, we care about what they did, and what they thought. Traditionally, research in moral psychology has relied on vignette studies, in which a protagonist’s actions and thoughts are explicitly communicated. While this research has revealed what variables are important for moral judgment, such as actions and intentions, it is limited in providing a more detailed understanding of exactly how these variables affect moral judgment. Using dynamic visual stimuli that allow for a more fine-grained experimental control, recent studies have proposed a direct mapping from visual features to moral judgments. We embrace the use of visual stimuli in moral psychology, but question the plausibility of a feature-based theory of moral judgment. We propose that the connection from visual features to moral judgments is mediated by an inference about what the observed action reveals about the agent’s mental states, and what causal role the agent’s action played in bringing about the outcome. We present a computational model that formalizes moral judgments of agents in visual scenes as computations over an intuitive theory of physics combined with an intuitive theory of mind. We test the model’s quantitative predictions in three experiments across a wide variety of dynamic interactions between agent and patient.

%B Cognition %V 217 %P 104890 %8 05/2021 %G eng %U https://linkinghub.elsevier.com/retrieve/pii/S0010027721003139 %! Cognition %R 10.1016/j.cognition.2021.104890 %0 Journal Article %J Nature Human Behaviour %D 2021 %T Multi-task reinforcement learning in humans %A Tomov, Momchil S. %A Eric Schulz %A Samuel J Gershman %X

The ability to transfer knowledge across tasks and generalize to novel ones is an important hallmark of human intelligence. Yet not much is known about human multitask reinforcement learning. We study participants’ behaviour in a two-step decision-making task with multiple features and changing reward functions. We compare their behaviour with two algorithms for multitask reinforcement learning, one that maps previous policies and encountered features to new reward functions and one that approximates value functions across tasks, as well as to standard model-based and model-free algorithms. Across three exploratory experiments and a large preregistered confirmatory experiment, our results provide evidence that partici-pants who are able to learn the task use a strategy that maps previously learned policies to novel scenarios. These results enrich our understanding of human reinforcement learning in complex environments with changing task demands.

%B Nature Human Behaviour %8 01/2021 %G eng %U http://www.nature.com/articles/s41562-020-01035-y %! Nat Hum Behav %R 10.1038/s41562-020-01035-y %0 Journal Article %J Cognitive Science %D 2021 %T What Is the Model in Model‐Based Planning? %A Pouncy, Thomas %A Tsividis, Pedro %A Samuel J Gershman %X

Flexibility is one of the hallmarks of human problem-solving. In everyday life, people adapt to changes in common tasks with little to no additional training. Much of the existing work on flexibility in human problem-solving has focused on how people adapt to tasks in new domains by drawing on solutions from previously learned domains. In real-world tasks, however, humans must generalize across a wide range of within-domain variation. In this work we argue that representational abstraction plays an important role in such within-domain generalization. We then explore the nature of this representational abstraction in realistically complex tasks like video games by demonstrating how the same model-based planning framework produces distinct generalization behaviors under different classes of task representation. Finally, we compare the behavior of agents with these task representations to humans in a series of novel grid-based video game tasks. Our results provide evidence for the claim that within-domain flexibility in humans derives from task representations composed of propositional rules written in terms of objects and relational categories.

%B Cognitive Science %V 45 %8 01/2021 %G eng %U https://onlinelibrary.wiley.com/toc/15516709/45/1 %N 1 %! Cogn Sci %R 10.1111/cogs.v45.110.1111/cogs.12928 %0 Journal Article %J Cognitive Science %D 2020 %T Analyzing Machine‐Learned Representations: A Natural Language Case Study %A Dasgupta, Ishita %A Guo, Demi %A Samuel J Gershman %A Goodman, Noah D. %X

As modern deep networks become more complex, and get closer to human‐like capabilities in certain domains, the question arises as to how the representations and decision rules they learn compare to the ones in humans. In this work, we study representations of sentences in one such artificial system for natural language processing. We first present a diagnostic test dataset to examine the degree of abstract composable structure represented. Analyzing performance on these diagnostic tests indicates a lack of systematicity in representations and decision rules, and reveals a set of heuristic strategies. We then investigate the effect of training distribution on learning these heuristic strategies, and we study changes in these representations with various augmentations to the training set. Our results reveal parallels to the analogous representations in people. We find that these systems can learn abstract rules and generalize them to new contexts under certain circumstances—similar to human zero‐shot reasoning. However, we also note some shortcomings in this generalization behavior—similar to human judgment errors like belief bias. Studying these parallels suggests new ways to understand psychological phenomena in humans as well as informs best strategies for building artificial intelligence with human‐like language understanding.

