Evidence for similar conceptual progress across diverse cultures in children’s understanding of emotion

Abstract

Research with adults has increasingly moved beyond the focus on a small set of allegedly basic emotions, each associated with a signature facial expression. That expansion has been accompanied by a greater emphasis on the potential variability of emotion concepts across different cultural settings. In this conceptual review of children’s understanding of emotion, we argue that it is also important in developmental research to look beyond the small set of emotions associated with distinctive facial expressions. At the same time, we caution against any premature rejection of a universalist approach to children’s understanding of emotion. We review three different lines of evidence in support of this stance: (1) children’s ability to appropriately cite situational elicitors for emotions beyond the basic set; (2) their developing understanding of the relations between emotions and other mental processes; and (3) their realization that a person’s facially expressed emotion may not indicate their felt emotion. In each of these three domains, we target studies that have included children from a variety of cultures to assess how far they respond similarly or differently. We conclude that there is robust evidence for similar conceptual progress in children’s understanding of emotion across a range of cultural settings.

Keywords

Children emotion understanding cross-cultural

Evidence for Universal Conceptual Progress in Children’s Understanding of Emotion

For several decades, psychological research with adults on the nature of emotion was dominated by the idea that, across diverse cultural settings, human beings experience and recognize a small set of basic emotions, with each emotion being identified on the basis of a distinctive facial expression. More recent theorizing has questioned this long-standing emphasis on basic emotions and facial expressions, with the plausible implication that the set of human emotions might be susceptible to important variation across different cultures and historical epochs (Barrett, 2006; Boddice, 2017).

With this theoretical debate as a backdrop, we review research on the development of children’s understanding of emotion. Echoing the findings with adults, we acknowledge that it is important in developmental research to look beyond children’s ability to identify the set of emotions that can be associated with distinctive facial expressions. At the same time, we argue that such an expansion of the developmental agenda will not necessarily show that children’s conceptualization of emotion varies markedly across different cultural settings. In that sense, we argue that it would be premature to reject the search for universals.

Given that we offer a conceptual review of selected, illustrative studies from various domains of research on children’s understanding of emotion, it will be helpful to provide a roadmap. We begin with a brief introductory review of recent research examining adults’ conceptualization of emotion, highlighting the shift away from an emphasis on the facial expression of basic emotions. We then proceed to a discussion of three aspects of children’s developing understanding of emotion. We focus on these three aspects because in each case they enable us to ask how far children’s understanding of emotion goes beyond their early emerging ability to identify and conceptualize emotions associated with a distinctive facial expression. More specifically, we review evidence confirming (1) children’s developing ability to conceptualize the situational elicitors for emotions beyond the basic set of emotions; (2) children’s emerging appreciation of the ways in which emotions are closely related to other mental states as measured by the Test of Emotion Comprehension (TEC; Pons & Harris, 2000); and (3) children’s developing realization that a person’s facially expressed emotion may be a misleading index of what that person is actually feeling.

In the context of each of these three aspects of children’s understanding of emotion, we highlight studies that have been conducted in different cultural settings. For example, we review evidence of the ability to generate situations for emotion among children in Western Europe as well as a remote area of Nepal. With respect to the TEC, we report in detail how far children’s performance on its various components exhibits a similar developmental ordering across a range of cultural settings. With regard to children’s emerging grasp of the distinction between experienced and overtly expressed emotion, we highlight its emergence among Western children and in a Melanesian culture that deems private experience to be opaque and unknowable. Taken together, the findings with respect to all three aspects of emotion understanding indicate that children (1) move beyond a conceptualization of emotion that is primarily focused on facially expressed emotions and (2) make similar conceptual progress in diverse cultures. Accordingly, we conclude that it remains fruitful to retain and further test a universalist approach to children’s understanding of emotion.

One important clarification is warranted. Our analysis focuses on the way that children conceptualize emotions. Accordingly, we make no claims about children’s emotional experience per se. Given their limited conceptual resources, children may be prone to misconstrue and misreport their emotional experience (Harris & Lipian, 1989). Indeed, there is experimental evidence that, just as young children are prone to misreport what they evidently believed a short time before, they are similarly prone to misreport what they evidently felt moments earlier (Bender et al., 2011).

In the concluding section, we highlight ways in which collaboration with cognate disciplines, notably history and anthropology, might enhance our understanding of how children’s understanding of emotion develops and also provide a stringent test of the current universalist proposals.

Adults’ Conceptualization of Emotion

Inspired by Darwin’s research on the expression of emotion, Ekman (1973) proposed that human beings universally experience, express, and recognize a small set of so-called basic emotions, notably happiness, sadness, fear, anger, surprise, and disgust. Within that framework, it was further argued that each of these basic emotions has a relatively distinct neural and physiological profile (Ekman, 1992). By contrast, more recent theories have proposed a “constructionist” approach to emotion, arguing that emotions are not biologically fixed but gradually constructed in the course of development (Barrett, 2017). This constructionist approach challenged the standard approach to basic emotions on four fronts. First, despite Ekman’s pioneering and influential cross-cultural research, there is persuasive evidence that the same facial expression can be interpreted differently depending on the larger context in which it is viewed (Barrett et al., 2019; Gendron et al., 2018; Russell, 1994). Second, efforts to identify a distinctive neural or physiological signature accompanying each of the basic expressions of emotion have not led to a clear consensus (Barrett, 2006; Lindquist et al., 2012). Third, from a developmental perspective, it is doubtful that distinctive emotions are readily detected and interpreted by human infants in the absence of guidance, especially linguistic guidance, from the surrounding culture (Barrett et al., 2019). Indeed, linguistic guidance is likely to be needed because language groups differ with respect to the particular, emotionally charged situations that are gathered into a single emotion category (Jackson et al., 2019). Thus, infants and young children are likely to group diverse instances of sadness or happiness together not because they are biologically equipped to pick out a recurrent and distinctive facial expression across those diverse instances, but because commentary by caregivers and others in the child’s immediate circle helps them to construct appropriate emotion categories (Hoemann et al., 2019). Fourth, as elaborated in more detail below, in the course of development, children’s conceptualization of emotions moves beyond the basic set of emotions emphasized by Ekman (1973). Hence, a comprehensive account of children’s developing understanding of emotion needs to go beyond that basic set.

It is important to acknowledge that, despite these challenges to research on basic emotions, some researchers argue for the existence of a larger set of discrete emotions across a wide range of cultures. Indeed, adults reported a set of more than 20 discrete varieties of emotional experience when presented with a large, heterogeneous sample of short videos (Cowen & Keltner, 2017). Moreover, provided analysis is not based on posed or static facial expressions, but on naturalistic video records of live human interaction from an extremely wide range of cultures in a variety of settings, it is possible, using deep neural networks to identify 16 specific facial expressions, each associated with particular types of social situations such as weddings and sporting events (Cowen et al., 2021).

Whatever the outcome of this ongoing debate about the way that emotion is expressed and conceptualized among adult, it is noteworthy that the recent shift in emphasis is congenial to two branches of emotion scholarship that highlight emotional development. First, despite the impact of Darwin and Ekman, social anthropologists have long emphasized the ways in which socialization practices, including pointed dialogue between children and their caregivers, might help children to acknowledge and indeed moderate their emotional states. For example, an ethnographic account of emotional development on the island of Ifaluk in the Western Pacific showed how adults seek to socialize the appropriate inhibition of excitement by children via expressions of song (justifiable anger)—which arouse metagu (apprehension) in an otherwise overly expressive or boisterous child (Lutz, 1987). Fieldwork among the Ilongot of Northern Luzon in the Philippines revealed a radically different pattern of socialization, in which the sustained and violent expression of liget (passionate anger), especially by headstrong young men, was viewed as a positive social force (Rosaldo, 1980). Indeed, prior to the mid-60s when governmental and missionary intervention contained the practice, elders among the Ilongot deliberately sought to nurture and channel expressions of liget among young men by helping them plan and execute head-hunting raids. Taken together, these very brief examples of ethnographic research illustrate how socialization practices are likely to moderate the development of emotional experience and expression, curbing certain forms of emotional expression and augmenting others.

Historians of the emotions have also been understandably reluctant to conclude on the basis of psychological theory that the history of human emotional experience is more or less invariant and, by implication, already written, insofar as the same small set of basic emotions has governed human experience from prehistory onward. Contrary to that timeless stance, historians have argued that analyses of paintings, letters, and archival documents highlight the emergence—and disappearance—of particular emotions or “emotional frontiers,” often tied to practices and moral assumptions about what should, or should not, be felt that no longer hold sway (Olsen, 2017). Accordingly, as might be expected, historians of emotion, including historians of childhood, have welcomed the more constructionist emphasis currently evident in psychological research, given that it lends support to the possibility of cultural and historical shifts in emotional experience (Boddice, 2017).

