Charting the development of emotion comprehension and abstraction from childhood to adulthood using observer-rated and linguistic measures

Erik C Nook; Caitlin M Stavish; Stephanie F Sasse; Hilary K Lambert; Patrick Mair; Katie A McLaughlin; Leah H Somerville

doi:10.1037/emo0000609

. Author manuscript; available in PMC: 2021 Aug 1.

Published in final edited form as: Emotion. 2019 Jun 13;20(5):773–792. doi: 10.1037/emo0000609

Charting the development of emotion comprehension and abstraction from childhood to adulthood using observer-rated and linguistic measures

Erik C Nook ¹, Caitlin M Stavish ¹, Stephanie F Sasse ¹, Hilary K Lambert ², Patrick Mair ¹, Katie A McLaughlin ², Leah H Somerville ¹

PMCID: PMC6908774 NIHMSID: NIHMS1037908 PMID: 31192665

Abstract

This study examined two facets of emotion development: emotion word comprehension (knowing the meaning of emotion words such as “anger” or “excitement”) and emotion concept abstraction (representing emotions in terms of internal psychological states that generalize across situations). Using a novel emotion vocabulary assessment, we captured how a cross-sectional sample of participants aged 4–25 (N=196) defined 24 emotions. Smoothing spline regression models suggested that emotion comprehension followed an emergent shape: knowledge of emotion words increased across childhood and plateaued around age 11. Human coders rated the abstractness of participants’ responses, and these ratings also followed an emergent shape but plateaued significantly later than comprehension, around age 18. An automated linguistic analysis of abstractness supported coders’ perceptions of increased abstractness across age. Finally, coders assessed the definitional “strategies” participants used to describe emotions. Young children tended to describe emotions using concrete strategies such as providing example situations that evoked those emotions or by referring to physiological markers of emotional experiences. Whereas use of these concrete strategies decreased with age, the tendency to use more abstract strategies such as providing general definitions that delineated the causes and characteristics of emotions or by providing synonyms of emotion words increased with age. Overall, this work (i) provides a tool for assessing definitions of emotion terms, (ii) demonstrates that emotion concept abstraction increases across age, and (iii) suggests that adolescence is a period in which emotion words are comprehended but their level of abstraction continues to mature.

Keywords: Emotion, development, abstraction, language

“Use your words” is a phrase frequently used by parents, teachers, and therapists as they ask children to verbalize and talk about how they feel. Although this phrase may appear to be a cliché, classic developmental theories suggest that language is a key self-regulatory tool (Luria, 1961; Vygotsky, 1962). According to these theories, internalized speech helps children regulate their behavior and emotions (Bretherton, Fritz, Zahn-Waxler, & Ridgeway, 1986; Kopp, 1989; Saarni, 1999). Decades of empirical research now support these notions, as a child’s general and emotional vocabulary are related to their ability to manage distressing situations, as well as their executive functioning, mental health, social likability, and academic outcomes (Cole, Armstrong, & Pemberton, 2010; Fabes, Eisenberg, Hanish, & Spinrad, 2001; Kuhn, Willoughby, Vernon-Feagans, & Blair, 2016; Matthews, Biney, & Abbot-Smith, 2018; Roben, Cole, & Armstrong, 2013; Salmon, O’Kearney, Reese, & Fortune, 2016; Trentacosta & Izard, 2007; Vallotton & Ayoub, 2011). Hence, developing a functional emotion lexicon appears to be an important ingredient to overall well-being.

Given the links between emotion vocabulary and psychosocial functioning, understanding how children develop mastery of emotion words is of critical importance. However, three key limitations to our understanding of emotion language development remain. First, even though creative methods have been developed for assessing emotion language in children and adults (see Castro, Cheng, Halberstadt, & Gruhn, 2016 and Zeman, Klimes-Dougan, Cassano, & Adrian, 2007 for reviews), research using an objective assessment of participants’ comprehension of emotion words is lacking. Second, because most research on emotion vocabulary has been conducted in young children, we know very little about developments in emotion language through mid and late adolescence, periods of substantial social and emotional change (Somerville & McLaughlin, 2018). Third, although prior work on emotion language development focuses on when individuals can recognize and use emotion words, a separate but related question concerns the ways in which the concepts underlying these emotion words differ across age. Here, we specifically examined the tendency to represent emotion concepts abstractly (i.e., as internal psychological states that can generalize across situations).

Measuring the Development of Emotion Word Comprehension

Emotion word comprehension refers to an individual’s knowledge of the meaning of words used to label emotional experiences (e.g., angry, excited, calm, or disappointed). Comprehending an emotion word means being able to connect that word with a culturally-agreed-upon concept of what characterizes or defines that emotion (Bloom, 2000; Yin & Csibra, 2015). In this definition, it is important to remember that words and concepts are distinct. In particular, words are symbols that are thought to help people organize (and potentially learn) conceptual information (Barsalou, 1999; Doyle & Lindquist, 2018; Fugate, Gouzoules, & Barrett, 2010; Gopnik, 2001; Lupyan, 2012b, 2012a; Shablack & Lindquist, 2019).

Comprehension of non-emotional words is typically assessed through well-validated and widely-used tests such as the vocabulary subtest of the Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 2011) or the Peabody Picture Vocabulary Test (PPVT; L. M. Dunn & Dunn, 1997). In the WASI, participants are shown a word and asked to generate a definition. If a vague or incorrect answer is provided, the experimenter further probes the participant to ensure they provide as full of a definition as possible. Responses are scored according to a rubric of agreed-upon definitions for each word. To date, there is only one measure of emotion vocabulary that takes a similar approach, the Kusche Affective Inventory— Revised (KAI; Greenberg, Kusche, Cook, & Quamma, 1995; Kusche, Belike, & Greenberg, 1988). The KAI is an experimenter-administered interview that provides a multi-faceted assessment of emotion understanding, and the “feelings vocabulary” component involves an assessment of participants’ abilities to define the emotion terms proud, guilty, jealous, nervous, and lonely.

The KAI has been used to show that emotion comprehension is lower in children with learning disabilities, that the general ability to define emotion terms increases during childhood, and that an intervention aimed at educating children about emotion terms can augment this developmental process (Bauminger, Schorr Edelsztein, & Morash, 2005; Greenberg et al., 1995; Kelly, Longbottom, Potts, & Williamson, 2004). However, many studies employing the KAI do not administer the specific subtest of emotion word comprehension described above (Beck, Kumschick, Eid, & Klann-Delius, 2012; Duncombe, Havighurst, Holland, & Frankling, 2013; Locke, Miller, Seifer, & Heinze, 2015; Southam-Gerow & Kendall, 2000). Thus to date, a measure like the KAI has not been used to broadly examine how emotion word comprehension develops across childhood and adolescence. Assessing how emotion definitions vary with age would (i) provide an objective assessment of which words participants comprehend and (ii) provide insight into how participants represent emotion terms. Theoretically, insight into the development of emotion word comprehension is important not only because emotion language is related to wellbeing (Cole et al., 2010; Salmon et al., 2016) but also because linguistic labels have recently been proposed to serve a central role in learning new concepts (Doyle & Lindquist, 2018; Fugate et al., 2010; Lupyan, 2012b, 2012a). In particular, some have suggested that emotion words might help children develop specific and multi-dimensional emotion concepts (Lindquist, MacCormack, & Shablack, 2015; Nook, Sasse, Lambert, McLaughlin, & Somerville, 2017; Shablack & Lindquist, 2019).

Furthermore, most research on the development of emotion word comprehension focuses on childhood and neglects protracted development through adolescence. Indeed, the only methods that have been used to assess emotion word comprehension in adolescence include recordings of unstructured free responses to vignettes, checklists in which participants (or their parents or teachers) report which words they know, age of acquisition recalled in adulthood, and tasks in which participants must match emotion words with scripts describing the causes and consequences of emotions (Baron-Cohen, Golan, Wheelwright, Granader, & Hill, 2010; Li & Yu, 2015; O’Kearney & Dadds, 2004; Ponari, Norbury, & Vigliocco, 2018; Widen, Pochedly, & Russell, 2015). Together, these studies show that self-reported emotion comprehension rises across childhood, that the tendency to spontaneously produce more nuanced emotion terms such as “disappointed” rather than “sad” increases across adolescence, that emotions are more accurately associated with scripts than facial expressions, and that overall comprehension of abstract words increases across childhood and into adolescence. Although these methods provide important first steps to understanding emotion language in adolescence, they are limited by their naturalistic and/or self-assessed (rather than experimenter-assessed) natures. As such, an interview-based assessment could validate these patterns.

Greater clarity concerning emotion development in adolescence is needed, as adolescence is a pivotal time of social and emotional change (Somerville & McLaughlin, 2018), with adolescents reporting stronger negative affect, greater difficulty differentiating co-experienced emotions, and increased risk of psychopathology (Kessler et al., 2005; Larson, Moneta, Richards, & Wilson, 2002; Nook, Sasse, Lambert, McLaughlin, & Somerville, 2018). Consequently, assessing emotion comprehension in adolescence could provide insight into the nature of emotion language and emotion representations during this dynamic developmental period.

Developing Abstract Emotion Concepts

As described above, emotion comprehension involves the ability to connect emotion words to culturally agreed-upon conceptual meanings. A separate but related question concerns how the concepts underlying these emotion terms vary across age. Studying emotion concepts themselves is of great interest because recent theories postulate that conceptual processes play a central role in the experience and perception of emotions (Barrett, 2006, 2017; Lindquist, 2017; Lindquist, MacCormack, et al., 2015; Lindquist, Satpute, & Gendron, 2015). Specifically, the constructionist theory of emotions posits that emotions arise when emotion concepts are used to parse ambiguous “core affect”—one’s internal sensations of valence (positivity and negativity) and arousal (activation and deactivation)—into specific emotion types (Barrett, 2006, 2017). A growing body of work supports this model by showing that priming, activating, or impeding emotion concepts can influence how emotions are experienced or perceived at both behavioral and neural levels of analysis (Doyle & Lindquist, 2018; Gendron, Lindquist, Barsalou, & Barrett, 2012; Lindquist, Barrett, Bliss-Moreau, & Russell, 2006; Nook, Lindquist, & Zaki, 2015; Oosterwijk, Lindquist, Adebayo, & Barrett, 2015; Satpute et al., 2016). In fact, a recent study showed that individual differences in emotion concept representations is related to perceptions of emotional facial expressions (J. A. Brooks & Freeman, 2018). Thus understanding emotion concepts is an important scientific goal, and there is reason to specifically explore how emotion concepts develop, as evidence suggests that the conceptual representations underlying emotion words are not static across age but rather become more multidimensional from childhood to adulthood, as one’s vocabulary increases (Nook, Sasse, et al., 2017).

Emotion theorists operating within the constructionist tradition often take a simulation-based approach to concepts, positing that concepts are ultimately grounded in simulations of prior sensory experiences (Barrett, 2017; Barsalou, 1999, 2003b, 2009; Wilson-Mendenhall, 2017; Wilson-Mendenhall, Barrett, & Barsalou, 2013). According to this theory, when thinking about the concept apple, the mind draws upon a network of relevant concepts (e.g., round, red, food, sweet) which are themselves tied to memories of prior sensory experiences to create an internal simulation of an apple. This internal simulation can then be used to produce other conceptual representations (e.g., picnic or pie). When applied to emotions, this theory proposes that people construct an experience of, for example, fear by drawing upon and organizing internal simulations of related concepts (e.g., threat, uncertainty, escape), which are ultimately grounded in perceptual memories. In other words, emotions arise through situated conceptualizations in which one’s emotional experience is constructed by simulating combinations of concepts relevant to a given situation (Barrett, 2017; Wilson-Mendenhall, 2017). This dynamic process of situated conceptualization is thought to explain why completely different situations could be constructed as instances of the same emotion (e.g., the feelings aroused by being alone in the dark woods or being in front of a large audience in a brightly lit room could both be labeled “fear”).

Historically, there has been a strict distinction between concrete and abstract concepts, with concrete concepts referring to things that have clear physical referents (e.g., objects such as apples and balls), and abstract concepts referring to general ideas that lack such clear physical referents (e.g., justice, truth, or freedom; Caramelli, Setti, & Maurizzi, 2004; Hale, 1988; Wang, Conder, Blitzer, & Shinkareva, 2010). Because emotions are not physically constrained objects, emotion concepts have historically been thought of as abstract concepts (see Altarriba & Bauer, 2004 for discussion). However, two key arguments have been made against a strict concrete vs. abstract dichotomy. First, the historical operational definition of abstraction is underspecified, merely defining it as “not concrete” (Barsalou, 2003a; Barsalou, Dutriaux, & Scheepers, 2018). Second, growing empirical and theoretical work suggest that this distinction is not binary but dimensional, such that all concepts appear to have some degree of both abstractness and concreteness (Andrews, Frank, & Vigliocco, 2014; Barsalou et al., 2018; Borghi & Binkofski, 2014; Borghi et al., 2017; Della Rosa, Catricala, Vigliocco, & Cappa, 2010; Lupyan & Winter, 2018; Mahon & Caramazza, 2008). For example, when thinking of the concept apple, both concrete qualities that have clear physical boundaries (e.g., its shape and color) and more abstract qualities that lack such physicality (e.g., the psychological states of desire or beauty) can come to mind. Thus even this historically “concrete” concept blends both concrete and abstract aspects.

Consequently, Barsalou et al. (2018) argue that the historical distinction between concrete and abstract concepts should be abandoned and instead the relative abstractness/concreteness of a representation should defined in terms of how much it focuses on external (i.e., physical, non-psychological) versus internal (psychological, non-physical) elements and how strongly it integrates these elements. In this paper we draw upon this formulation to postulate that concepts vary continuously in their levels of concreteness— abstractness, and we define relatively concrete representations as those that focus on external (i.e., physical and situationally-bounded) elements and relatively abstract representations as those that focus instead on internal (i.e., psychological and not-situationally-bounded) elements.

Interestingly, the argument that all concepts include concrete and abstract qualities converges with data showing that emotion concepts do not fit cleanly into either abstract or concrete categories but instead show features of both concept types (Altarriba & Bauer, 2004; Caramelli & Setti, 2005; Mazzuca, Barca, & Borghi, 2017). One potential reason for this is that emotions can be conceptualized with either an emphasis on concrete external qualities (i.e., situational details that gave rise to specific instances of an emotion) or more abstract internal qualities that generalize across situations (i.e., the psychological principles that give rise to an emotion). For example, one could represent anger in terms of concrete external situational details (e.g., anger is the feeling caused by getting cut off in traffic) or by more abstract internal qualities that integrate across situations (e.g., anger is the feeling caused by one’s goals being blocked). Relatedly, neuroimaging data show that emotion processing routinely activates brain regions thought to process both basic sensation and higher-order conceptualization (Kober et al., 2008; Lindquist et al., 2012; Wilson-Mendenhall, Barrett, Simmons, & Barsalou, 2011). Even though psychologists have long known that a given situation can be represented concretely or abstractly (Trope & Liberman, 2010), how this flexibility of representation relates to emotion conceptualization has received little attention in developmental, affective, or cognitive theory.

From a developmental perspective, it is reasonable to hypothesize that across childhood and adolescence the conceptual representations underlying emotion terms shift from a relative focus on external situational details to more internal psychological principles that generalize across situations (i.e., emotion concepts become more abstract). Such a hypothesis is an extension of classic Piagetian theories positing that psychological development proceeds from a concrete sensorimotor focus to a more abstract hypothetico-deductive focus (Demetriou et al., 2018; Inhelder & Piaget, 1958). However, there are additional reasons to hypothesize that emotion concepts become more abstract across age. First, data show that mastery of words corresponding to abstract non-physical internal phenomena increases across childhood into adolescence (Caramelli et al., 2004; Joelson & Herrmann, 1978; Ponari et al., 2018). Second, the cognitive ability to extract general principles and integrate abstract information across specific situations—even at the non-verbal level—increases across age (Crone et al., 2009; Dumontheil, 2014; Ferrer, O’Hare, & Bunge, 2009; Raven, 2000; Whitaker, Vendetti, Wendelken, & Bunge, 2018). And third, as we age, the number and diversity of situations that produce emotions increases (e.g., only after growing to an age at which we give public speeches do we understand that this situation can evoke fear). This experiential diversity likely facilitates expanding emotion concepts from a focus on a narrow set of situations to deeper principles that generalize across situations. For these reasons, we hypothesize that emotion abstraction increases across development, and we use participants’ emotion definitions in an interview-based emotion vocabulary assessment to test this hypothesis.

An Automated Linguistic Measure of Abstractness

Researchers frequently measure the abstractness of conceptual representations by asking human coders to rate how strongly they conform to a given definition of abstraction (e.g., Brysbaert, Warriner, & Kuperman, 2014; Della Rosa et al., 2010; Forgas, 2007; Gray, Parkinson, & Dunbar, 2015; Joelson & Herrmann, 1978; Paivio, Yuille, & Madigan, 1968). This method has been used to generate abstractness norms for thousands of words that are now widely-used in the literature (Brysbaert et al., 2014; Paivio et al., 1968). It can be employed with high interrater reliability, and resulting data are face-valid (e.g., children comprehend fewer words that are rated as abstract compared to adolescents and adults; Ponari et al., 2018). However, this method also suffers from some weaknesses: human raters may be biased by stimulus qualities other than its actual abstractness (e.g., the length of a word or the number of words in a response), the way that raters understand abstractness can differ within and across studies, and human coding is a laborious and time-consuming process that scales with the size of a study’s dataset. These challenges limit the reliability and practicability of measuring abstractness via human raters.

Fortunately, automated psycholinguistic techniques can help curtail these limitations. Quantifying abstractness purely based on the words in a response mitigates human bias, standardizes the mathematical computation of abstractness across studies, and can be automated to substantially reduce the time needed to compute abstractness. Semin and Fiedler (1988, 1991) developed a psycholinguistic approach to measuring abstractness called the Linguistic Category Model (LCM; see also Carnaghi et al., 2008; Seih, et al., 2017; Semin, Gorts, Nandram, & Semin-Goossens, 2002 for contemporary updates to this theory). Although this model was developed to quantify how abstractly participants represented other people, extant work suggests that this approach produces face valid measures of abstractness in text more broadly (e.g., asking participants to take a distanced perspective on a situation increased linguistic abstractness scores, as would be expected; Seih et al., 2017; Trope & Liberman, 2010). In this model, five word classes are ranked in their levels of abstractness, with use of three kinds of verbs denoting more concrete representations and use of adjectives and nouns denoting more abstract representations (see Methods and Carnaghi et al., 2008 for details). For example, defining sadness as “what you feel when you cry” (a descriptive action verb) suggests a more concrete representation than defining it as “feeling sorrowful in response to a loss” (an adjective and a noun).

