Research in the Mind and Development Lab focuses on social cognitive development, particularly the role of inference and explanation in social and causal reasoning. We're currently exploring how infants and children learn about other people, and how they understand the goals of others' actions.
Our work has also focused on music cognition, particularly the origins of our capacity to move in time with a beat. We often use music cognition as a window into novel aspects of social cognition, leveraging musical phenomena to answer questions about mental state inference.
Similar to tools or visual art, music is an artifact - something designed and produced by intentional actions. From early in development, understanding of material artifacts like tools depends on social cognition, particularly inferences about the artifacts' animate, intentional designer (Kelemen, 1999). Does the perception of auditory artifacts similarly rely on social cognition, even early in development? What cognitive mechanisms underlie this link to social cognition?
Infants' early developmentIn this line of research, we are exploring infants' reasoning about auditory artifacts, asking whether (and how) year-old infants infer that non-vocal sounds were caused by an animate agent. Using looking-time methods, we have found that year-old infants spontaneously infer whether agents caused the sounds they observe, an aspect of reasoning crucial to understanding sounds as artifacts. We are currently exploring the extent to which infants are able to rationally infer the causes of sounds they observe (i.e. Gopnik & Schulz, 2004), by combining their prior knowledge of the unique capacities of agents (i.e. self-propelled motion) with their understanding of physical causality (i.e. Gelman & Opfer, 2002). These studies lay the foundation for an exciting arm of our research program, characterizing children's early social reasoning about auditory artifacts like music, a ubiquitous and ancient feature of human culture.
Adults' mature representationsIn a line of related studies with adults, we are investigating the specific hypothesis that people intuitively invoke living, animate agents as the explanation or cause of regular, orderly, patterned structure -- in music-like sounds, and also in other stimuli. Understanding how we reason about orderly stimuli is crucial for multiple psychological domains, including orderly objects (e.g. tools), orderly sounds (e.g. music), and order in nature (e.g. 'intelligent design'). In this work, we ask questions like:
Do infants use others' past behavior to choose preferred social partners? Do infants think of people as members of social groups, and expect members of a group to act similarly to one another?
Infant-directed speech as a social cue
Around the world, adults speak to infants in a specific manner termed infant-directed speech - with exaggerated emotion, slower speed, more repetition and higher pitch. Spontaneous use of infant-directed speech when interacting with infants appears to be an honest, informative signal of caregiver interest and ability. We asked whether infants use this informative social signal to guide their preferences for people, even after the speech behavior has ended. We found that even 5-month-old infants use this cue: After seeing someone speak in an infant-directed manner, 5-month-old infants preferred this person to a novel person; in contrast, after seeing someone speak in an adult-directed manner, infants preferred the novel individual - even after the speech behavior had ended. Thus, use of infant-directed speech may provide an informative social signal that allows infants to select appropriate social partners, focusing their attention on individuals who will provide optimal care and opportunity for learning.
Infants' understanding of social groups: Imitation as a social cue
Adults expect members of social groups to act alike, and imitate others to promote social affiliation. Do infants have similar expectations? In ongoing work, we and our collaborators find that year-old infants expect members of the same social group to produce the same the non-causal, dance-like actions (like jumping or sliding), but have no such expectations for actions that achieve a clear external goal (jumping that results in changing the color of a box). Imitation of non-causal actions appears to be more closely linked to social group membership than imitation of causal actions. Our work on goal inference provides a useful account of this finding (Schachner & Carey, 2013), and these data suggest that year-old infants have a nuanced understanding of the social implications of imitation of different types of actions. This lays the foundation for future work connecting goal inferences with the phenomena of imitation and reasoning about social groups.
As adults, we differentiate between the different domains of the natural world, including artifacts like tools and art, animate agents like people, and inanimate objects like mountains. We have rich, interconnected mental theories of each domain which allow us to reason about past origins and current properties, and make predictions about future states. How do these domain-specific mental theories develop in childhood?
Past work has characterized young children's understanding of the natural world, which in some ways appears notably different than adults'. In particular, young children in the US and UK often endorse purpose-based explanations even for non-living natural phenomena - they say mountains were created for hiking, just like hats were created for warmth. Is this bias innate and universal in young children, or is it a product of Western religious culture?
In two experimental studies involving 128 Chinese children and adults, we and our colleagues find evidence of a bias for purpose-based explanations of nature in Chinese children, and not in Chinese adults. This suggests that the bias for purpose-based explanations may be universal in early life, not solely a product of Western culture.
Why can humans universally move to a beat, while other primates cannot? How did this ability evolve, and how does it develop? Does moving in synchrony with others promote prosocial attitudes and behavior?
Evolutionary origins of entrainment
In past work, we and our colleagues tested the frequent claim that the ability to entrain, or move along with an external auditory pulse, is uniquely human. We showed that two vocal mimicking nonhuman animals (parrots) were able to entrain to music, spontaneously producing synchronized movements resembling human dance. Next, we tested the hypothesis that entrainment evolved as a by-product of vocal mimicry. In particular, we examined an extensive comparative data set from a global video database, including hundreds of species both capable and incapable of vocal mimicry. Despite the higher representation of vocal non-mimics in the database and comparable exposure of mimics and nonmimics to humans and music, only vocal mimics showed evidence of entrainment. Thus, entrainment is not unique to humans, and the distribution of entrainment across species supports the hypothesis that entrainment evolved as a by-product of selection for vocal mimicry.
Social consequences of entrainment: The role of synchrony
In related work, we have asked whether there are social benefits to auditory-motor entrainment, perhaps explaining why humans so often engage in entrainment to sound while other species rarely do. Many studies have claimed that moving in synchrony with other people results in higher prosocial behavior and prosocial attitudes, including greater cooperation and conformity. However, in our work we have failed to replicate several of these effects with both exact and conceptual replications when additional necessary controls are instituted. In particular, we do not find social effects of synchrony when experimenters are kept unaware of participants' experimental condition, preventing possible experimenter bias; or when the effects of synchrony are isolated from other factors. This work provides important methodological recommendations to ensure clarity of findings in this literature, and suggests that the social effects of synchrony may result from having engaged in cooperative behavior toward a shared goal (i.e. the goal of coordinating one's movements), not due to synchrony per se.
In this line of meta-research, we ask questions like: How can we increase the importance of replicability, and incentivize replication? How can we promote a nuanced understanding of replication data, and use meta-analysis to avoid over-weighting non-replications? We also do our best to present our own replications and non-replications at conferences and in journals.
One step toward incentivizing replications that we and our colleagues have proposed is a system for post-publication replication tracking, designed to improve the reliability and accuracy of published research. We believe that the incentive structure of scientific publishing must change for the quality of data and scientific conclusions to be improved. Under the current system, the quality of individual scientists is often judged on the basis of their number of publications and citations, with journals similarly judged via numbers of citations. Neither of these measures takes into account the replicability of the published findings, as false or controversial results are often particularly widely cited. We propose tracking replications as a means of post-publication evaluation, both to help researchers identify reliable findings and to incentivize the publication of reliable results. In particular, we propose creating a replicability score for each journal to go along with their impact factor, hopefully incentivizing the publishing of not only exciting research but replicable research as well.
We have presented - and published - research that examines the reliability of important areas of research. In particular, we have presented work showing that when experimenters are blind to condition, effects of synchrony on social variables are not robust. We have also shown, with our collaborators, that at least some kinds of preschool music enrichment provide no consistent benefits to nonmusical cognitive tasks. In both cases, we presented and/or published this data to ensure the completeness of the scientific record on these issues.