A rational account of pedagogical reasoning: Teaching by, and learning from, examples

Children's developing ability to recognize deceptive use of true information

The strategic use of deliberate omissions, conveying true but selective information for deceptive purposes, is a prevalent and pernicious disinformation tactic. Crucially, its recognition requires engaging in a sophisticated, multi-part social cognitive reasoning process. In two preregistered studies, we investigated the development of children's ability to engage in this process and successfully recognize this form of deception, finding that children even as young as 5 years are capable of doing so, but only with sufficient scaffolding. This work highlights the key role that social cognition plays in the ability to recognize the manipulation techniques that underpin disinformation. It suggests that the interrelated development of pragmatic competence and epistemic vigilance can be harnessed in the design of tools and strategies to help bolster psychological resistance against disinformation in even our youngest citizens-children at the outset of formal education.

A Hierarchical Bayesian Model of Adaptive Teaching

How do teachers learn about what learners already know? How do learners aid teachers by providing them with information about their background knowledge and what they find confusing? We formalize this collaborative reasoning process using a hierarchical Bayesian model of pedagogy. We then evaluate this model in two online behavioral experiments (N = 312 adults). In Experiment 1, we show that teachers select examples that account for learners' background knowledge, and adjust their examples based on learners' feedback. In Experiment 2, we show that learners strategically provide more feedback when teachers' examples deviate from their background knowledge. These findings provide a foundation for extending computational accounts of pedagogy to richer interactive settings.

Teaching Without Thinking: Negative Evaluations of Rote Pedagogy

When people make decisions, they act in a way that is either automatic ("rote"), or more thoughtful ("reflective"). But do people notice when others are behaving in a rote way, and do they care? We examine the detection of rote behavior and its consequences in U.S. adults, focusing specifically on pedagogy and learning. We establish repetitiveness as a cue for rote behavior (Experiment 1), and find that rote people are seen as worse teachers (Experiment 2). We also find that the more a person's feedback seems similar across groups (indicating greater rote-ness), the more negatively their teaching is evaluated (Experiment 3). A word-embedding analysis of an open-response task shows people naturally cluster rote and reflective teachers into different semantic categories (Experiment 4). We also show that repetitiveness can be decoupled from perceptions of rote-ness given contextual explanation (Experiment 5). Finally, we establish two additional cues to rote behavior that can be tied to quality of teaching (Experiment 6). These results empirically show that people detect and care about scripted behaviors in pedagogy, and suggest an important extension to formal frameworks of social reasoning.

Imagination vs. routines: festive time, weekly time, and the predictive brain

This paper examines the relationship between societal structures shaped by traditions, norms, laws, and customs, and creative expressions in arts and media through the lens of the predictive coding framework in cognitive science. The article proposes that both dimensions of culture can be viewed as adaptations designed to enhance and train the brain's predictive abilities in the social domain. Traditions, norms, laws, and customs foster shared predictions and expectations among individuals, thereby reducing uncertainty in social environments. On the other hand, arts and media expose us to simulated experiences that explore alternative social realities, allowing the predictive machinery of the brain to hone its skills through exposure to a wider array of potentially relevant social circumstances and scenarios. We first review key principles of predictive coding and active inference, and then explore the rationale of cultural traditions and artistic culture in this perspective. Finally, we draw parallels between institutionalized normative habits that stabilize social worlds and creative and imaginative acts that temporarily subvert established conventions to inject variability.

The Collaborative Nature of Testimonial Learning

Children's testimonial learning often occurs in epistemic collaborations with others. In this paper, we will discuss ways in which cultural learning emerges in social and interpersonal contexts, and is intrinsically supported and guided by children's collaborative capacities. Much work in cultural learning has focused on children's examination of speaker and model characteristics, but more recent research has investigated the interactive aspects of testimonial exchanges. We will review evidence that children (1) participate in the interpersonal commitments that are shared in testimonial transactions by way of direct address and epistemic buck passing, (2) participate in social groups that affect their selective learning in nuanced ways, and (3) may detect epistemic harms by listeners who refuse to believe sincere and accurate speakers. Implications for conceptualizing children's testimonial learning as an interactive mechanism of collaboration will be discussed.

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.

Storytelling as Inverse Inverse Planning

Great storytelling takes us on a journey the way ordinary reality rarely does. But what exactly do we mean by this "journey?" Recently, literary theorist Karin Kukkonen proposed that storytelling is "probability design:" the art of giving an audience pieces of information bit by bit, to craft the journey of their changing beliefs about the fictional world. A good "probability design" choreographs a delicate dance of certainty and surprise in the reader's mind as the story unfolds from beginning to end. In this paper, we computationally model this conception of storytelling. Building on the classic Bayesian inverse planning model of human social cognition, we treat storytelling as inverse inverse planning: the task of choosing actions to manipulate an inverse planner's inferences, and therefore a human audience's beliefs. First, we use an inverse inverse planner to depict social and physical situations, and present behavioral studies indicating that inverse inverse planning produces more expressive behavior than ordinary "naïve planning." Then, through a series of examples, we demonstrate how inverse inverse planning captures many storytelling elements from first principles: character, narrative arcs, plot twists, irony, flashbacks, and deus ex machina are all naturally encoded in the flexible language of probability design.