Here, we used this psycholinguistic method (as well as human ratings) to measure the abstractness of emotion concept representations. Not only does this linguistic approach provide a method for validating human coders’ ratings of emotion abstractness, it also lays the foundation for emotion abstractness to be quantified automatically (i.e., without human input). Such a tool would reduce the burden of empirical research on emotion concept abstraction and allow for rapid processing of large-scale datasets, an approach that has been fruitful in other areas of research (e.g., Doré, Ort, Braverman, & Ochsner, 2015; Fan et al., 2019; Franz, Nook, Mair, & Nock, 2019).

The Current Study

The current project seeks to address the theoretical gaps outlined above by investigating the developmental changes in emotion comprehension and abstraction from early childhood to early adulthood. A broad, performance-based assessment of emotion comprehension (called the Emotion Vocabulary Assessment) was developed by extending the structure of the WASI vocabulary and KAI assessments (Kusché et al., 1988; Wechsler, 2011). This assessment was administered to a cross-sectional sample of participants spanning childhood, adolescence, and early adulthood (i.e., ages 4–25). Trained examiners asked participants to define 24 emotion words, probed their comprehension of each emotion word, and scored their responses on a 0–2 scale. Performance on this assessment constituted participants’ emotion comprehension scores.

To allow for direct comparisons of emotion comprehension and abstraction, these measures were assessed in the same dataset. Following data collection, participants’ responses were transcribed, and trained coders rated the abstractness of these responses. Coders also identified the concrete and abstract “strategies” participants used to describe the meaning of emotion terms. Finally, a purely linguistic measure of abstraction based on the LCM was used to corroborate coders’ abstractness ratings and provide a first step towards the automatic (i.e., non-human) computation of emotion concept abstraction.

Methods

Participants

Two hundred three participants enrolled in the study. Data from seven participants were unusable and thus excluded (one did not complete the emotion vocabulary assessment, five did not understand and/or cooperate with task instructions, and audio recordings were inaudible for one participant). Hence, analyses included data from 196 participants (age range = 4.13–25.91, M_age = 14.53, SD _age = 5.81, 51.02% female, 64.29% Caucasian, three participants did not disclose race, range of income-to-needs ratio = 0.07–13.32, M_{income-to-needs} = 4.93, SD_{income-to-needs} = 3.09, 9.18% below poverty line, 25 participants did not report data required for computing income-to-needs ratio¹). Linguistic abstractness scores were excluded from seven participants due to insufficient linguistic data (i.e., usable definitions did not include enough words to produce valid linguistic abstractness scores, see Supplemental Materials). Participants were recruited from areas surrounding Harvard University and the University of Washington. All participants were fluent in English with English as their first language. Participants did not have cognitive impairments that would limit their ability to provide consent/assent or complete tasks. Adult participants and guardians of minor participants provided informed consent prior to participation, and minor participants assented to participation. All methods were approved by the Institutional Review Boards of Harvard University (IRB#15–2214: “Development of fundamental emotion processes”) and the University of Washington (IRB#50239: “Typical emotional development”).

Given the absence of published data on the development of emotion abstraction, an a priori power analysis was not possible. Hence, we ensured that the sample size was sufficiently powered to detect small-to-medium effects (i.e., β > .20; Cohen, 1988). A power analysis suggested that 194 participants would be required to detect an effect of this size at 80% power. A post hoc power analysis verified that our final sample (N = 196) was sufficiently powered (power = 90%) to detect the study’s smallest age-related effect (i.e., the overall linear decrease in physiological marker use across age,β = .23). Data for this study can be accessed at https://osf.io/m6ue7/?viewonly=1a06da0cfd564f55bc80cb7a6054232f.

Emotion Vocabulary Assessment

A performance-based assessment of emotion vocabulary was created by adapting the format of the vocabulary test of the Wechsler Abbreviated Scale of Intelligence (WASI; Wechsler, 2011). In this assessment (see Supplemental Materials), participants were asked to define 27 emotion terms, and their responses were scored for comprehension. For each emotion word, the experimenter stated the term aloud, presented a card displaying the written emotion word to the participant, and asked “what does_mean?” Responses were self-paced. To thoroughly probe participants’ understanding of emotion terms, participants were prompted to provide more information if their initial response did not demonstrate full comprehension. Responses were audio-recorded and later transcribed.

Participants were asked to define the emotions amazed, angry, annoyed, bored, calm, disappointed, disgusted, embarrassed, excited, grumpy, happy, hate, jealous, lonely, love, nervous, pleased, proud, relaxed, sad, safe, scared, sorry, surprised, thankful, upset, and worried. These emotions were selected to include (i) basic emotions (Ekman & Friesen, 1971), (ii) emotions that correspond to all four quadrants of the circumplex dimensions of valence and arousal (Russell, 1980), and (iii) a mix of emotions that prior work suggested young children can identify (e.g., angry, happy, sad) as well as some that may emerge later in development (e.g., disappointed; Baron-Cohen et al., 2010). Data from three words—sorry, thankful, and pleased—were not included in measures of emotion comprehension or abstraction. The decision to exclude these three words was made following data collection but before human coders produced measures of emotion abstraction. This decision was made (i) to reduce burden of human coding and (ii) because these three words were considered less central to classical conceptualizations of what constitutes an emotion (i.e., in contrast to classic labels of emotion states, sorry and thankful refer more to interpersonal expressions of affect, and pleased refers more or one’s broad level of positively-valenced affect). Emotion words were presented in random order for each participant.

Trained experimenters scored participants’ responses during the testing session. A score of 2 points indicated full comprehension, 1 point indicated partial comprehension, and 0 points indicated no comprehension. To earn 2 points, the participant’s response must have included (i) a general definition of the emotion word (i.e., an abstract explanation of the general causes or characteristics of the emotion that was not bound to a specific situation; e.g., “people feel sad when they experience loss”), (ii) a synonym of the emotion word (e.g., “sad means feeling sorrowful”), or (iii) an example situation that would likely give rise to that emotion and not others (e.g., “I felt sad when my pet died”). During the assessment, experimenters could refer to a scoring guide that included example definitions and synonyms for each emotion word (see Supplemental Materials). Responses earned 1 point if they were of the correct valence but were overly vague and did not specifically describe the emotion in question. For example, the responses “calm means you think things are good” is too vague to specifically define calmness, and “I felt angry when my sister got something that I wanted” is a situation that could produce either anger or jealousy. Hence, these responses would both receive 1 point. Reponses earned 0 points if they described emotions of the incorrect valence or if the participant responded by saying “I don’t know.” An emotion comprehension score was computed for each participant by summing their scores across all trials and converting this sum to a percentage of the total possible score. Because all participants included in analyses completed the entire emotion vocabulary assessment, the total possible score for every participant was 48 points (2 points for 24 emotions). We converted participants’ scores to percentages by dividing participants’ scores by this total possible score and multiplying by 100.

Human Coder Assessment of Emotion Abstractness

Following transcription, trained coders rated the abstractness of participants’ responses on the emotion vocabulary assessment. First, coders scored each response on three 10-point scales that provided numerical assessments of abstractness. These three scales drew upon extant definitions of abstraction (Burgoon, Henderson, & Markman, 2013; Joelson & Herrmann, 1978) as the extent to which responses focused on generalizable, internal principles rather than external situational details (see also Barsalou et al., 2018). The scales included the questions: (i) “how abstract is this response” (1 = extremely concrete, 10 = extremely abstract, where abstract was defined for raters as “the extent to which responses referred to general concepts, thoughts, or ideas, and not bound to specific contexts or stimuli”), (ii) “how much does this response depend on thoughts and ideas about the general principles that produce this emotion?” (1 = not at all, 10 = extremely), and (iii) “how much does this response focus on concrete situation(s) or characteristics that produce this emotion?” (1 = not at all, 10 = extremely). The third scale was reverse-scored, and ratings on these three scales were averaged to produce a single coder abstractness score for each response.

Second, coders indicated which of four “strategies” the participant used to describe each emotion. Three of these strategies came from the a priori list that participants could use to obtain a 2-point response (i.e., providing a general definition, example situation, or synonym). To assess how strongly participants conceptualized emotions in terms of bodily sensations, coders also indicated whether or not each response referenced a “physiological marker” (i.e., it described a bodily sensation or physiological change associated with a given emotion; e.g., “people cry when they are sad”). Responses could employ multiple strategies.

To ensure that ratings were reliable, the second author coded all responses, and two research assistants provided a second rating for all responses (i.e., each provided a second rating of approximately half of the responses). Coders were not informed of participants’ ages during coding, and vocal cues as to a participant’s age could not bias coding because coders only had access to transcriptions of participants’ responses. Coders also underwent extensive training before beginning the coding process to increase reliability. Measures of inter-rater reliability for the three abstraction score ratings demonstrate strong reliability. Across all responses, correlations between coders for the three human rated abstractness scales were r = .81 (“how abstract is this response?”), r = .79 (“how much does this response depend on thoughts and ideas about the general principles that produce this emotion?”), and r = .90 (“how much does this response focus on concrete situation(s) or characteristics that produce this emotion?”). The correlation between coders for the average of these three scales was r = .89, and Cronbach’s alpha across the three scales was α = .93. Hence, the coder abstractness score for each response was averaged across the two coders to provide a single score for each response. Coders came to full agreement on which strategies were used in each response through discussion.

It was determined a priori that only definitions that earned 2 points would be included in abstractness analyses. This ensured that estimates of emotion abstraction only included responses that accurately described each emotion type. Crucially, however, full comprehension could be demonstrated both through abstract (e.g., providing a general definition or synonym) and concrete (e.g., providing an example situation) strategies, so restricting abstraction analyses to emotion words that participants fully comprehended did not inherently bias these analyses. Coder abstractness scores were averaged across all available trials within each participant (i.e., participant-level coder abstractness scores were computed without scores for any emotions on which the participant did not receive a comprehension score of 2). We similarly computed the percentage of trials for which each participant used each of the four strategies. We call these percentages general definition use, synonym use, example situation use, and physiological marker use. Because younger participants comprehended fewer emotions than older participants (see Results), different emotions could have been included in estimates of emotion abstraction across age. Although we conceptualized this as a part of the phenomenon that this study investigates, we conducted supplemental analyses to ensure that dependent variables that showed emergent or linear patterns underwent significant developmental changes even when the sample was restricted to participants who fully comprehended emotion terms (see Supplemental Materials). Hence, age-related differences in emotion comprehension do not confound analyses of emotion abstraction reported in the main text.

Linguistic Assessment of Emotion Abstractness

To corroborate human coded ratings of abstractness, we also produced a linguistic measure of abstractness by drawing upon the Linguistic Category Model (LCM; Carnaghi et al., 2008; Seih et al., 2017; Semin & Fiedler, 1988, 1991; Semin et al., 2002). In this model, word classes are sorted into 1 of 5 levels of abstraction. Abstraction level 1 includes descriptive action verbs (i.e., verbs describing a specific action that can be constrained to a single instance; e.g., “die” in a definition of sad). Abstraction level 2 includes interpretive action verbs (i.e., verbs describing general behaviors that cross specific instances; e.g., “withdraw” in a definition of sad). Abstraction level 3 includes state verbs (i.e., verbs involving mental states without clear beginning or end; e.g., “mourn” in a definition of sad). Abstraction level 4 includes adjectives (i.e., references to qualities that require considering mental states or abstracting across several instances; e.g., “depressed” in a definition of sad). Abstraction level 5 includes nouns (references to classes of things; e.g., “loss” in a definition of sad). A mathematical formula is used to compute a passage’s abstractness by comparing the proportion of words that fall within each abstraction level (see Seih et al., 2017 for details). Scores range from 1 to 5, with higher values indicating higher levels of abstractness.

Following prior work (Pennebaker, Booth, & Francis, 2007; Schmid, 1994; Seih et al., 2017), we used the Linguistic Inquiry and Word Count (LIWC) and TreeTagger software to extract the number of words in each response that fell within each level of linguistic abstraction. We then computed a linguistic abstractness score for each response using the formula provided by Seih et al., (2017). To parallel the other measures of emotion abstractness, we computed LIWC measures for responses that earned 2 points. Linguistic abstractness scores for all qualifying responses were then averaged within participants to produce participant-level linguistic abstractness scores. Only two trials did not have any words in abstraction levels 1–5, so a linguistic abstraction score could not be computed. These trials were treated as missing values and not included in linguistic abstractness scores for those participants.

A series of exploratory analyses suggested that linguistic abstractness scores may not have been valid for very young children in our sample (see Supplemental Materials). Coherence between linguistic abstractness scores and coder abstractness scores broke down at young ages, and this lack of coherence was explained by the fact that some young participants had few words on which a linguistic abstractness score could be computed (likely because only words from responses that showed full comprehension could be analyzed, and younger children comprehended fewer emotions). Hence, we excluded linguistic abstractness scores from seven participants who had fewer than 100 total words in their usable verbal responses during the assessment. Excluding these participants ensured that the correlation between coder and linguistic abstractness scores was stable across age.

Hypotheses and Analyses

Interrelations between dependent variables.

In all, this study produced one measure of emotion comprehension, two measures of emotion abstraction (i.e., coder abstractness ratings and linguistic abstractness scores), and four measures of definitional strategies (i.e., general definition use, synonym use, example situation use, and physiological marker use). We hypothesized that people who comprehended more emotion words would also represent emotions more abstractly. We hypothesized that the two measures of emotion abstractness would converge: coder abstractness scores and linguistic abstractness scores should positively correlate. We hypothesized that general definition use and synonym use would track increased abstractness (i.e., they should correlate positively with coder and linguistic abstractness scores) because providing a general definition for an emotion indicates understanding of the internal principles that give rise to emotions across situations and providing synonyms denotes an understanding of the taxonomic relations between emotion words (i.e., knowing that two words can be used to refer to the same concept). Finally, we hypothesized that example situation use and physiological marker use would track increased concreteness (i.e., they should correlate negatively with coder and linguistic abstractness scores) because providing example situations and physiological markers involves attending to concrete physical details of emotional experiences.

To test these hypotheses, Spearman’s correlations assessed the significance and direction of all pairwise relations between these variables at the participant level. Note that because measures of emotion abstraction were only valid for responses on which the participant showed full comprehension of emotion terms, these analyses involved correlating participant-level emotion comprehension scores (i.e., how well participants comprehended the 24 emotion terms we assessed) with participant-level emotion abstraction scores (i.e., how abstractly participants represented the subset of emotions that they fully comprehended). Spearman’s correlations were used for all analyses because of the skewed distribution of all dependent variables other than linguistic abstractness scores. Given the number of correlation analyses conducted, we used the Benjamini & Hochberg (1995) false discovery rate (FDR) method to correct for multiple comparisons.

Overall developmental changes in emotion comprehension and abstraction.

Based on prior work (e.g., Baron-Cohen et al., 2010; Ponari et al., 2018), we hypothesized that both emotion comprehension and emotion abstraction would follow emergent age-related patterns (i.e., they would increase across early development before reaching a plateau; Somerville et al., 2013). We also hypothesized that emotion comprehension would plateau earlier in development than emotion abstraction. We did not have strong hypotheses concerning potential differences between age-related patterns of the two measures of emotion abstractness and use of the four strategies except that the two measures in which higher scores indicated less abstraction (i.e., example situation use and physiological marker use) should show “inverse emergent patterns” (or diminishing patterns) that decreased across childhood and adolescence before plateauing at their nadir.

To test whether measures of emotion comprehension and abstraction indeed followed emergent or diminishing patterns of change (specific kinds of non-linear shapes), we analyzed each dependent variable following the non-linear analytic methods of Rodman, Powers, & Somerville (2017). This approach tests whether “traditional” polynomial patterns (i.e., linear, quadratic, and cubic models) or more complex nonlinear patterns (i.e., spline-based models) best fit age-related patterns in the data. Each dependent variable was subjected to four models that assessed for (i) linear, (ii) quadratic, (iii) cubic, and (iv) other non-linear relations with age. Linear, quadratic, and cubic analyses were tested within linear regression frameworks using the poly function in R’s stats package (R Core Team, 2016). The poly function transforms age into orthogonalized linear, quadratic, and cubic regressors. Linear, quadratic, and cubic regressions were each tested in separate regression models, but lower-order regressors were included in higher-order models (i.e., the linear regressor was included in the quadratic analysis, and both the linear and quadratic regressors were included in the cubic analysis). Finally, we used generalized additive modeling approaches to test for age-related patterns that did not fit linear, quadratic, or cubic patterns. Specifically, thin plate regression smoothing spline analyses were used to produce regression equations that fit the data using cross-validation procedures but were also penalized for the number of parameters to prevent overfitting. The result of these models is a stable smooth curve that describes the data’s age-related patterns but is not constrained to stereotyped linear, quadratic, or cubic shapes. Spline analyses were conducted using the mgcv package in R (Wood, 2003, 2017).

The four models (linear, quadratic, cubic, and spline) were compared to each other as well as a null model containing no age predictors using Akaike Information Criterion (AIC) values (Akaike, 1974). AIC values provide a measure of goodness of fit that takes into account the number of parameters, and lower AIC values indicate better model fit after penalizing for the number of parameters. Adjusted R² values for each model are also reported as additional measures of goodness of fit for each model. These values summarize the proportion of variance explained by each model after adjusting for the number of predictors.