%B Cognitive Science %V 44 %8 12/2020 %G eng %U https://onlinelibrary.wiley.com/toc/15516709/44/12 %N 12 %! Cogn Sci %R 10.1111/cogs.12925 %0 Journal Article %J Open Mind %D 2020 %T Communicating Compositional Patterns %A Schulz, Eric %A Quiroga, Francisco %A Samuel J Gershman %B Open Mind %V 4 %P 25 - 39 %8 08/2020 %G eng %U https://direct.mit.edu/opmi/article/95939 %! Open Mind %R 10.1162/opmi_a_00032 %0 Journal Article %J Proceedings of the National Academy of Sciences %D 2020 %T Hierarchical structure is employed by humans during visual motion perception %A Bill, Johannes %A Pailian, Hrag %A Samuel J Gershman %A Drugowitsch, Jan %B Proceedings of the National Academy of Sciences %V 117 %P 24581 - 24589 %8 09/2022 %G eng %U http://www.pnas.org/lookup/doi/10.1073/pnas.2008961117 %N 39 %! Proc Natl Acad Sci USA %R 10.1073/pnas.2008961117 %0 Journal Article %J eLife %D 2020 %T Hippocampal remapping as hidden state inference %A Honi Sanders %A Matthew A. Wilson %A Samuel J Gershman %X

Cells in the hippocampus tuned to spatial location (place cells) typically change their tuning when an animal changes context, a phenomenon known as remapping. A fundamental challenge to understanding remapping is the fact that what counts as a ‘‘context change’’ has never been precisely defined. Furthermore, different remapping phenomena have been classified on the basis of how much the tuning changes after different types and degrees of context change, but the relationship between these variables is not clear. We address these ambiguities by formalizing remapping in terms of hidden state inference. According to this view, remapping does not directly reflect objective, observable properties of the environment, but rather subjective beliefs about the hidden state of the environment. We show how the hidden state framework can resolve a number of puzzles about the nature of remapping.

%B eLife %V 9 %8 06/2020 %G eng %U https://elifesciences.org/articles/51140 %R 10.7554/eLife.51140 %0 Journal Article %J Cognition %D 2020 %T Origin of perseveration in the trade-off between reward and complexity %A Samuel J Gershman %K Decision making %K Information theory %K reinforcement learning %X

When humans and other animals make repeated choices, they tend to repeat previously chosen actions independently of their reward history. This paper locates the origin of perseveration in a trade-off between two computational goals: maximizing rewards and minimizing the complexity of the action policy. We develop an information-theoretic formalization of policy complexity and show how optimizing the trade-off leads to perseveration. Analysis of two data sets reveals that people attain close to optimal trade-offs. Parameter estimation and model comparison supports the claim that perseveration quantitatively agrees with the theoretically predicted functional form (a softmax function with a frequency-dependent action bias).

%B Cognition %V 204 %P 104394 %8 11/2020 %G eng %U https://linkinghub.elsevier.com/retrieve/pii/S0010027720302134 %! Cognition %R 10.1016/j.cognition.2020.104394 %0 Journal Article %J Psychological Review %D 2020 %T A theory of learning to infer. %A Ishita Dasgupta %A Eric Schulz %A Joshua B. Tenenbaum %A Samuel J Gershman %X