Children Use Teachers' Beliefs About Their Abilities to Calibrate Explore-Exploit Decisions

Models of the explore-exploit problem have explained how children's decision making is weighed by a bias for information (directed exploration), randomness, and generalization. These behaviors are often tested in domains where a choice to explore (or exploit) is guaranteed to reveal an outcome. An often overlooked but critical component of the assessment of explore-exploit decisions lies in the expected success of taking actions in the first place-and, crucially, how such decisions might be carried out when learning from others. Here, we examine how children consider an informal teacher's beliefs about the child's competence when deciding how difficult a task they want to pursue. We present a simple model of this problem that predicts that while learners should follow the recommendation of an accurate teacher, they should exploit easier games when a teacher overestimates their abilities, and explore harder games when she underestimates them. We tested these predictions in two experiments with adults (Experiment 1) and 6- to 8-year-old children (Experiment 2). In our task, participants' performance on a picture-matching game was either overestimated, underestimated, or accurately represented by a confederate (the "Teacher"), who then presented three new matching games of varying assessed difficulty (too easy, too hard, just right) at varying potential reward (low, medium, high). In line with our model's predictions, we found that both adults and children calibrated their choices to the teacher's representation of their competence. That is, to maximize expected reward, when she underestimated them, participants chose games the teacher evaluated as being too hard for them; when she overestimated them, they chose games she evaluated as being too easy; and when she was accurate, they chose games she assessed as being just right. This work provides insight into the early-emerging ability to calibrate explore-exploit decisions to others' knowledge when learning in informal pedagogical contexts.

People can use the placement of objects to infer communicative goals

Beyond words and gestures, people have a remarkable capacity to communicate indirectly through everyday objects: A hat on a chair can mean it is occupied, rope hanging across an entrance can mean we should not cross, and objects placed in a closed box can imply they are not ours to take. How do people generate and interpret the communicative meaning of objects? We hypothesized that this capacity is supported by social goal inference, where observers recover what social goal explains an object being placed in a particular location. To test this idea, we study a category of common ad-hoc communicative objects where a small cost is used to signal avoidance. Using computational modeling, we first show that goal inference from indirect physical evidence can give rise to the ability to use object placement to communicate. We then show that people from the U.S. and the Tsimane'-a farming-foraging group native to the Bolivian Amazon-can infer the communicative meaning of object placement in the absence of a pre-existing convention, and that people's inferences are quantitatively predicted by our model. Finally, we show evidence that people can store and retrieve this meaning for use in subsequent encounters, revealing a potential mechanism for how ad-hoc communicative objects become quickly conventionalized. Our model helps shed light on how humans use their ability to interpret other people's behavior to embed social meaning into the physical world.

Teaching categories via examples and explanations

People often learn categories through interaction with knowledgeable others who may use verbal explanations, visual exemplars, or both, to share their knowledge. Verbal and nonverbal means of pedagogical communication are commonly used in conjunction, but their respective roles are not fully understood. In this work, we studied how well these modes of communication work with different category structures. We conducted two experiments to investigate the effect of perceptual confusability and stimulus dimensionality on the effectiveness of verbal, exemplar-based, and mixed communication. One group of participants - teachers - learned a categorization rule and prepared learning materials for the students. Students studied the materials prepared for them and then demonstrated their knowledge on test stimuli. All communication modes were generally successful, but not equivalent, with mixed communication consistently showing best results. When teachers were free to generate as many visual exemplars or words as they wish, verbal and exemplar-based communication showed similar performance, although the verbal channel was slightly less reliable in situations requiring high perceptual precision. At the same time, verbal communication was better suited to handling high-dimensional stimuli when communication volume was restricted. We believe that our work serves as an important step towards studying language as a means for pedagogical category leaning.

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.

Socially intelligent machines that learn from humans and help humans learn

A hallmark of human intelligence is the ability to understand and influence other minds. Humans engage in inferential social learning (ISL) by using commonsense psychology to learn from others and help others learn. Recent advances in artificial intelligence (AI) are raising new questions about the feasibility of human-machine interactions that support such powerful modes of social learning. Here, we envision what it means to develop socially intelligent machines that can learn, teach, and communicate in ways that are characteristic of ISL. Rather than machines that simply predict human behaviours or recapitulate superficial aspects of human sociality (e.g. smiling, imitating), we should aim to build machines that can learn from human inputs and generate outputs for humans by proactively considering human values, intentions and beliefs. While such machines can inspire next-generation AI systems that learn more effectively from humans (as learners) and even help humans acquire new knowledge (as teachers), achieving these goals will also require scientific studies of its counterpart: how humans reason about machine minds and behaviours. We close by discussing the need for closer collaborations between the AI/ML and cognitive science communities to advance a science of both natural and artificial intelligence. This article is part of a discussion meeting issue 'Cognitive artificial intelligence'.