When spline regressions were the best fitting model, we visualized the resulting regression lines and found that they conformed to the “emergent” pattern defined by prior literature (Casey, 2015; Somerville et al., 2013) for four dependent variables. Emotion comprehension scores, coder abstractness scores, general definition use, and example situation use showed rapid change in the early portion of the study’s age range before stabilizing (see Results). We used a data-driven method to identify the age at which these variables reached their plateaus (see Supplemental Materials for details on method development). Two criteria identified plateaus: (i) reduced rate of change in the dependent variable and (ii) the dependent variable reached the maximum/minimum of its development. First, we extracted the best-fit line summarizing age-related change for each variable and computed its first derivative (Simpson, 2014). Because the first derivative quantifies the slope of a curve, the plateau of an emergent curve can be defined as the earliest age at which the first derivative approaches 0 (i.e., the curve flattens). Hence, we set the thresholds for a plateau as occurring when the change in the dependent variable slowed to less than 0.5%/year (for emotion comprehension scores and strategy use, which are measured on a 100-point scale) or 0.05/year (for coder abstractness scores, which are measured on a 10-point scale).

Second, to prevent “plateaus” from being incorrectly identified at points where the slope of the curve slowed temporarily, we ensured that plateaus reflected the age at which the dependent variable could be considered fully developed. In other words, we wanted to ensure the plateau occurred at a point where the dependent variable had finished increasing to its maximum (or falling to its minimum) and then remained stable. We consequently added a second criterion such that the value of the dependent variable at the plateau must have been near the maximum value (or minimum value, for example situation use, which fell across age). “Near” the maximum or minimum value was operationalized as falling within the 95% CI of the smoothing spline’s estimate of the maximum (or minimum) value of the curve.

Using coder abstractness scores as an illustration (see Figure 1b), the curve slowed to < .5%/year at age 17.69, satisfying the first criterion. To check the second criterion, we next evaluated whether this plateau point reflected the conclusion of age-related change (rather than a temporary flattening). The 95% CI of coder abstractness scores around this point were 8.98–9.56 (i.e., the grey shaded region around the maximal point of the curve). Because the value of the curve at age 17.69 (i.e., 9.27) was within this 95% CI, this age was deemed the plateau point.

We used nonparametric bootstrapping methods to test for significant differences in the ages at which each of these variables plateaued. We conducted 10,000 bootstrapped simulations of our data using the boot package in R (Canty & Ripley, 2017; Davison & Hinkley, 1997). Within each simulated sample, we (i) conducted spline analyses of emotion comprehension scores, coder abstractness scores, general definition use, and example situation use; (ii) used the first derivative method described above to identify the age of each variable’s plateau; and (iii) computed pairwise differences between the 4 plateau ages. The 95% bias-corrected and accelerated (BCa) confidence intervals (CIs) of these differences were computed, and those that did not include 0 were considered significant (DiCiccio & Efron, 1996). We also used these bootstrap simulations to compute estimates of the 95% BC_a CIs of each plateau point for descriptive purposes.

Development of emotion comprehension and abstraction for each emotion.

To provide descriptive estimates of emotion comprehension and abstraction for each of the 24 emotion words assessed in this study, we subjected emotion comprehension scores and coder abstractness scores for each emotion to smoothing spline analyses. We extracted the lines of best fit for these spline models and visually presented estimates of these values. These descriptive data are provided to allow for speculative interpretations of age-related changes for each emotion and to offer norms that can guide future research.

Control analyses.

We conducted two sets of analyses to control for potential confounds. First, it was possible that abstractness measures were biased by the length of participants’ responses. To address this concern, we used LIWC to extract the word count of each response and tested (i) whether average word count varied across age and (ii) whether controlling for word count eliminated age-related variation in other dependent variables. Second, it is possible that providing a synonym does not actually reflect comprehension of an emotion term but instead is merely evidence that participants could produce terms semantically associated with that emotion word. In other words, providing synonyms might reflect shallow semantic priming in which participants produced words they knew were associated with the target emotion word but did not actually comprehend the concept underlying either word. To address this concern, we computed the number of trials on which participants only provided synonyms, and we investigated whether the study’s results differed when these trials were treated as incorrect responses.

Results

Interrelations Between Dependent Measures

Spearman’s correlations indicated that participants who showed higher emotion word comprehension had more abstract emotion representations (Table 1). Within emotion abstraction measures, modest convergence emerged between human ratings of emotion abstractness and automated linguistic measures of abstractness (Table 1). Additionally, the direction of correlations supported the hypotheses that general definition use and synonym use tracked increased abstractness, whereas example situation use and physiological marker use tracked decreased abstractness. All of these relations were significant and survived FDR correction for multiple comparisons except: (i) emotion comprehension scores showed only a trending relationship with physiological marker use (p = .076) that did not survive correction, and (ii) synonym use did not correlate significantly with physiological marker use.

Table 1.

Spearman’s correlations showing interrelations between dependent measures.

	1.	2.	3.	4.	5.	6.

1. Emotion comprehension scores	-
2. Coder abstractness scores	.52^***	-
3. Linguistic abstractness scores	.20^**	.45^***	-
4. General definition strategy use	.50^***	.74^***	.25^***	-
5. Synonym strategy use	.23^**	.33^***	.29^***	.17^*	-
6. Situation strategy use	−.49^***	−.90^***	−.50^***	−.63^***	−.32^***	-
7. Physiological marker strategy use	−.13^#	−.35^***	−.40^***	−.16^*	−.11	.36^***

Open in a new tab

Notes: See main text for tests of linear and non-linear relations between age and other variables. All significant relations (i.e., ps < .05) survive FDR correction for multiple comparisons, but other relations (i.e., ps > .05) do not.

^***

p < .001,

^**

p < .01,

p < .05,

p < .10.

Age-related Change in Emotion Comprehension and Abstraction

Results of analyses using linear, quadratic, and cubic polynomial models, as well as thin plate regression smoothing spline models are presented in Table 2. AlCs of null models are provided for comparison.

Table 2.

Results of null, polynomial, and thin plate smoothing spline models for each dependent variable.

Dependent variable	Model	β	p	Adj R²	AIC

Emotion comprehension scores	Null				1608.10
	Linear	.48	< .001^***	.23	1558.51
	Quadratic	−.50	< .001^***	.47	1484.31
	Cubic	.32	< .001^***	.58	1442.01
	Spline			.62	1426.36
Coder abstractness scores	Null				833.48
	Linear	.76	< .001^***	.58	663.07
	Quadratic	−.37	< .001^***	.72	586.51
	Cubic	−.01	.745	.72	588.40
	Spline			.73	580.05
Linguistic abstractness scores	Null				−142.50
	Linear	.46	< .001^***	.21	−185.25
	Quadratic	−.001	.991	.20	−183.25
	Cubic	−.05	.467	.20	−181.79
	Spline			.21	−185.25⁺
General definition use	Null				1859.18
	Linear	.70	< .001^***	.49	1728.89
	Quadratic	−.46	< .001^***	.70	1627.36
	Cubic	.07	.057	.70	1625.66
	Spline			.70	1623.66
Synonym use	Null				1738.33
	Linear	.32	< .001^***	.10	1719.66
	Quadratic	−.05	.458	.09	1721.10
	Cubic	−.01	.848	.09	1723.06
	Spline			.10	1719.66⁺
Example situation use	Null				1866.19
	Linear	−.67	< .001^***	.45	1750.57
	Quadratic	.32	< .001^***	.55	1712.61
	Cubic	.02	.687	.55	1714.44
	Spline			.56	1710.20
Physiological marker use	Null				1392.78
	Linear	−.23	.001^**	.05	1384.27
	Quadratic	−.01	.882	.04	1386.25
	Cubic	.13	.062	.05	1384.68
	Spline			.10	1378.66

Open in a new tab

Note: Bold text indicates best fitting model for each dependent variable, as determined by having the lowest AIC.

⁺

In these two cases, spline model algorithms revealed that the best fits were identical to linear models; hence we interpreted the linear models. β = standardized beta, Adj R² = adjusted R², AIC = Akaike Information Criterion,

^***

p < .001,

^**

p < .01

Emotion comprehension scores.

The spline model produced the best fit for emotion comprehension scores (Table 2), revealing that emotion comprehension scores followed an emergent pattern across age (Figure 1a). The first derivative of the spline curve suggested that development of emotion comprehension plateaued at age 10.95, 95% CI = [8.76, 12.11].

Coder abstractness scores.

The spline model produced the best fit to coder abstractness scores (Table 2). Coder abstractness scores also followed an emergent age-related pattern, but these scores plateaued at age 17.69, 95% CI = [16.06, 19.17] (Figure 1b). A nonparametric bootstrapping analysis suggested that this plateau was significantly later than the emotion comprehension plateau, 95% CI = [4.83, 9.25].

Linguistic abstractness scores.

Both the linear and spline model suggested that linguistic abstractness scores followed a linear age-related pattern (Table 2, Figure 2) after removing 7 participants with little linguistic data (see Supplemental Materials). Hence, this purely linguistic measure of abstractness also suggested that emotion representations became more abstract from childhood to early adulthood. However, the best-fitting shape of this relation was linear rather than emergent, and the strength of the relation between age and linguistic abstractness scores was smaller than the relation between age and coder abstractness scores.

General definition use.

General definitions were the most frequent strategy participants used to describe emotions (81.21% of trials). A spline model provided the best fit to these data, revealing that general definition use followed an emergent age-related pattern (Table 2, Figure 3a). General definition use plateaued at age 17.19, 95% CI = [13.40, 18.50]. This plateau was significantly later than the plateau of emotion comprehension, 95% CI = [4.08, 9.00], but it did not significantly differ from the plateau of coder abstractness scores, 95% CI = [−1.55, 4.50].

Figure 3. — Age-related patterns of 4 different strategies for describing emotions. A) Use of general definitions followed an emergent pattern that increased across childhood and adolescence before plateauing. B) Use of synonyms increased linearly with increasing age. C) Use of example situations followed a diminishing pattern that decreased across childhood and adolescence before plateauing. D) Use of physiological markers followed a non-linear pattern that was elevated in childhood. Solid black lines depict spline model fits (dark grey regions represent 95% CI from spline models). Vertical dashed lines show developmental plateaus computed using first derivatives of spline fits (vertical light grey regions represent 95% CIs from bootstrap simulations).

Synonym use.

Participants used synonyms to describe emotions on 33.63% of trials. Both linear and spline models suggested that synonym use followed a linear increase across development (Table 2, Figure 3b).

Example situation use.

Example situations were used to describe emotions on 26.22% of trials. A spline model provided the best fit to the data, revealing that use of this strategy followed a diminishing pattern (Table 2). Example situation use fell across childhood and adolescence before reaching a plateau at age 18.19, 95% CI = [16.67, 22.06]² (Figure 3c). This plateau was significantly later than the plateau of emotion comprehension, 95% CI = [5.32, 11.78], but it did not significantly differ from the plateaus of coder abstractness scores, 95% CI = [−5.18, 0.59], or general definition use, 95% CI = [−1.17, 6.06]. Thus, all plateau ages for measures of emotion abstraction occurred significantly later than the emotion comprehension plateau but did not differ significantly from each other.

Physiological marker use.

Physiological markers were used to describe 7.29% of trials. A significant linear regression suggested that use of this strategy decreased across age, but the spline model provided the best fit to the data (Table 2). This model suggested that physiological marker use followed a non-linear but also not emergent pattern of age-related change (Figure 3d). Use of physiological markers was elevated in mid-childhood but low at other ages. First derivative analyses suggested that physiological marker use increased from age 4.13 to age 10.00, then decreased until age 15.05, and then remained low through the rest of the study’s age range.

Controlling for word count.

A linear regression at the participant level showed that average word count of participants’ responses increased with age, β = .21, p = .004. However, differences in the length of participants’ responses across age did not confound measures of emotion abstraction. Linear regressions examining coder abstractness scores, general definition use, and example situation use (from childhood until the coder abstraction score plateau of age 17.69) showed that age remained significantly related to these dependent variables even after controlling for the average word count of participants’ responses, ps < .001. Similarly, age was significantly related to linguistic abstractness scores and synonym use across the entire sample even after controlling for average word count, ps < .001. A series of parallel analyses at the trial level (i.e., mixed-effects models that nested data within participants and examined whether each definition’s abstraction metrics were related to participants’ age after controlling for the length of each response) also showed that age remained a significant predictor of these dependent variables after controlling for word count, ps < .001. As such, the length of participants’ responses did not confound any of the study’s key findings.

Control analyses investigating trials on which participants only provided synonyms.

A second set of control analyses ensured that results were not significantly affected by the possibility that only providing a synonym for an emotion word may reflect shallow semantic priming rather than full comprehension of an emotion word’s underlying concept. Across all trials, 4.49% of trials involved only providing a synonym (4.90% of 2-point responses), and removing these trials did not affect conclusions presented in the main text (see Supplemental Materials for details).

Age-related Change for Each Emotion

Visual representations of fits provided by spline models of emotion comprehension scores and coder abstractness scores are shown in Figure 4. These analyses are presented for descriptive purposes. Observable trends suggest that emotion comprehension at age 4 varied substantially across emotions (e.g., emotion comprehension scores were > 80% for love but < 10% for amazed). The age at which emotion comprehension approached ceiling (i.e., exceeded 90%) also varied across emotions (e.g., around age 6 for scared and safe but around age 10 for calm). However, inter-emotion variability was even greater for age-related patterns of emotion abstraction. Whereas coder abstractness scores approached ceiling (i.e., exceeded 9 of 10) as early as age 13 for hate, disappointed, and love, comparable levels only emerged around age 20 for proud and annoyed. This threshold was never reached for nervous.

Discussion

The current study assessed the development of emotion word comprehension and emotion concept abstraction from childhood to adulthood. Emotion comprehension and abstraction both followed emergent age-related patterns of change that increased with age before reaching a plateau. However, emotion comprehension plateaued significantly earlier (i.e., around age 11) than emotion abstraction (i.e., around age 18). A linguistic measure of abstractness was computed from the words participants used to describe emotions, and this measure also suggested that emotion concepts became increasingly abstract across age. Finally, age was related to the strategies participants used to describe emotions. The tendency to use abstract strategies increased across age, with general definition use following an emergent pattern and synonym use following a linear pattern. Conversely, use of concrete strategies decreased across age, with example situation use following a diminishing pattern and physiological marker use showing a non-linear pattern that was slightly elevated in childhood. We discuss the implications of these findings below.

An Interview-Based Method for Measuring Emotion Comprehension

This study offers a novel tool for assessing emotion comprehension, the emotion vocabulary assessment. This interview adapts techniques from the WASI vocabulary assessment (Wechsler, 2011) and updates the approach used by the emotion definition task of the KAI (Kusché et al., 1988). Other than the KAI, several innovative methods have been used to study emotion language in childhood and adolescence. These methods include (i) coding conversations between children and their parents or their peers (Aznar & Tenenbaum, 2013; Bretherton & Beeghly, 1982; Cervantes & Callanan, 1998; J. Dunn, Bretherton, & Munn, 1987; Lagattuta & Wellman, 2002; MacWhinney & Snow, 1990), (ii) asking participants to verbalize how they or other people would feel in emotional situations described by vignettes (O’Kearney & Dadds, 2004, 2005), (iii) asking participants to name as many emotion terms as they can think of until they have exhausted all known terms (Beck et al., 2012; Duncombe et al., 2013; Kusché et al., 1988; Locke et al., 2015), (iv) asking children, adolescents, or their parents to self-report on their emotion word comprehension (Baron-Cohen et al., 2010; Li & Yu, 2015), (v) asking adults to recall the age at which they acquired words (Kuperman, Stadthagen-Gonzalez, & Brysbaert, 2012; Stadthagen-Gonzalez & Davis, 2006), (vi) using written tests in which participants identify which emotion would most likely be produced by specific situations (MacCann & Roberts, 2008), (vii) using lexical decision tasks to test whether participants can identify emotion words vs. non-words (Ponari et al., 2018), (viii) asking participants to match emotion words with scripts describing situational causes and consequences of emotions (Balconi & Carrera, 2007; Reichenbach & Masters, 1983; Russell & Widen, 2002; Smith & Walden, 2001; Widen et al., 2015; Widen & Russell, 2010a, 2010b), and (ix) asking participants to define non-emotional and emotional terms in writing (Caramelli & Setti, 2005).

There are pros and cons to each these methods. For instance, some provide naturalistic assessments of emotion word use (e.g., methods i and ii) whereas others offer more focused examination of whether or not participants comprehend a specific set of emotion words (e.g., method vi). Some rely on participants’ own self-assessment of whether they know emotion words (e.g., method iv), and others involve experimenter-assessed tests of comprehension (e.g., methods vi-ix). Finally, some of them offer insight into how participants define emotions (e.g., method ix) whereas others do not (e.g., methods i-viii).

The emotion vocabulary assessment developed here expands upon these prior methods to assess comprehension of emotion words through an interview format, which allows experimenters to ask follow-up questions in real time and thereby fully probe participants’ comprehension. The breadth and thoroughness of this assessment offers several strengths. For example, prior work on emotion word comprehension across childhood and adolescence relied on participants’ self-assessment of their emotion comprehension, which is susceptible to demand characteristics (Baron-Cohen et al., 2010; Li & Yu, 2015). Hence, the present study used an experimenter-assessed method to validate that emotion word comprehension emerges across early childhood and plateaus around age 11. Additionally, verbally probing and recording participants’ emotion definitions allows for deeper insight into how people represent the concepts underlying emotion words. Although we studied the abstraction of these concepts, this method could be used to study other facets of emotion concept development.

When comprehension patterns for individual emotions were examined, we found that comprehension of all 24 emotions assessed in this project reached 90% comprehension within a 4-year window (ages 6–10). Hence, these years represent a particularly active period of maturation in emotion word comprehension. However, we also found variability across emotions both in levels of comprehension at age 4 and in the age at which they approached full comprehension. As such, the overall plateau point we identified in this study was influenced by the specific emotions included in this assessment, and this value should not be taken as absolute.

