Michael C. Frank’s research while affiliated with University of California, San Diego and other places

Color word learning is an important case study for the relationship between language and perception. While English color word learning is well-documented, there is relatively limited evidence on the developmental trajectory for color words, especially in languages from non-Western populations. We study color words and their acquisition in the Shipibo-Konibo (SK), an indigenous group within the Peruvian Amazon. In Study 1, we measure the color vocabulary in SK adults, updating findings from the World Color Survey. We then study receptive and productive knowledge of color words in children, conducted in both SK (Study 2) and Spanish (Study 3). Children learning the SK system show a protracted developmental trajectory towards adult-like color term knowledge compared to contemporary studies of English-speaking children. Further, when SK children lack precise color term knowledge, they appeared to follow different strategies for SK and Spanish, using Spanish vocabulary in SK and overgeneralizing in Spanish. For both children and adults, bilingual vocabulary is used adap-tively to facilitate task performance, broadly supporting communicative views of color vocabulary. ORCID ID(s): https://orcid.org/0000-0003-3282-8120 (Fortier, M.); https://orcid.org/0000-0001-7811-3468 (Kellier, D.J.); https://orcid.org/0000-0002-2699-2509 (Fernández-Flecha, M.), https://or-cid.org/0000-0002-7551-4378 (Frank, M. C.) Citation: Fortier, M., Kellier, D. J., Fernández-Flecha, M., Frank, M. C. (2024). The development of color terms in Shipibo-Konibo children.

How do we decide what to look at and when to stop looking? Even very young infants engage in active visual selection, looking less and less as stimuli are repeated (habituation) and regaining interest when novel stimuli are subsequently introduced (dishabituation). The mechanisms underlying these looking time changes remain uncertain, however, due to limits on both the scope of existing formal models and the empirical precision of measurements of infant behavior. To address this, we developed the Rational Action, Noisy Choice for Habituation (RANCH) model, which operates over raw images and makes quantitative predictions of participants’ looking behaviors. In a series of pre-registered experiments, we exposed infants and adults to stimuli for varying durations and measured looking time to familiar and novel stimuli. We found that these data were well captured by RANCH. Using RANCH’s stimulus-computability, we also tested its out-of-sample predictions about the magnitude of dishabituation in a new experiment in which we manipulated the similarity between the familiar and novel stimulus. By framing looking behaviors as rational decision-making, this work identified how the dynamics of learning and exploration guide our visual attention from infancy through adulthood.

How do we decide what to look at and when to stop looking? Even very young infants engage in active visual selection, looking less and less as stimuli are repeated (habituation) and regaining interest when novel stimuli are subsequently introduced (dishabituation). The mechanisms underlying these looking time changes remain uncertain, however, due to limits on both the scope of existing formal models and the empirical precision of measurements of infant behavior. To address this, we developed the Rational Action, Noisy Choice for Habituation (RANCH) model, which operates over raw images and makes quantitative predictions of participants’ looking behaviors. In a series of pre-registered experiments, we exposed infants and adults to stimuli for varying durations and measured looking time to familiar and novel stimuli. We found that these data were well captured by RANCH. Using RANCH’s stimulus-computability, we also tested its out-of-sample predictions about the magnitude of dishabituation in a new experiment in which we manipulated the similarity between the familiar and novel stimulus. By framing looking behaviors as rational decision-making, this work identified how the dynamics of learning and exploration guide our visual attention from infancy through adulthood.

Evaluating whether someone's behavior is praiseworthy or blameworthy is a fundamental human trait. A seminal study by Hamlin and colleagues in 2007 suggested that the ability to form social evaluations based on third-party interactions emerges within the first year of life: infants preferred a character who helped, over hindered, another who tried but failed to climb a hill. This sparked a new line of inquiry into the origins of social evaluations; however, replication attempts have yielded mixed results. We present a preregistered, multi-laboratory, standardized study aimed at replicating infants' preference for Helpers over Hinderers. We intended to (1) provide a precise estimate of the effect size of infants' preference for Helpers over Hinderers, and (2) determine the degree to which preferences are based on social information. Using the ManyBabies framework for big team-based science, we tested 1018 infants (567 included, 5.5-10.5 months) from 37 labs across five continents. Overall, 49.34% of infants preferred Helpers over Hinderers in the social condition, and 55.85% preferred characters who pushed up, versus down, an inanimate object in the nonsocial condition; neither proportion differed from chance or from each other. This study provides evidence against infants' prosocial preferences in the hill paradigm, suggesting the effect size is weaker, absent, and/or develops later than previously estimated. As the first of its kind, this study serves as a proof-of-concept for using active behavioral measures (e.g., manual choice) in large-scale, multi-lab projects studying infants.