To summarize, multiple lines of enquiry into emotion, including contemporary psychological studies with adults, anthropological fieldwork, and historical research all suggest that the search for pan-cultural forms of development in the domain of emotion and emotion understanding is a challenging and contentious enterprise. Nevertheless, we aim to show that such an enterprise can be fruitful and revealing even as we acknowledge, in sympathy with constructionist approaches to emotion, that the focus on a small set of emotional expressions is restrictive and problematic.

Children’s Conceptualization of Specific Situations and Emotions

As noted above, the research by Ekman on emotion recognition led researchers to focus on the face and on facial expressions as a primary index of emotion. Adult participants were asked to indicate the facial expression of emotion that would be likely in particular situations, such as the unexpected arrival of a friend, or an encounter with a rotting carcass on a footpath. To the extent that adults from different cultures picked out the “appropriate” expression, the results suggested that a given facial expression is universally associated with a particular emotion. Equally important, but less frequently discussed, is the implication that particular situations evoke a similar emotion across different cultures—for example, a visit by a friend typically evokes happiness and an encounter with a rotting carcass typically evokes disgust.

If there are recurrent links between particular situations and particular emotions, we might reasonably expect young children to start noting those links from an early age. Indeed, early studies of children’s understanding of emotion demonstrated exactly that. Thus, when preschool and kindergarten children were told about various familiar situations (going to a party, having a quarrel, getting lost), they successfully indicated the particular emotional reaction that those situations would be likely to provoke (Borke, 1971). Conversely, when presented with various emotional reactions, they could nominate plausible situational elicitors. Thus, when preschoolers were asked what makes someone feel happy, sad, or afraid, they could describe situation that would typically do just that (Trabasso et al., 1981).

In addition, cross-cultural studies indicated that young children from different cultures frequently agree on the links between particular situations and basic emotions. For example, building on studies by Borke (1973) and Lewis (1989), Chinese and American 3- to 6-year-old children were presented with a variety of situations (e.g., winning a competition, a pet getting sick, having a nightmare about a monster; being forced to wear a disliked item of clothing) and asked to indicate whether the situation would provoke happiness, sadness, fear, or anger (Wang, 2003). The majority choice of emotion was the same across the two cultural groups for almost all (18 out of 20) of the situations.

Because of young children’s apparent facility in identifying emotional expressions and in linking them to particular situations, subsequent analysis of children’s understanding of emotion has tended to focus on the development of that early competence (e.g., Denham, 1986) and to examine family socialization practices, particularly emotion-related discourse, which might promote the understanding of such links (Denham et al., 1994; Grazzani et al., 2016). Admittedly, some researchers have questioned whether children’s early conceptualization of particular emotions, such as happiness or anger, is primarily cued by their associated facial expressions (Widen, 2016; Widen & Russell, 2010). Nevertheless, the research agenda has been dominated by a focus on such emotions.

However, it is likely that children’s conceptualization of emotion increasingly expands beyond the set of emotions that can be associated with a distinctive facial expression. To demonstrate that larger scope, children’s facility in describing the types of situations that would provoke various named emotions was examined (Harris et al., 1987). Children in the United Kingdom and the Netherlands, ranging from 5 to 14 years, were presented with a set of 20 emotion terms, some frequent and simple (e.g., happy, sad) and others less frequent and potentially more complex (e.g., relieved, disappointed). Children were invited to describe a situation that would elicit each particular emotion. To assess the specificity with which children responded, the situations that they described were shown to a set of raters whose task was to figure out which among the set of 20 emotion terms the child had been responding to. This meant that a generic response by the child was likely to be misassigned to the wrong eliciting term by the raters, whereas a detailed and specific response could be assigned more accurately. The study produced two notable results. First, according to the relatively stringent criterion just described, identifiable situations were generated for 5–6 emotion terms by 5-year-olds, for 12–14 terms by 7-year-olds, for 15 terms by 10-year-olds, and for 18–19 emotion terms by 14-year-olds. This steady increment with age was equally evident in both the United Kingdom and the Netherlands. Second, although the 5-year-olds described recognizably specific and appropriate situations chiefly for emotions associated with a distinctive facial expression (e.g., afraid, happy, sad, and angry), this was not true for the older children. Seven-year-olds proposed situations for several, additional emotions (e.g., proud, jealous, grateful, worried, guilty, and excited). Of course, these latter emotions may be accompanied by a facial expression, but such expressions do not readily permit a distinction between, for example, feeling happy versus feeling proud or feeling afraid versus feeling worried. By implication, from as early as 7 years, children’s emotional horizon extends beyond the basic set of emotions proposed by Ekman. Indeed, as might be expected, this expansion is even more evident among 10- and 14-year-olds.

The pattern of development was quite similar in both the United Kingdom and the Netherlands, with respect to both how many emotions and also which particular emotions were understood at any given age. However, it could be argued that such cross-national parallels are scarcely surprising. First, English and Dutch are cognate languages. Second, both countries are industrialized, Western European democracies with universal access to schooling and a Christian heritage. Third, beyond these demographic similarities, it is feasible that patterns of socialization, both within the family and in the context of peer relationships, are similar in the two countries.

Accordingly, to find out how far children’s understanding of specific emotions would be similar or different in a radically different culture, a follow-up study, using the same technique, was carried out in Pangma, a remote village in Eastern Nepal (Harris et al., 1987). Two groups of children were interviewed: younger children ranging from 6 to 10 years (M = 8 years 10 months) and older children ranging from 12 to 14 years (M = 13 years 4 months). The children were all native speakers of Lohorung, a language of Tibeto-Burman descent. Almost all of the villagers were engaged in subsistence agriculture. Schooling was available and valued by the community, but domestic and agricultural tasks often interrupted children’s school attendance or cut short their schooling altogether. Thus, the younger children had received, on average, 18 months of schooling and the older children had received 4 years. At the time when the interviews took place (in the summer of 1981), Pangma was remote from any urban center: the children had rarely, if ever, encountered foreigners. Roads, cars, bicycles, electricity, television, cinema, and telephones were non-existent. The world beyond the local, neighboring regions mainly reached the village only via a few radio sets. In sum, the children were growing up in a small, isolated agricultural community, scarcely touched by Western technology and culture, by literacy, or by any Abrahamic religion.

To identify suitable emotion terms, Charlotte Hardman, a social anthropologist who had previously conducted fieldwork in Pangma for 2 years and was fluent in Lohorung, consulted with her former adult informants to arrive at a set of 16 common emotions terms and phrases in Lohorung that could be tentatively glossed in English. For example, niwa chaikhe’mi could be rendered as worried, ngongsik’mi as jealous, and niwa enta’e as relieved. As in the initial study, children were interviewed individually—presented with each term or phrase and invited to describe a situation that would be likely to elicit the emotion in question. The various descriptions that children proposed were subsequently stripped of tell-tale clues, mixed together, and two raters sought to assess which particular term or phrase each child had been responding to.

The younger group proposed identifiable situations for 11 emotions and the older group for 14 emotions, thereby replicated the findings in Europe that there is an expansion of children’s emotional horizon with age. Moreover, even among the younger group of children, the scope of that horizon already exceeded the set of basic emotions identified by Ekman. Thus, beyond proposing recognizable situations for experiencing basic emotions, younger children proposed identifiable situations for feeling jealous, relieved, shy, irritable, hopeful, and pity.

Notwithstanding the marked change of cultural setting, the emotions that children were invited to think about and the situations that they described as eliciting them could be paired by the raters. For example, when an 8-year-old said, “If mother doesn’t come back from the bazaar, and only arrives the next day, then in the night I feel . . . ,” it was plausible that the child had been responding to the phrase niwa chaikhe’mi (i.e., worried); when a 9-year-old said, “When friends have something like clothes and sweets that you don’t have,” it was plausible that the child had been responding to the term ngongsik’mi (i.e., jealous); and when a 13-year-old said, “When I’ve done something—work has gone wrong—I think whether or not my father knows. I wonder whether they might hit me. When they don’t (hit me), I feel . . . ,” it was plausible that the child had been responding to the phrase niwa enta’e (i.e., relieved). Thus, the emotionally charged situations described by the Lohorung children were identifiable, even if the particular elicitors (e.g., a mother’s unannounced overnight stay at the bazaar or the threat of physical punishment for work gone awry) were emblematic of the local culture. More generally, there was little in the children’s conceptualization of their emotional experience that was incommensurable with that of their peers in Western Europe. Admittedly, such alien categories of emotional experience might exist—and their role in the Lohorung community might have been overlooked or misconstrued during the initial consultations with informants and more extensive analysis of the Lohorung’s indigenous psychological concepts is available (Hardman, 2000). Still, the study with Lohorung children offers an important existence proof that there is considerable cross-cultural stability in children’s conceptualization of their emotional experience. That stability is found even when we look beyond those emotions that can be closely tied to a distinctive facial expression, to consider more complex emotions such as worry, jealousy, and relief. Finally, that stability emerges even when we study children’s emotional experience in a community that has been little impacted by Western culture and technology.