Although we caution against over-interpreting patterns for each emotion, they suggest that some emotion words (e.g., love, scared, excited, angry, sad, happy) may be comprehended before more “social” (e.g., proud, embarrassed, jealous) or “nuanced” emotion words (e.g., nervous, annoyed, amazed). These patterns accord with the notion that a core set of emotion concepts arise early in development and differentiate into more nuanced categories across age (Bridges, 1930; Nook, Sasse, et al., 2017; Widen, 2013; Wu, Muentener, & Schulz, 2017). Interestingly, this pattern aligns with Rosch’s overarching cognitive theory that “basic” levels of categories (i.e., those that are most representative of a category and most frequently used by caregivers; e.g., “fear”) are learned early in development, and subordinate category members (i.e., those that describe more nuanced levels of a category; e.g., “nervousness”) are mastered with greater experience across age (Mervis & Rosch, 1981; Rosch, Mervis, Gray, Johnson, & Boyes-Braem, 1976). Additionally, the fact that classic “social” emotions were later-emerging than non-social emotions also converges with work showing that theory of mind shows protracted maturation in social emotional development (Blakemore, 2008; Burnett, Bird, Moll, Frith, & Blakemore, 2009; Choudhury, Blakemore, & Charman, 2006; Sebastian et al., 2012). Future research that specifically manipulates these factors (e.g., by comparing social and non-social emotions) using lists of emotion words that are matched for word length and other salient dimensions could further test this notion.

Although emotion vocabulary assessment provided several benefits beyond prior methods, two weaknesses of this interview-based approach should be noted. First, we originally counted synonyms as demonstrating full comprehension because (i) Wechsler vocabulary assessments (on which this test was modeled) accept synonyms as correct responses (Wechsler, 1999) and (ii) providing synonyms demonstrates at least some understanding of how words within a taxonomy are organized (i.e., that two words refer to the same concept and not different concepts). However, it is possible to provide a synonym but not fully comprehend the concept underlying the target word. Even though control analyses suggested that this possibility did not invalidate this study’s results, researchers may nonetheless wish to remove this criterion in future work. Second, success on interview-based assessments likely relies on several cognitive and social factors beyond emotion comprehension and abstraction. These factors include developmental increases in general vocabulary and working memory (which could potentially allow older participants to more easily organize and verbally express their thoughts; Farkas & Beron, 2004; Siegel & Ryan, 1989), as well as increased awareness of the “social script” that one should provide abstract definitions during vocabulary assessments. Although these weaknesses exist for all developmental studies that use interview-based methods, future work could control for these factors by measuring them all in one study or using non-emotion terms as control items.

Emotion Abstraction: Implications for Affective, Developmental, and Cognitive Theory

Because the emotion vocabulary assessment collected participants’ verbal definitions of emotions, data from this study could also be used to investigate the phenomenon of emotion concept abstraction. Scholars have long known that non-verbal and verbal abstraction skills grow across development and that multiple components of emotion understanding also develop (Ferrer et al., 2009; Nook, Sasse, et al., 2017; Pons, Harris, & de Rosnay, 2004; Widen, 2013). However, insight into whether these two faculties interact has been lacking (though see Waggoner & Palermo, 1989). The current findings suggest that these processes indeed converge: emotion concepts become more abstract from childhood to adolescence. We observed that emotion representations became more abstract through childhood and adolescence. In particular, emotion word definitions shifted from understanding emotions in terms of external concrete situations to understanding them in terms of internal principles that generalized across situations (Barsalou et al., 2018). Further, we found that emotion abstraction was a later-emerging phenomenon than emotion word comprehension. Control analyses showed that these results were not driven by differences in response length across age.

The notion that emotion words can be associated with relatively more concrete or abstract representations extends theories of emotion. Burgeoning theoretical and empirical work argues that emotion concepts shape how individuals understand their own and others’ emotions (Doyle & Lindquist, 2018; Gendron et al., 2012; Lindquist, MacCormack, et al., 2015; Nook et al., 2015; Satpute et al., 2016). However, these theories have not considered how the abstraction of one’s emotion concepts might vary. Based on simulation-based theories of concept representation (Barrett, 2017; Barsalou, 1999, 2009; Wilson-Mendenhall, 2017), these data suggest that when defining emotion words, children and adolescents draw upon more “external” simulations (i.e., simulations directly connected to perceptual memories; e.g., an episode of crying), whereas adults draw upon simulations that are more “internal” and integrated (i.e., simulations involving other conceptual representations; e.g., the abstract concept of loss). This reveals that development of emotion concepts involves building an inner network of conceptual representations that progressively become more abstract with age. Future research could extend this work by charting the fine-grained development of this network and by examining if and how the abstraction of emotion concepts impacts actual emotional experiences. Theories of situated conceptualization (Barrett, 2017; Wilson-Mendenhall et al., 2013, 2011) suggest that this would be the case, but because this study focused on conceptual understanding of emotion words outside of actual emotional experiences, this remains an open question.

Developmentally, these results grant additional insight into how emotions vary across childhood and adolescence. Researchers have found that several aspects of emotion understanding emerge across childhood (e.g., Pons et al., 2004), and the current findings reveal that building abstract emotion representations is another task for the developing mind. These findings also extend understanding of the unique affective processes that occur during adolescence. These data suggest that adolescence is a period in which many emotion words are fully comprehended but the concepts underlying these words have not yet reached mature levels of abstraction. Recent research has shown that adolescents less readily differentiate negative affect into specific types and that multi-dimensional emotion representations continue to develop through adolescence into young adulthood (Nook, Sasse, et al., 2017; Nook et al., 2018). The current results extend these findings to also include the notion that emotion abstractness develops during this period. It remains an open question whether the gap between these two milestones might be related to the elevated risk of mental illness in adolescence (Kessler et al., 2005), but this possibility should motivate future research testing whether and how emotion abstraction relates to emotion regulation and well-being across development.

These results also have implications for cognitive theories of abstract concepts. Some of these theories argue that emotions “ground” abstract concepts (Kousta, Vigliocco, Vinson, Andrews, & Del Campo, 2011; Ponari et al., 2018; Vigliocco et al., 2014; but see Borghi et al., 2017). These theories posit that affective experiences may be the first cues that non-physical phenomena exist, leading individuals to build concepts for other non-physical abstract phenomena. As such, these scholars argue that emotion might provide a crucial role in building abstract concepts altogether. This argument calls for increased attention to the intersection between affective science and cognitive theories of concept representations, and our findings provide an initial foray. Because we did not compare age-related patterns of change for emotional and non-emotional concepts, data from the current study cannot directly test if or how emotion concept development relates to more general developments in concept abstraction. Nonetheless, future research could use the emotion vocabulary assessment to examine whether increased emotion concept abstraction (or attention to the affective nature of abstract experiences—like the pleasing tone of justice) facilitate the development of abstract nonemotional concepts.

Although age-related variation in the strategies participants used to define emotions overall supported the notion that emotion representations became more abstract across age, use of these strategies did not all share the same age-related pattern. Whereas example situation and general definition use followed emergent patterns, synonym use increased linearly across age, and physiological marker use overall fell across age but was slightly elevated in childhood. Because identifying synonyms requires the taxonomic understanding that two words share roughly equivalent conceptual referents, increased synonym use across age accords with general increases in emotion abstraction. However, the lack of an emergent pattern within our observed age window suggests that synonym use may arise younger than 4 years of age. Future research could investigate this possibility. Interestingly, the physiological marker pattern suggests that the external simulations children use to understand emotions are tightly linked to their physiological experiences of emotion. A speculative interpretation for this finding is that children are taught about emotion words via specific physiological cues (e.g., smiles, tears, racing heart), and so these cues are salient when they are asked to define emotion words. However, this unexpected result merits further research.

Although we refrain from overinterpreting data from individual emotions, descriptive data suggest that age-related patterns of emotion concept abstraction are more heterogeneous across emotions than emotion word comprehension. This observation shows that emotion comprehension and abstraction demonstrate interesting qualitative differences across emotions. However, this finding also raises the possibility that representations of some emotions may inherently be more abstract than others (e.g., definitions of nervousness appear to remain more concrete than definitions of other emotions even in young adulthood). If this finding replicates in future work, researchers should investigate what qualities of different emotion concepts lead participants to represent them as more or less abstract.

To connect this and the preceding sections, it is worth noting that including assessments of emotion comprehension and abstraction in one study provided several important benefits. First, this design allowed for tight control of emotion abstraction measures. It is meaningless to analyze the abstractness of participants’ emotion definitions if they are not in fact providing a valid definition of the emotion in question. Thus, measuring emotion comprehension and abstraction in one study allows for inaccurate definitions to be discarded. Second, this design allowed for direct comparisons between measures of emotion comprehension and emotion abstraction. In particular, comparing these phenomena within-subject allowed us to empirically demonstrate that abstraction plateaus significantly later than comprehension. This comparison underlies some of the study’s key conclusions, including (i) that the concepts underlying emotion words undergo continued abstraction even after they are comprehended, (ii) that children primarily demonstrate comprehension of emotion words by providing example situations (rather than general definitions), and (iii) that adolescence is a period in which emotion terms are comprehended but there is ongoing development in abstractness.

Toward an Automated Linguistic Measure of Abstraction

A final methodological contribution of the current work is the use of a purely linguistic approach to measure abstraction of emotion definitions. We found that linguistic abstractness scores defined by the LCM showed moderate correlations with abstractness ratings provided by human coders and increased linearly across age. These results support the notions that linguistic and human methods can converge as measures of abstractness and that emotion concept abstraction increases across age. As such, these data provide a first step towards developing a fully automated linguistic measure of emotion concept abstraction.

However, four concerns emerge when considering the strength of the relationship between these two approaches: (i) the two measures did not reveal the same patterns of developmental change (i.e., one was linear and the other emergent), (ii) the strength of the correlation between coder and linguistic measures was modest (i.e., r = .45), (iii) age was more strongly associated with coder abstractness scores than linguistic abstractness scores, and (iv) coherence between linguistic abstractness scores and coder abstractness scores broke down at young ages. Although lack of coherence was explained by the fact that some young participants had few words on which a linguistic abstractness score could be computed (see Supplemental Materials) and excluding these participants stabilized the correlation between coder and linguistic abstractness scores across age, this suggests that the current psycholinguistic approach was particularly noisy in computing the abstractness of young children’s speech.

These discrepancies could either imply that these measures tap different constructs or that the measures suffer from too much measurement error to reliably assess the underlying construct of “abstractness.” It is our intuition that these discrepancies arose because the linguistic measure was less sensitive to detecting and quantifying abstract features than human coders (i.e., the measurement error of the LCM approach reduced its correlation with the human measure). However, this is purely speculative. Future work could take a data-driven approach to this problem and use machine-learning methods to determine the optimal mathematical formula for determining emotion abstraction from linguistic data. Yet this work will need to contend with the difficult question of how to benchmark linguistic measures (i.e., what should be considered the “true” measure of abstractness against which linguistic measures are compared?). Many researchers use human ratings to measure abstractness, but these perceptions may be subject to bias. Hence, a more thorough treatment of construct validity in abstractness measures— including a definition of how to define “true” measures of abstractness—is needed for the field to refine this automated linguistic measure.

Limitations and Future Directions

In addition to the limitations and future directions outlined throughout the discussion, further research can address three additional limitations of the current work. First, the cross-sectional nature of the current design limits inferences about the causal influence of age on the development of emotion comprehension and abstraction. A longitudinal design could rule out potential third variables or cohort effects that limit causal inferences concerning relations between age and the dependent variables studied here.

Second, although the current project characterizes age-related changes of emotion comprehension and abstraction, it is limited in that these patterns are not compared to those of non-emotional words and concepts. Hence, the project does not address the question of whether these patterns are unique to emotion words or rather the product of more domain-general developmental shifts. Although prior theoretical and empirical work (as well as the principle of parsimony) suggests that emotion development likely scaffolds on the development of other faculties (like non-emotion vocabulary and non-verbal abstraction; e.g., Nook, Sasse, et al., 2017; Ponari et al., 2018), this question should be empirically tested. Put more broadly, the current study does not identify the mechanisms of increased emotion abstraction across age. This development could arise because of domain-general (i.e., both verbal and non-verbal) increases in the ability to engage in abstract representations (Caramelli et al., 2004; Dumontheil, 2014; Ponari et al., 2018; Whitaker et al., 2018). Thus, the extent to which the phenomenon assessed here scaffolds on broader developments in the ability to engage verbal and non-verbal abstractions remains an important question for future research. However, increased emotion abstraction across age could also arise because of increased diversity in one’s affective experiences, allowing for simulations of the same emotion concept to draw upon abstract features that generalize across situations (e.g., feeling afraid both when alone in the woods or when giving a speech because of a common sense of threat). This explanatory hypothesis also merits further research.

Third, the current study did not address the downstream consequences of emotion comprehension and abstraction. Several lines of work suggest that having mastery over a large set of emotion words facilitates adaptive emotion regulation (A. W. Brooks, 2014; Jamieson, Mendes, Blackstock, & Schmader, 2010; Kashdan, Barrett, & McKnight, 2015; Torre & Lieberman, 2018), implying that increased emotion comprehension might help people label their affective states in ways that allow them to effectively down-regulate negative affect. However, it is less clear whether increased emotion abstraction is helpful or harmful for mental wellness. On the one hand, emotion abstraction might facilitate emotion regulation. Prior work demonstrates that (i) psychological distancing facilitates emotion regulation (Ayduk & Kross, 2010; Kross et al., 2014; Moser et al., 2017; Nook, Schleider, & Somerville, 2017) and (ii) psychological distancing involves representing stimuli at higher levels of abstraction (Liberman & Forster, 2009; Soderberg, Callahan, Kochersberger, Amit, & Ledgerwood, 2015). Combining these ideas suggests that being able to view one’s emotional experiences abstractly might allow one to more adaptively engage in behaviors that are in one’s long-term best interests when faced with emotionally evocative situations (e.g., Fujita & Carnevale, 2012).

On the other hand, increased abstraction could actually lead to increased susceptibility to distress and mental disorders. As Davey, Yucel, & Allen (2008) argue, the rewards that young children pursue tend to be concrete in nature (e.g., food, immediate expressions of care), but these rewards shift to be more abstract in adolescence and adulthood (e.g., social belonging, career success). These more abstract sources of joy and contentment are also more difficult to attain (e.g., subtle forms of social rejection or minor errors during career evaluation can disrupt social belonging and career success). Increased abstraction across development could increase one’s feelings of rejection, frustration, and anxiety, ultimately increasing one’s vulnerability to depression (Davey et al., 2008). Thus if, when, and how emotion abstraction confers psychological benefits remains an important question for future research.

Conclusion

The current project utilizes a novel method for assessing emotion comprehension and abstraction, takes advantage of cutting-edge statistical and linguistic methods, and reveals that the nature of emotion concept changes across development. These findings extend how we understand emotion concepts and emotion development and may foster additional insight into the implications of abstract emotion representations.

Supplementary Material

Supplemental Material

NIHMS1037908-supplement-Supplemental_Material.pdf^{(190.8KB, pdf)}

Acknowledgments

The authors thank China Rose Marie Bolden, Aridenne Dews, Eliot Fearey, Megan Garrad, Tierra Gogue-Garcia, Kevin Kent, Alka Sareen, Manna Sirak, Taylor Stacy, and Christina Uhrig for assistance with data collection; Charlotte Anrig, Hannah Broos, Sandy Li, Azure Reid-Russell, and José Santiago for assistance with transcriptions; and Sandy Li and Sadhana Ponnaluri for assistance with ratings. This project was supported by a Harvard University startup grant to LHS, a National Institutes of Mental Health grant to KAM (R01-MH103291), and a National Science Foundation Graduate Research Fellowship (DGE1144152) to ECN.

Footnotes

Data concerning race and socioeconomic status were provided by adult participants themselves and from parents of participants who were minors

A plateau could not be identified in 0.84% of bootstrapped simulations for example situation use. In these simulations, the slope decreased across the entire sample and did not slow to the rate that would qualify as a plateau, suggesting that the plateau age was not reached before age 25.91. Although these simulations were excluded from confidence interval estimates, we ensured that replacing these missing values with ages above 26 did not affect the significance of inferences drawn from confidence interval comparisons (i.e., these exclusions do not affect the conclusion that the plateaus of the three abstraction measures do not differ significantly from each other, and they all differ significantly from the comprehension plateau).

Related publications

Portions of these results have been shared via presentations at the Social Communication Across the Lifespan conference (2018) and the Nebraska Symposium on Motivation (2018).

Conflicts of Interest

Authors declare no conflicts of interest.