Teachers recruit mentalizing regions to represent learners' beliefs

Teaching enables humans to impart vast stores of culturally specific knowledge and skills. However, little is known about the neural computations that guide teachers' decisions about what information to communicate. Participants (N = 28) played the role of teachers while being scanned using fMRI; their task was to select examples that would teach learners how to answer abstract multiple-choice questions. Participants' examples were best described by a model that selects evidence that maximizes the learner's belief in the correct answer. Consistent with this idea, participants' predictions about how well learners would do closely tracked the performance of an independent sample of learners (N = 140) who were tested on the examples they had provided. In addition, regions that play specialized roles in processing social information, namely the bilateral temporoparietal junction and middle and dorsal medial prefrontal cortex, tracked learners' posterior belief in the correct answer. Our results shed light on the computational and neural architectures that support our extraordinary abilities as teachers.

Visual explanations prioritize functional properties at the expense of visual fidelity

Visual explanations play an integral role in communicating mechanistic knowledge about how things work. What do people think distinguishes such pictures from those that are intended to convey how things look? To explore this question, we used a drawing paradigm to elicit both visual explanations and depictions of novel machine-like objects, then conducted a detailed analysis of the semantic information conveyed in each drawing. We found that visual explanations placed greater emphasis on parts of the machines that move or interact to produce an effect, while visual depictions emphasized parts that were visually salient, even if they were static. Moreover, we found that these differences in visual emphasis impacted what information naive viewers could extract from these drawings: explanations made it easier to infer which action was needed to operate the machine, but more difficult to identify which machine it represented. Taken together, our findings suggest that people spontaneously prioritize functional information when producing visual explanations but that this strategy may be double-edged, facilitating inferences about physical mechanism at the expense of preserving visual fidelity.

Show or tell? Exploring when (and why) teaching with language outperforms demonstration

People use a wide range of communicative acts across different modalities, from concrete demonstrations to abstract language. While these modalities are typically studied independently, we take a comparative approach and ask when and why one modality might outperform another. We present a series of real-time, multi-player experiments asking participants to teach concepts using either demonstrations or language. Our first experiment (N=416) asks when language might outperform demonstration. We manipulate the complexity of the concept being taught and find that language communicates complex concepts more effectively than demonstration. We then ask why language succeeds in this setting. We hypothesized that language allowed teachers to reference abstract object features (e.g., shapes and colors), while demonstration teachers could only provide concrete examples (specific positive or negative objects). To test this hypothesis, our second experiment (N=568) ablated object features from the teacher's interface. This manipulation severely impaired linguistic (but not demonstrative) teaching. Our findings suggest that language communicates complex concepts by directly transmitting abstract rules. In contrast, demonstrations transmit examples, requiring the learner to infer the rules.

When Naïve Pedagogy Breaks Down: Adults Rationally Decide How to Teach, but Misrepresent Learners' Beliefs

From early in childhood, humans exhibit sophisticated intuitions about how to share knowledge efficiently in simple controlled studies. Yet, untrained adults often fail to teach effectively in real-world situations. Here, we explored what causes adults to struggle in informal pedagogical exchanges. In Experiment 1, we first showed evidence of this effect, finding that adult participants failed to communicate their knowledge to naïve learners in a simple teaching task, despite reporting high confidence that they taught effectively. Using a computational model of rational teaching, we found that adults assigned to our teaching condition provided highly informative examples but failed to teach effectively because their examples were tailored to learners who were only considering a small set of possible explanations. In Experiment 2, we then found experimental evidence for this possibility, showing that knowledgeable participants systematically misunderstand the beliefs of naïve participants. Specifically, knowledgeable participants assumed naïve agents would primarily consider hypotheses close to the correct one. Finally, in Experiment 3, we aligned learners' beliefs to knowledgeable agents' expectations and showed learners the same examples selected by participants assigned to teach in Experiment 1. We found that these same examples were significantly more informative once learners' hypothesis spaces were constrained to match teachers' expectations. Our findings show that, in informal settings, adult pedagogical failures result from an inaccurate representation of what naïve learners believe is plausible and not an inability to select informative data in a rational way.

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.

A Pragmatic Account of the Weak Evidence Effect

Language is not only used for neutral information; we often seek to persuade by arguing in favor of a particular view. Persuasion raises a number of challenges for classical accounts of belief updating, as information cannot be taken at face value. How should listeners account for a speaker’s “hidden agenda” when incorporating new information? Here, we extend recent probabilistic models of recursive social reasoning to allow for persuasive goals and show that our model provides a pragmatic account for why weakly favorable arguments may backfire, a phenomenon known as the weak evidence effect. Critically, this model predicts a systematic relationship between belief updates and expectations about the information source: weak evidence should only backfire when speakers are expected to act under persuasive goals and prefer the strongest evidence. We introduce a simple experimental paradigm called the Stick Contest to measure the extent to which the weak evidence effect depends on speaker expectations, and show that a pragmatic listener model accounts for the empirical data better than alternative models. Our findings suggest further avenues for rational models of social reasoning to illuminate classical decision-making phenomena.