Bayesian theories of cognition assume that people can integrate probabilities rationally. However, several empirical findings contradict this proposition: human probabilistic inferences are prone to systematic deviations from optimality. Puzzlingly, these deviations sometimes go in opposite directions. Whereas some studies suggest that people underreact to prior probabilities (base rate neglect), other studies find that people underreact to the likelihood of the data (conservatism). We argue that these deviations arise because the human brain does not rely solely on a general-purpose mechanism for approximating Bayesian inference that is invariant across queries. Instead, the brain is equipped with a recognition model that maps queries to probability distributions. The parameters of this recognition model are optimized to get the output as close as possible, on average, to the true posterior. Because of our limited computational resources, the recognition model will allocate its resources so as to be more accurate for high probability queries than for low probability queries. By adapting to the query distribution, the recognition model learns to infer. We show that this theory can explain why and when people underreact to the data or the prior, and a new experiment demonstrates that these two forms of underreaction can be systematically controlled by manipulating the query distribution. The theory also explains a range of related phenomena: memory effects, belief bias, and the structure of response variability in probabilistic reasoning. We also discuss how the theory can be integrated with prior sampling-based accounts of approximate inference.

%B Psychological Review %V 127 %P 412 - 441 %8 04/2020 %G eng %U http://doi.apa.org/getdoi.cfm?doi=10.1037/rev0000178 %N 3 %! Psychological Review %R 10.1037/rev0000178 %0 Journal Article %J Frontiers in Artificial Intelligence %D 2019 %T The Generative Adversarial Brain %A Samuel J Gershman %X

The idea that the brain learns generative models of the world has been widely promulgated. Most approaches have assumed that the brain learns an explicit density model that assigns a probability to each possible state of the world. However, explicit density models are difficult to learn, requiring approximate inference techniques that may find poor solutions. An alternative approach is to learn an implicit density model that can sample from the generative model without evaluating the probabilities of those samples. The implicit model can be trained to fool a discriminator into believing that the samples are real. This is the idea behind generative adversarial algorithms, which have proven adept at learning realistic generative models. This paper develops an adversarial framework for probabilistic computation in the brain. It first considers how generative adversarial algorithms overcome some of the problems that vex prior theories based on explicit density models. It then discusses the psychological and neural evidence for this framework, as well as how the breakdown of the generator and discriminator could lead to delusions observed in some mental disorders.

%B Frontiers in Artificial Intelligence %V 2 %8 09/2019 %G eng %U https://www.frontiersin.org/article/10.3389/frai.2019.00018/full %! Front. Artif. Intell. %R 10.3389/frai.2019.00018 %0 Conference Paper %B Cognitive Science Society %D 2019 %T Hard choices: Children’s understanding of the cost of action selection. %A Shari Liu %A Fiery A Cushman %A Samuel J Gershman %A Kool, Wouter %A Elizabeth S Spelke %B Cognitive Science Society %G eng %0 Generic %D 2019 %T Hippocampal Remapping as Hidden State Inference %A Honi Sanders %A Matthew A. Wilson %A Samuel J Gershman %X

Cells in the hippocampus tuned to spatial location (place cells) typically change their tuning when an animal changes context, a phenomenon known as remapping. A fundamental challenge to understanding remapping is the fact that what counts as a “context change” has never been precisely defined. Furthermore, different remapping phenomena have been classified on the basis of how much the tuning changes after different types and degrees of context change, but the relationship between these variables is not clear. We address these ambiguities by formalizing remapping in terms of hidden state inference. According to this view, remapping does not directly reflect objective, observable properties of the environment, but rather subjective beliefs about the hidden state of the environment. We show how the hidden state framework can resolve a number of puzzles about the nature of remapping.

%8 08/2019 %1

https://www.biorxiv.org/content/10.1101/743260v1

%2

https://hdl.handle.net/1721.1/122040

%R https://doi.org/10.1101/743260 %0 Journal Article %J Psychonomic Bulletin & Review %D 2019 %T How to never be wrong %A Samuel J Gershman %X

Human beliefs have remarkable robustness in the face of disconfirmation. This robustness is often explained as the product of heuristics or motivated reasoning. However, robustness can also arise from purely rational principles when the reasoner has recourse to ad hoc auxiliary hypotheses. Auxiliary hypotheses primarily function as the linking assumptions connecting different beliefs to one another and to observational data, but they can also function as a "protective belt" that explains away disconfirmation by absorbing some of the blame. The present article traces the role of auxiliary hypotheses from philosophy of science to Bayesian models of cognition and a host of behavioral phenomena, demonstrating their wide-ranging implications.