REFERENCES

Akaike H (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 79(6), 716–723. 10.1109/TAC.1974.1100705 [DOI] [Google Scholar]
Altarriba J, & Bauer LM (2004). The distinctiveness of emotion concepts: A comparison between emotion, abstract, and concrete words. The American Journal of Psychology, 117(3), 389 10.2307/4149007 [DOI] [PubMed] [Google Scholar]
Andrews M, Frank S, & Vigliocco G (2014). Reconciling embodied and distributional accounts of meaning in language. Topics in Cognitive Science, 6(3), 359–370. 10.1111/tops.12096 [DOI] [PubMed] [Google Scholar]
Ayduk Ö, & Kross E (2010). From a distance: Implications of spontaneous self-distancing for adaptive self-reflection. Journal of Personality and Social Psychology, 98(5), 809–829. 10.1037/a0019205 [DOI] [PMC free article] [PubMed] [Google Scholar]
Aznar A, & Tenenbaum HR (2013). Spanish parents’ emotion talk and their children’s understanding of emotion. Frontiers in Psychology, 4, 1–11. 10.3389/fpsyg.2013.00670 [DOI] [PMC free article] [PubMed] [Google Scholar]
Balconi M, & Carrera A (2007). Emotional representation in facial expression and script. A comparison between normal and autistic children. Research in Developmental Disabilities, 28(4), 409–422. 10.1016/j.ridd.2006.05.001 [DOI] [PubMed] [Google Scholar]
Baron-Cohen S, Golan O, Wheelwright S, Granader Y, & Hill J (2010). Emotion word comprehension from 4 to 16 years old: A developmental survey. Frontiers in Evolutionary Neuroscience, 2, 109 10.3389/fnevo.2010.00109 [DOI] [PMC free article] [PubMed] [Google Scholar]
Barrett LF (2006). Solving the emotion paradox: Categorization and the experience of emotion. Personality and Social Psychology Review, 10(1), 20–46. 10.1207/s15327957pspr1001_2 [DOI] [PubMed] [Google Scholar]
Barrett LF (2017). The theory of constructed emotion: an active inference account of interoception and categorization. Social Cognitive and Affective Neuroscience, 12(1), 1–23. 10.1093/scan/nsw154 [DOI] [PMC free article] [PubMed] [Google Scholar]
Barsalou LW (1999). Perceptual symbol systems. The Behavioral and Brain Sciences, 22(4), 577–660. [DOI] [PubMed] [Google Scholar]
Barsalou LW (2003a). Abstraction in perceptual symbol systems. Philosophical Transactions of the Royal Society B: Biological Sciences, 355(1435), 1177–1187. 10.1098/rstb.2003.1319 [DOI] [PMC free article] [PubMed] [Google Scholar]
Barsalou LW (2003b). Situated simulation in the human conceptual system. Language and Cognitive Processes, 15(5–6), 513–562. 10.1080/01690960344000026 [DOI] [Google Scholar]
Barsalou LW (2009). Simulation, situated conceptualization, and prediction. Philosophical Transactions of the Royal Society B: Biological Sciences, 364, 1281–1289. 10.1098/rstb.2008.0319 [DOI] [PMC free article] [PubMed] [Google Scholar]
Barsalou LW, Dutriaux L, & Scheepers C (2018). Moving beyond the distinction between concrete and abstract concepts. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1752), 20170144. 10.1098/rstb.2017.0144 [DOI] [PMC free article] [PubMed] [Google Scholar]
Bauminger N, Schorr Edelsztein H, & Morash J (2005). Social information processing and emotional understanding in children with LD. Journal of Learning Disabilities, 35(1), 45–61. 10.1177/00222194050380010401 [DOI] [PubMed] [Google Scholar]
Beck L, Kumschick IR, Eid M, & Klann-Delius G (2012). Relationship between language competence and emotional competence in middle childhood. Emotion, 12(3), 503–514. 10.1037/a0026320 [DOI] [PubMed] [Google Scholar]
Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B, 57(1), 289–300. 10.2307/2346101 [DOI] [Google Scholar]
Blakemore SJ (2008). The social brain in adolescence. Nature Reviews Neuroscience, 9(4), 267–277. 10.1038/nrn2353 [DOI] [PubMed] [Google Scholar]
Bloom P (2000). How children learn the meanings of words. Cambridge, MA: MIT Press. [Google Scholar]
Borghi AM, & Binkofski F (2014). Words as social tools: An embodied view on abstract concepts. New York, NY: Springer. [Google Scholar]
Borghi AM, Binkofski F, Castelfranchi C, Cimatti F, Scorolli C, & Tummolini L (2017). The challenge of abstract concepts. Psychological Bulletin, 143(3), 263–292. 10.1037/bul0000089 [DOI] [PubMed] [Google Scholar]
Bretherton I, & Beeghly M (1982). Talking about internal states: The acquisition of an explicit theory of mind. Developmental Psychology, 18(6), 906–921. 10.1037/0012-1649.18.6.906 [DOI] [Google Scholar]
Bretherton I, Fritz J, Zahn-Waxler C, & Ridgeway D (1986). Learning to talk about emotions: A functionalist perspective. Child Development, 57(3), 529 10.2307/1130334 [DOI] [Google Scholar]
Bridges KMB (1930). A genetic theory of the emotions. Pedagogical Seminary and Journal of Genetic Psychology, 37, 514–527. [Google Scholar]
Brooks AW (2014). Get excited: Reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143(3), 1144–1158. 10.1037/a0035325 [DOI] [PubMed] [Google Scholar]
Brooks JA, & Freeman JB (2018). Conceptual knowledge predicts the representational structure of facial emotion perception. Nature Human Behaviour, 2(8), 581–591. 10.1038/s41562-018-0376-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
Brysbaert M, Warriner AB, & Kuperman V (2014). Concreteness ratings for 40 thousand generally known English word lemmas. Behavior Research Methods, 46(3), 904–911. 10.3758/s13428-013-0403-5 [DOI] [PubMed] [Google Scholar]
Burgoon EM, Henderson MD, & Markman AB (2013). There are many ways to see the forest for the trees. Perspectives on Psychological Science, 5(5), 501–520. 10.1177/1745691613497964 [DOI] [PubMed] [Google Scholar]
Burnett S, Bird G, Moll J, Frith CD, & Blakemore S-J (2009). Development during adolescence of the neural processing of social emotion. Journal of Cognitive Neuroscience, 21(9), 1736–1750. 10.1162/jocn.2009.21121 [DOI] [PMC free article] [PubMed] [Google Scholar]
Canty A, & Ripley B (2017). Boot: Bootstrap R (S-Plus) functions. R package version 13–20. [Google Scholar]
Caramelli N, & Setti A (2005). Different domains in abstract concepts. In Bara B, Barsalou L, & Bucciarelli M (Eds.), Proceedings of the XXVII Annual Conference of the Cognitive Science Society (pp. 1997–2002). Mahwah, NJ: Erlbaum. [Google Scholar]
Caramelli N, Setti A, & Maurizzi DD (2004). Concrete and abstract concepts in school age children. Psychology of Language and Communication, 5(2), 19–34. [Google Scholar]
Carnaghi A, Maass A, Gresta S, Bianchi M, Cadinu M, & Arcuri L (2008). Nomina sunt omina: On the inductive potential of nouns and adjectives in person perception. Journal of Personality and Social Psychology, 94(5), 839–859. 10.1037/0022-3514.94.5.839 [DOI] [PubMed] [Google Scholar]
Casey BJ (2015). Beyond simple models of self-control to circuit-based accounts of adolescent behavior. Annual Review of Psychology, 66(1), 295–319. [DOI] [PubMed] [Google Scholar]
Castro VL, Cheng Y, Halberstadt AG, & Gruhn D (2016). EUReKA! A conceptual model of emotion understanding. Emotion Review, 8(3), 258–268. 10.1177/1754073915580601 [DOI] [PMC free article] [PubMed] [Google Scholar]
Cervantes CA, & Callanan MA (1998). Labels and explanations in mother-child emotion talk: Age and gender differentiation. Developmental Psychology, 34(1), 88–98. 10.1037/0012-1649.34.L88 [DOI] [PubMed] [Google Scholar]
Choudhury S, Blakemore SJ, & Charman T (2006). Social cognitive development during adolescence. Social Cognitive and Affective Neuroscience, 1(3), 165–174. 10.1093/scan/nsl024 [DOI] [PMC free article] [PubMed] [Google Scholar]
Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum; 10.4324/9780203771587 [DOI] [Google Scholar]
Cole PM, Armstrong LM, & Pemberton CK (2010). The role of language in the development of emotion regulation In Calkins SD & Bell MA (Eds.), Human brain development. Child development at the intersection of emotion and cognition (pp. 59–77). Washington, DC: American Psychological Association. [Google Scholar]
Crone EA, Wendelken C, van Leijenhorst L, Honomichl RD, Christoff K, & Bunge SA (2009). Neurocognitive development of relational reasoning. Developmental Science, 12(1), 55–66. 10.1111/j.1467-7687.2008.00743.x [DOI] [PMC free article] [PubMed] [Google Scholar]
Davey CG, Yücel M, & Allen NB (2008). The emergence of depression in adolescence: Development of the prefrontal cortex and the representation of reward. Neuroscience and BiobehavioralReviews, 32(1), 1–19. https://doi.org/10.1016Zj.neubiorev.2007.04.016 [DOI] [PubMed] [Google Scholar]
Davison AC, & Hinkley DV (1997). Bootstrap methods and their applications. Cambridge: Cambridge University Press. [Google Scholar]
Della Rosa PA, Catricalà E, Vigliocco G, & Cappa SF (2010). Beyond the abstract—concrete dichotomy: Mode of acquisition, concreteness, imageability, familiarity, age of acquisition, context availability, and abstractness norms for a set of 417 Italian words. Behavior Research Methods, 42(4), 1042–1048. 10.3758/BRM.42.4.1042 [DOI] [PubMed] [Google Scholar]
Demetriou A, Makris N, Spanoudis G, Kazi S, Shayer M, & Kazali E (2018). Mapping the dimensions of general intelligence: An integrated differential-developmental theory. Human Development, 61(1), 4–42. 10.1159/000484450 [DOI] [Google Scholar]
DiCiccio TJ, & Efron B (1996). Bootstrap confidence intervals. Statistical Science, 11(3), 189–228. 10.1214/ss/1032280214 [DOI] [Google Scholar]
Doré B, Ort L, Braverman O, & Ochsner KN (2015). Sadness shifts to anxiety over time and distance from the national tragedy in Newtown, Connecticut. Psychological Science, 26(4), 363–373. 10.1177/0956797614562218 [DOI] [PMC free article] [PubMed] [Google Scholar]
Doyle CM, & Lindquist KA (2018). When a word is worth a thousand pictures: Language shapes perceptual memory for emotion. Journal of Experimental Psychology: General, 147(1), 62–73. 10.1037/xge0000361 [DOI] [PubMed] [Google Scholar]
Dumontheil I (2014). Development of abstract thinking during childhood and adolescence: The role of rostrolateral prefrontal cortex. Developmental Cognitive Neuroscience, 10, 57–76. 10.1016/_j.dcn.2014.07.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
Duncombe M, Havighurst SS, Holland KA, & Frankling EJ (2013). Relations of emotional competence and effortful control to child disruptive behavior problems. Early Education & Development, 24(5), 599–615. 10.1080/10409289.2012.701536 [DOI] [Google Scholar]
Dunn J, Bretherton I, & Munn P (1987). Conversations about feeling states between mothers and their young children. Developmental Psychology, 23(1), 132–139. 10.1037/0012-1649.23.L132 [DOI] [Google Scholar]
Dunn LM, & Dunn LM (1997). Peabody Picture Vocabulary Test (3rd ed.). Circle Pines, MN: American Guidance Service. [Google Scholar]
Ekman P, & Friesen WV (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17(2), 124–129. 10.1037/h0030377 [DOI] [PubMed] [Google Scholar]
Fabes RA, Eisenberg N, Hanish LD, & Spinrad TL (2001). Preschoolers’ spontaneous emotion vocabulary: Relations to likability. Early Education & Development, 12(1), 11–27. 10.1207/s15566935eed1201_2 [DOI] [Google Scholar]
Fan R, Varol O, Varamesh A, Barron A, van de Leemput IA, Scheffer M, & Bollen J (2019). The minute-scale dynamics of online emotions reveal the effects of affect labeling. Nature Human Behaviour, 3(1), 92–100. 10.1038/s41562-018-0490-5 [DOI] [PubMed] [Google Scholar]
Farkas G, & Beron K (2004). The detailed age trajectory of oral vocabulary knowledge: Differences by class and race. Social Science Research, 33(3), 464–497. https://doi.org/10.1016Zj.ssresearch.2003.08.001 [Google Scholar]
Ferrer E, O’Hare ED, & Bunge SA (2009). Fluid reasoning and the developing brain. Frontiers in Neuroscience, 3(1), 46–51. 10.3389/neuro.01.003.2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
Forgas JP (2007). When sad is better than happy: Negative affect can improve the quality and effectiveness of persuasive messages and social influence strategies. Journal of Experimental Social Psychology, 43(4), 513–528. 10.1016/jjesp.2006.05.006 [DOI] [Google Scholar]
Franz P, Nook EC, Mair P, & Nock MK (2019). Using topic modeling to detect and describe self-injurious and related content on a large-scale digital platform. Revision under Review. [DOI] [PubMed] [Google Scholar]
Fugate JMB, Gouzoules H, & Barrett LF (2010). Reading chimpanzee faces: Evidence for the role of verbal labels in categorical perception of emotion. Emotion, 10(4), 544–554. 10.1037/a0019017 [DOI] [PMC free article] [PubMed] [Google Scholar]
Fujita K, & Carnevale JJ (2012). Transcending temptation through abstraction. Current Directions in Psychological Science, 21(4), 248–252. 10.1177/0963721412449169 [DOI] [Google Scholar]
Gendron M, Lindquist KA, Barsalou L, & Barrett LF (2012). Emotion words shape emotion percepts. Emotion, 12(2), 314–325. 10.1037/a0026007 [DOI] [PMC free article] [PubMed] [Google Scholar]
Gopnik A (2001). Theories, language, and culture: Whorf without wincing In Bowerman M & Levinson S (Eds.), Language acquisition and conceptual development (pp. 45–69). Cambridge: Cambridge University Press; 10.1017/CB09780511620669.004 [DOI] [Google Scholar]
Gray AW, Parkinson B, & Dunbar RI (2015). Laughter’s influence on the intimacy of self disclosure. Human Nature, 26(1), 28–43. 10.1007/s12110-015-9225-8 [DOI] [PubMed] [Google Scholar]
Greenberg MT, Kusche CA, Cook ET, & Quamma JP (1995). Promoting emotional competence in school-aged children: The effects of the PATHS curriculum. Development and Psychopathology, 7(01), 117 10.1017/S0954579400006374 [DOI] [Google Scholar]
Hale SC (1988). Spacetime and the abstract / concrete distinction. Philosophical Studies, 53(1), 85–102. [Google Scholar]
Inhelder B, & Piaget J (1958). The growth of logical thinking from childhood to adolescence: An essay on the construction of formal operational structures. New York, NY: Basic Books. [Google Scholar]
Jamieson JP, Mendes WB, Blackstock E, & Schmader T (2010). Turning the knots in your stomach into bows: Reappraising arousal improves performance on the GRE. Journal of Experimental Social Psychology, 46(1), 208–212. 10.1016/jjesp.2009.08.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
Joelson JM, & Herrmann DJ (1978). Properties of categories in semantic memory. The American Journal of Psychology, 91(1), 101 10.2307/1421826 [DOI] [Google Scholar]
Kashdan TB, Barrett LF, & McKnight PE (2015). Unpacking emotion differentiation: Transforming unpleasant experience by perceiving distinctions in negativity. Current Directions in Psychological Science, 24(1), 10–16. 10.1177/0963721414550708 [DOI] [Google Scholar]
Kelly B, Longbottom J, Potts F, & Williamson J (2004). Applying emotional intelligence: Exploring the Promoting Alternative Thinking Strategies curriculum. Educational Psychology in Practice, 20(3), 221–240. 10.1080/0266736042000251808 [DOI] [Google Scholar]
Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, & Walters EE (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 593 10.1001/archpsyc.62.6.593 [DOI] [PubMed] [Google Scholar]
Kober H, Barrett LF, Joseph J, Bliss-Moreau E, Lindquist K, & Wager TD (2008). Functional grouping and cortical-subcortical interactions in emotion: A meta-analysis of neuroimaging studies. NeuroImage, 42(2), 998–1031. https://doi.org/10.1016Zj.neuroimage.2008.03.059 [DOI] [PMC free article] [PubMed] [Google Scholar]
Kopp CB (1989). Regulation of distress and negative emotions: A developmental view. Developmental Psychology, 25(3), 343–354. 10.1037//0012-1649.25.3.343 [DOI] [Google Scholar]
Kousta S-T, Vigliocco G, Vinson DP, Andrews M, & Del Campo E (2011). The representation of abstract words: Why emotion matters. Journal of Experimental Psychology: General, 140(1), 14–34. 10.1037/a0021446 [DOI] [PubMed] [Google Scholar]
Kross E, Bruehlman-Senecal E, Park J, Burson A, Dougherty A, Shablack H, ... Ayduk O (2014). Self-talk as a regulatory mechanism: How you do it matters. Journal of Personality and Social Psychology, 106(2), 304–324. 10.1037/a0035173 [DOI] [PubMed] [Google Scholar]
Kuhn LJ, Willoughby MT, Vernon-Feagans L, & Blair CB (2016). The contribution of children’s time-specific and longitudinal expressive language skills on developmental trajectories of executive function. Journal of Experimental Child Psychology, 145, 20–34. 10.1016/jjecp.2016.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
Kuperman V, Stadthagen-Gonzalez H, & Brysbaert M (2012). Age-of-acquisition ratings for 30,000 English words. Behavior Research Methods, 44(4), 978–990. 10.3758/s13428-012-0210-4 [DOI] [PubMed] [Google Scholar]
Kusché CA, Belike RL, & Greenberg MT (1988). Kusché Affective Inventory-Revised. Unpublished measure. Seattle, WA: University of Washington. [Google Scholar]
Lagattuta KH, & Wellman HM (2002). Differences in early parent-child conversations about negative versus positive emotions: Implications for the development of psychological understanding. Developmental Psychology, 35(4), 564–580. 10.1037/0012-1649.38.4.564 [DOI] [PubMed] [Google Scholar]
Larson RW, Moneta G, Richards MH, & Wilson S (2002). Continuity, stability, and change in daily emotional experience across adolescence. Child Development, 73(4), 1151–1165. 10.1111/1467-8624.00464 [DOI] [PubMed] [Google Scholar]
Li Y, & Yu D (2015). Development of emotion word comprehension in Chinese children from 2 to 13 years old: Relationships with valence and empathy. PLOS ONE, 10(12), e0143712. 10.1371/journal.pone.0143712 [DOI] [PMC free article] [PubMed] [Google Scholar]
Liberman N, & Förster J (2009). The effect of psychological distance on perceptual level of construal. Cognitive Science, 33(7), 1330–1341. 10.1111/j.1551-6709.2009.01061.x [DOI] [PubMed] [Google Scholar]
Lindquist KA (2017). The role of language in emotion: existing evidence and future directions. Current Opinion in Psychology, 17, 135–139. 10.1016/jxopsyc.2017.07.006 [DOI] [PubMed] [Google Scholar]
Lindquist KA, Barrett LF, Bliss-Moreau E, & Russell JA (2006). Language and the perception of emotion. Emotion, 6(1), 125–138. 10.1037/1528-3542.6.1.125 [DOI] [PubMed] [Google Scholar]
Lindquist KA, MacCormack JK, & Shablack H (2015). The role of language in emotion: predictions from psychological constructionism. Frontiers in Psychology, 6(444). 10.3389/fpsyg.2015.00444 [DOI] [PMC free article] [PubMed] [Google Scholar]
Lindquist KA, Satpute AB, & Gendron M (2015). Does language do more than communicate emotion? Current Directions in Psychological Science, 24(2), 99–108. 10.1177/0963721414553440 [DOI] [PMC free article] [PubMed] [Google Scholar]
Lindquist KA, Wager TD, Kober H, Bliss-Moreau E, Feldman Barrett L, & Barrett LF (2012). The brain basis of emotion: A meta-analytic review. The Behavioral and Brain Sciences, 35(3), 121–143. 10.1017/S0140525X11000446 [DOI] [PMC free article] [PubMed] [Google Scholar]
Locke RL, Miller AL, Seifer R, & Heinze JE (2015). Context-inappropriate anger, emotion knowledge deficits, and negative social experiences in preschool. Developmental Psychology, 51(10), 1450–1463. 10.1037/a0039528 [DOI] [PMC free article] [PubMed] [Google Scholar]
Lupyan G (2012a). Linguistically modulated perception and cognition: The label-feedback hypothesis. Frontiers in Psychology, 3 10.3389/fpsyg.2012.00054 [DOI] [PMC free article] [PubMed] [Google Scholar]
Lupyan G (2012b). What do words do? Toward a theory of language-augmented thought In Ross BH (Ed.), Psychology of Learning and Motivation (Vol. 57, pp. 255–297). Waltham, MA: Academic Press; 10.1016/B978-0-12-394293-7.00007-8 [DOI] [Google Scholar]
Lupyan G, & Winter B (2018). Language is more abstract than you think, or, why aren’t languages more iconic? Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1752). 10.1098/rstb.2017.0137 [DOI] [PMC free article] [PubMed] [Google Scholar]
Luria A (1961). The role of speech in the regulation of normal and abnormal behaviors. New York, NY: Liveright. [Google Scholar]
MacCann C, & Roberts RD (2008). New paradigms for assessing emotional intelligence: Theory and data. Emotion, 5(4), 540–551. 10.1037/a0012746 [DOI] [PubMed] [Google Scholar]
MacWhinney B, & Snow C (1990). The Child Language Data Exchange System: an update. Journal of Child Language, 17(2), 457–472. 10.1017/S0305000900013866 [DOI] [PMC free article] [PubMed] [Google Scholar]
Mahon BZ, & Caramazza A (2008). A critical look at the embodied cognition hypothesis and a new proposal for grounding conceptual content. Journal of Physiology-Paris, 102(1–3), 59–70. 10.1016/jjphysparis.2008.03.004 [DOI] [PubMed] [Google Scholar]
Matthews D, Biney H, & Abbot-Smith K (2018). Individual differences in children’s pragmatic ability: A review of associations with formal language, social cognition, and executive functions. Language Learning and Development, 14(3), 186–223. 10.1080/15475441.2018.1455584 [DOI] [Google Scholar]
Mazzuca C, Barca L, & Borghi AM (2017). The particularity of emotional words: A grounded approach. Rivista Internazionale Di Filosofia e Psicologia, 8(2), 124–133. 10.4453/rifp.2017.0010 [DOI] [Google Scholar]
Mervis C, & Rosch E (1981). Categorization of natural objects. Annual Review of Psychology, 32(1), 89–115. 10.1146/annurev.ps.32.020181.000513 [DOI] [Google Scholar]
Moser JS, Dougherty A, Mattson WI, Katz B, Moran TP, Guevarra D, ... Kross E (2017). Third-person self-talk facilitates emotion regulation without engaging cognitive control: Converging evidence from ERP and fMRI. Scientific Reports, 7(1), 4519 10.1038/s41598-017-04047-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
Nook EC, Lindquist KA, & Zaki J (2015). A new look at emotion perception: Concepts speed and shape facial emotion recognition. Emotion, 15(5), 569–578. 10.1037/a0039166 [DOI] [PubMed] [Google Scholar]
Nook EC, Sasse SF, Lambert HK, McLaughlin KA, & Somerville LH (2017). Increasing verbal knowledge mediates development of multidimensional emotion representations. Nature Human Behaviour, 1(12), 881–889. 10.1038/s41562-017-0238-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
Nook EC, Sasse SF, Lambert HK, McLaughlin KA, & Somerville LH (2018). The nonlinear development of emotion differentiation: Granular emotional experience Is low in adolescence. Psychological Science, 29(8), 1346–1357. 10.1177/0956797618773357 [DOI] [PMC free article] [PubMed] [Google Scholar]
Nook EC, Schleider JL, & Somerville LH (2017). A linguistic signature of psychological distancing in emotion regulation. Journal of Experimental Psychology: General, 146(3), 337–346. 10.1037/xge0000263 [DOI] [PubMed] [Google Scholar]
O’Kearney R, & Dadds M (2004). Developmental and gender differences in the language for emotions across the adolescent years. Cognition & Emotion, 18(7), 913–938. 10.1080/02699930341000356 [DOI] [Google Scholar]
O’Kearney R, & Dadds MR (2005). Language for emotions in adolescents with externalizing and internalizing disorders. Development and Psychopathology, 17(2), 529–548. 10.1017/S095457940505025X [DOI] [PubMed] [Google Scholar]
Oosterwijk S, Lindquist KA, Adebayo M, & Barrett LF (2015). The neural representation of typical and atypical experiences of negative images: Comparing fear, disgust and morbid fascination. Social Cognitive and Affective Neuroscience, (July), nsv088. 10.1093/scan/nsv088 [DOI] [PMC free article] [PubMed] [Google Scholar]
Paivio A, Yuille JC, & Madigan SA (1968). Concreteness, imagery, and meaningfulness values for 925 nouns. Journal of Experimental Psychology, 76(1), 1–25. 10.1037/h0025327 [DOI] [PubMed] [Google Scholar]
Pennebaker JW, Booth RJ, & Francis ME (2007). Linguistic Inquiry and Word Count: LIWC [Computer software]. Austin, TX: LIWC.net. [Google Scholar]
Ponari M, Norbury CF, & Vigliocco G (2018). Acquisition of abstract concepts is influenced by emotional valence. Developmental Science, 21(2), e12549. 10.1111/desc.12549 [DOI] [PubMed] [Google Scholar]
Pons F, Harris PL, & de Rosnay M (2004). Emotion comprehension between 3 and 11 years: Developmental periods and hierarchical organization. European Journal of Developmental Psychology, 1(2), 127–152. 10.1080/17405620344000022 [DOI] [Google Scholar]
R Core Team. (2018). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; Retrieved from https://www.r-project.org/ [Google Scholar]
Raven J (2000). The Raven’s Progressive Matrices: Change and stability over culture and time. Cognitive Psychology, 41(1), 1–48. 10.1006/cogp.1999.0735 [DOI] [PubMed] [Google Scholar]
Reichenbach L, & Masters JC (1983). Children’s use of expressive and contextual cues in judgments of emotion. Child Development, 54(4), 993 10.2307/1129903 [DOI] [Google Scholar]
Roben CKP, Cole PM, & Armstrong LM (2013). Longitudinal relations among language skills, anger expression, and regulatory strategies in early childhood. Child Development, 84(3), 891–905. 10.1111/cdev.12027 [DOI] [PMC free article] [PubMed] [Google Scholar]
Rodman AM, Powers KE, & Somerville LH (2017). Development of self-protective biases in response to social evaluative feedback. Proceedings of the National Academy of Sciences, 114(50), 13158–13163. 10.1073/pnas.1712398114 [DOI] [PMC free article] [PubMed] [Google Scholar]
Rosch E, Mervis CB, Gray WD, Johnson DM, & Boyes-Braem P (1976). Basic objects in natural categories. Cognitive Psychology, 8(3), 382–439. 10.1016/0010-0285(76)90013-X [DOI] [Google Scholar]
Russell JA (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. [Google Scholar]
Russell JA, & Widen SC (2002). Words versus faces in evoking preschool children’s knowledge of the causes of emotions. International Journal of Behavioral Development, 26(2), 97–103. 10.1080/01650250042000582 [DOI] [Google Scholar]
Saarni C (1999). The development of emotional competence. New York, NY: Guilford Press. [Google Scholar]
Salmon K, O’Kearney R, Reese E, & Fortune C-A (2016). The role of language skill in child psychopathology: Implications for intervention in the early years. Clinical Child and Family Psychology Review, 19(4), 352–367. 10.1007/s10567-016-0214-1 [DOI] [PubMed] [Google Scholar]
Satpute AB, Nook EC, Narayanan S, Shu J, Weber J, & Ochsner KN (2016). Emotions in “black and white” or shades of gray? How we think about emotion shapes our perception and neural representation of emotion. Psychological Science, 27(11), 1428–1442. 10.1177/0956797616661555 [DOI] [PMC free article] [PubMed] [Google Scholar]
Schmid H (1994). Probabilistic part-of-speech tagging using decision trees. In Jones DB & Somers HL (Eds.), Proceedings of the International Conference on New Methods in Language Processing (pp. 44–49). Manchester, England: University of Manchester Institute of Science and Technology. [Google Scholar]
Sebastian CL, Fontaine NMG, Bird G, Blakemore S-J, De Brito SA, McCrory EJP, & Viding E (2012). Neural processing associated with cognitive and affective Theory of Mind in adolescents and adults. Social Cognitive and Affective Neuroscience, 7(1), 53–63. 10.1093/scan/nsr023 [DOI] [PMC free article] [PubMed] [Google Scholar]
Seih Y-T, Beier S, & Pennebaker JW (2017). Development and examination of the Linguistic Category Model in a computerized text analysis method. Journal of Language and Social Psychology, 36(3), 343–355. 10.1177/0261927X16657855 [DOI] [Google Scholar]
Semin GR, & Fiedler K (1988). The cognitive functions of linguistic categories in describing persons: Social cognition and language. Journal of Personality and Social Psychology, 54(4), 558–568. 10.1037/0022-3514.54A558 [DOI] [Google Scholar]
Semin GR, & Fiedler K (1991). The Linguistic Category Model, its bases, applications and range. European Review of Social Psychology, 2(1), 1–30. 10.1080/14792779143000006 [DOI] [Google Scholar]
Semin GR, Görts CA, Nandram S, & Semin-Goossens A (2002). Cultural perspectives on the linguistic representation of emotion and emotion events. Cognition & Emotion, 16(1), 11–28. 10.1080/02699930143000112 [DOI] [Google Scholar]
Shablack H, & Lindquist KA (2019). The role of language in emotional development. In LoBue V, Perez-Edgar K, & Buss K (Eds.), Handbook of Emotion Development. New York, NY: Springer. [Google Scholar]
Siegel LS, & Ryan EB (1989). The development of working memory in normally achieving and subtypes of learning disabled children. Child Development, 60(4), 973 10.2307/1131037 [DOI] [PubMed] [Google Scholar]
Simpson G (2014). First derivative of a spline term. GitHub Repository. Retrieved from https://gist.github.com/gavinsimpson/e73f011fdaaab4bb5a30 [Google Scholar]
Smith M, & Walden T (2001). Understanding feelings and coping with emotional situations: A comparison of maltreated and nonmaltreated preschoolers. Social Development, 5(1), 93–116. 10.1111/1467-9507.00082 [DOI] [Google Scholar]
Soderberg CK, Callahan SP, Kochersberger AO, Amit E, & Ledgerwood A (2015). The effects of psychological distance on abstraction: Two meta-analyses. Psychological Bulletin, 141(3), 525–548. 10.1037/bul0000005 [DOI] [PubMed] [Google Scholar]
Somerville LH, Jones RM, Ruberry EJ, Dyke JP, Glover G, & Casey BJ (2013). The medial prefrontal cortex and the emergence of self-conscious emotion in adolescence. Psychological Science, 24(8), 1554–1562. 10.1177/0956797613475633 [DOI] [PMC free article] [PubMed] [Google Scholar]
Somerville LH, & McLaughlin KA (2018). What develops during emotional development? Normative trajectories and sources of psychopathology risk in adolescence In Fox Andrew S., Lapate Regina C., Shackman Alexander J., & Davidson Richard J. (Eds.), The Nature of Emotion: Fundamental Questions (2nd ed.). [Google Scholar]
Southam-Gerow MA, & Kendall PC (2000). A preliminary study of the emotion understanding of youths referred for treatment of anxiety disorders. Journal of Clinical Child Psychology, 29(3), 319–327. 10.1207/S15374424JCCP2903_3 [DOI] [PubMed] [Google Scholar]
Stadthagen-Gonzalez H, & Davis CJ (2006). The Bristol norms for age of acquisition, imageability, and familiarity. Behavior Research Methods, 38(4), 598–605. 10.3758/BF03193891 [DOI] [PubMed] [Google Scholar]
Torre JB, & Lieberman MD (2018). Putting feelings into words: Affect labeling as implicit emotion regulation. Emotion Review, 10(2), 116–124. 10.1177/1754073917742706 [DOI] [Google Scholar]
Trentacosta CJ, & Izard CE (2007). Kindergarten children’s emotion competence as a predictor of their academic competence in first grade. Emotion, 7(1), 77–88. 10.1037/1528-3542.7.1.77 [DOI] [PubMed] [Google Scholar]
Trope Y, & Liberman N (2010). Construal-level theory of psychological distance. Psychological Review, 117(2), 440–463. 10.1037/a0018963 [DOI] [PMC free article] [PubMed] [Google Scholar]
Vallotton C, & Ayoub C (2011). Use your words: The role of language in the development of toddlers’ self-regulation. Early Childhood Research Quarterly, 26(2), 169–181. 10.1016/_j.ecresq.2010.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
Vigliocco G, Kousta S-T, Della Rosa PA, Vinson DP, Tettamanti M, Devlin JT, & Cappa SF (2014). The neural representation of abstract words: The role of emotion. Cerebral Cortex, 24(7), 1767–1777. 10.1093/cercor/bht025 [DOI] [PubMed] [Google Scholar]
Vygotsky LS (1962). Thought and language. Cambridge, MA: MIT Press. [Google Scholar]
Waggoner JE, & Palermo DS (1989). Betty is a bouncing bubble: Children’s comprehension of emotion-descriptive metaphors. Developmental Psychology, 25(1), 152–163. 10.1037/0012-1649.25.L152 [DOI] [Google Scholar]
Wang J, Conder JA, Blitzer DN, & Shinkareva SV (2010). Neural representation of abstract and concrete concepts: A meta-analysis of neuroimaging studies. Human Brain Mapping, 31(10), 1459–1468. 10.1002/hbm.20950 [DOI] [PMC free article] [PubMed] [Google Scholar]
Wechsler D (1999). Wechsler Abbreviated Scale of Intelligence (WASI). San Antonio, TX: Psychological Corporation. [Google Scholar]
Wechsler D (2011). Wechsler Abbreviated Scale of Intelligence-Second Edition Manual. Bloomington, MN: Pearson. [Google Scholar]
Whitaker KJ, Vendetti MS, Wendelken C, & Bunge SA (2018). Neuroscientific insights into the development of analogical reasoning. Developmental Science, 21(2), e12531. 10.1111/desc.12531 [DOI] [PMC free article] [PubMed] [Google Scholar]
Widen SC (2013). Children’s interpretation of facial expressions: The long path from valence-based to specific discrete categories. Emotion Review, 5(1), 72–77. 10.1177/1754073912451492 [DOI] [Google Scholar]
Widen SC, Pochedly JT, & Russell JA (2015). The development of emotion concepts: A story superiority effect in older children and adolescents. Journal of Experimental Child Psychology, 131, 186–192. 10.1016/jjecp.2014.10.009 [DOI] [PubMed] [Google Scholar]
Widen SC, & Russell JA (2010a). Children’s scripts for social emotions: Causes and consequences are more central than are facial expressions. British Journal of Developmental Psychology, 28(3), 565–581. 10.1348/026151009X457550d [DOI] [PubMed] [Google Scholar]
Widen SC, & Russell JA (2010b). Differentiation in preschooler’s categories of emotion. Emotion, 10(5), 651–661. 10.1037/a0019005 [DOI] [PubMed] [Google Scholar]
Wilson-Mendenhall CD (2017). Constructing emotion through simulation. Current Opinion in Psychology, 17, 189–194. https://doi.org/10.10167j.copsyc.2017.07.015 [DOI] [PubMed] [Google Scholar]
Wilson-Mendenhall CD, Barrett LF, & Barsalou LW (2013). Situating emotional experience. Frontiers in Human Neuroscience, 7, 764 10.3389/fnhum.2013.00764 [DOI] [PMC free article] [PubMed] [Google Scholar]
Wilson-Mendenhall CD, Barrett LF, Simmons WK, & Barsalou LW (2011). Grounding emotion in situated conceptualization. Neuropsychologia, 49(5), 1105–1127. 10.1016/j.neuropsychologia.2010.12.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
Wood SN (2003). Thin plate regression splines. Journal of the Royal Statistical Society B, 65(1), 95–114. [Google Scholar]
Wood SN (2017). Generalized additive models: An introduction with R (2nd ed.). CRC Press. [Google Scholar]
Wu Y, Muentener P, & Schulz LE (2017). One-to four-year-olds connect diverse positive emotional vocalizations to their probable causes. Proceedings of the National Academy of Sciences, 114(45), 11896–11901. 10.1073/pnas.1707715114 [DOI] [PMC free article] [PubMed] [Google Scholar]
Yin J, & Csibra G (2015). Concept-based word learning in human infants. Psychological Science, 26(8), 1316–1324. 10.1177/0956797615588753 [DOI] [PMC free article] [PubMed] [Google Scholar]
Zeman J, Klimes-Dougan B, Cassano M, & Adrian M (2007). Measurement issues in emotion research with children and adolescents. Clinical Psychology: Science and Practice, 14(4), 377–401. 10.1111/j.1468-2850.2007.00098.x [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material