In situ bidirectional human-robot value alignment

A prerequisite for social coordination is bidirectional communication between teammates, each playing two roles simultaneously: as receptive listeners and expressive speakers. For robots working with humans in complex situations with multiple goals that differ in importance, failure to fulfill the expectation of either role could undermine group performance due to misalignment of values between humans and robots. Specifically, a robot needs to serve as an effective listener to infer human users’ intents from instructions and feedback and as an expressive speaker to explain its decision processes to users. Here, we investigate how to foster effective bidirectional human-robot communications in the context of value alignment—collaborative robots and users form an aligned understanding of the importance of possible task goals. We propose an explainable artificial intelligence (XAI) system in which a group of robots predicts users’ values by taking in situ feedback into consideration while communicating their decision processes to users through explanations. To learn from human feedback, our XAI system integrates a cooperative communication model for inferring human values associated with multiple desirable goals. To be interpretable to humans, the system simulates human mental dynamics and predicts optimal explanations using graphical models. We conducted psychological experiments to examine the core components of the proposed computational framework. Our results show that real-time human-robot mutual understanding in complex cooperative tasks is achievable with a learning model based on bidirectional communication. We believe that this interaction framework can shed light on bidirectional value alignment in communicative XAI systems and, more broadly, in future human-machine teaming systems.

Leveraging artificial intelligence to improve people's planning strategies

SignificanceMany bad decisions and their devastating consequences could be avoided if people used optimal decision strategies. Here, we introduce a principled computational approach to improving human decision making. The basic idea is to give people feedback on how they reach their decisions. We develop a method that leverages artificial intelligence to generate this feedback in such a way that people quickly discover the best possible decision strategies. Our empirical findings suggest that a principled computational approach leads to improvements in decision-making competence that transfer to more difficult decisions in more complex environments. In the long run, this line of work might lead to apps that teach people clever strategies for decision making, reasoning, goal setting, planning, and goal achievement.

Knowledge in Sight: Toddlers Plan Efficient Epistemic Actions by Anticipating Learning Gains

Anticipating the learning consequences of actions is crucial to plan efficient information seeking. Such a capacity is needed for learners to determine which actions are most likely to result in learning. Here, we tested the early ontogeny of the human capacity to anticipate the amount of learning gained from seeing. In study 1, we tested infants' capacity to anticipate the availability of sight. Fourteen-month-old infants (N = 72) were invited to search for a toy hidden inside a container. The participants were faster to attempt at opening a shutter when this action allowed them to see inside the container. Moreover, this effect was specifically observed when seeing inside the container was potentially useful to the participants' goals. Thus, infants anticipated the availability of sight, and they calibrated their information-seeking behaviors accordingly. In studies 2 and 3, we tested toddlers' capacity to anticipate whether data would be cognitively useful for their goals. Two-and-a-half-year-olds (N = 72) had to locate a target character hidden among distractors. The participants flipped the characters more often, and were comparatively faster to initiate this action when it yielded access to visual data allowing them to locate the target. Thus, toddlers planned their information-seeking behaviors by anticipating the cognitive utility of sight. In contrast, toddlers did not calibrate their behaviors to the cognitive usefulness of auditory data. These results suggest that cognitive models of learning guide toddlers' search for information. The early developmental onset of the capacity to anticipate future learning gains is crucial for active learning.

One model for the learning of language

A major goal of linguistics and cognitive science is to understand what class of learning systems can acquire natural language. Until recently, the computational requirements of language have been used to argue that learning is impossible without a highly constrained hypothesis space. Here, we describe a learning system that is maximally unconstrained, operating over the space of all computations, and is able to acquire many of the key structures present in natural language from positive evidence alone. We demonstrate this by providing the same learning model with data from 74 distinct formal languages which have been argued to capture key features of language, have been studied in experimental work, or come from an interesting complexity class. The model is able to successfully induce the latent system generating the observed strings from small amounts of evidence in almost all cases, including for regular (e.g., aⁿ , [Formula: see text], and [Formula: see text]), context-free (e.g., [Formula: see text], and [Formula: see text]), and context-sensitive (e.g., [Formula: see text], and xx) languages, as well as for many languages studied in learning experiments. These results show that relatively small amounts of positive evidence can support learning of rich classes of generative computations over structures. The model provides an idealized learning setup upon which additional cognitive constraints and biases can be formalized.

A cognitive-evolutionary model for the evolution of teaching

Mechanisms for social learning have rightly been the focus of much work in cultural evolution. But mechanisms for teaching-mechanisms that determine what information is available for learners to learn in the first place-are equally important to cultural evolution, especially in the case of humans. Here, we propose a simple model of teaching in the context of skill transmission. Our model derives the evolutionary cost and benefit of teaching by explicitly representing cognitive aspects of skill transmission as a dual-inheritance process. We then show that teaching cannot evolve when its direct cost is too high. We also show that there is an "explain-exploit" trade-off inherent to teaching: when payoffs from sharing information are not constant, there can be an indirect cost to teaching. This gives rise to an opportunity cost that goes beyond any direct cost that it may also entail. Finally, we show that evolution limits the strength of teaching provided that the direct cost of teaching is an increasing function of teaching effort. We then discuss how these factors might explain why teaching mechanisms are self-limiting, suggesting that such mechanisms may nevertheless play an important role in the evolution of cumulative culture in humans.

