Where is the baby in core knowledge?

What we know about what babies know - as represented by the core knowledge proposal - is perhaps missing a place for the baby itself. By studying the baby as an actor rather than an observer, we can better understand the origins of human intelligence as an interface between perception and action, and how humans think and learn about themselves in a complex world.

Toddlers Prefer Agents Who Help Those Facing Harder Tasks.

Capacities to understand and evaluate others' actions are fundamental to human social life. Infants and toddlers are sensitive to the costs of others' actions, infer others' values from the costs of the actions they take, and prefer those who help others to those who hinder them, but it is largely unknown whether and how cost considerations inform early understanding of third-party prosocial actions. In three experiments (N = 94), we asked whether 16-month-old toddlers value agents who selectively help those who need it most. Presented with two agents who attempted two tasks, toddlers preferentially looked to and touched someone who helped the agent in greater need, both when one agent's task required more effort and when the tasks were the same but one agent was weaker. These results provide evidence that toddlers engage in need-based evaluations of helping, applying their understanding of action utilities to their social evaluations.

Expecting the Unexpected: Infants Use Others' Surprise to Revise Their Own Expectations.

Human infants show systematic responses to events that violate their expectations. Can they also revise these expectations based on others' expressions of surprise? Here we ask whether infants (N = 156, mean = 15.2 months, range: 12.0-18.0 months) can use an experimenter's expression of surprise to revise their own expectations about statistically probable vs. improbable events. An experimenter sampled a ball from a box of red and white balls and briefly displayed either a surprised or an unsurprised expression at the outcome before revealing it to the infant. Following an unsurprised expression, the results were consistent with prior work; infants looked longer at a statistically improbable outcome than a probable outcome. Following a surprised expression, however, this standard pattern disappeared or was even reversed. These results suggest that even before infants can observe the unexpected events themselves, they can use others' surprise to expect the unexpected. Starting early in life, human learners can leverage social information that signals others' prediction error to update their own predictions.

This too shall pass, but when? Children's and adults' beliefs about the time duration of emotions, desires, and preferences.

This research investigated children's and adults' understanding of the mind by assessing beliefs about the temporal features of mental states. English-speaking North American participants, varying in socioeconomic status (Study 1: N = 50 adults; Study 2: N = 112, 8- to 10-year-olds and adults; and Study 3: N = 116, 5- to 7-year-olds and adults; tested 2017-2022), estimated the duration (seconds to a lifetime) of emotions, desires (wanting), preferences (liking), and control trials (e.g., napping and having eyes). Participants were 56% female and 44% male; 32% Asian, 1% Black, 13% Hispanic/Latino, 38% White (non-Hispanic/Latino), and 16% multiracial or another race/ethnicity. Children and adults judged that preferences last longer than emotions and desires, with age differences in distinguishing specific emotions by duration ( η p 2 s > .03 ). By 5 to 7 years, ideas about the mind include consideration of time.

Open dataset of theory of mind reasoning in early to middle childhood.

Theory of mind (ToM) reasoning refers to the process by which we reason about the mental states (beliefs, desires, emotions) of others. Here, we describe an open dataset of responses from children who completed a story booklet task for assessing ToM reasoning (n = 321 3-12-year-old children, including 64 (neurotypical) children assessed longitudinally and 68 autistic children). Children completed one of two versions of the story booklet task (Booklet 1 or 2). Both versions include two-alternative forced choice and free response questions that tap ToM concepts ranging in difficulty from reasoning about desires and beliefs to reasoning about moral blameworthiness and mistaken referents. Booklet 2 additionally includes items that assess understanding of sarcasm, lies, and second-order belief-desire reasoning. Compared to other ToM tasks, the booklet task provides relatively dense sampling of ToM reasoning within each child (Booklet 1: 41 items; Booklet 2: 65 items). Experimental sessions were video recorded and data were coded offline; the open dataset consists of children's accuracy (binary) on each item and, for many children (n = 171), transcriptions of free responses. The dataset also includes children's scores on standardized tests of receptive language and non-verbal IQ, as well as other demographic information. As such, this dataset is a valuable resource for investigating the development of ToM reasoning in early and middle childhood.

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'.

Children seek help based on how others learn.