NIHMS1037908-supplement-Supplemental_Material.pdf^{(190.8KB, pdf)}

[R1] Akaike H (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 79(6), 716–723. 10.1109/TAC.1974.1100705 [DOI] [Google Scholar]

[R2] Altarriba J, & Bauer LM (2004). The distinctiveness of emotion concepts: A comparison between emotion, abstract, and concrete words. The American Journal of Psychology, 117(3), 389 10.2307/4149007 [DOI] [PubMed] [Google Scholar]

[R3] Andrews M, Frank S, & Vigliocco G (2014). Reconciling embodied and distributional accounts of meaning in language. Topics in Cognitive Science, 6(3), 359–370. 10.1111/tops.12096 [DOI] [PubMed] [Google Scholar]

[R4] Ayduk Ö, & Kross E (2010). From a distance: Implications of spontaneous self-distancing for adaptive self-reflection. Journal of Personality and Social Psychology, 98(5), 809–829. 10.1037/a0019205 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] Aznar A, & Tenenbaum HR (2013). Spanish parents’ emotion talk and their children’s understanding of emotion. Frontiers in Psychology, 4, 1–11. 10.3389/fpsyg.2013.00670 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] Balconi M, & Carrera A (2007). Emotional representation in facial expression and script. A comparison between normal and autistic children. Research in Developmental Disabilities, 28(4), 409–422. 10.1016/j.ridd.2006.05.001 [DOI] [PubMed] [Google Scholar]