%B Psychonomic Bulletin & Review %V 26 %P 13 - 28 %8 02/2019 %G eng %U http://link.springer.com/10.3758/s13423-018-1488-8 %N 1 %! Psychon Bull Rev %R 10.3758/s13423-018-1488-8 %0 Journal Article %J Biological Psychiatry %D 2019 %T Incentives Boost Model-Based Control Across a Range of Severity on Several Psychiatric Constructs %A Patzelt, Edward H. %A Kool, Wouter %A Millner, Alexander J. %A Samuel J Gershman %K Computational psychiatry %K Habits and goals Incentives %K Model-based control %K Psychiatric constructs %K reinforcement learning %X

Background

Human decision making exhibits a mixture of model-based and model-free control. Recent evidence indicates that arbitration between these two modes of control (“metacontrol”) is based on their relative costs and benefits. While model-based control may increase accuracy, it requires greater computational resources, so people invoke model-based control only when potential rewards exceed those of model-free control. We used a sequential decision task, while concurrently manipulating performance incentives, to ask if symptoms and traits of psychopathology decrease or increase model-based control in response to incentives.

Methods

We recruited a nonpatient population of 839 online participants using Amazon Mechanical Turk who completed transdiagnostic self-report measures encompassing symptoms, traits, and factors. We fit a dual-controller reinforcement learning model and obtained a computational measure of model-based control separately for small incentives and large incentives.

Results

None of the constructs were related to a failure of large incentives to boost model-based control. In fact, for the sensation seeking trait and anxious-depression factor, higher scores were associated with a larger incentive effect, whereby greater levels of these constructs were associated with larger increases in model-based control. Many constructs showed decreases in model-based control as a function of severity, but a social withdrawal factor was positively correlated; alcohol use and social anxiety were unrelated to model-based control.

Conclusions

Our results demonstrate that model-based control can reliably be improved independent of construct severity for most measures. This suggests that incentives may be a useful intervention for boosting model-based control across a range of symptom and trait severity.

%B Biological Psychiatry %V 85 %P 425 - 433 %8 03/2019 %G eng %U https://linkinghub.elsevier.com/retrieve/pii/S0006322318316329 %N 5 %! Biological Psychiatry %R 10.1016/j.biopsych.2018.06.018 %0 Journal Article %J Scientific Reports %D 2019 %T The transdiagnostic structure of mental effort avoidance %A Patzelt, Edward H. %A Kool, Wouter %A Millner, Alexander J. %A Samuel J Gershman %X

The law of least mental effort states that, everything else being equal, the brain tries to minimize mental effort expenditure during task performance by avoiding decisions that require greater cognitive demands. Prior studies have shown associations between disruptions in effort expenditure and specific psychiatric illnesses (e.g., schizophrenia and depression) or clinically-related symptoms and traits (e.g., anhedonia and apathy), yet no research has explored this issue transdiagnostically. Specifically, this research has largely focused on a single diagnostic category, symptom, or trait. However, abnormalities in effort expression could be related to several different psychiatrically-relevant constructs that cut across diagnostic boundaries. Therefore, we examined the relationship between avoidance of mental effort and a diverse set of clinically-related symptoms and traits, and transdiagnostic latent factors in a large sample (n = 811). Only lack of perseverance, a dimension of impulsiveness, was associated with increased avoidance of mental effort. In contrast, several constructs were associated with less mental effort avoidance, including positive urgency, distress intolerance, obsessive-compulsive symptoms, disordered eating, and a factor consisting of compulsive behavior and intrusive thoughts. These findings demonstrate that deviations from normative effort expenditure are associated with a number of constructs that are common to several forms of psychiatric illness.