To avoid potential misunderstanding, the claim of universality proposed here needs clarification in one important respect. It should not be taken to imply that children everywhere will always propose a similar list of particular, concrete situations when invited to say what would provoke a given emotion. For example, when an 8-year-old boy in Pangma was asked to say what would make him feel relieved, he said, “If somebody’s cattle eat all the crops, but then the owner of the cattle pays it back, you feel relief.” Not surprisingly, none of the children interviewed in the United Kingdom or the Netherlands gave such a reply. However, the identification of commonalities across cultures may sometimes call for an abstraction away from specific situations to recurrent underlying themes (Lewis, 1989). Thus, in describing what made them relieved, children in the United Kingdom, the Netherlands, and Nepal typically identified a potentially negative outcome that failed to materialize even if the types of negative outcomes they described were often different.

Recent work on children’s emotion understanding provides further strong support for the proposal that children steadily expand their conceptual repertoire for emotions beyond the basic set and succeed in doing so by identifying the types of situations that recur with respect to a given lexical category of emotional experience. Nook et al. (2020) presented US participants ranging from 4 to 25 years with 24 emotion terms and asked for a definition. Coders then rated the adequacy of the definition. Consistent with the findings reviewed so far, there was a steady expansion with age in participants’ ability to understand and define emotion terms, and again that expansion went beyond those terms that can be readily tied to a particular facial expression. Indeed, by the age of 10, children were able to provide adequate definitions for all 24 terms.

There was also a notable developmental shift in the way that children provided a definition. At 5 years of age, they mostly proposed an example of a specific situation that had given rise to the emotion, for example, “I felt sad when my pet died.” But with age, they were increasingly likely to shift to a more general, definition-like strategy, for example, “People feel sad when they experience loss.” Indeed, from around 10 years of age, this more generic strategy predominated. Nevertheless, as these two examples indicate, children of all ages readily conceptualized each emotion either in terms of a particular situation, or, alternatively, the class of situations, that would provoke it—echoing and reinforcing earlier findings (Harris et al., 1987). Children rarely mentioned bodily sensations or physiological changes at any age.

To underline the key points so far—although young children are adept at connecting emotions to certain facial expressions, children are also good at conceptualizing emotion terms in relation to the situations that provoke them. Indeed, among 7- and 8-year-olds, this understanding of emotion extends beyond those basic emotions that can be easily associated with distinctive facial expressions. When explicitly asked, children can describe situations that would provoke those emotions and when asked to define a given emotion, they spontaneously describe a situation or a class of situations that would provoke it. Finally, this aspect of children’s developing understanding of emotion is not culture specific. Whether children are interviewed in urban areas of Western Europe, in North America, or in a remote, rural area of Nepal, they readily talk about situations that can elicit basic emotions such as sadness, fear, and happiness and also about situations that can elicit more complex but nonetheless distinct emotions, such as jealousy, worry, and relief. By implication, even if we move away from the assumption that there is a small set of human emotions, each closely linked to a distinctive facial expression, we need not abandon the search for universals of emotional experience that can be defined and recognized in light of the class of situations that provoke them.

Children’s Understanding of Generalized Aspects of Emotion

In this section, we move beyond a discussion of children’s understanding of particular emotions (such as jealousy, worry, or relief) to a discussion of the development of children’s understanding of aspects of emotions that generalize across most, if not all, emotions—for example, their developing understanding that emotions often depend on mental states, especially desires and beliefs; that emotions may or may not be expressed overtly; and that conflicting emotions can co-occur. Building on experimental studies in the theory-of-mind tradition probing children’s developing grasp of these more over-arching aspects of emotion (Harris, 1989), Pons & Harris (2000); Pons et al. (2004) created the TEC. Below, we review the findings from studies that have used the TEC to assess how far children’s developing understanding of emotion advances in a similar fashion across a range of cultural settings.

Before turning to that review, it is appropriate to explain and justify our selective focus on the TEC. A comprehensive review identified 56 unique tools for measuring children’s understanding of emotion (Castro et al., 2016). However, most of those tools (80%) focus either on emotion recognition ability, notably the ability to identify and label particular emotions, or alternatively on broader knowledge about emotions, including an understanding of the causes, consequences, and management of emotions. Thus, only a minority of measurement tools assess both emotion recognition and emotion knowledge. Furthermore, among those tools that do assess both aspects of emotion understanding, most are designed for a single age period, typically early childhood. Indeed, only two of the 56 measurement tools identified by Castro et al. (2016) provide an assessment of both emotion recognition and emotion knowledge across more than one age period: the TEC (Pons et al., 2004) and the emotional intelligence test (Mayer et al., 1999, 2003). However, the latter is designed for adolescents and adults rather than for children. Accordingly, we opted to review studies that have used the TEC, given that (1) it assesses children’s recognition of emotions as well as their broader knowledge about emotions and (2) is suitable for children ranging from 3 to 11 years.

The TEC includes nine different components that can be arranged into three successive phases based on the age at which children pass the relevant components as well as the particular themes targeted by components within a given phase (Pons et al., 2004). More specifically, by the age of 5 years, most children are able to identify key external aspects of emotion: as noted earlier, they can recognize facial expressions of happiness, sadness, anger, and fear, and they can also say what emotion an encounter with various concrete situations (e.g., a birthday, the loss of a pet) would provoke. By around 7 years of age, children are increasingly able to understand important, mentalistic aspects of emotions—notably the fact that emotions are elicited not by the objective features of a given external situation, but rather by the particular way that a situation is appraised (whether accurately or inaccurately) by someone, especially in light of their beliefs and desires. In addition, children increasingly realize that the emotion that someone expresses overtly may not correspond to their private experience of emotion. Finally, around 10 years of age, children are able to understand more reflective aspects of emotion: for example, they realize that sadness at a loss can be mitigated by turning thoughts away from the loss, that the same situation can evoke conflicting thoughts and hence mixed emotions, and that a failure to confess a misdeed can lead to guilty rumination.

Studies of the TEC have established that both the English (Pons & Harris, 2005; Pons et al., 2002) and Italian versions (Ornaghi et al., 2011) have acceptable test–retest reliability. They have also lent support to the three-phase, developmental progression described above: children’s understanding of emotion is initially focused on its visible and external aspects, but increasingly incorporates key mentalistic aspects of emotion, and eventually incorporates reflective aspects. For example, when the Italian version of the TEC was administered to ~1,500 typically developing 3- to 10-year-old Italian children from a variety of backgrounds, children’s scores for external items increased continuously up to 7 years but then leveled off, in that most children answered all items correctly (Cavioni et al., 2020). Improvement on mentalistic items was more protracted with the oldest children aged 10 years eventually obtaining maximum scores. Finally, despite low initial performance, children steadily improved with age on reflective items but did not obtain maximum scores even at 10 years. These developmental advances were clearly seen among both males and females. Studies with a narrower age range than those just described have reported a similar overall improvement with age (and no effect of gender) in a variety of other countries, including Albania (Farina & Belacchi, 2014), Germany (Göbel et al., 2016; Janke, 2008; Voltmer & von Salisch, 2019), Norway (Kårstad et al., 2015), Portugal (Franco et al., 2017), and Russia (Veraksa et al., 2020).

Beyond looking at the improvement with age in children’s overall score on the TEC, a considerable number of studies undertaken in Europe, China, and South America have followed up earlier findings with the TEC (Pons et al., 2004) by examining the relative difficulty of the nine components of the TEC, thereby permitting a more fine-grained comparison of the course of conceptual development across different countries. Accordingly, to identify pertinent studies for review, we identified all papers that cited Pons et al. (2004). We then excluded papers that (1) were not in English; (2) were dissertations; (3) examined only a single age group instead of an age range; (4) did not report the percentage of children passing each of the nine TEC components; and (5) examined children from an English-speaking country only (e.g., the United States, the United Kingdom, Canada, Australia)—with the sole exception of Pons et al. (2004) which was included as an initial benchmark. These selection criteria yielded a total of 11 studies for review. We show relevant findings for younger children in Table 1 and for older children in Table 2. Note that three studies included a wide age range: Pons et al. (2004), Cavioni et al. (2020) and Franco et al. (2017). Accordingly, for comparison purposes, the relevant sample in each of these three studies was divided, with data for younger children included in Table 1 and data for older children included in Table 2.

Table 1.

Rank and Percent Correct on Each Component of the Test of Emotion Comprehension for Younger British, Norwegian, German, Italian, Portuguese, Chinese, and Brazilian Children.