[R7] Baron-Cohen S, Golan O, Wheelwright S, Granader Y, & Hill J (2010). Emotion word comprehension from 4 to 16 years old: A developmental survey. Frontiers in Evolutionary Neuroscience, 2, 109 10.3389/fnevo.2010.00109 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] Barrett LF (2006). Solving the emotion paradox: Categorization and the experience of emotion. Personality and Social Psychology Review, 10(1), 20–46. 10.1207/s15327957pspr1001_2 [DOI] [PubMed] [Google Scholar]

[R9] Barrett LF (2017). The theory of constructed emotion: an active inference account of interoception and categorization. Social Cognitive and Affective Neuroscience, 12(1), 1–23. 10.1093/scan/nsw154 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] Barsalou LW (1999). Perceptual symbol systems. The Behavioral and Brain Sciences, 22(4), 577–660. [DOI] [PubMed] [Google Scholar]

[R11] Barsalou LW (2003a). Abstraction in perceptual symbol systems. Philosophical Transactions of the Royal Society B: Biological Sciences, 355(1435), 1177–1187. 10.1098/rstb.2003.1319 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] Barsalou LW (2003b). Situated simulation in the human conceptual system. Language and Cognitive Processes, 15(5–6), 513–562. 10.1080/01690960344000026 [DOI] [Google Scholar]

[R13] Barsalou LW (2009). Simulation, situated conceptualization, and prediction. Philosophical Transactions of the Royal Society B: Biological Sciences, 364, 1281–1289. 10.1098/rstb.2008.0319 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] Barsalou LW, Dutriaux L, & Scheepers C (2018). Moving beyond the distinction between concrete and abstract concepts. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1752), 20170144. 10.1098/rstb.2017.0144 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] Bauminger N, Schorr Edelsztein H, & Morash J (2005). Social information processing and emotional understanding in children with LD. Journal of Learning Disabilities, 35(1), 45–61. 10.1177/00222194050380010401 [DOI] [PubMed] [Google Scholar]

[R16] Beck L, Kumschick IR, Eid M, & Klann-Delius G (2012). Relationship between language competence and emotional competence in middle childhood. Emotion, 12(3), 503–514. 10.1037/a0026320 [DOI] [PubMed] [Google Scholar]

[R17] Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B, 57(1), 289–300. 10.2307/2346101 [DOI] [Google Scholar]

[R18] Blakemore SJ (2008). The social brain in adolescence. Nature Reviews Neuroscience, 9(4), 267–277. 10.1038/nrn2353 [DOI] [PubMed] [Google Scholar]

[R19] Bloom P (2000). How children learn the meanings of words. Cambridge, MA: MIT Press. [Google Scholar]

[R20] Borghi AM, & Binkofski F (2014). Words as social tools: An embodied view on abstract concepts. New York, NY: Springer. [Google Scholar]

[R21] Borghi AM, Binkofski F, Castelfranchi C, Cimatti F, Scorolli C, & Tummolini L (2017). The challenge of abstract concepts. Psychological Bulletin, 143(3), 263–292. 10.1037/bul0000089 [DOI] [PubMed] [Google Scholar]

[R22] Bretherton I, & Beeghly M (1982). Talking about internal states: The acquisition of an explicit theory of mind. Developmental Psychology, 18(6), 906–921. 10.1037/0012-1649.18.6.906 [DOI] [Google Scholar]

[R23] Bretherton I, Fritz J, Zahn-Waxler C, & Ridgeway D (1986). Learning to talk about emotions: A functionalist perspective. Child Development, 57(3), 529 10.2307/1130334 [DOI] [Google Scholar]

[R24] Bridges KMB (1930). A genetic theory of the emotions. Pedagogical Seminary and Journal of Genetic Psychology, 37, 514–527. [Google Scholar]

[R25] Brooks AW (2014). Get excited: Reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143(3), 1144–1158. 10.1037/a0035325 [DOI] [PubMed] [Google Scholar]

[R26] Brooks JA, & Freeman JB (2018). Conceptual knowledge predicts the representational structure of facial emotion perception. Nature Human Behaviour, 2(8), 581–591. 10.1038/s41562-018-0376-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] Brysbaert M, Warriner AB, & Kuperman V (2014). Concreteness ratings for 40 thousand generally known English word lemmas. Behavior Research Methods, 46(3), 904–911. 10.3758/s13428-013-0403-5 [DOI] [PubMed] [Google Scholar]

[R28] Burgoon EM, Henderson MD, & Markman AB (2013). There are many ways to see the forest for the trees. Perspectives on Psychological Science, 5(5), 501–520. 10.1177/1745691613497964 [DOI] [PubMed] [Google Scholar]

[R29] Burnett S, Bird G, Moll J, Frith CD, & Blakemore S-J (2009). Development during adolescence of the neural processing of social emotion. Journal of Cognitive Neuroscience, 21(9), 1736–1750. 10.1162/jocn.2009.21121 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] Canty A, & Ripley B (2017). Boot: Bootstrap R (S-Plus) functions. R package version 13–20. [Google Scholar]

[R31] Caramelli N, & Setti A (2005). Different domains in abstract concepts. In Bara B, Barsalou L, & Bucciarelli M (Eds.), Proceedings of the XXVII Annual Conference of the Cognitive Science Society (pp. 1997–2002). Mahwah, NJ: Erlbaum. [Google Scholar]

[R32] Caramelli N, Setti A, & Maurizzi DD (2004). Concrete and abstract concepts in school age children. Psychology of Language and Communication, 5(2), 19–34. [Google Scholar]

[R33] Carnaghi A, Maass A, Gresta S, Bianchi M, Cadinu M, & Arcuri L (2008). Nomina sunt omina: On the inductive potential of nouns and adjectives in person perception. Journal of Personality and Social Psychology, 94(5), 839–859. 10.1037/0022-3514.94.5.839 [DOI] [PubMed] [Google Scholar]

[R34] Casey BJ (2015). Beyond simple models of self-control to circuit-based accounts of adolescent behavior. Annual Review of Psychology, 66(1), 295–319. [DOI] [PubMed] [Google Scholar]

[R35] Castro VL, Cheng Y, Halberstadt AG, & Gruhn D (2016). EUReKA! A conceptual model of emotion understanding. Emotion Review, 8(3), 258–268. 10.1177/1754073915580601 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] Cervantes CA, & Callanan MA (1998). Labels and explanations in mother-child emotion talk: Age and gender differentiation. Developmental Psychology, 34(1), 88–98. 10.1037/0012-1649.34.L88 [DOI] [PubMed] [Google Scholar]

[R37] Choudhury S, Blakemore SJ, & Charman T (2006). Social cognitive development during adolescence. Social Cognitive and Affective Neuroscience, 1(3), 165–174. 10.1093/scan/nsl024 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum; 10.4324/9780203771587 [DOI] [Google Scholar]

[R39] Cole PM, Armstrong LM, & Pemberton CK (2010). The role of language in the development of emotion regulation In Calkins SD & Bell MA (Eds.), Human brain development. Child development at the intersection of emotion and cognition (pp. 59–77). Washington, DC: American Psychological Association. [Google Scholar]

[R40] Crone EA, Wendelken C, van Leijenhorst L, Honomichl RD, Christoff K, & Bunge SA (2009). Neurocognitive development of relational reasoning. Developmental Science, 12(1), 55–66. 10.1111/j.1467-7687.2008.00743.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] Davey CG, Yücel M, & Allen NB (2008). The emergence of depression in adolescence: Development of the prefrontal cortex and the representation of reward. Neuroscience and BiobehavioralReviews, 32(1), 1–19. https://doi.org/10.1016Zj.neubiorev.2007.04.016 [DOI] [PubMed] [Google Scholar]

[R42] Davison AC, & Hinkley DV (1997). Bootstrap methods and their applications. Cambridge: Cambridge University Press. [Google Scholar]

[R43] Della Rosa PA, Catricalà E, Vigliocco G, & Cappa SF (2010). Beyond the abstract—concrete dichotomy: Mode of acquisition, concreteness, imageability, familiarity, age of acquisition, context availability, and abstractness norms for a set of 417 Italian words. Behavior Research Methods, 42(4), 1042–1048. 10.3758/BRM.42.4.1042 [DOI] [PubMed] [Google Scholar]

[R44] Demetriou A, Makris N, Spanoudis G, Kazi S, Shayer M, & Kazali E (2018). Mapping the dimensions of general intelligence: An integrated differential-developmental theory. Human Development, 61(1), 4–42. 10.1159/000484450 [DOI] [Google Scholar]

[R45] DiCiccio TJ, & Efron B (1996). Bootstrap confidence intervals. Statistical Science, 11(3), 189–228. 10.1214/ss/1032280214 [DOI] [Google Scholar]

[R46] Doré B, Ort L, Braverman O, & Ochsner KN (2015). Sadness shifts to anxiety over time and distance from the national tragedy in Newtown, Connecticut. Psychological Science, 26(4), 363–373. 10.1177/0956797614562218 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] Doyle CM, & Lindquist KA (2018). When a word is worth a thousand pictures: Language shapes perceptual memory for emotion. Journal of Experimental Psychology: General, 147(1), 62–73. 10.1037/xge0000361 [DOI] [PubMed] [Google Scholar]

[R48] Dumontheil I (2014). Development of abstract thinking during childhood and adolescence: The role of rostrolateral prefrontal cortex. Developmental Cognitive Neuroscience, 10, 57–76. 10.1016/_j.dcn.2014.07.009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] Duncombe M, Havighurst SS, Holland KA, & Frankling EJ (2013). Relations of emotional competence and effortful control to child disruptive behavior problems. Early Education & Development, 24(5), 599–615. 10.1080/10409289.2012.701536 [DOI] [Google Scholar]

[R50] Dunn J, Bretherton I, & Munn P (1987). Conversations about feeling states between mothers and their young children. Developmental Psychology, 23(1), 132–139. 10.1037/0012-1649.23.L132 [DOI] [Google Scholar]

[R51] Dunn LM, & Dunn LM (1997). Peabody Picture Vocabulary Test (3rd ed.). Circle Pines, MN: American Guidance Service. [Google Scholar]

[R52] Ekman P, & Friesen WV (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17(2), 124–129. 10.1037/h0030377 [DOI] [PubMed] [Google Scholar]

[R53] Fabes RA, Eisenberg N, Hanish LD, & Spinrad TL (2001). Preschoolers’ spontaneous emotion vocabulary: Relations to likability. Early Education & Development, 12(1), 11–27. 10.1207/s15566935eed1201_2 [DOI] [Google Scholar]

[R54] Fan R, Varol O, Varamesh A, Barron A, van de Leemput IA, Scheffer M, & Bollen J (2019). The minute-scale dynamics of online emotions reveal the effects of affect labeling. Nature Human Behaviour, 3(1), 92–100. 10.1038/s41562-018-0490-5 [DOI] [PubMed] [Google Scholar]

[R55] Farkas G, & Beron K (2004). The detailed age trajectory of oral vocabulary knowledge: Differences by class and race. Social Science Research, 33(3), 464–497. https://doi.org/10.1016Zj.ssresearch.2003.08.001 [Google Scholar]

[R56] Ferrer E, O’Hare ED, & Bunge SA (2009). Fluid reasoning and the developing brain. Frontiers in Neuroscience, 3(1), 46–51. 10.3389/neuro.01.003.2009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R57] Forgas JP (2007). When sad is better than happy: Negative affect can improve the quality and effectiveness of persuasive messages and social influence strategies. Journal of Experimental Social Psychology, 43(4), 513–528. 10.1016/jjesp.2006.05.006 [DOI] [Google Scholar]

[R58] Franz P, Nook EC, Mair P, & Nock MK (2019). Using topic modeling to detect and describe self-injurious and related content on a large-scale digital platform. Revision under Review. [DOI] [PubMed] [Google Scholar]

[R59] Fugate JMB, Gouzoules H, & Barrett LF (2010). Reading chimpanzee faces: Evidence for the role of verbal labels in categorical perception of emotion. Emotion, 10(4), 544–554. 10.1037/a0019017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] Fujita K, & Carnevale JJ (2012). Transcending temptation through abstraction. Current Directions in Psychological Science, 21(4), 248–252. 10.1177/0963721412449169 [DOI] [Google Scholar]

[R61] Gendron M, Lindquist KA, Barsalou L, & Barrett LF (2012). Emotion words shape emotion percepts. Emotion, 12(2), 314–325. 10.1037/a0026007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] Gopnik A (2001). Theories, language, and culture: Whorf without wincing In Bowerman M & Levinson S (Eds.), Language acquisition and conceptual development (pp. 45–69). Cambridge: Cambridge University Press; 10.1017/CB09780511620669.004 [DOI] [Google Scholar]

[R63] Gray AW, Parkinson B, & Dunbar RI (2015). Laughter’s influence on the intimacy of self disclosure. Human Nature, 26(1), 28–43. 10.1007/s12110-015-9225-8 [DOI] [PubMed] [Google Scholar]

[R64] Greenberg MT, Kusche CA, Cook ET, & Quamma JP (1995). Promoting emotional competence in school-aged children: The effects of the PATHS curriculum. Development and Psychopathology, 7(01), 117 10.1017/S0954579400006374 [DOI] [Google Scholar]

[R65] Hale SC (1988). Spacetime and the abstract / concrete distinction. Philosophical Studies, 53(1), 85–102. [Google Scholar]

[R66] Inhelder B, & Piaget J (1958). The growth of logical thinking from childhood to adolescence: An essay on the construction of formal operational structures. New York, NY: Basic Books. [Google Scholar]

[R67] Jamieson JP, Mendes WB, Blackstock E, & Schmader T (2010). Turning the knots in your stomach into bows: Reappraising arousal improves performance on the GRE. Journal of Experimental Social Psychology, 46(1), 208–212. 10.1016/jjesp.2009.08.015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R68] Joelson JM, & Herrmann DJ (1978). Properties of categories in semantic memory. The American Journal of Psychology, 91(1), 101 10.2307/1421826 [DOI] [Google Scholar]

[R69] Kashdan TB, Barrett LF, & McKnight PE (2015). Unpacking emotion differentiation: Transforming unpleasant experience by perceiving distinctions in negativity. Current Directions in Psychological Science, 24(1), 10–16. 10.1177/0963721414550708 [DOI] [Google Scholar]

[R70] Kelly B, Longbottom J, Potts F, & Williamson J (2004). Applying emotional intelligence: Exploring the Promoting Alternative Thinking Strategies curriculum. Educational Psychology in Practice, 20(3), 221–240. 10.1080/0266736042000251808 [DOI] [Google Scholar]

[R71] Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, & Walters EE (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 593 10.1001/archpsyc.62.6.593 [DOI] [PubMed] [Google Scholar]

[R72] Kober H, Barrett LF, Joseph J, Bliss-Moreau E, Lindquist K, & Wager TD (2008). Functional grouping and cortical-subcortical interactions in emotion: A meta-analysis of neuroimaging studies. NeuroImage, 42(2), 998–1031. https://doi.org/10.1016Zj.neuroimage.2008.03.059 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R73] Kopp CB (1989). Regulation of distress and negative emotions: A developmental view. Developmental Psychology, 25(3), 343–354. 10.1037//0012-1649.25.3.343 [DOI] [Google Scholar]

[R74] Kousta S-T, Vigliocco G, Vinson DP, Andrews M, & Del Campo E (2011). The representation of abstract words: Why emotion matters. Journal of Experimental Psychology: General, 140(1), 14–34. 10.1037/a0021446 [DOI] [PubMed] [Google Scholar]

[R75] Kross E, Bruehlman-Senecal E, Park J, Burson A, Dougherty A, Shablack H, ... Ayduk O (2014). Self-talk as a regulatory mechanism: How you do it matters. Journal of Personality and Social Psychology, 106(2), 304–324. 10.1037/a0035173 [DOI] [PubMed] [Google Scholar]

[R76] Kuhn LJ, Willoughby MT, Vernon-Feagans L, & Blair CB (2016). The contribution of children’s time-specific and longitudinal expressive language skills on developmental trajectories of executive function. Journal of Experimental Child Psychology, 145, 20–34. 10.1016/jjecp.2016.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] Kuperman V, Stadthagen-Gonzalez H, & Brysbaert M (2012). Age-of-acquisition ratings for 30,000 English words. Behavior Research Methods, 44(4), 978–990. 10.3758/s13428-012-0210-4 [DOI] [PubMed] [Google Scholar]

[R78] Kusché CA, Belike RL, & Greenberg MT (1988). Kusché Affective Inventory-Revised. Unpublished measure. Seattle, WA: University of Washington. [Google Scholar]

[R79] Lagattuta KH, & Wellman HM (2002). Differences in early parent-child conversations about negative versus positive emotions: Implications for the development of psychological understanding. Developmental Psychology, 35(4), 564–580. 10.1037/0012-1649.38.4.564 [DOI] [PubMed] [Google Scholar]

[R80] Larson RW, Moneta G, Richards MH, & Wilson S (2002). Continuity, stability, and change in daily emotional experience across adolescence. Child Development, 73(4), 1151–1165. 10.1111/1467-8624.00464 [DOI] [PubMed] [Google Scholar]

[R81] Li Y, & Yu D (2015). Development of emotion word comprehension in Chinese children from 2 to 13 years old: Relationships with valence and empathy. PLOS ONE, 10(12), e0143712. 10.1371/journal.pone.0143712 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R82] Liberman N, & Förster J (2009). The effect of psychological distance on perceptual level of construal. Cognitive Science, 33(7), 1330–1341. 10.1111/j.1551-6709.2009.01061.x [DOI] [PubMed] [Google Scholar]

[R83] Lindquist KA (2017). The role of language in emotion: existing evidence and future directions. Current Opinion in Psychology, 17, 135–139. 10.1016/jxopsyc.2017.07.006 [DOI] [PubMed] [Google Scholar]