%B Scientific Reports %V 9 %8 02/2019 %G eng %U http://www.nature.com/articles/s41598-018-37802-1 %N 1 %! Sci Rep %R 10.1038/s41598-018-37802-1 %0 Journal Article %J Journal of Cognitive Neuroscience %D 2018 %T Planning Complexity Registers as a Cost in Metacontrol %A Kool, Wouter %A Samuel J Gershman %A Fiery A Cushman %X

Decision-making algorithms face a basic tradeoff between accuracy and effort (i.e., computational demands). It is widely agreed that humans can choose between multiple decision-making processes that embody different solutions to this tradeoff: Some are computationally cheap but inaccurate, whereas others are computationally expensive but accurate. Recent progress in understanding this tradeoff has been catalyzed by formalizing it in terms of model-free (i.e., habitual) versus model-based (i.e., planning) approaches to reinforcement learning. Intuitively, if two tasks offer the same rewards for accuracy but one of them is much more demanding, we might expect people to rely on habit more in the difficult task: Devoting significant computation to achieve slight marginal accuracy gains would not be "worth it." We test and verify this prediction in a sequential reinforcement learning task. Because our paradigm is amenable to formal analysis, it contributes to the development of a computational model of how people balance the costs and benefits of different decision-making processes in a task-specific manner; in other words, how we decide when hard thinking is worth it.

%B Journal of Cognitive Neuroscience %V 30 %P 1391 - 1404 %8 10/2018 %G eng %U https://www.mitpressjournals.org/doi/abs/10.1162/jocn_a_01263 %N 10 %! Journal of Cognitive Neuroscience %R 10.1162/jocn_a_01263 %0 Journal Article %J Behavioral and Brain Sciences %D 2017 %T Building machines that learn and think like people. %A Brenden M Lake %A Ullman, Tomer D %A Joshua B. Tenenbaum %A Samuel J Gershman %X

Recent progress in artificial intelligence has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats that of humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn and how they learn it. Specifically, we argue that these machines should (1) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (2) ground learning in intuitive theories of physics and psychology to support and enrich the knowledge that is learned; and (3) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes toward these goals that can combine the strengths of recent neural network advances with more structured cognitive models.

%B Behavioral and Brain Sciences %V 40 %P e253 %8 2017 Jan %G eng %U https://www.cambridge.org/core/journals/behavioral-and-brain-sciences/article/building-machines-that-learn-and-think-like-people/A9535B1D745A0377E16C590E14B94993/core-reader %( Published online: 24 November 2016 %R https://doi.org/10.1017/S0140525X16001837 %0 Journal Article %J Cogn Psychol %D 2017 %T Compositional inductive biases in function learning. %A Eric Schulz %A Joshua B. Tenenbaum %A David Duvenaud %A Maarten Speekenbrink %A Samuel J Gershman %X

How do people recognize and learn about complex functional structure? Taking inspiration from other areas of cognitive science, we propose that this is achieved by harnessing compositionality: complex structure is decomposed into simpler building blocks. We formalize this idea within the framework of Bayesian regression using a grammar over Gaussian process kernels, and compare this approach with other structure learning approaches. Participants consistently chose compositional (over non-compositional) extrapolations and interpolations of functions. Experiments designed to elicit priors over functional patterns revealed an inductive bias for compositional structure. Compositional functions were perceived as subjectively more predictable than non-compositional functions, and exhibited other signatures of predictability, such as enhanced memorability and reduced numerosity. Taken together, these results support the view that the human intuitive theory of functions is inherently compositional.

%B Cogn Psychol %V 99 %P 44-79 %8 2017 Dec %G eng %U https://www.sciencedirect.com/science/article/pii/S0010028517301743?via%3Dihub %R 10.1016/j.cogpsych.2017.11.002 %0 Journal Article %J Psychol Sci %D 2017 %T Cost-Benefit Arbitration Between Multiple Reinforcement-Learning Systems. %A Kool, Wouter %A Samuel J Gershman %A Fiery A Cushman %X

Human behavior is sometimes determined by habit and other times by goal-directed planning. Modern reinforcement-learning theories formalize this distinction as a competition between a computationally cheap but inaccurate model-free system that gives rise to habits and a computationally expensive but accurate model-based system that implements planning. It is unclear, however, how people choose to allocate control between these systems. Here, we propose that arbitration occurs by comparing each system's task-specific costs and benefits. To investigate this proposal, we conducted two experiments showing that people increase model-based control when it achieves greater accuracy than model-free control, and especially when the rewards of accurate performance are amplified. In contrast, they are insensitive to reward amplification when model-based and model-free control yield equivalent accuracy. This suggests that humans adaptively balance habitual and planned action through on-line cost-benefit analysis.

