About

Dr. Douglas A. Frye is an Associate Professor at the University of Pennsylvania Graduate School of Education, with a background that includes teaching and research appointments abroad at Cambridge University, as well as research in computer science at Yale University. He returned to the United States to pursue further academic work and has been a faculty member at Penn GSE since 1998. His educational credentials include a Ph.D. in Psychology from Yale University and a B.A. in Psychology and Philosophy from New York University. Dr. Frye's research concentrates on two primary areas within cognitive development: children’s theories of mind and early mathematical development. His work on theory of mind explores the developmental processes that influence children’s social competence, examining the relationship between theory of mind, executive control, and early childhood development. He investigates how these cognitive developments benefit children’s social-emotional understanding and their comprehension of teaching and learning. His second research focus aims to understand how children acquire mathematical reasoning skills, assessing cognitive changes during early math learning and designing interactive classroom and computer settings to enhance math understanding. Currently, he collaborates with urban Head Start centers to support emergent numeracy and social competence, ensuring young children are prepared for the transition to school.

Research topics

• Developmental psychology
• Social psychology
• Psychology
• Internal medicine
• Surgery
• Cancer research
• Oncology
• Medicine

Selected publications

• Editorial: Children's teaching

  Frontiers in Developmental Psychology · 2025-10-09

  editorialOpen accessSenior authorCorresponding

  When we think of teaching, our thoughts quickly jump to the occupation of the teacher: we imagine an adult instructor in a classroom passing on knowledge to a group of pupils. With such a narrow view of teaching, it is not apparent why fundamental research in child development should concern itself with the emergence of a pedagogical mindset—after all, only a small minority of children grow up to become professional teachers. To see the merit of studying the development of pedagogy, we need a notion of teaching that covers a much wider terrain and recognizes just how varied human teaching interactions are. In "The Culture of Education", Jerome Bruner (1996, p. xi) reminds us that, "Education does not only occur in classrooms, but around the dinner table when family members try to make joint sense of what happened that day, or when kids try to help each other make sense of the adult world, or when a master and apprentice interact on the job." Teaching, we can say, is a transaction between individuals which aims at increasing the knowledge (practical or theoretical) or understanding of some of its participants. This broader conception helps us to see that teaching, unlike institutionalized schooling, is a human universal, and also that it occurs in varied settings (e.g., playgrounds, grocery stores) and interpersonal constellations. It is this last point that motivates investigations of the kind compiled in this Research Topic. Knowledge does not only flow "top down", from adult to child. It also spreads horizontally between children and even runs "bottom up," from child to adult (Qiu & Moll, 2022). A child of adult immigrants helping their parents to understand the host culture's rituals and customs is as much engaged in teaching as is a mother showing her child how to properly hold and twist a screwdriver. The articles in this collection give clear evidence that children, from a young age, have a remarkable ability to identify good teaching and are themselves skilled at teaching. Karadağ et al. examined selective information-transmission in the understudied age of toddlerhood. Two-year-olds preferred showing a naïve learner object functions they had learned from an adult over ones they had discovered by themselves. A comparison group