Component	British: N = 60, age = 3–7, M = 5.0		Norwegian: N = 882, age = 4–5, M = 4.4		German: N = 108, age = 3–6, M = 5.0		Italian: N = 114, age = 3–6, M = 4.8		Italian: N = 1,177, age = 3–7, M = 5.3		Portuguese: N = 267, age = 4–7, M = 6.3		Chinese: N = 192, age = 4–6, M = 5.2		Brazilian: N = 68, age = 5–9, M = 7.5
	Rank	%	Rank	%	Rank	%	Rank	%	Rank	%	Rank	%	Rank	%	Rank	%
I—Recognition	1	73	1	89	1	76	1	72	1	89	1.5	88	1	92	1.5	97
II—External	3	65	2	49	2	57	2	50	2	68	1.5	88	2.5	79	1.5	97
III—Desire	4	50	4	38	3	44	4	43	3	60	6	33	2.5	79	3	93
IV—Belief	5	48	8	20	6	28	6	37	7	43	5	35	5	40	9	59
V—Reminder	2	68	5	35	4	35	5	42	5	46	3	51	6	36	5	74
VI—Regulation	8	18	6	30	7.5	27	8	30	8	32	4	37	7	35	6	72
VII—Hiding	6	40	3	43	5	30	3	48	4	49	8	27	4	67	7	71
VIII—Mixed	9	13	7	25	9	10	9	13	9	21	7	31	8	13	8	62
IX—Morality	7	20	9	9	7.5	27	7	35	6	45	9	23	9	2	4	77

Note. Age range and mean age are in years. Data are based on the work of Pons et al. (2004) for British 3- to 7-year-olds; Kårstad et al. (2014) for Norwegian 4- to 5-year-olds; Molina et al. (2014) for German and Italian 3- to 6-year-olds; Cavioni et al. (2020) for Italian 3- to 7-year-olds; Franco et al. (2017) for Portuguese 4- to 7-year-olds; Tang et al. (2018) for Chinese 4- to 6-year-olds; and Viana et al. (2020) for Brazilian 5- to 9-year-olds.

Table 2.

Rank and Percent Correct on Each Component of the Test of Emotion Comprehension for Older British, German, Italian, and Portuguese Children.

Component	British: N = 40, age = 8–11, M = 10.0		German: N = 135, age = 7–10, M = 9.0		Italian: N = 283, age = 8–10, M = 9.4		Portuguese: N = 182, age = 8–11, M = 8.8		Portuguese: N = 139, age = 8–10, M = 9.4
	Rank	%	Rank	%	Rank	%	Rank	%	Rank	%
I—Recognition	2	100	1.5	100	1.5	100	1	99	2	97
II—External	2	100	1.5	100	1.5	100	2	97	1	99
III—Desire	5	90	3	90	4	94	5.5	76	7	53
IV—Belief	4	98	4	84	7	69	3	82	8	46
V—Reminder	2	100	5	82	3	97	4	80	3	73
VI—Regulation	8.5	70	6.5	79	8	68	5.5	76	4	65
VII—Hiding	6	88	6.5	79	5	83	7.5	69	6	58
VIII—Mixed	7	78	8	61	9	67	7.5	69	5	59
IX—Morality	8.5	70	9	47	6	75	9	43	9	33

Note. Age range and mean age are in years. Data are based on the works of Pons et al. (2004) for British 8- to 11-year-olds; Göbel et al. (2016) for German 7- to 10-yearolds; Cavioni et al. (2020) for Italian 8- to 10-year-olds; Rocha et al. (2015) for Portuguese, 8- to 11-year-olds; and Franco et al. (2017) for Portuguese 8- to 10-year-olds.

Two of the 11 studies tested children in South America. One study involved 100 Brazilian 3- to 5-year-olds, half from lower-class families and half from middle-class families (Kårstad et al., 2016). As compared to the 3- and 5-year-olds tested by Pons et al. (2004), performance was quite poor on two components—the recognition and external cause components—especially among the lower-class children. The second study was based on a small sample of Peruvian children (18 4- to 7-year-olds and 21 8- to 11-year-olds) in an agro-pastoralist Quechua community (Tenenbaum et al., 2004). Again, as compared to the 3- to 11-year-olds tested by Pons et al. (2004), performance was poor on the recognition and external cause components. The majority of TEC components require that children select the expression representing the appropriate emotion in a given scenario. Hence, children’s poor performance on the recognition and external cause components raises doubts about the measurement of their performance on the remaining, more challenging, components. Accordingly, we did not retain these two studies for further analysis.

To provide a systematic comparison of the relative difficulty of the nine components of the TEC across a range of cultural settings, Spearman’s rank-order correlations and confidence intervals were calculated within the set of studies conducted with younger children (Table 3) as well as within the set of studies conducted with older children (Table 4). As shown in Table 3, all coefficients were significant and most confidence intervals acceptable when comparing the performance of younger children from five of the seven countries listed: Britain (Pons et al., 2004), Norway (Kårstad et al., 2014), Germany (Molina et al., 2014), Italy (Cavioni et al., 2020; Molina et al., 2014), and China (Tang et al., 2018). The pattern of correlation and the confidence intervals were more variable for the two remaining countries, Portugal and Brazil. More specifically, the rank-order for younger Portuguese children (Franco et al., 2017) correlated significantly with that for British (Pons et al., 2004) and German (Molina et al., 2014) children and marginally with that for Norwegian (Kårstad et al., 2014) and Chinese (Tang et al., 2018) children. The rank-order of TEC components among Brazilian children (Viana et al., 2020) significantly correlated with that for German (Molina et al., 2014) and Italian (Cavioni et al., 2020) children and marginally with that for British (Pons et al., 2004), Norwegian (Kårstad et al., 2014), and Italian (Molina et al., 2014) children. In sum, these findings with younger children point to a similar sequence of EU development across diverse cultures, with most but not all rank-order correlations proving significant.

Table 3.

Spearman’s Rho Coefficients Comparing the Rank-Order of Components on the Test of Emotion Comprehension among Younger British, Norwegian, German, Italian, Chinese, Portuguese, and Brazilian Children.

	Age	1	2	3	4	5	6	7
1. British	3–7
2. Norwegian	4–5	.67* [.03, .92]
3. German	3–6	.93** [.69, .99]	.85** [.42, .97]
4. Italian	3–6	.83** [.39, .96]	.87** [.48, .97]	.95** [.75, .99]
5. Italian	3–7	.83** [.39, .96]	.83** [.39, .96]	.95** [.79, .99]	.97** [.84, .99]
6. Chinese	4–6	.76* [.22, .95]	.87** [.50, .97]	.92**[.64, .98]	.90** [.60, .98]	.86** [.47, .97]
7. Portuguese	4–7	.70* [.10, .93]	.61^† [−.06, .9 ]	.67* [.04, .92]	.51 [−.20, .87]	.48 [−.24, .86]	.59^† [−.09, .90]
8. Brazilian	5–9	.64^† [−.02, .91]	.63^† [−.03, .91]	.75* [.19, .94]	.66^† [.02, .92]	.81** [.34, .96]	.55 [−.15, .88]	.50 [−.22, .86]

Note. Age range is in years. Confidence intervals of Spearman’s rho coefficients are in brackets. Estimation of standard error is based on the formula of Caruso and Cliff (1997). Data are based on the works of Pons et al. (2004) for British 3- to 7-year-olds; Kårstad et al. (2014) for Norwegian 4- to 5-year-olds; Molina et al. (2014) for German and Italian 3- to 6-year-olds; Cavioni et al. (2020) for Italian 3- to 7-year-olds; Tang et al. (2018) for Chinese 4- to 6-year-olds; Franco et al. (2017) for Portuguese 4- to 7-year-olds; and Viana et al. (2020) for Brazilian 5- to 9-year-olds.

†

p < .1; *p < .05; **p < .01.

Table 4.

Spearman’s Rho Coefficients Comparing the Rank-Order of Test of Emotion Comprehension Components Among Older British, German, Italian, and Portuguese Children.

Age	1	2	3	4
1. British 8–11
2. German 7–10	.84** [.40, .96]
3. Italian 8–10	.80** [.31, .96]	.76* [.21, .94]
4. Portuguese 8–11	.85** [.42, .97]	.91** [.61, .98]	.63^† [–.02, .91]
5. Portuguese 8–10	.60^† [–.07, .90]	.56 [–.13, .89]	.56 [–.13, .89]	.66^† [.01, .92]

Note. Age range is in years. Confidence intervals of Spearman’s rho coefficients are in brackets. Estimation of standard error is based on the formula of Caruso and Cliff (1997). Data are based on the works of Pons et al. (2004) for British 8- to 11-year-olds; Göbel et al. (2016) for German 7- to 10-year-olds; Cavioni et al. (2020) for Italian 8- to 10-year-olds; Rocha et al. (2015) for Portuguese 8- to 11-year-olds; and Franco et al. (2017) for Portuguese 8- to 10-year-olds.

†

p < .1; *p < .05; **p < .01.