%B Psychol Sci %V 28 %P 1321-1333 %8 2017 Sep %G eng %N 9 %R 10.1177/0956797617708288 %0 Conference Paper %B AAAI Spring Symposium Series %D 2017 %T Human Learning in Atari %A Pedro Tsividis %A Thomas Pouncy %A Jacqueline L. Xu %A Joshua B. Tenenbaum %A Samuel J Gershman %X

Atari games are an excellent testbed for studying intelligent behavior, as they offer a range of tasks that differ widely in their visual representation, game dynamics, and goals presented to an agent. The last two years have seen a spate of research into artificial agents that use a single algorithm to learn to play these games. The best of these artificial agents perform at better-than-human levels on most games, but require hundreds of hours of game-play experience to produce such behavior. Humans, on the other hand, can learn to perform well on these tasks in a matter of minutes. In this paper we present data on human learning trajectories for several Atari games, and test several hypotheses about the mechanisms that lead to such rapid learning. 

%B AAAI Spring Symposium Series %G eng %0 Journal Article %J Annual Review of Psychology %D 2017 %T Reinforcement learning and episodic memory in humans and animals: an integrative framework %A Samuel J Gershman %A Nathaniel D Daw %B Annual Review of Psychology %V 68 %G eng %& 101 %0 Generic %D 2017 %T Thinking fast or slow? A reinforcement-learning approach %A Kool, W %A Samuel J Gershman %A Fiery A Cushman %B Society for Personality and Social Psychology %C San Antonio, TX %0 Generic %D 2016 %T Building machines that learn and think like people %A Brenden M Lake %A Tomer Ullman %A Joshua B. Tenenbaum %A Samuel J Gershman %X

Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn, and how they learn it. Specifically, we argue that these machines should (a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning in intuitive theories of physics and psychology, to support and enrich the knowledge that is learned; and (c) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes towards these goals that can combine the strengths of recent neural network advances with more structured cognitive models.

%8 04/2016 %1

arXiv:1604.00289

%2

http://hdl.handle.net/1721.1/102089

%0 Generic %D 2016 %T Probing the compositionality of intuitive functions %A Eric Schulz %A Joshua B. Tenenbaum %A David Duvenaud %A Maarten Speekenbrink %A Samuel J Gershman %X

How do people learn about complex functional structure? Taking inspiration from other areas of cognitive science, we propose that this is accomplished by harnessing compositionality: complex structure is decomposed into simpler building blocks. We formalize this idea within the framework of Bayesian regression using a grammar over Gaussian process kernels. We show that participants prefer compositional over non-compositional function extrapolations, that samples from the human prior over functions are best described by a compositional model, and that people perceive compositional functions as more predictable than their non-compositional but otherwise similar counterparts. We argue that the compositional nature of intuitive functions is consistent with broad principles of human cognition.

%8 05/2016 %2

http://hdl.handle.net/1721.1/102698

%0 Journal Article %J PLoS Comput Biol %D 2016 %T When Does Model-Based Control Pay Off? %A Kool, Wouter %A Fiery A Cushman %A Samuel J Gershman %X

Many accounts of decision making and reinforcement learning posit the existence of two distinct systems that control choice: a fast, automatic system and a slow, deliberative system. Recent research formalizes this distinction by mapping these systems to "model-free" and "model-based" strategies in reinforcement learning. Model-free strategies are computationally cheap, but sometimes inaccurate, because action values can be accessed by inspecting a look-up table constructed through trial-and-error. In contrast, model-based strategies compute action values through planning in a causal model of the environment, which is more accurate but also more cognitively demanding. It is assumed that this trade-off between accuracy and computational demand plays an important role in the arbitration between the two strategies, but we show that the hallmark task for dissociating model-free and model-based strategies, as well as several related variants, do not embody such a trade-off. We describe five factors that reduce the effectiveness of the model-based strategy on these tasks by reducing its accuracy in estimating reward outcomes and decreasing the importance of its choices. Based on these observations, we describe a version of the task that formally and empirically obtains an accuracy-demand trade-off between model-free and model-based strategies. Moreover, we show that human participants spontaneously increase their reliance on model-based control on this task, compared to the original paradigm. Our novel task and our computational analyses may prove important in subsequent empirical investigations of how humans balance accuracy and demand.