of 5-year-olds were equally likely to demonstrate taught and discovered information–indicating previously unrecognized ontogenetic changes in children's pedagogical choices. Bass et al. tested 3-to 8-year-olds' reasoning about the quality of teaching, in particular their detection of mechanical or rote teaching. Theory of mind predicted children's ability to identify such teaching in adults, and children who skillfully spotted mindless teachers were more discriminant in following a tutor's deliberate demonstrations but not their accidental behaviors. This study contributes to our understanding of learners' sensitivity to the quality, and in particular the intentionality, of acts of teaching. Ye et al. demonstrated that the development of children's teaching is related to mental state reasoning and reflects culturally bound pedagogical styles. The authors compared the use of verbal, contrastive ("do x, not y") and contingent (addressing a learner's specific mistake) teaching methods by 3-to 7-year-olds from American, Chinese traditional and Westernized preschools. Reflecting the West's learner-centered approach to pedagogy, children from American and Westernized Chinese preschools used more verbal and contingent teaching styles than students from the traditional Chinese school system. Bowman-Smith et al. investigated children's teaching of social robots. Learner mistakes not only elicited more teaching behavior, but children who taught robots that made unexpected or atypical mistakes (errors on previously learned material) benefited more from the teaching episode than those who taught robots that made typical mistakes (errors on novel material) or none at all. This study brings important insights into children's sensitivity to the learner's knowledge state, demonstrates how the "learning-by-teaching" process extends even to nonhuman agents, and is particularly timely given current advancements in artificial intelligence. The findings from these papers provoke a re-evaluation of how we conceive education and cultural evolution. Dominant theories from the learning sciences emphasize how educable and "culture-absorbent" children are, but they typically fail to recognize children's productive, over and above their receptive, contribution to education and cultural evolution (Lew-Levy & Amir, 2025). Children's bilateral involvement with teaching harmonizes with the shared intentionality thesis (Tomasello, 2020). According to this thesis, participation in cooperatively structured activities (give-take, speak-listen, throw-catch, etc.) leads to "perspectival cognitive representations", which represent the transaction not just from one's own role but simultaneously from the complementary role of one's interaction partner. Applied to teaching, children internalize the teacher's stance toward them and understand the exchange from both points of view, that of the learner and the teacher—allowing them to reverse roles and return the favor of teaching. We close by pointing out three promising directions for future research. One concerns the significance of children's knowledge of teaching for their cognitive growth. Do we know that children's insights into pedagogy are not just trivial information but meaningful for their learning? Bowman-Smith et al.'s study and work by Jeong and Frye (2018) with Korean students suggests this is so, but more research is needed to demonstrate whether and how children's understanding of teaching impacts their learning. Second, the study of children as teachers remains incomplete without rigorous longitudinal work documenting developmental change in children's use of strategies, the expansion of their behavioral repertoire, and selectivity in information transmission. Third, children's thoughts and feelings about AI in education ought to be explored in depth. The difficult task of responsibly guiding, and limiting, the reliance on AI in childhood education will be easier if we better understand children's attitudes and expectations regarding AI assistance.