As shown in Table 4, the Spearman rho coefficients were also significant and the confidence intervals acceptable among older children in the United Kingdom (Pons et al., 2004), Germany (Göbel et al., 2016), and Italy (Cavioni et al., 2020). Echoing the pattern for younger Portuguese children, the pattern of correlation and the confidence intervals for older Portuguese children were more variable. The rank-order of TEC components among 8- to 11-year-old Portuguese children in the work of Rocha et al. (2015) correlated significantly with that for British (Pons et al., 2004) and German (Göbel et al., 2016) children and marginally with that for Italian (Cavioni et al., 2020) children. Unexpectedly, the rank-order for 8- to 10-year-old Portuguese children in the work of Franco et al. (2017) was only marginally significantly correlated with that for Portuguese children in the work of Rocha et al. (2015) and for British children in the work of Pons et al. (2004), and did not correlate significantly with that for German and Italian children. Despite this variation, all correlations among the older children were positive.

Overall, these studies indicate a similar sequence of conceptual development on the TEC among both younger and older children across a variety of cultures in Europe, China, and South America. By implication, children construct a similar conception of emotion across these setting—focusing initially on external aspects of emotion, increasingly grasping mentalistic aspects of emotion, and eventually realizing how emotions can be re-directed via reflective processes. Nevertheless, caution is appropriate. The majority of the children tested hitherto have been growing up in urban or semi-urban settings, with access to schooling as well as books and television. Thus, it remains an open question whether the same sequence would emerge in other settings. Indeed, the TEC relies on the schematic, cartoon-like depiction of facially expressed emotions (happiness, sadness, fear, and anger) and the description of episodes involving artifacts (e.g., a bus stop, a birthday cake, a bicycle, a photo album) that may not be familiar to children in remote, rural communities. As noted earlier, the studies conducted by Kårstad et al. (2016) with Brazilian children and by Tenenbaum et al. (2004) with Quechua-speaking Peruvian children reported such challenges. Accordingly, careful selection and adjustment of test items may be needed for valid testing. We return to the challenges of cross-cultural research with the TEC in our concluding remarks.

In the next section, we look more closely at one particular component of the TEC (Component VII) which targets children’s realization that overt indices of emotion, especially a person’s facial expression, may be a misleading index of that person’s actual feelings. The relevant studies again demonstrate that children’s understanding of emotion exhibits similar conceptual progress with respect to this distinction in different cultural settings and they further highlight the limitations of a focus on facial expressions of emotion.

Children’s Understanding of Hidden Emotions

Young children increasingly understand that it is sometimes expected or appropriate to hide one’s actual emotion. Whereas 4-year-olds struggle to understand the conflicting elements in such social encounters, 6-year-olds are more insightful. Thus, when presented with appropriate vignettes about a protagonist who has reason to conceal his or her emotion from onlookers, 6-year-olds can typically infer (1) the protagonist’s actual emotion as well as its cause (e.g., “She’s sad because she fell over”); (2) the protagonist’s expression of a different emotion, together with the motive for such dissembling (“Happy—because she didn’t want the other children to laugh at her”); and (3) the misattribution that onlookers will likely make as a result of the protagonist’s misleading expression (Harris & Gross, 1988).

Experimental studies revealed such an age change among young children in the United Kingdom (Harris et al., 1986; Tenenbaum et al., 2008), the United States (Gross & Harris, 1988), and Japan (Gardner et al., 1988). Moreover, as noted in the preceding section, an assessment of children’s understanding of hidden emotions is included in the TEC. More specifically, for the Hiding task children are presented with the following probe:

This is Tom and this is Daniel. Daniel is teasing Tom because Daniel has lots of marbles and Tom does not have any. Tom is smiling because he does not want to show Daniel how he is feeling inside. How is Tom really feeling inside?

Children are credited with a correct answer if they acknowledge that Tom feels badly about being teased—even if he is smiling. As would be expected on the basis of the earlier experimental findings, a comparison of Tables 1 and 2 confirms that there is a considerable improvement with age in children’s performance on this component of the TEC (Component VII). This age change is especially clear in the two studies that included a wide age range and therefore straddle Tables 1 and 2. Thus, among British children (Pons et al., 2004), 40% of 5-year-olds passed the Hiding component, whereas 88% of 10-year-olds did so. Similarly, among Italian children (Cavioni et al., 2020), 49% of 5-year-olds passed the Hiding component, whereas 83% of 9-year-olds did so.

A similar probe of children’s understanding of hidden emotions has been incorporated into the Theory-of-Mind scale (Wellman & Liu, 2004). Children are asked to indicate how a boy really feels when everyone laughs at a mean joke that another child has told about him and also how he looks on his face. They are given credit if they acknowledge that the boy feels worse than he looks. A steady improvement with age has been observed in studies conducted in Australia, Iran, and China (Duh et al., 2016; Shahaeian et al., 2014; Wellman et al., 2006)—with only a small minority of preschoolers succeeding and more than three-quarters of 7- to 9-year-olds. Thus, despite differences in the social situation that children are invited to think about as well as the framing of the questions posed, findings from the TEC and from the Theory-of-Mind scale reveal a similar age change across a range of cultures.

Granted the stability and ubiquity of this conceptual advance, three issues warrant further discussion: (1) its implications for research on children’s attribution of emotion; (2) the cues that enable children to understand the difference between felt and expressed emotion; and (3) the possibility that such a conceptual distinction is universal.

With respect to the first issue, as noted earlier, a long line of research has assessed the ability of children and adults to determine how a person feels from their facial expression. However, there are many social situations in which that is not possible. Indeed, a focus on the emotion expressed can be misleading. The findings from both the TEC and the Theory-of-Mind scale highlight young children’s increasing recognition of this fact. Moreover, fully in line with the findings discussed earlier regarding children’s understanding of the situations that provoke specific emotions, children give priority to a different cue: they infer what someone is likely to be feeling from the situation that they are facing and not from the emotion that they display. In sum, research on children’s increasing ability to differentiate between privately felt and outwardly expressed emotion again underlines the limitations of conceiving of children’s understanding of emotion in terms of their ability to recognize and label facially expressed emotions, even if, as noted earlier, that ability has been the focus of many studies of children’s understanding of emotion (Castro et al., 2016).

Turning to the second issue, how do children gradually come to realize that felt and expressed emotion do not necessarily coincide? More specifically, how do they come to realize that facially expressed emotion may be a misleading index of someone’s actual feelings? One possibility is that children enlarge the scope of their non-verbal processing. For example, they might pay more careful attention to someone’s gestures, tone of voice, or posture and infer from such cues that the person’s facial expression does not signal what they actually feel. However, the available evidence shows that children are quite poor at differentiating between genuine and fake expressions of emotion (DePaulo & Jordan, 1982). Even when presented with paired pictures of fake versus genuine smiles, 5-year-olds proved unable to recognize which picture showed someone who was “really smiling” versus “just pretending to smile”—except under relatively artificial conditions, namely when two pictures of the same person were presented side-by-side (Song et al., 2016). In any case, this line of explanation, implying a more subtle analysis of expressive cues, begs an important conceptual question: how are children to determine which particular expressive cues are valid and which are misleading? To make such a determination implies that they have some pre-existing and reliable knowledge of what another person genuinely feels. But, if so, where does that knowledge come from? Arguably, it is based on input from agents of socialization. Presumably, children often learn what feelings they should hide and what they may express on the basis of adult teaching. However, this does not speak to the thornier issue of how children decide what their real feelings are: the feelings that they express overtly or those that they feel privately. In other words, when children experience negative feelings but are told to express positive feelings—or the reverse—how do children know what they actually feel? It seems unlikely that parents will help them with that conceptual dilemma. If anything, parents, and indeed other adults, might mislead children by responding to them as if their expressed feelings were their actual feelings.

Accordingly, it is worth considering a radically different way in which children’s conceptual development might proceed. Children might gain an initial purchase on the possibility that felt and expressed emotion do not always coincide, not by closer or more refined scrutiny of other people but by reflection on their own experience. It is plausible that children will often encounter social situations in which they experience a particular emotion, such as fear, disappointment, or joy, but inhibit the overt expression of those emotions. Such occurrences would be instructive for children if the following two conditions were met: (1) children know, given how they appraise the situation that they find themselves in, what they are actually feeling, despite their efforts at concealment, and (2) other people, by contrast, fail to recognize what they are actually feeling. Faced with this duality, children would have evidence that their emotions can remain hidden from other people—and more generally that they can know what they themselves feel even if other people do not. It is plausible that repetition of such experiences would lead children to realize that they are not alone in this respect. More specifically, they will come to the realization that people in general can hide their true feelings—just as they do themselves.