Causal theory error in college students' understanding of science studies

When reasoning about science studies, people often make causal theory errors by inferring or accepting a causal claim based on correlational evidence. While humans naturally think in terms of causal relationships, reasoning about science findings requires understanding how evidence supports—or fails to support—a causal claim. This study investigated college students’ thinking about causal claims presented in brief media reports describing behavioral science findings. How do science students reason about causal claims from correlational evidence? And can their reasoning be improved through instruction clarifying the nature of causal theory error? We examined these questions through a series of written reasoning exercises given to advanced college students over three weeks within a psychology methods course. In a pretest session, students critiqued study quality and support for a causal claim from a brief media report  suggesting an association between two variables. Then, they created diagrams depicting possible alternative causal theories. At the beginning of the second session, an instructional intervention introduced students to an extended example of a causal theory error through guided questions about possible alternative causes. Then, they completed the same two tasks with new science reports immediately and again 1 week later. The results show students’ reasoning included fewer causal theory errors after the intervention, and this improvement was maintained a week later. Our findings suggest that interventions aimed at addressing reasoning about causal claims in correlational studies are needed even for advanced science students, and that training on considering alternative causal theories may be successful in reducing casual theory error.

The effects of information utility and teachers' knowledge on evaluations of under-informative pedagogy across development

Teaching is a powerful way to transmit knowledge, but with this power comes a hazard: When teachers fail to select the best set of evidence for the learner, learners can be misled to draw inaccurate inferences. Evaluating others' failures as teachers, however, is a nontrivial problem; people may fail to be informative for different reasons, and not all failures are equally blameworthy. How do learners evaluate the quality of teachers, and what factors influence such evaluations? Here, we present a Bayesian model of teacher evaluation that considers the utility of a teacher's pedagogical sampling given their prior knowledge. In Experiment 1 (N=1168), we test the model predictions against adults' evaluations of a teacher who demonstrated all or a subset of the functions on a novel device. Consistent with the model predictions, participants' ratings integrated information about the number of functions taught, their values, as well as how much the teacher knew. Using a modified paradigm for children, Experiments 2 (N=48) and 3 (N=40) found that preschool-aged children (2a, 3) and adults (2b) make nuanced judgments of teacher quality that are well predicted by the model. However, after an unsuccessful attempt to replicate the results with preschoolers (Experiment 4, N=24), in Experiment 5 (N=24) we further investigate the development of teacher evaluation in a sample of seven- and eight-year-olds. These older children successfully distinguished teachers based on the amount and value of what was demonstrated, and their ability to evaluate omissions relative to the teacher's knowledge state was related to their tendency to spontaneously reference the teacher's knowledge when explaining their evaluations. In sum, our work illustrates how the human ability to learn from others supports not just learning about the world but also learning about the teachers themselves. By reasoning about others' informativeness, learners can evaluate others' teaching and make better learning decisions.

Expression unleashed: The evolutionary and cognitive foundations of human communication

Human expression is open-ended, versatile, and diverse, ranging from ordinary language use to painting, from exaggerated displays of affection to micro-movements that aid coordination. Here we present and defend the claim that this expressive diversity is united by an interrelated suite of cognitive capacities, the evolved functions of which are the expression and recognition of informative intentions. We describe how evolutionary dynamics normally leash communication to narrow domains of statistical mutual benefit, and how expression is unleashed in humans. The relevant cognitive capacities are cognitive adaptations to living in a partner choice social ecology; and they are, correspondingly, part of the ordinarily developing human cognitive phenotype, emerging early and reliably in ontogeny. In other words, we identify distinctive features of our species' social ecology to explain how and why humans, and only humans, evolved the cognitive capacities that, in turn, lead to massive diversity and open-endedness in means and modes of expression. Language use is but one of these modes of expression, albeit one of manifestly high importance. We make cross-species comparisons, describe how the relevant cognitive capacities can evolve in a gradual manner, and survey how unleashed expression facilitates not only language use, but also novel behaviour in many other domains too, focusing on the examples of joint action, teaching, punishment, and art, all of which are ubiquitous in human societies but relatively rare in other species. Much of this diversity derives from graded aspects of human expression, which can be used to satisfy informative intentions in creative and new ways. We aim to help reorient cognitive pragmatics, as a phenomenon that is not a supplement to linguistic communication and on the periphery of language science, but rather the foundation of the many of the most distinctive features of human behaviour, society, and culture.

Quantitative mental state attributions in language understanding

Human social intelligence relies on our ability to infer other people’s mental states such as their beliefs, desires, and intentions. While people are proficient at mental state inference from physical action, it is unknown whether people can make inferences of comparable granularity from simple linguistic events. Here, we show that people can make quantitative mental state attributions from simple referential expressions, replicating the fine-grained inferential structure characteristic of nonlinguistic theory of mind. Moreover, people quantitatively adjust these inferences after brief exposures to speaker-specific speech patterns. These judgments matched the predictions made by our computational model of theory of mind in language, but could not be explained by a simpler qualitative model that attributes mental states deductively. Our findings show how the connection between language and theory of mind runs deep, with their interaction showing in one of the most fundamental forms of human communication: reference.