  PublisherOA PDFDOI

• New insights and approaches to early learning

  British Journal of Developmental Psychology · 2025-01-24

  editorialOpen accessSenior author

  New insights and approaches to early learning New insights and approaches to early learningOne of the most notable characteristics of young children is how quickly and how much they learn.Perhaps because they are at the beginning of acquiring the culture's accumulated knowledge, their first steps and persistent curiosity are evident.Research and practice in early childhood education have long debated how to support young children's learning.Proposed approaches to early education range from playful learning (Weisberg et al., 2016) to direct instruction.Each approach takes a different position on how much and in what ways others should be involved in young children's learning, but they all seem to recognize that there is something different about learning in this period of development that should be acknowledged.Approaches to instruction with young children depend in part on the characteristics of their learning.There is evidence that the nature of their learning changes over early childhood.For example, research on children's understanding of learning or how knowledge and skill change is related to the development of theory of mind (Wang et al., 2017).If young children develop an understanding of learning, it could open the possibilities of monitoring their own learning and perhaps being able to guide it.These possibilities allow new studies of what young children know about learning and how that knowledge affects their actual learning.The five papers that follow examined different aspects of learning in the early childhood period.The first two tested the relevance of the zone of proximal development (ZPD) (Vygotsky, 1978) to young children's learning.Jeong and Frye (2024) ask whether children recognize a learner's ZPD and begin to make judgements about what is easy or difficult to learn.Huang et al. (2023) examine how parents' scaffolding in their children's ZPD benefits their math learning and in what ways.Jeong and Frye's (2024) paper explored children's understanding of the ZPD by incorporating the learner's ability level, their desire to learn, and the task difficulty relevant to the learner's ability.In two studies, it was found that only 4-and 5-year-olds, but not 3-year-olds, could match the learner's ability and desire when choosing tasks for the learner.Four-and 5-year-old understood that someone with low ability should have an easy task for a game, while another person with high ability might be able to manage a difficult task.More important, when the goal was to learn, 4-and 5-year-olds chose a difficult task beyond the learner's ability.Furthermore, 5-and 6-year-olds were more careful not to give extremely difficult tasks to a learner with low ability or a learner who just wanted to play a game without wanting to learn new skills.Taken together, the findings suggest that with age, children could estimate the appropriate level of challenge, or the ZPD, according to the learner's desire and ability, and understand the proper degree of challenge needed for learning.This ability could function as the stepping stone for children's own teaching and possibly their self-regulated learning.Based on observations of mothers' and fathers' scaffolding in math activities, Huang et al.'s (2023) paper highlighted the importance of authentic contexts in learning.Parents of 5-to 6-year-old children intuitively applied more scaffolding in real-life activities, such as talking about prices when shopping, than in math games or worksheets.In addition, only parental scaffolding in real-life applications predicted preschool children's formal math ability.Compared to worksheet activities that might provoke more direct instruction, real-life activities might activate parents' scaffolding motivation and solicit the skills attained in everyday life.Although not explicit, this study also implied that children might interpret games as competition and worksheets as goal-oriented tasks.As such, it was likely that in their eyes, parents were either playing a game with them or helping them finish a task.The real-life activity, on the