Is there empirical evidence that the above conditions are indeed fulfilled? Do children recognize what they are actually feeling, even in contexts where they seek to hide those feelings? And do children presume that, as a result, other people are likely to be ignorant of what they actually feel? Experimental work with the disappointing gift paradigm, pioneered by Saarni (1984), provides persuasive answers to both these questions. In this paradigm, children designate which of several toys they would prefer to receive later as a reward for helping the experimenter, but, disappointingly, they are eventually given a drab or broken toy instead. Young children, including preschoolers, express more positive feelings in the presence of the donor than when left alone (Cole, 1986). By implication, children are apt to mask their feelings of disappointment. Follow-up studies have confirmed that such masking is a robust phenomenon and throw further light on children’s construal of the situation. For example, the reactions of 5-year-olds at low, moderate, or high risk for developing disruptive behavior disorders were recorded both immediately on being given their non-preferred toy by an experimenter and also once that experimenter had left them alone with that disappointing gift (Cole et al., 1994). Children expressed more positive emotion (as indexed by their facial expression and/or their tone of voice) when the experimenter was present as compared to when they were alone. Moreover, this difference in expression was quite robust—it emerged irrespective of children’s gender and of which risk group they belonged to. Thus, many young children, including those at risk of disruptive behavior, are prone to mask their disappointment from another person.

When subsequently interviewed, children acknowledged their disappointment in two ways. First, they all said that they had not been given the gift they wanted and chose to exchange it for their original choice. Second, when asked how they had felt, the majority reported feeling a negative emotion (such as sad, mad, or yukky) and the remaining minority reported feeling neutral, ambiguous, or no emotion. Thus, no child reported positive emotion. By implication, even if children expressed positive emotion immediately after the experimenter gave them the disappointing gift, that expressive behavior had no misleading impact on the feelings that they reported later—which were reported as predominantly negative rather than positive. Finally, when explicitly asked, the majority of children said that the experimenter who had given them the disappointing toy did not know how they felt about it.

A similar pattern has emerged in other cultural settings (Ip et al., 2021). When 5-year-olds in the United States, China, and Japan were presented a disappointing gift, children in all three settings displayed a relatively consistent reaction—they expressed more positive emotion in the presence of the experimenter than when left alone with the gift. Nevertheless, consistent with earlier findings (Cole et al., 1994), when directly asked how they had felt, the majority reported either negative or neutral feelings about the gift—fewer than 20% of children in each culture reported positive feelings—and between approximately half and three-quarters claimed that the experimenter who had offered them the disappointing gift would not know how they felt about it.

Overall, the data from these two studies are impressively consistent. They show that young children remain aware of their own feelings even when they engage in non-verbal masking of those feelings. Moreover, children presume that other people will not be aware of their masked feelings. This overall pattern is evident, irrespective of gender, risk status for disruptive behavior and cultural setting. These robust findings provide encouraging evidence that children could come to understand the difference between felt and expressed emotion by registering the gap in knowledge between self and other, that is, by remaining aware of what they actually feel while inferring that other people are not. More generally, the pattern of results indicates that children conceptualize their actual feelings not in terms of the emotional signals that they overtly express but in terms of the negative situation they have experienced—namely receiving a disappointing gift.

Is the distinction between actual and expressed emotion universally understood by young children? Clearly, it is not practicable to sample children in all the world’s cultures, but, as a strong test case, we can ask whether it is understood even in a culture where hidden feelings are rarely discussed or acknowledged. Robbins reviews ethnographic evidence showing that various Melanesian cultures endorse a distinctive stance toward mental states—a stance that is typically called “the opacity of mind doctrine” by social anthropologists (Robbins, 2021). Members of such cultures adopt the view that mental states, especially, the mental states of other people, are opaque and cannot be known. Accordingly, there is an emphasis on people’s overt words and deeds rather than on the feelings and motives behind them. Indeed, enquiries about such opaque mental states are regarded as inappropriate. Detailed analyses of the application and manifestations of this doctrine are available for the Urapmin (Robbins, 2008) and for the Korowai (Stasch, 2008).

How do young children growing up in such a Melanesian culture respond to the opacity doctrine? Two quite different responses seem feasible. On the one hand, to the extent that references to mental states are likely to be limited and even discouraged in such a culture, it is conceivable that children’s acquisition of a Theory-of-Mind will be delayed and that, in particular, their tendency to differentiate between the overt expression of emotion and the private experience of emotion will be reduced. After all, the opacity of mind doctrine implies that private experience is hard to know and not to be asked about. On the other hand, to the extent that children remain aware of their own private feelings but also—in line with the opacity doctrine—alert to the possibility that other people may not be, it is plausible that they will acknowledge the difference between felt and expressed emotion—in much the same way as children in other cultures.

Pertinent findings, based on 400 children living in the Melanesian archipelago of Vanuatu, have been reported (Dixson et al., 2018). Consistent with the opacity of mind doctrine, ni-Vanuatu adults claim that the mental states of other people are—and should remain—unknowable (Dixson, 2016). Thus, we may ask whether ni-Vanuatu children effectively overlook or downplay the presence of hidden feelings—in line with the opacity of mind doctrine or alternatively acknowledge that what is privately felt may be different from what is overtly expressed. The findings reported by Dixson et al. (2018) lend firm support to the latter prediction. Ni-Vanuatu children performed comparatively well on a hidden emotions task as indexed by their performance on the Theory-of-Mind scale (Wellman & Liu, 2004). Thus, when presented with the vignette about a boy being teased who did not want his friends to call him a baby, children differentiated between how he felt privately (i.e., sad) and the feeling that he expressed overtly on his face (i.e., happy or OK). By implication, and despite the thrust of the opacity of the mind doctrine, children were willing to make a claim about the boy’s private experience—a claim that was clearly not grounded in his overt facial expression.

Conclusion

Across three different aspects of emotion, we have made two claims that might, prima facie, appear to be in tension with each other. We have argued that psychological research on children’s developing understanding of emotion should not be limited to those so-called basic emotions that can be linked to distinctive facial expressions. Indeed, whether we study children’s grasp of the situations that provoke emotion or their diagnosis of the emotions that someone might be feeling, they increasingly attach no special priority to facial expressions as a diagnostic sign. First, in the course of development, children display a growing understanding of emotions, such as relief, pride, and worry that are not associated with a distinctive facial expressions but can nevertheless be linked to particular eliciting situations. Second, children come to realize, as documented by their progress on the TEC, that the diagnosis of what someone feels calls for an understanding of their beliefs, desires, and reflective processes. Third, with age, children show a steadily increasing insight into the fact that facial expressions can sometimes be a misleading index of what someone actually feels.

Nevertheless, despite these arguments for loosening the presumed conceptual links between the facial expression of emotion and emotion itself, we have also held out the possibility that children’s understanding of emotion proceeds along similar lines across different cultures. First, across radically different cultures, children conceptualize particular emotions in relation to triggering situations and the types of situations that they describe reveal intriguing and recognizable cross-cultural parallels—despite major differences in language, level of education, religion, patterns of socialization, and access to Western technology. Second, when probed via the TEC for their understanding of more general components of emotion, especially the mental processes that impact how a situation is appraised and what is felt as a result, psychometric data reveal comparable conceptual progress in children’s understanding, across a wide range of cultural settings. Third, children’s developing insight into hidden emotions provides an especially persuasive example of this universalist proposal. Even in a culture that downplays speculation about hidden mental states, young children claim that someone might display one emotion but feel a different emotion. By implication, even if, based on cultural beliefs, adults profess that private mental states are unknowable and focus instead on observable speech and behavior, children are less doctrinaire.

One important caveat is warranted. It remains entirely possible that particular emotions are linguistically marked by one culture and not by another. Indeed, as mentioned earlier, historical evidence suggests that particular areas of emotional experience or the regulation of such experiences, may wax, and wane commensurate with historical fluctuations in the relevant, emotionally charged situations. However, the existence of such cultural and historical specificity with respect to either the experience or the regulation of certain emotions—even if it were firmly established—does not preclude the possibility that there is, in addition, a set of emotions that are conceptualized across many different cultures and across millennia. More specifically, the existence of cultural and historical variation is not incompatible with a universal core to the way that emotions are conceptualized.

Accordingly, in future research on children’s understanding of emotions, it will be informative to pursue four concurrent objectives: (1) to systematically extend the range of cultures investigated beyond the more customary Western and/or urban settings; (2) to build on the ethnographic expertise of social anthropologists so that the set of emotions that children are invited to think about is as extensive as possible; (3) to establish how far members of that set of emotions recur across the world’s cultures, especially when we look beyond the basic emotions that research on emotion has traditionally emphasized; and (4) to further assess whether the developmental changes in the way that emotions are conceptualized—as revealed by the TEC—also emerge among children living in remote or traditional communities with minimal schooling and limited media access. Here too, collaboration between developmental psychologists and social anthropologists can help to ensure that test items and interview methods are appropriate.

Footnotes

Acknowledgements

The authors thank Charlotte Hardman, Francisco Pons, and Karine Viana for comments on this manuscript.