The Intentional Selection Assumption

There exists a rich literature describing how social context influences decision making. Here, we propose a novel framing of social influences, the Intentional Selection Assumption. This framework proposes that, when a person is presented with a set of options by another social agent, people may treat the set of options as intentionally selected, reflecting the chooser's inferences about the presenter and the presenter's goals. To describe our proposal, we draw analogies to the cognition literature on sampling inferences within concept learning. This is done to highlight how the Intentional Selection Assumption accounts for both normative (e.g., comparing perceived utilities) and subjective (e.g., consideration of context relevance) principles in decision making, while also highlighting how analogous findings in the concept learning literature can aid in bridging these principles by drawing attention to the importance of potential sampling assumptions within decision making paradigms. We present the two behavioral experiments that provide support to this proposal and find that social-contextual cues influence choice behavior with respect to the induction of sampling assumptions. We then discuss a theoretical framework of the Intentional Selection Assumption alongside the possibility of its potential relationships to contemporary models of choice. Overall, our results emphasize the flexibility of decision makers with respect to social-contextual factors without sacrificing systematicity regarding the preference for specific options with a higher value or utility.

Balancing Performance and Human Autonomy With Implicit Guidance Agent

The human-agent team, which is a problem in which humans and autonomous agents collaborate to achieve one task, is typical in human-AI collaboration. For effective collaboration, humans want to have an effective plan, but in realistic situations, they might have difficulty calculating the best plan due to cognitive limitations. In this case, guidance from an agent that has many computational resources may be useful. However, if an agent guides the human behavior explicitly, the human may feel that they have lost autonomy and are being controlled by the agent. We therefore investigated implicit guidance offered by means of an agent's behavior. With this type of guidance, the agent acts in a way that makes it easy for the human to find an effective plan for a collaborative task, and the human can then improve the plan. Since the human improves their plan voluntarily, he or she maintains autonomy. We modeled a collaborative agent with implicit guidance by integrating the Bayesian Theory of Mind into existing collaborative-planning algorithms and demonstrated through a behavioral experiment that implicit guidance is effective for enabling humans to maintain a balance between improving their plans and retaining autonomy.

Cooperative communication as belief transport

Recent research formalizes cooperative communication as belief transport using the mathematical theory of optimal transport. This formalization allows rigorous a priori analysis of the statistical and ecological properties of models of cooperative communication, unification of prior models and analysis of their differences, and promising directions for future research.

Inferential social learning: cognitive foundations of human social learning and teaching

Social learning is often portrayed as a passive process of copying and trusting others. This view, however, does not fully capture what makes human social learning so powerful: social information is often 'curated' by helpful teachers. I argue that both learning from others (social learning) and helping others learn (teaching) can be characterized as probabilistic inferences guided by an intuitive understanding of how people think, plan, and act. Consistent with this idea, even young children draw rich inferences from evidence provided by others and generate informative evidence that helps others learn. By studying social learning and teaching through a common theoretical lens, inferential social learning provides an integrated account of how human cognition supports acquisition and communication of abstract knowledge.

Linking young children's teaching to their reasoning of mental states: Evidence from Singapore

To fully participate in the human information-sharing ecosystem that allows for efficient knowledge dissemination and creation, children need to be able to teach others effectively. The current research is the first to investigate links between children's teaching abilities and their developing theory of mind abilities in a non-Western sample. In a sample of 4- to 6-year-old Singaporean children (N = 49), we examined relations between specific components of theory of mind abilities and teaching ability on a social cognitive task. We found that both false belief understanding and the ability to make mental state inferences in a teaching context were associated with effective teaching even after controlling for age and language ability. These findings provide a nuanced picture of the links between mental state reasoning and teaching ability. More broadly, they provide evidence that these links extend beyond Western cultures and generalize to social-cognitive teaching contexts.

Mitigating belief projection in explainable artificial intelligence via Bayesian teaching

State-of-the-art deep-learning systems use decision rules that are challenging for humans to model. Explainable AI (XAI) attempts to improve human understanding but rarely accounts for how people typically reason about unfamiliar agents. We propose explicitly modelling the human explainee via Bayesian teaching, which evaluates explanations by how much they shift explainees' inferences toward a desired goal. We assess Bayesian teaching in a binary image classification task across a variety of contexts. Absent intervention, participants predict that the AI's classifications will match their own, but explanations generated by Bayesian teaching improve their ability to predict the AI's judgements by moving them away from this prior belief. Bayesian teaching further allows each case to be broken down into sub-examples (here saliency maps). These sub-examples complement whole examples by improving error detection for familiar categories, whereas whole examples help predict correct AI judgements of unfamiliar cases.

If Mathematical Psychology Did Not Exist We Might Need to Invent It: A Comment on Theory Building in Psychology

It is commonplace, when discussing the subject of psychological theory, to write articles from the assumption that psychology differs from the physical sciences in that we have no theories that would support cumulative, incremental science. In this brief article I discuss one counterexample: Shepard's law of generalization and the various Bayesian extensions that it inspired over the past 3 decades. Using Shepard's law as a running example, I argue that psychological theory building is not a statistical problem, mathematical formalism is beneficial to theory, measurement and theory have a complex relationship, rewriting old theory can yield new insights, and theory growth can drive empirical work. Although I generally suggest that the tools of mathematical psychology are valuable to psychological theorists, I also comment on some limitations to this approach.