  PublisherOA PDFDOI

• Editorial: Theory of mind in relation to other cognitive abilities, volume II

  Frontiers in Psychology · 2025-06-19

  editorialOpen accessSenior authorCorresponding

  development. However, there is still much debate as to how it should be defined and even as to whether it is a single entity. In particular, there is controversy as to the extent to which it should be seen as a specific cognitive module, or rather as dependent on, or mutually developing with, other cognitive abilities and characteristics, such as language, metacognition, executive function, and cognitive and perceptual styles that emphasize gist versus detail ('strong' versus 'weak' central coherence). It is also possible that the theory of mind itself has several different components, which may be related to different degrees, different cognitive abilities and characteristics. Any relations between the theory of mind and other cognitive characteristics may also vary with age, and may differ between typically developing children and those with autism and other atypical conditions.Gaining a greater understanding of these issues is important for increasing our understanding of theory of mind itself, the nature of cognitive development, the similarities and differences between typically and atypically developing children, and The articles in this Research Topic can be divided into three broad groups: the nature, correlates and predictors of theory of mind in children; the nature, correlates and predictors of theory of mind in adults; and the role of theory of mind difficulties in disorders.With regard to theory of mind in children, Esra Ünlüer investigated whether preschool (4-and 5-year-old) children's theory of mind skills and peer relationships predicted their subsequent school adjustment. There did indeed turn out to be significant relationships. Theory of mind significantly predicted school adjustment as a whole (positively predicting liking for school and negatively predicting school avoidance, while prosocial and aggressive behaviour toward peers specifically predicted liking for school.The other articles on children in this Research Topic look at characteristics that predict and may contribute to theory of mind, rather than those that follow on from it. Both of these articles suggest that certain language skills are important predictors.Jill de Villiers and Peter de Villiers studied 258 children between 3 and 5 years over few months and tested them on three occasions on false belief reasoning and on the possible contributory factors of general language development, complement syntax, vocabulary, and inhibitory control. Cross-sectional and longitudinal regressions showed that all these factors contributed significantly to false belief reasoning, but that by time 3, the major proximal contribution was the child's comprehension of syntactic complements. The authors conclude that, as suggested by their earlier training studies, complement syntax makes an important specific contribution to false belief reasoning, but that vocabulary and executive function skills also form pathways to it.Honglan Li and Man-Tak Leung assessed the language skills, executive functions and first-order and second-order false belief reasoning in 150 Mandarin-speaking preschoolers and early primary school children. They found that language was a significant independent predictor of both first-order and second-order false belief reasoning. Executive function predicted first-order false belief reasoning after controlling for age, but not after also controlling for language skills. However, it did continue to be an independent significant predictor of second-order theory of mind even after controlling for both age and language skills. With regard to theory of mind in adults, one study, like that of Li and Leung with children, looked at the possible predictive roles of language and executive function.Derek Montgomery, Virginia Tompkins and Xin Feng investigated adults' theory of mind and its possible relation to language and executive function. They gave 207 adults a battery of advanced theory of mind tasks as well as tests of vocabulary and executive function. They found that the Strange Stories, Higher-Order False Belief, and Frith-Happé Animation tasks, though relatively weakly correlated, all loaded onto a common factor (?), which they considered to involve perspective-taking, within a narrative context, to represent a protagonist's mental state and to use it to predict and explain their actions. This factor was related more closely to vocabulary than to executive function .Florence Mayrand, Francesca Cappozzi and Jelena Ristic carried out a rather different type of study, looking at adults' interpretation of the information communicated by gaze.They investigated how spatially dissociated versus spatially combined effects of gaze (i.e., cases where an agent's inferred mental content implied by gaze is discrepant with the directional information communicated by gaze, versus cases where the two types of cue give concordant information) influence participants' target performance. They found that that performance was worse when cue direction and mental content were discordant than when they were concordant. This effect was more marked when a social avatar served as a cue than when a comparison arrow was the cue. The findings suggest that a typical gaze communicates information both about what a person is attending to and about the location where they are attending.Other studies looked at theory of mind in relation to disorders. The disorder that has received most study over the years with regard to associated limitations in theory of mind is autistic spectrum disorder; and this was reflected in the articles in this Research Topic.Fu-Qiang Qiao and colleagues conducted a comprehensive review of the literature over the past 30 years on the broader autism phenotype. They first used the Web of Science Core Collection database to find articles published between 1994 and 2024 on the autism phenotype in general. They then used the CiteSpace and VOS viewer software to visualize and analyze the citations further. They found a total of 1,075 articles about the broader autism phenotype. The annual number of publications on the subject has been rising over the 30-year period. The largest number of publications came from the United States, followed by England and Canada. The United States also came first with regard to the extent to which its publications were cited .Meng-Jung Liu used photographs of social scenes to compare adolescents with autism spectrum disorder and controls on their ability to reason and make inferences about people's intentions. The adolescents with ASD performed significantly less well than the controls in making inferences about intentions. However, their ability to make physical causal inferences was unimpaired. Liu also investigated relations between performance on these tasks and performance on tests of working memory and attention. Among individuals with ASD, working memory predicted physical causal inference, while divided attention predicted inferences about intention.

  PublisherOA PDFDOI

• Approximating the <scp>ZPD</scp> ? Young children's judgements of appropriate task level for learning

  British Journal of Developmental Psychology · 2024-08-28 · 2 citations

  articleOpen accessSenior author

  This research examined how 3-6-year-olds judge appropriate levels of counting games based on a person's ability, desire for learning and degree of difficulty. Study 1 found that 3-year-olds did not consider a character's ability or desire, whereas older children gave high ability characters large number games and low ability characters small number games when the characters wanted to play a manageable game. However, older children gave large number games to characters who wanted to learn counting, regardless of their ability. In Study 2, in addition to a similar developmental change of jointly considering a character's ability and desire, it was found that 5-and 6-year-olds were more sensitive to the degree of difficulty. They were more careful than younger children to choose exceedingly large number games given the character's ability and desire. Implications for children's understanding of the Zone of Proximal Development (ZPD) and goal orientation are discussed.