Do children consider how others learned when seeking help? Across three experiments, German children (N = 536 3-to-8 year olds, 49% female, majority White, tested 2017-2019) preferred to learn from successful active learners selectively by context: They sought help solving a problem from a learner who had independently discovered the solution to a previous problem over those who had learned through instruction or observation, but only when the current problem was novel, yet related, to the learners' problem (Experiment 1). Older, but not younger, children preferred the active learner even when she was offered help (Experiment 2), though only when her discovery was deliberate (Experiment 3). Although a preference to learn from successful active learners emerges early, a genuine appreciation for process beyond outcome increases across childhood.

Young children infer and manage what others think about them.

We care about what others think of us and often try to present ourselves in a good light. What cognitive capacities underlie our ability to think (or even worry) about reputation, and how do these concerns manifest as strategic self-presentational behaviors? Even though the tendency to modify one's behaviors in the presence of others emerges early in life, the degree to which these behaviors reflect a rich understanding of what others think about the self has remained an open question. Bridging prior work on reputation management, communication, and theory of mind development in early childhood, here we investigate young children's ability to infer and revise others' mental representation of the self. Across four experiments, we find that 3- and 4-y-old children's decisions about to whom to communicate (Experiment 1), what to communicate (Experiments 2 and 3), and which joint activity to engage in with a partner (Experiment 4) are systematically influenced by the partner's observations of the children's own past performance. Children in these studies chose to present self-relevant information selectively and strategically when it could revise the partner's outdated, negative representation of the self. Extending research on children's ability to engage in informative communication, these results demonstrate the sophistication of early self-presentational behaviors: Even young children can draw rich inferences about what others think of them and communicate self-relevant information to revise these representations.

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.

Beyond knowledge versus belief: The contents of mental-state representations and their underlying computations.

Moving beyond distinguishing knowledge and beliefs, we propose two lines of inquiry for the next generation of theory of mind (ToM) research: (1) characterizing the contents of different mental-state representations and (2) formalizing the computations that generate such contents. Studying how children reason about what others think of the self provides an illuminating window into the richness and flexibility of human social cognition.

Moderated Online Data-Collection for Developmental Research: Methods and Replications.

Online data collection methods are expanding the ease and access of developmental research for researchers and participants alike. While its popularity among developmental scientists has soared during the COVID-19 pandemic, its potential goes beyond just a means for safe, socially distanced data collection. In particular, advances in video conferencing software has enabled researchers to engage in face-to-face interactions with participants from nearly any location at any time. Due to the novelty of these methods, however, many researchers still remain uncertain about the differences in available approaches as well as the validity of online methods more broadly. In this article, we aim to address both issues with a focus on moderated (synchronous) data collected using video-conferencing software (e.g., Zoom). First, we review existing approaches for designing and executing moderated online studies with young children. We also present concrete examples of studies that implemented choice and verbal measures (Studies 1 and 2) and looking time (Studies 3 and 4) across both in-person and online moderated data collection methods. Direct comparison of the two methods within each study as well as a meta-analysis of all studies suggest that the results from the two methods are comparable, providing empirical support for the validity of moderated online data collection. Finally, we discuss current limitations of online data collection and possible solutions, as well as its potential to increase the accessibility, diversity, and replicability of developmental science.

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.

Preschool-Aged Children Jointly Consider Others' Emotional Expressions and Prior Knowledge to Decide When to Explore.

Emotional expressions are abundant in children's lives. What role do they play in children's causal inference and exploration? This study investigates whether preschool-aged children use others' emotional expressions to infer the presence of unknown causal functions and guide their exploration accordingly. Children (age: 3.0-4.9; N = 112, the United States) learned about one salient causal function of a novel toy and then saw an adult play with it. Children explored the toy more when the adult expressed surprise than when she expressed happiness (Experiment 1), but only when the adult already knew about the toy's salient function (Experiment 2). These results suggest that children consider others' knowledge and selectively interpret others' surprise as vicarious prediction error to guide their own exploration.

The Division of Labor in Communication: Speakers Help Listeners Account for Asymmetries in Visual Perspective.