Authors’ Note

Paul Harris delivered an invited keynote address for the 2020 biennial meeting of ISSBD. This paper was invited based on that keynote address. Invited articles are reviewed, but the review process differs from that of unsolicited articles, in that the goal is to improve the impact and appeal of the submission..

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Paul L. Harris

References

Barrett

L. F.

(2006). Are emotions natural kinds? Perspectives on Psychological Science, 1, 28–58. https://doi.org/10.1111/j.1745-6916.2006.00003.x

Barrett

L. F.

(2017). How emotions are made. Houghton Mifflin Harcourt.

Barrett

L. F.

Adolphs

Marsella

Martinez

A. M.

Pollak

S. D.

(2019). Emotional expressions reconsidered: Challenges to inferring emotion from human facial movements. Psychological Science in the Public Interest, 20, 1–68. https://doi.org/10.1177/1529100619832930

Bender

P. K.

Pons

Harris

P. L.

de Rosnay

(2011). Do young children misunderstand their own emotions? European Journal of Developmental Psychology, 8, 331–348. https://doi.org/10.1080/17405629.2010.495615

Boddice

(2017). The history of emotions: Past, present and future. Revista de Estudios Sociales, 62, 10–15. https://doi.org/10.7440/res62.2017.02

Borke

(1971). Interpersonal perception of young children: Egocentrism or empathy. Developmental Psychology, 5, 263–269. https://doi.org/10.1037/h0031267

Borke

(1973). The development of empathy in Chinese and American children between three and six years of age: A cross-cultural study. Developmental Psychology, 9, 102–108. https://doi.org/10/1037/h0035080

Caruso

J. C.

Cliff

(1997). Empirical size, coverage and power of confidence intervals for Spearman’s rho. Educational and Psychological Measurement, 57, 637–654. https://doi.org/10.1177/0013164497057004009

Castro

V. L.

Cheng

Halberstadt

A. G.

Grühn

(2016). EUReKA! A conceptual model of emotion understanding. Emotion Review, 8, 258–268. https://doi.org/10.1177/1754073915580601

10.

Cavioni

Grazzani

Ornaghi

Pepe

Pons

(2020). Assessing the factor structure and measurement invariance of the Test of Emotion Comprehension (TEC): A large cross-sectional study with children aged 3-10 Years. Journal of Cognition and Development, 21, 406–424. https://doi.org/10.1080/15248372.2020.1741365

11.

Cole

P. M.

(1986). Children’s spontaneous control of facial expression. Child Development, 57, 1309–1321. https://doi.org/10.2307/1130411

12.

Cole

P. M.

Zahn-Waxler

Smith

K. D.

(1994). Expressive control during a disappointment: Variations related to preschoolers’ behavior problems. Developmental Psychology, 30, 835–846. https://doi.org/10.1037/0012-1649.30.6.835

13.

Cowen

A. S.

Keltner

(2017). Self-report captures 27 distinct categories of emotion bridged by continuous gradients. Proceedings of the National Academy of Sciences, 114(38), E7900–E7909. https://doi.org/10.1073/pnas.1702247114

14.

Cowen

A. S.

Keltner

Schroff

Jou

Adam

Prasad

(2021). Sixteen facial expressions occur in similar contexts worldwide. Nature, 589, 250–257. https://doi.org/10.1038/s41586-020-3037-7

15.

Denham

S. A.

(1986). Social cognition, social behavior, and emotion in preschoolers: Contextual validation. Child Development, 57, 194–201. https://doi.org/10.2307/1130651

16.

Denham

S. A.

Zoller

Couchoud

E. A.

(1994). Socialization of preschoolers’ emotion understanding. Developmental Psychology, 30, 928–936. https://doi.org/10.1037/0012-1649.30.6.928

17.

DePaulo

B. M.

Jordan

(1982). Age changes in deceiving and detecting deceit. In Feldman

R. S.

(Ed.), Development of nonverbal behavior in children (pp. 151–180). Springer-Verlag.

18.

Dixson

H. G. W.

(2016). Generosity, third party punishment, and theory of mind among ni-Vanuatu children [PhD Thesis, Australian National University].

19.

Dixson

H. G. W.

Komugabe-Dixson

A. F.

Dixson

B. J.

Low

(2018). Scaling theory of mind in a small-scale society: A case study from Vanuatu. Child Development, 89, 2157–2175. https://doi.org/10.1111/cdev.12919

20.

Duh

Paik

J. H.

Miller

P. H.

Gluck

S. C.

(2016). Theory of mind and executive function in Chinese preschool children. Developmental Psychology, 52, 582–591. https://doi.org/10.1037/a0040068

21.

Ekman

(1973). Cross-cultural studies of facial expression. In Ekman

(Ed.), Darwin and facial expression (pp. 169–222). Academic Press.

22.

Ekman

(1992). Are there basic emotions? Psychological Review, 99, 50–53. http://doi.org/10.1037/0033-295X.99.3.550

23.

Farina

Belacchi

(2014). The relationship between emotional competence and hostile/prosocial behavior in Albanian preschoolers: An exploratory study. School Psychology International, 35, 475–484. https://doi.org/10.1177/0143034313511011

24.

Franco

Beja

M. J.

Candeias

Santos

(2017). Emotion understanding, social competence and school achievement in children from primary school in Portugal. Frontiers in Psychology, 8, Article 1376. https://doi.org/10.3389/fpsyg.2017.01376

25.

Gardner

Harris

P. L.

Ohmoto

Hamazaki

(1988). Japanese children’s understanding of the distinction between real and apparent emotion. International Journal of Behavioral Development, 11, 203–218. https://doi.org/10.1177/016502548801100204

26.

Gendron

Crivelli

Barrett

L. F.

(2018). Universality reconsidered: Diversity in making meaning of facial expressions. Current Directions in Psychological Science, 27, 211–219. https://doi.org/10.1177/0963721417746794

27.

Göbel

Henning

Möller

Aschersleben

(2016). The relationship between emotion comprehension and internalizing and externalizing behavior in 7- to 10-year-old children. Frontiers in Psychology, 7, Article 1917. https://doi.org/10.3389/fpsyg.2016.01917

28.

Grazzani

Ornaghi

Agliati

Brazzelli

(2016). How to foster toddlers’ mental-state talk, emotion understanding, and prosocial behavior: A conversation-based intervention at nursery school. Infancy, 21, 199–227. https://doi.org/10.1111/infa.12107

29.

Gross

Harris

P. L.

(1988). Understanding false beliefs about emotion. International Journal of Behavioral Development, 11, 475–488. https://doi.org/10.1177/016502548801100406

30.

Hardman

C. E.

(2000). Other worlds: Notions of self and emotion among the Lohorung Rai. Oxford University Press.

31.

Harris

P. L.

(1989). Children and emotion. Blackwell.

32.

Harris

P. L.

Donnelly

Guz

G. R.

Pitt-Watson

(1986). Children’s understanding of the distinction between real and apparent emotion. Child Development, 57, 895–909. https://doi.org/10.2307/1130366

33.

Harris

P. L.

Gross

(1988). Children’s understanding of real and apparent emotion. In Astington

J. W.

Harris

P. L.

Olson

D. R.

(Eds.), Developing theories of mind (pp. 295–314). Cambridge University Press.

34.

Harris

P. L.

Lipian

(1989). Understanding emotion and experiencing emotion. In Saarni

Harris

P. L.

(Eds.), Children’s understanding of emotion (pp. 241–258). Cambridge University Press.

35.

Harris

P. L.

Olthof

Meerum Terwogt

Hardman

C. E.

(1987). Children’s knowledge of the situations that provoke emotion. International Journal of Behavioral Development, 10, 319–343. https://doi.org/10.1177/016502548701000304

36.

Hoemann

Barrett

L. F.

(2019). Emotion words, emotion concepts, and emotional development in children: A constructionist hypothesis. Developmental Psychology, 55, 1830–1849. https://doi.org/10.1037/dev0000686

37.

K. I.

Miller

A. L.

Karasawa

Hirabayashi

Kazama

Wang

Olson

S. L.

Kessler

Tardif

(2021). Emotion expression and regulation in three cultures: Chinese, Japanese, and American preschoolers’ reactions to disappointment. Journal of Experimental Child Psychology, 201, Article 104972. https://doi.org/10.1016/j.jecp.2020.104972

38.

Jackson

J. C.

Watts

Henry

T. R.

List

J.-M.

Forkel

Mucha

P. J.

Greenhill

S. J.

Gray

R. D.

Lindquist

K. A.

(2019). Emotion semantics show both cultural variation and universal structure. Science, 366, 1517–1522. https://doi.org/10.1126/science.aaw8160

39.

Janke

(2008). Emotionswisse und Sozial Kompetenz von Kindern im Alter von drei bis zehn Jahre. [Development of emotion knowledge and social competence between 3 and 10]. Empirische Pädagogik, 22, 127–144.

40.

Kårstad

S. B.

Kvello

Ø.