  PublisherOA PDFDOI

• Editorial: Theory of mind in relation to other cognitive abilities

  Frontiers in Psychology · 2023-01-18 · 3 citations

  editorialOpen accessCorresponding

  EDITORIAL article Front. Psychol., 18 January 2023Sec. Cognitive Science Volume 13 - 2022 | https://doi.org/10.3389/fpsyg.2022.1123321

  PublisherOA PDFDOI

• Development of Intention: The Relation of Executive Function to Theory of Mind

  Psychology Press eBooks · 2023-05-12 · 5 citations

  book-chapter1st authorCorresponding

  Historically, the construct of executive function has suffered from its ready use as a covering term for a heterogeneous collection of different abilities. Clarifying the relation between executive function and theory of mind could help to reveal the relative importance of these different abilities. However, the necessity for planning and problem solving becomes apparent when the relevance of intentional action to executive function is acknowledged. Embedded rules make explicit how theory of mind could be related to executive function. The two are distinct and neither underlies the other, yet in a sense they are two sides of the same thing, because they both rely on different use of the same set of rules. The role of intention in preschool executive function has less often been recognized, presumably because of the prevailing emphasis on response inhibition. When executive function is cast in terms of intentional action, its relation to theory of mind becomes much more comprehensible.

  PublisherDOI

• Local and distant brain control in melanoma and NSCLC brain metastases with concurrent radiosurgery and immune checkpoint inhibition

  Journal of Neuro-Oncology · 2022 · 21 citations

  • Medicine
  • Oncology
  • Internal medicine
  PublisherDOI

• Affective Theory of Mind in Late Adulthood: The Role of Emotion Complexity and Social Relatedness

  Experimental Aging Research · 2022-10-25 · 8 citations

  articleSenior author

  BACKGROUND: Age-related declines in adult affective theory of mind (AToM) have been discovered. However, AToM measures have not accounted for emotional state complexity involved in AToM. Measures have also not accounted for different types of relationships - friends versus strangers - for which AToM is employed, which is important considering the limited social networks of aging adults. OBJECTIVE: We address these issues and examine the emotion complexity, social-relatedness, and contextual relevance in AToM across adult ages (18-89 years) using a new task and two well-established measures. RESULTS: The new task displayed good structural fit and internal construct validity. Overall, an age-related decline in AToM was found along with an interaction between age and emotion complexity. For all ages, AToM performance was best for complex emotions. However, as age increased, there was more rapid decline in AToM for more complex emotions than for less complex ones. Surprisingly, AToM performance for strangers was better than for social companions. CONCLUSION: The findings suggest age-related AToM declines are more nuanced than previously understood given that adult age differences are related to emotional state complexity. They indicate that the emotion complexity levels of basic, complex, and self-conscious should be included in AToM assessments. Implications for AToM tasks and development are discussed.

  PublisherDOI

• When a Circle Becomes the Letter O: Young Children’s Conceptualization of Learning and Its Relation With Theory of Mind Development

  Frontiers in Psychology · 2021 · 6 citations

  Senior authorCorresponding
  • Psychology
  • Developmental psychology
  • Social psychology

  = 11.83, 31 girls) participated in study 2, where children were asked to judge whether the story protagonists intended to learn and whether they eventually learned. The results suggested that children over-attributed learning intention to discovery and implicit learning. Stories with conflict between the learning intention and outcome appeared to be most challenging for children. Children's intention judgment was correlated with their ToM understanding, and ToM marginally predicted intention judgment when the effect of age was accounted for. The implication of the findings for school readiness was discussed. Training studies and longitudinal designs in the future are warranted to better understand the relation between ToM development and children's learning understanding.

  PublisherDOI

• Self-regulated learning: Is understanding learning a first step?

  Early Childhood Research Quarterly · 2019-02-10 · 25 citations

  articleSenior author
  PublisherDOI

Frequent coauthors

• Elliot Soloway

  57 shared

• Henry Braun

  53 shared

• Randy Elliot Bennett

  Educational Testing Service

  53 shared

• Philip David Zelazo

  University of Minnesota

  35 shared

• Stuart Marcovitch

  University of North Carolina at Greensboro

  16 shared

• Chris Moore

  Dalhousie University

  12 shared

• Homan Mohammadi

  Mayo Clinic in Florida

  8 shared

• Ulrich Müller

  University of Victoria

  7 shared