On the Moral Functions of Language

Previous comparisons of language and morality have taken a cognitively internalist (i.e., within-minds) perspective. We take a socially externalist (i.e., between-minds) perspective, viewing both language and morality as forms of social action. During human evolution, social cognitive adaptations for cooperation evolved, including cooperative communication (social acts to mentally coordinate with others for common goals) and social normativity (social acts to regulate cooperative social relationships). As human cooperation scaled up in complexity, cooperative communication and social normativity scaled up as well, leading to the development of culturally elaborated forms of language and morality. Language facilitates all aspects of morality and is even necessary for certain aspects. Humans use language to (1) initiate, (2) preserve, (3) revise, and (4) act on morality in ways such as forming joint commitments, teaching norms, modifying social realities, and engaging in moral reason-giving.

Evolution and impact of bias in human and machine learning algorithm interaction

Traditionally, machine learning algorithms relied on reliable labels from experts to build predictions. More recently however, algorithms have been receiving data from the general population in the form of labeling, annotations, etc. The result is that algorithms are subject to bias that is born from ingesting unchecked information, such as biased samples and biased labels. Furthermore, people and algorithms are increasingly engaged in interactive processes wherein neither the human nor the algorithms receive unbiased data. Algorithms can also make biased predictions, leading to what is now known as algorithmic bias. On the other hand, human's reaction to the output of machine learning methods with algorithmic bias worsen the situations by making decision based on biased information, which will probably be consumed by algorithms later. Some recent research has focused on the ethical and moral implication of machine learning algorithmic bias on society. However, most research has so far treated algorithmic bias as a static factor, which fails to capture the dynamic and iterative properties of bias. We argue that algorithmic bias interacts with humans in an iterative manner, which has a long-term effect on algorithms' performance. For this purpose, we present an iterated-learning framework that is inspired from human language evolution to study the interaction between machine learning algorithms and humans. Our goal is to study two sources of bias that interact: the process by which people select information to label (human action); and the process by which an algorithm selects the subset of information to present to people (iterated algorithmic bias mode). We investigate three forms of iterated algorithmic bias (personalization filter, active learning, and random) and how they affect the performance of machine learning algorithms by formulating research questions about the impact of each type of bias. Based on statistical analyses of the results of several controlled experiments, we found that the three different iterated bias modes, as well as initial training data class imbalance and human action, do affect the models learned by machine learning algorithms. We also found that iterated filter bias, which is prominent in personalized user interfaces, can lead to more inequality in estimated relevance and to a limited human ability to discover relevant data. Our findings indicate that the relevance blind spot (items from the testing set whose predicted relevance probability is less than 0.5 and who thus risk being hidden from humans) amounted to 4% of all relevant items when using a content-based filter that predicts relevant items. A similar simulation using a real-life rating data set found that the same filter resulted in a blind spot size of 75% of the relevant testing set.

Using Occam's razor and Bayesian modelling to compare discrete and continuous representations in numerosity judgements

Previous research has established that numeric estimates are based not just on perceptual data but also past experience, and so may be influenced by the form of this stored information. It remains unclear, however, how such experience is represented: numerical data can be processed by either a continuous analogue number system or a discrete symbolic number system, with each predicting different generalisation effects. The present paper therefore contrasts discrete and continuous prior formats within the domain of numerical estimation using both direct comparisons of computational models of this process using these representations, as well as empirical contrasts exploiting different predicted reactions of these formats to uncertainty via Occam's razor. Both computational and empirical results indicate that numeric estimates commonly rely on a continuous prior format, mirroring the analogue approximate number system, or 'number sense'. This implies a general preference for the use of continuous numerical representations even where both stimuli and responses are discrete, with learners seemingly relying on innate number systems rather than the symbolic forms acquired in later life. There is however remaining uncertainty in these results regarding individual differences in the use of these systems, which we address in recommendations for future work.

The role of communication in acquisition, curation, and transmission of culture

Veissière et al.'s proposal aims to explain how cognition enables cultural learning, but fails to acknowledge a distinctively human behavior critical to this process: communication. Recent advances in developmental and computational cognitive science suggest that the social-cognitive capacities central to TTOM also support sophisticated yet remarkably early-emerging inferences and communicative behaviors that allow us to learn and share abstract knowledge.

Witnessing, Remembering, and Testifying: Why the Past Is Special for Human Beings

The past is undeniably special for human beings. To a large extent, both individuals and collectives define themselves through history. Moreover, humans seem to have a special way of cognitively representing the past: episodic memory. As opposed to other ways of representing knowledge, remembering the past in episodic memory brings with it the ability to become a witness. Episodic memory allows us to determine what of our knowledge about the past comes from our own experience and thereby what parts of the past we can give testimony about. In this article, we aim to give an account of the special status of the past by asking why humans have developed the ability to give testimony about it. We argue that the past is special for human beings because it is regularly, and often principally, the only thing that can determine present social realities such as commitments, entitlements, and obligations. Because the social effects of the past often do not leave physical traces behind, remembering the past and the ability to bear testimony it brings is necessary for coordinating social realities with other individuals.