Having an abstract understanding of communication means being able to reason not just about its success but also its failure: when and why it fails, what happens as a consequence, and how to fix it. Auditory noise in particular can radically alter the fidelity of verbal communication, yet little is known about how language users reason about its role in communication and when such reasoning emerges in development. The current work investigates the development of children’s reasoning about auditory noise. As observers of others' communication, we found that a sample of US children as young as age three understood that auditory noise impedes others’ hearing, though the ability to reason about its effect on others’ knowledge and behaviors may not develop until slightly later. As communicators, a separate sample of four-year-olds tailored how they themselves communicated by using more gestures when their partner couldn't hear. Taken together, these results suggest that even young children may possess an abstract understanding of communication that is integrated with their ability to reason about others’ perception and knowledge. Such an understanding enables children to reason about how noise disrupts communication and how to communicate effectively in the presence of noise.

How do children learn words so rapidly? A powerful source of information about a new word’s meaning is the set of social cues provided by its speaker (e.g. eye-gaze). Studies of children’s use of social cues have tended to focus on the emergence of this ability in early infancy. We show, however, that this early-emerging ability has a long developmental trajectory: Slow, continuous improvements in speed of social information processing occur over the course of the first five years of life. This developing ability to allocate social attention is a significant bottleneck on early word learning—continuous changes in social information processing predict continuous changes in children’s ability to learn new words. Further, we show that this bottleneck generalizes to children diagnosed with autism spectrum disorder, whose social information processing is atypical. These results describe a route by which increases in social expertise can lead to changes in language learning ability, and more generally highlight the dependence of developmental outcomes not on just the existence of particular competencies, but on their proficient use in complex contexts.

In recent years, psychological researchers have been heavily criticizedfor generalizing broadly from narrow samples, a concern that inter-sects with questions about the validity, reproducibility, replicability,and generalizability of the psychological literature. One issue is thelimited reporting of participants’ identities, backgrounds, and livedexperiences. To address this issue, several journals have begun torequire greater reporting of participants’ sociodemographic informa-tion. In this article, we address both challenges and considerationswith respect to sociodemographic reporting for researchers, reviewers,and journal editors. We provide guidance for recording, evaluating,protecting, and interpreting sociodemographic data.

Evaluating whether someone's behavior is praiseworthy or blameworthy is a fundamental human trait. A seminal study by Hamlin and colleagues in 2007 suggested that the ability to form social evaluations based on third-party interactions emerges within the first year of life: infants preferred a character who helped, over hindered, another who tried but failed to climb a hill. This sparked a new line of inquiry into the origins of social evaluations; however, replication attempts have yielded mixed results. We present a preregistered, multi-laboratory, standardized study aimed at replicating infants’ preference for Helpers over Hinderers. We intended to (1) provide a precise estimate of the effect size of infants’ preference for Helpers over Hinderers, and (2) determine the degree to which preferences are based on social information. Using the ManyBabies framework for big team-based science, we tested 1018 infants (567 included, 5.5–10.5 months) from 37 labs across five continents. Overall, 49.34% of infants preferred Helpers over Hinderers in the social condition, and 55.85% preferred characters who pushed up, versus down, an inanimate object in the nonsocial condition; neither proportion differed from chance or from each other. This study provides evidence against infants’ prosocial preferences in the hill paradigm, suggesting the effect size is weaker, absent, and/or develops later than previously estimated. As the first of its kind, this study serves as a proof-of-concept for using active behavioral measures (e.g., manual choice) in large-scale, multi-lab projects studying infants.

When a replication fails, scientists have to decide whether to make a second attempt or move on. Psychology researchers who attempt to replicate studies often face this decision, given the empirical rate of replication success in psychology, which is lower than desired. Here, we report 17 re-replications of experiments for which an original replication had failed. In 5/17 of these “rescue” projects (29%), the “rescue” study mostly or fully replicated the original results, albeit with a smaller effect size; in the other 12, the second replication was also judged to have failed. We speculate that successful rescue projects were due to larger sample sizes or methodological changes such as attention checks. In the absence of obvious weaknesses in a failed replication study’s sample or procedure, however, it may be most efficient to stop pursuing an effect after a single failed replication.

The current article describes the Remote Infant Studies of Early Learning, a battery intended to provide robust looking time measures of cognitive development that can be administered remotely to inform our understanding of individual developmental trajectories in typical and atypical populations, particularly infant siblings of autistic children. This battery was developed to inform our understanding of early cognitive and language development in infants who will later receive a diagnosis of autism. Using tasks that have been successfully implemented in lab-based paradigms, we included assessments of attention, memory, prediction, word recognition, numeracy, multimodal processing, and social evaluation. This study reports results on the feasibility and validity of administration of this task battery in 55 infants who were recruited from the general population at age 6 months (n = 29; 14 female, 15 male) or 12 months (n = 26; 14 female, 12 male; 62% White, 13% Asian, 1% Black, 1% Pacific Islander, 22% more than one race; 6% Hispanic). Infant looking behavior was recorded during at-home administration of the battery on the family’s home computer and automatically coded for attention to stimuli using iCatcher+, an open-access software that assesses infant gaze direction. Results indicate that while some tasks replicated lab-based findings (attention, memory, prediction, and numeracy), others did not (word recognition, multimodal processing, and social evaluation). These findings will inform efforts to refine the battery as we continue to develop a robust set of tasks to improve the understanding of early cognitive development at the individual level in general and clinical populations.