Wichstrøm

Berg-Nielsen

T. S.

(2014). What do parents know about their children’s comprehension of emotions? Accuracy of parental estimates in a community sample of preschoolers. Child: Care, Health and Development, 40, 346–353. https://doi.org/10.1111/cch.12071

41.

Kårstad

S. B.

Vikan

Berg-Nielsen

T. S.

de Lucena Moreira

de Abreu

E. L.

Rique

(2016). Young Brazilian children’s emotion understanding: A comparison within and across cultures. Journal of Educational and Developmental Psychology, 6, 113–124. https://doi.org/10.5539/jedp.v6n2p113

42.

Kårstad

S. B.

Wichstrøm

Reinfjell

Belsky

Berg-Nielsen

T. S.

(2015). What enhances the development of emotion understanding in young children? A longitudinal study of interpersonal predictors. British Journal of Developmental Psychology, 33, 340–354. https://doi.org/10.1111/bjdp.12095

43.

Lewis

(1989). Cultural differences in children’s knowledge of emotional scripts. In Saarni

Harris

P. L.

(Eds.), Children’s understanding of emotion (pp. 350–373). Cambridge University Press.

44.

Lindquist

K. A.

Wager

T. D.

Kober

Bliss-Moreau

Barrett

L. F.

(2012). The brain basis of emotion: A meta-analytic review. Behavioral and Brain Sciences, 35, 121–143. https://doi.org/10.1017/S0140525X11000446

45.

Lutz

(1987). Goals, events, and understanding in Ifaluk emotion theory. In Holland

Quinn

(Eds.), Cultural models in language and thought (pp. 290–312). Cambridge University Press.

46.

Mayer

J. D.

Caruso

D. R.

Salovey

(1999). Emotional intelligence meets traditional standards for an intelligence. Intelligence, 27, 267–298. https://doi.org/10.1016/S0160-2896(99)00016-1

47.

Mayer

J. D.

Salovey

Caruso

D. R.

Sitarenios

(2003). Measuring emotional intelligence with the MSCEIT V2. 0. Emotion, 3, 97–105. https://doi.org/10.1037/1528-3542.3.1.97

48.

Molina

Bulgarelli

Henning

Aschersleben

(2014). Emotion understanding: A cross-cultural comparison between Italian and German preschoolers. European Journal of Developmental Psychology, 11, 592–607. https://doi.org/10.1080/17405629.2014.890585

49.

Nook

E. C.

Stavish

C. M.

Lambert

H. K.

Mair

McLaughlin

K. A.

Somerville

L. H.

(2020). Charting the development of emotion comprehension and abstraction from childhood to adulthood using observer-rated and linguistic measures. Emotion, 20, 773–792. https://doi.org/10.1037/emo0000609

50.

Olsen

(2017). The history of childhood and the emotional turn. History Compass, 15, Article e12410. https://doi.org/10.1111/hic3.12410

51.

Ornaghi

Brockmeier

Grazzani Gavazzi

(2011). The role of language games in children’s understanding of mental states: A training study. Journal of Cognition and Development, 12, 239–259. https://doi.org/10.1080/15248372.2011.563487

52.

Pons

Harris

P. L.

(2000). Test of emotion comprehension—TEC. Oxford University.

53.

Pons

Harris

P. L.

(2005). Longitudinal change and longitudinal stability of individual differences in children’s emotion understanding. Cognition and Emotion, 19, 1158–1174. https://doi.org/10.1080/02699930500282108

54.

Pons

Harris

P. L.

de Rosnay

(2004). Emotion comprehension between 3 and 11 years: Developmental periods and hierarchical organization. European Journal of Developmental Psychology, 1, 127–152. https://doi.org/10.1080/17405620344000022

55.

Pons

Harris

P. L.

Doudin

P. A.

(2002). Teaching emotion understanding. European Journal of Psychology of Education, 17, 293–304. https://doi.org/10.1007/BF03173538

56.

Robbins

(2008). On not knowing other minds: Confession, intention, and linguistic exchange in a Papua New Guinea community. Anthropological Quarterly, 81, 421–429. https://doi.org/10.1353/anq.0.0007

57.

Robbins

(2021). Mental opacity. In Stanlaw

(Ed.), The international encyclopedia of linguistic anthropology. Wiley. https://doi.org/10.1002/9781118786093.iela0257

58.

Rocha

A. A.

Roazzi

da Silva

A. L.

Candeias

A. A.

Minervino

C. A.

Roazzi

M. M.

, et al, . (2015). Test of emotion comprehension?: Exploring the underlying structure through confirmatory factor analysis and similarity structure analysis. In Roazzi

de Souza

B. C.

Bilsky

(Eds.), Facet theory: Searching for structure in complex social, cultural and psychological phenomena (pp. 66–84). Editora UFPE. https://https://doi.org/10.13140/RG.2.1.2457.4483

59.

Rosaldo

M. Z.

(1980). Knowledge and passion. Cambridge University Press.

60.

Russell

J. A.

(1994). Is there universal recognition of emotion from facial expression? A review of the cross-cultural studies. Psychological Bulletin, 115, 102–141. http://doi.org/10.1037/0033-2909.115.1.102

61.

Saarni

(1984). An observational study of children’s attempts to monitor their expressive behavior. Child Development, 55, 1504–1513. https://doi.org/10.2307/1130020

62.

Shahaeian

Nielsen

Peterson

C. C.

Slaughter

(2014). Cultural and family influences on children’s theory of mind development: A comparison of Australian and Iranian school-age children. Journal of Cross-Cultural Psychology, 45, 555–568. https://doi.org/10.1177/0022022113513921

63.

Song

Over

Carpenter

(2016). Young children discriminate genuine from fake smiles and expect people displaying genuine smiles to be more prosocial. Journal of Experimental Child Psychology, 37, 490–501. https://doi.org/10.1016/j.evolhumbehav.2016.05.002

64.

Stasch

(2008). Knowing minds is a matter of authority: Political dimensions of opacity statements in Korowai moral psychology. Anthropological Quarterly, 81, 443–453. https://doi.org/10.1353/anq.0.0009

65.

Tang

Harris

P. L.

Pons

Zou

Zhang

(2018). The understanding of emotion among young Chinese children. International Journal of Behavioral Development, 42, 512–517. https://doi.org/10.1177/0165025417741366

66.

Tenenbaum

H. R.

Alfieri

Brooks

P. J.

Dunne

(2008). The effects of explanatory conversations on children’s emotion understanding. British Journal of Developmental Psychology, 26, 249–263. https://doi.org/10.1348/026151007X231057

67.

Tenenbaum

H. R.

Visscher

Pons

Harris

P. L.

(2004). Emotional understanding in Quechua children from an agro-pastoralist village. International Journal of Behavioral Development, 28, 471–478. https://doi.org/10.1080/01650250444000225

68.

Trabasso

Stein

N. L.

Johnson

L. R.

(1981). Children’s knowledge of events: A causal analysis of story structure. In Bower

(Ed.), Learning and motivation (Vol. 15, pp. 237–282). Academic Press.

69.

Veraksa

A. N.

Gavrilova

M. N.

Pons

(2020). The impact of classroom quality on young children’s emotion understanding. European Early Childhood Education Research Journal, 28, 690–700. https://doi.org/10.1080/1350293X.2020.1817240

70.

Viana

K. M. P.

Zambrana

I. M.

Karevold

E. B.

Pons

(2020). Emotions in motion: Impact of emotion understanding on children’s peer action coordination. Cognition and Emotion, 34, 831–838. https://doi.org/10.1080/02699931.2019.1669535

71.

Voltmer

von Salisch

(2019). Native-born German and immigrant children’s development of emotion knowledge: A latent growth curve analysis. British Journal of Developmental Psychology, 37, 112–129. https://doi.org/10.1111/bjdp.12262

72.

Wang

(2003). Emotion situation knowledge in American and Chinese preschool children and adults. Cognition and Emotion, 17, 725–746. http://doi.org/10.1080/02699930302285

73.

Wellman

H. M.

Fang

Liu

Zhu

Liu

(2006). Scaling of theory-of-mind understanding in Chinese children. Psychological Science, 17, 1075–1081. https://doi.org/10.1111/j.1467-9280.2006.01830.x

74.

Wellman

H. M.

Liu

(2004). Scaling of theory-of-mind tasks. Child Development, 75, 523–541. https://doi.org/10.1111/j.1467-8624.2004.00691.x

75.

Widen

S. C.

(2016). The development of children’s concepts of emotion. In Barrett

L. F.

Lewis

Haviland-Jones

J. M.

(Eds.), Handbook of emotions (4th ed., pp. 307–318). Guilford Press.

76.

Widen

S. C.

Russell

(2010). Children’s scripts for social emotions: Causes and consequences are more central than facial expressions. British Journal of Developmental Psychology, 28, 565–581. https://doi.org/10.1348/026151009X457550d