Young children consider the expected utility of others' learning to decide what to teach

Direct instruction facilitates learning without the costs of exploration, yet teachers must be selective because not everything can nor needs to be taught. How do we decide what to teach and what to leave for learners to discover? Here we investigate the cognitive underpinnings of the human ability to prioritize what to teach. We present a computational model that decides what to teach by maximizing the learner's expected utility of learning from instruction and from exploration, and we show that children (aged 5-7 years) make decisions that are consistent with the model's predictions (that is, minimizing the learner's costs and maximizing the rewards). Children flexibly considered either the learner's utility or their own, depending on the context, and even considered costs they had not personally experienced, to decide what to teach. These results suggest that utility-based reasoning may play an important role in curating cultural knowledge by supporting selective transmission of high-utility information.

Statistical distributions of consonant variants in infant-directed speech: evidence that /t/ may be exceptional

Statistical distributions of phonetic variants in spoken language influence speech perception for both language learners and mature users. We theorized that patterns of phonetic variant processing of consonants demonstrated by adults might stem in part from patterns of early exposure to statistics of phonetic variants in infant-directed (ID) speech. In particular, we hypothesized that ID speech might involve greater proportions of canonical /t/ pronunciations compared to adult-directed (AD) speech in at least some phonological contexts. This possibility was tested using a corpus of spontaneous speech of mothers speaking to other adults, or to their typically-developing infant. Tokens of word-final alveolar stops - including /t/, /d/, and the nasal stop /n/ - were examined in assimilable contexts (i.e., those followed by a word-initial labial and/or velar); these were classified as canonical, assimilated, deleted, or glottalized. Results confirmed that there were significantly more canonical pronunciations in assimilable contexts in ID compared with AD speech, an effect which was driven by the phoneme /t/. These findings suggest that at least in phonological contexts involving possible assimilation, children are exposed to more canonical /t/ variant pronunciations than adults are. This raises the possibility that perceptual processing of canonical /t/ may be partly attributable to exposure to canonical /t/ variants in ID speech. Results support the need for further research into how statistics of variant pronunciations in early language input may shape speech processing across the lifespan.

Selective sampling and inductive inference: Drawing inferences based on observed and missing evidence

We propose and test a Bayesian model of property induction with evidence that has been selectively sampled leading to "censoring" or exclusion of potentially relevant data. A core model prediction is that identical evidence samples can lead to different patterns of inductive inference depending on the censoring mechanisms that cause some instances to be excluded. This prediction was confirmed in four experiments examining property induction following exposure to identical samples that were subject to different sampling frames. Each experiment found narrower generalization of a novel property when the sample instances were selected because they shared a common property (property sampling) than when they were selected because they belonged to the same category (category sampling). In line with model predictions, sampling frame effects were moderated by the addition of explicit negative evidence (Experiment 1), sample size (Experiment 2) and category base rates (Experiments 3-4). These data show that reasoners are sensitive to constraints on the sampling process when making property inferences; they consider both the observed evidence and the reasons why certain types of evidence has not been observed.

Editors' Introduction: Computational Approaches to Social Cognition

What place should formal or computational methods occupy in social psychology? We consider this question in historical perspective, survey the current state of the field, introduce the several new contributions to this special issue, and reflect on the future.

The diversity effect in inductive reasoning depends on sampling assumptions

A key phenomenon in inductive reasoning is the diversity effect, whereby a novel property is more likely to be generalized when it is shared by an evidence sample composed of diverse instances than a sample composed of similar instances. We outline a Bayesian model and an experimental study that show that the diversity effect depends on the assumption that samples of evidence were selected by a helpful agent (strong sampling). Inductive arguments with premises containing either diverse or nondiverse evidence samples were presented under different sampling conditions, where instructions and filler items indicated that the samples were selected intentionally (strong sampling) or randomly (weak sampling). A robust diversity effect was found under strong sampling, but was attenuated under weak sampling. As predicted by our Bayesian model, the largest effect of sampling was on arguments with nondiverse evidence, where strong sampling led to more restricted generalization than weak sampling. These results show that the characteristics of evidence that are deemed relevant to an inductive reasoning problem depend on beliefs about how the evidence was generated.

Is the most representative skunk the average or the stinkiest? Developmental changes in representations of biological categories

People often think of categories in terms of their most representative examples (e.g., robin for BIRD). Thus, determining which exemplars are most representative is a fundamental cognitive process that shapes how people use concepts to navigate the world. The present studies (N = 669; ages 5 years - adulthood) revealed developmental change in this important component of cognition. Studies 1-2 found that young children view exemplars with extreme values of characteristic features (e.g., the very fastest cheetah) as most representative of familiar biological categories; the tendency to view average exemplars in this manner (e.g., the average-speeded cheetah) emerged slowly across age. Study 3 examined the mechanisms underlying these judgments, and found that participants of all ages viewed extreme exemplars as representative of novel animal categories when they learned that the variable features fulfilled category-specific adaptive needs, but not otherwise. Implications for developmental changes in conceptual structure and biological reasoning are discussed.