Recent debates over adults' theory of mind use have been fueled by surprising failures of perspective-taking in communication, suggesting that perspective-taking may be relatively effortful. Yet adults routinely engage in effortful processes when needed. How, then, should speakers and listeners allocate their resources to achieve successful communication? We begin with the observation that the shared goal of communication induces a natural division of labor: The resources one agent chooses to allocate toward perspective-taking should depend on their expectations about the other's allocation. We formalize this idea in a resource-rational model augmenting recent probabilistic weighting accounts with a mechanism for (costly) control over the degree of perspective-taking. In a series of simulations, we first derive an intermediate degree of perspective weighting as an optimal trade-off between expected costs and benefits of perspective-taking. We then present two behavioral experiments testing novel predictions of our model. In Experiment 1, we manipulated the presence or absence of occlusions in a director-matcher task. We found that speakers spontaneously modulated the informativeness of their descriptions to account for "known unknowns" in their partner's private view, reflecting a higher degree of speaker perspective-taking than previously acknowledged. In Experiment 2, we then compared the scripted utterances used by confederates in prior work with those produced in interactions with unscripted directors. We found that confederates were systematically less informative than listeners would initially expect given the presence of occlusions, but listeners used violations to adaptively make fewer errors over time. Taken together, our work suggests that people are not simply "mindblind"; they use contextually appropriate expectations to navigate the division of labor with their partner. We discuss how a resource-rational framework may provide a more deeply explanatory foundation for understanding flexible perspective-taking under processing constraints.

Leaving a Choice for Others: Children's Evaluations of Considerate, Socially-Mindful Actions.

People value those who act with others in mind even as they pursue their own goals. Across three studies (N = 566; 4- to 6-year-olds), we investigated children's developing understanding of such considerate, socially-mindful actions. By age 6, both U.S. and Chinese children positively evaluate a character who takes a snack for herself in a way that leaves a snack choice for others over a character who leaves no choice (Study 1), but only when the actors had alternative possible actions (Study 2) and when a clear beneficiary was present (Study 3). These results suggest an emerging ability to infer underlying social intentions from self-oriented actions, providing insights into the role of social-cognitive capacities versus culture-specific norms in children's moral evaluations.

Learning from other minds: An optimistic critique of reinforcement learning models of social learning.

Reinforcement learning models have been productively applied to identify neural correlates of the value of social information. However, by operationalizing social information as a lean, reward-predictive cue, this literature underestimates the richness of human social learning: Humans readily go beyond action-outcome mappings and can draw flexible inferences from a single observation. We argue that computational models of social learning need minds, i.e, a generative model of how others' unobservable mental states cause their observable actions. Recent advances in inferential social learning suggest that even young children learn from others by using an intuitive, generative model of other minds. Bridging developmental, Bayesian, and reinforcement learning perspectives can enrich our understanding of the neural bases of distinctively human social learning.

Online Developmental Science to Foster Innovation, Access, and Impact.

We propose that developmental cognitive science should invest in an online CRADLE, a Collaboration for Reproducible and Distributed Large-Scale Experiments that crowdsources data from families participating on the internet. Here, we discuss how the field can work together to further expand and unify current prototypes for the benefit of researchers, science, and society.

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.

Response patterns in the developing social brain are organized by social and emotion features and disrupted in children diagnosed with autism spectrum disorder.

Adults and children recruit a specific network of brain regions when engaged in "Theory of Mind" (ToM) reasoning. Recently, fMRI studies of adults have used multivariate analyses to provide a deeper characterization of responses in these regions. These analyses characterize representational distinctions within the social domain, rather than comparing responses across preferred (social) and non-preferred stimuli. Here, we conducted opportunistic multivariate analyses in two previously collected datasets (Experiment 1: n = 20 5-11 year old children and n = 37 adults; Experiment 2: n = 76 neurotypical and n = 29 5-12 year old children diagnosed with Autism Spectrum Disorder (ASD)) in order to characterize the structure of representations in the developing social brain, and in order to discover if this structure is disrupted in ASD. Children listened to stories that described characters' mental states (Mental), non-mentalistic social information (Social), and causal events in the environment (Physical), while undergoing fMRI. We measured the extent to which neural responses in ToM brain regions were organized according to two ToM-relevant models: 1) a condition model, which reflected the experimenter-generated condition labels, and 2) a data-driven emotion model, which organized stimuli according to their emotion content. We additionally constructed two control models based on linguistic and narrative features of the stories. In both experiments, the two ToM-relevant models outperformed the control models. The fit of the condition model increased with age in neurotypical children. Moreover, the fit of the condition model to neural response patterns was reduced in the RTPJ in children diagnosed with ASD. These results provide a first glimpse into the conceptual structure of information in ToM brain regions in childhood, and suggest that there are real, stable features that predict responses in these regions in children. Multivariate analyses are a promising approach for sensitively measuring conceptual and neural developmental change and individual differences in ToM.

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.