%B PLoS Comput Biol %V 12 %P e1005090 %8 2016 Aug %G eng %N 8 %R 10.1371/journal.pcbi.1005090 %0 Generic %D 2016 %T Where do hypotheses come from? %A Ishita Dasgupta %A Eric Schulz %A Samuel J Gershman %X

Why are human inferences sometimes remarkably close to the Bayesian ideal and other times systematically biased? One notable instance of this discrepancy is that tasks where the candidate hypotheses are explicitly available result in close to rational inference over the hypothesis space, whereas tasks requiring the self-generation of hypotheses produce systematic deviations from rational inference. We propose that these deviations arise from algorithmic processes approximating Bayes' rule. Specifically in our account, hypotheses are generated stochastically from a sampling process, such that the sampled hypotheses form a Monte Carlo approximation of the posterior. While this approximation will converge to the true posterior in the limit of infinite samples, we take a small number of samples as we expect that the number of samples humans take is limited by time pressure and cognitive resource constraints. We show that this model recreates several well-documented experimental findings such as anchoring and adjustment, subadditivity, superadditivity, the crowd within as well as the self-generation effect, the weak evidence, and the dud alternative effects. Additionally, we confirm the model's prediction that superadditivity and subadditivity can be induced within the same paradigm by manipulating the unpacking and typicality of hypotheses, in 2 experiments.

%8 10/2016 %2

http://hdl.handle.net/1721.1/105158

%0 Journal Article %J Science %D 2015 %T Computational rationality: A converging paradigm for intelligence in brains, minds, and machines %A Samuel J Gershman %A Eric J. Horvitz %A Joshua B. Tenenbaum %X
After growing up together, and mostly growing apart in the second half of the 20th century,the fields of artificial intelligence (AI), cognitive science, and neuroscience arereconverging on a shared view of the computational foundations of intelligence thatpromotes valuable cross-disciplinary exchanges on questions, methods, and results. We chart advances over the past several decades that address challenges of perceptionand action under uncertainty through the lens of computation. Advances include thedevelopment of representations and inferential procedures for large-scale probabilisticinference and machinery for enabling reflection and decisions about tradeoffs in effort, precision, and timeliness of computations. These tools are deployed toward the goal of computational rationality: identifying decisions with highest expected utility, while taking into consideration the costs of computation in complex real-world problems inwhich most relevant calculations can only be approximated. We highlight key concepts with examples that show the potential for interchange between computer science, cognitive science, and neuroscience.
%B Science %V 349 %P 273-278 %8 07/17/2015 %G eng %U http://www.sciencemag.org/content/349/6245/273.abstract %N 6245 %9 Review; Special Section: Artificial Intelligence %R 10.1126/science.aac6076 %0 Journal Article %J Vision Research %D 2015 %T Discovering hierarchical motion structure %A Samuel J Gershman %A Joshua B. Tenenbaum %A Frank Jaekel %X

Scenes filled with moving objects are often hierarchically organized: the motion of a migrating goose is nested within the flight pattern of its flock, the motion of a car is nested within the traffic pattern of other cars on the road, the motion of body parts are nested in the motion of the body. Humans perceive hierarchical structure even in stimuli with two or three moving dots. An influential theory of hierarchical motion perception holds that the visual system performs a "vector analysis" of moving objects, decomposing them into common and relative motions. However, this theory does not specify how to resolve ambiguity when a scene admits more than one vector analysis. We describe a Bayesian theory of vector analysis and show that it can account for classic results from dot motion experiments, as well as new experimental data. Our theory takes a step towards understanding how moving scenes are parsed into objects.

 

 

%B Vision Research %V Available online 26 March 2015 %8 03/2015 %G eng %U http://www.sciencedirect.com/science/article/pii/S0042698915000814#sthash.vpJfuWmr.dpuf %R doi:10.1016/j.visres.2015.03.004 %0 Generic %D 2015 %T Information Selection in Noisy Environments with Large Action Spaces %A Pedro Tsividis %A Samuel J Gershman %A Joshua B. Tenenbaum %A Laura Schulz %B 9th Biennial Conference of the Cognitive Development Society %V Columbus, OH %G eng