[R84] Lindquist KA, Barrett LF, Bliss-Moreau E, & Russell JA (2006). Language and the perception of emotion. Emotion, 6(1), 125–138. 10.1037/1528-3542.6.1.125 [DOI] [PubMed] [Google Scholar]

[R85] Lindquist KA, MacCormack JK, & Shablack H (2015). The role of language in emotion: predictions from psychological constructionism. Frontiers in Psychology, 6(444). 10.3389/fpsyg.2015.00444 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R86] Lindquist KA, Satpute AB, & Gendron M (2015). Does language do more than communicate emotion? Current Directions in Psychological Science, 24(2), 99–108. 10.1177/0963721414553440 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R87] Lindquist KA, Wager TD, Kober H, Bliss-Moreau E, Feldman Barrett L, & Barrett LF (2012). The brain basis of emotion: A meta-analytic review. The Behavioral and Brain Sciences, 35(3), 121–143. 10.1017/S0140525X11000446 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R88] Locke RL, Miller AL, Seifer R, & Heinze JE (2015). Context-inappropriate anger, emotion knowledge deficits, and negative social experiences in preschool. Developmental Psychology, 51(10), 1450–1463. 10.1037/a0039528 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R89] Lupyan G (2012a). Linguistically modulated perception and cognition: The label-feedback hypothesis. Frontiers in Psychology, 3 10.3389/fpsyg.2012.00054 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R90] Lupyan G (2012b). What do words do? Toward a theory of language-augmented thought In Ross BH (Ed.), Psychology of Learning and Motivation (Vol. 57, pp. 255–297). Waltham, MA: Academic Press; 10.1016/B978-0-12-394293-7.00007-8 [DOI] [Google Scholar]

[R91] Lupyan G, & Winter B (2018). Language is more abstract than you think, or, why aren’t languages more iconic? Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1752). 10.1098/rstb.2017.0137 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R92] Luria A (1961). The role of speech in the regulation of normal and abnormal behaviors. New York, NY: Liveright. [Google Scholar]

[R93] MacCann C, & Roberts RD (2008). New paradigms for assessing emotional intelligence: Theory and data. Emotion, 5(4), 540–551. 10.1037/a0012746 [DOI] [PubMed] [Google Scholar]

[R94] MacWhinney B, & Snow C (1990). The Child Language Data Exchange System: an update. Journal of Child Language, 17(2), 457–472. 10.1017/S0305000900013866 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R95] Mahon BZ, & Caramazza A (2008). A critical look at the embodied cognition hypothesis and a new proposal for grounding conceptual content. Journal of Physiology-Paris, 102(1–3), 59–70. 10.1016/jjphysparis.2008.03.004 [DOI] [PubMed] [Google Scholar]

[R96] Matthews D, Biney H, & Abbot-Smith K (2018). Individual differences in children’s pragmatic ability: A review of associations with formal language, social cognition, and executive functions. Language Learning and Development, 14(3), 186–223. 10.1080/15475441.2018.1455584 [DOI] [Google Scholar]

[R97] Mazzuca C, Barca L, & Borghi AM (2017). The particularity of emotional words: A grounded approach. Rivista Internazionale Di Filosofia e Psicologia, 8(2), 124–133. 10.4453/rifp.2017.0010 [DOI] [Google Scholar]

[R98] Mervis C, & Rosch E (1981). Categorization of natural objects. Annual Review of Psychology, 32(1), 89–115. 10.1146/annurev.ps.32.020181.000513 [DOI] [Google Scholar]

[R99] Moser JS, Dougherty A, Mattson WI, Katz B, Moran TP, Guevarra D, ... Kross E (2017). Third-person self-talk facilitates emotion regulation without engaging cognitive control: Converging evidence from ERP and fMRI. Scientific Reports, 7(1), 4519 10.1038/s41598-017-04047-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R100] Nook EC, Lindquist KA, & Zaki J (2015). A new look at emotion perception: Concepts speed and shape facial emotion recognition. Emotion, 15(5), 569–578. 10.1037/a0039166 [DOI] [PubMed] [Google Scholar]

[R101] Nook EC, Sasse SF, Lambert HK, McLaughlin KA, & Somerville LH (2017). Increasing verbal knowledge mediates development of multidimensional emotion representations. Nature Human Behaviour, 1(12), 881–889. 10.1038/s41562-017-0238-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R102] Nook EC, Sasse SF, Lambert HK, McLaughlin KA, & Somerville LH (2018). The nonlinear development of emotion differentiation: Granular emotional experience Is low in adolescence. Psychological Science, 29(8), 1346–1357. 10.1177/0956797618773357 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R103] Nook EC, Schleider JL, & Somerville LH (2017). A linguistic signature of psychological distancing in emotion regulation. Journal of Experimental Psychology: General, 146(3), 337–346. 10.1037/xge0000263 [DOI] [PubMed] [Google Scholar]

[R104] O’Kearney R, & Dadds M (2004). Developmental and gender differences in the language for emotions across the adolescent years. Cognition & Emotion, 18(7), 913–938. 10.1080/02699930341000356 [DOI] [Google Scholar]

[R105] O’Kearney R, & Dadds MR (2005). Language for emotions in adolescents with externalizing and internalizing disorders. Development and Psychopathology, 17(2), 529–548. 10.1017/S095457940505025X [DOI] [PubMed] [Google Scholar]

[R106] Oosterwijk S, Lindquist KA, Adebayo M, & Barrett LF (2015). The neural representation of typical and atypical experiences of negative images: Comparing fear, disgust and morbid fascination. Social Cognitive and Affective Neuroscience, (July), nsv088. 10.1093/scan/nsv088 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R107] Paivio A, Yuille JC, & Madigan SA (1968). Concreteness, imagery, and meaningfulness values for 925 nouns. Journal of Experimental Psychology, 76(1), 1–25. 10.1037/h0025327 [DOI] [PubMed] [Google Scholar]

[R108] Pennebaker JW, Booth RJ, & Francis ME (2007). Linguistic Inquiry and Word Count: LIWC [Computer software]. Austin, TX: LIWC.net. [Google Scholar]

[R109] Ponari M, Norbury CF, & Vigliocco G (2018). Acquisition of abstract concepts is influenced by emotional valence. Developmental Science, 21(2), e12549. 10.1111/desc.12549 [DOI] [PubMed] [Google Scholar]

[R110] Pons F, Harris PL, & de Rosnay M (2004). Emotion comprehension between 3 and 11 years: Developmental periods and hierarchical organization. European Journal of Developmental Psychology, 1(2), 127–152. 10.1080/17405620344000022 [DOI] [Google Scholar]

[R111] R Core Team. (2018). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; Retrieved from https://www.r-project.org/ [Google Scholar]

[R112] Raven J (2000). The Raven’s Progressive Matrices: Change and stability over culture and time. Cognitive Psychology, 41(1), 1–48. 10.1006/cogp.1999.0735 [DOI] [PubMed] [Google Scholar]

[R113] Reichenbach L, & Masters JC (1983). Children’s use of expressive and contextual cues in judgments of emotion. Child Development, 54(4), 993 10.2307/1129903 [DOI] [Google Scholar]

[R114] Roben CKP, Cole PM, & Armstrong LM (2013). Longitudinal relations among language skills, anger expression, and regulatory strategies in early childhood. Child Development, 84(3), 891–905. 10.1111/cdev.12027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R115] Rodman AM, Powers KE, & Somerville LH (2017). Development of self-protective biases in response to social evaluative feedback. Proceedings of the National Academy of Sciences, 114(50), 13158–13163. 10.1073/pnas.1712398114 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R116] Rosch E, Mervis CB, Gray WD, Johnson DM, & Boyes-Braem P (1976). Basic objects in natural categories. Cognitive Psychology, 8(3), 382–439. 10.1016/0010-0285(76)90013-X [DOI] [Google Scholar]

[R117] Russell JA (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. [Google Scholar]

[R118] Russell JA, & Widen SC (2002). Words versus faces in evoking preschool children’s knowledge of the causes of emotions. International Journal of Behavioral Development, 26(2), 97–103. 10.1080/01650250042000582 [DOI] [Google Scholar]

[R119] Saarni C (1999). The development of emotional competence. New York, NY: Guilford Press. [Google Scholar]

[R120] Salmon K, O’Kearney R, Reese E, & Fortune C-A (2016). The role of language skill in child psychopathology: Implications for intervention in the early years. Clinical Child and Family Psychology Review, 19(4), 352–367. 10.1007/s10567-016-0214-1 [DOI] [PubMed] [Google Scholar]

[R121] Satpute AB, Nook EC, Narayanan S, Shu J, Weber J, & Ochsner KN (2016). Emotions in “black and white” or shades of gray? How we think about emotion shapes our perception and neural representation of emotion. Psychological Science, 27(11), 1428–1442. 10.1177/0956797616661555 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R122] Schmid H (1994). Probabilistic part-of-speech tagging using decision trees. In Jones DB & Somers HL (Eds.), Proceedings of the International Conference on New Methods in Language Processing (pp. 44–49). Manchester, England: University of Manchester Institute of Science and Technology. [Google Scholar]

[R123] Sebastian CL, Fontaine NMG, Bird G, Blakemore S-J, De Brito SA, McCrory EJP, & Viding E (2012). Neural processing associated with cognitive and affective Theory of Mind in adolescents and adults. Social Cognitive and Affective Neuroscience, 7(1), 53–63. 10.1093/scan/nsr023 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R124] Seih Y-T, Beier S, & Pennebaker JW (2017). Development and examination of the Linguistic Category Model in a computerized text analysis method. Journal of Language and Social Psychology, 36(3), 343–355. 10.1177/0261927X16657855 [DOI] [Google Scholar]

[R125] Semin GR, & Fiedler K (1988). The cognitive functions of linguistic categories in describing persons: Social cognition and language. Journal of Personality and Social Psychology, 54(4), 558–568. 10.1037/0022-3514.54A558 [DOI] [Google Scholar]

[R126] Semin GR, & Fiedler K (1991). The Linguistic Category Model, its bases, applications and range. European Review of Social Psychology, 2(1), 1–30. 10.1080/14792779143000006 [DOI] [Google Scholar]

[R127] Semin GR, Görts CA, Nandram S, & Semin-Goossens A (2002). Cultural perspectives on the linguistic representation of emotion and emotion events. Cognition & Emotion, 16(1), 11–28. 10.1080/02699930143000112 [DOI] [Google Scholar]

[R128] Shablack H, & Lindquist KA (2019). The role of language in emotional development. In LoBue V, Perez-Edgar K, & Buss K (Eds.), Handbook of Emotion Development. New York, NY: Springer. [Google Scholar]

[R129] Siegel LS, & Ryan EB (1989). The development of working memory in normally achieving and subtypes of learning disabled children. Child Development, 60(4), 973 10.2307/1131037 [DOI] [PubMed] [Google Scholar]

[R130] Simpson G (2014). First derivative of a spline term. GitHub Repository. Retrieved from https://gist.github.com/gavinsimpson/e73f011fdaaab4bb5a30 [Google Scholar]

[R131] Smith M, & Walden T (2001). Understanding feelings and coping with emotional situations: A comparison of maltreated and nonmaltreated preschoolers. Social Development, 5(1), 93–116. 10.1111/1467-9507.00082 [DOI] [Google Scholar]

[R132] Soderberg CK, Callahan SP, Kochersberger AO, Amit E, & Ledgerwood A (2015). The effects of psychological distance on abstraction: Two meta-analyses. Psychological Bulletin, 141(3), 525–548. 10.1037/bul0000005 [DOI] [PubMed] [Google Scholar]

[R133] Somerville LH, Jones RM, Ruberry EJ, Dyke JP, Glover G, & Casey BJ (2013). The medial prefrontal cortex and the emergence of self-conscious emotion in adolescence. Psychological Science, 24(8), 1554–1562. 10.1177/0956797613475633 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R134] Somerville LH, & McLaughlin KA (2018). What develops during emotional development? Normative trajectories and sources of psychopathology risk in adolescence In Fox Andrew S., Lapate Regina C., Shackman Alexander J., & Davidson Richard J. (Eds.), The Nature of Emotion: Fundamental Questions (2nd ed.). [Google Scholar]

[R135] Southam-Gerow MA, & Kendall PC (2000). A preliminary study of the emotion understanding of youths referred for treatment of anxiety disorders. Journal of Clinical Child Psychology, 29(3), 319–327. 10.1207/S15374424JCCP2903_3 [DOI] [PubMed] [Google Scholar]

[R136] Stadthagen-Gonzalez H, & Davis CJ (2006). The Bristol norms for age of acquisition, imageability, and familiarity. Behavior Research Methods, 38(4), 598–605. 10.3758/BF03193891 [DOI] [PubMed] [Google Scholar]

[R137] Torre JB, & Lieberman MD (2018). Putting feelings into words: Affect labeling as implicit emotion regulation. Emotion Review, 10(2), 116–124. 10.1177/1754073917742706 [DOI] [Google Scholar]

[R138] Trentacosta CJ, & Izard CE (2007). Kindergarten children’s emotion competence as a predictor of their academic competence in first grade. Emotion, 7(1), 77–88. 10.1037/1528-3542.7.1.77 [DOI] [PubMed] [Google Scholar]

[R139] Trope Y, & Liberman N (2010). Construal-level theory of psychological distance. Psychological Review, 117(2), 440–463. 10.1037/a0018963 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R140] Vallotton C, & Ayoub C (2011). Use your words: The role of language in the development of toddlers’ self-regulation. Early Childhood Research Quarterly, 26(2), 169–181. 10.1016/_j.ecresq.2010.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R141] Vigliocco G, Kousta S-T, Della Rosa PA, Vinson DP, Tettamanti M, Devlin JT, & Cappa SF (2014). The neural representation of abstract words: The role of emotion. Cerebral Cortex, 24(7), 1767–1777. 10.1093/cercor/bht025 [DOI] [PubMed] [Google Scholar]

[R142] Vygotsky LS (1962). Thought and language. Cambridge, MA: MIT Press. [Google Scholar]

[R143] Waggoner JE, & Palermo DS (1989). Betty is a bouncing bubble: Children’s comprehension of emotion-descriptive metaphors. Developmental Psychology, 25(1), 152–163. 10.1037/0012-1649.25.L152 [DOI] [Google Scholar]

[R144] Wang J, Conder JA, Blitzer DN, & Shinkareva SV (2010). Neural representation of abstract and concrete concepts: A meta-analysis of neuroimaging studies. Human Brain Mapping, 31(10), 1459–1468. 10.1002/hbm.20950 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R145] Wechsler D (1999). Wechsler Abbreviated Scale of Intelligence (WASI). San Antonio, TX: Psychological Corporation. [Google Scholar]

[R146] Wechsler D (2011). Wechsler Abbreviated Scale of Intelligence-Second Edition Manual. Bloomington, MN: Pearson. [Google Scholar]

[R147] Whitaker KJ, Vendetti MS, Wendelken C, & Bunge SA (2018). Neuroscientific insights into the development of analogical reasoning. Developmental Science, 21(2), e12531. 10.1111/desc.12531 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R148] Widen SC (2013). Children’s interpretation of facial expressions: The long path from valence-based to specific discrete categories. Emotion Review, 5(1), 72–77. 10.1177/1754073912451492 [DOI] [Google Scholar]

[R149] Widen SC, Pochedly JT, & Russell JA (2015). The development of emotion concepts: A story superiority effect in older children and adolescents. Journal of Experimental Child Psychology, 131, 186–192. 10.1016/jjecp.2014.10.009 [DOI] [PubMed] [Google Scholar]

[R150] Widen SC, & Russell JA (2010a). Children’s scripts for social emotions: Causes and consequences are more central than are facial expressions. British Journal of Developmental Psychology, 28(3), 565–581. 10.1348/026151009X457550d [DOI] [PubMed] [Google Scholar]

[R151] Widen SC, & Russell JA (2010b). Differentiation in preschooler’s categories of emotion. Emotion, 10(5), 651–661. 10.1037/a0019005 [DOI] [PubMed] [Google Scholar]

[R152] Wilson-Mendenhall CD (2017). Constructing emotion through simulation. Current Opinion in Psychology, 17, 189–194. https://doi.org/10.10167j.copsyc.2017.07.015 [DOI] [PubMed] [Google Scholar]

[R153] Wilson-Mendenhall CD, Barrett LF, & Barsalou LW (2013). Situating emotional experience. Frontiers in Human Neuroscience, 7, 764 10.3389/fnhum.2013.00764 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R154] Wilson-Mendenhall CD, Barrett LF, Simmons WK, & Barsalou LW (2011). Grounding emotion in situated conceptualization. Neuropsychologia, 49(5), 1105–1127. 10.1016/j.neuropsychologia.2010.12.032 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R155] Wood SN (2003). Thin plate regression splines. Journal of the Royal Statistical Society B, 65(1), 95–114. [Google Scholar]

[R156] Wood SN (2017). Generalized additive models: An introduction with R (2nd ed.). CRC Press. [Google Scholar]

[R157] Wu Y, Muentener P, & Schulz LE (2017). One-to four-year-olds connect diverse positive emotional vocalizations to their probable causes. Proceedings of the National Academy of Sciences, 114(45), 11896–11901. 10.1073/pnas.1707715114 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R158] Yin J, & Csibra G (2015). Concept-based word learning in human infants. Psychological Science, 26(8), 1316–1324. 10.1177/0956797615588753 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R159] Zeman J, Klimes-Dougan B, Cassano M, & Adrian M (2007). Measurement issues in emotion research with children and adolescents. Clinical Psychology: Science and Practice, 14(4), 377–401. 10.1111/j.1468-2850.2007.00098.x [DOI] [Google Scholar]

PERMALINK

Charting the development of emotion comprehension and abstraction from childhood to adulthood using observer-rated and linguistic measures

Erik C Nook

Caitlin M Stavish

Stephanie F Sasse

Hilary K Lambert

Patrick Mair

Katie A McLaughlin

Leah H Somerville

Abstract

Measuring the Development of Emotion Word Comprehension

Developing Abstract Emotion Concepts

An Automated Linguistic Measure of Abstractness

The Current Study

Methods

Participants

Emotion Vocabulary Assessment

Human Coder Assessment of Emotion Abstractness

Linguistic Assessment of Emotion Abstractness

Hypotheses and Analyses

Interrelations between dependent variables.

Overall developmental changes in emotion comprehension and abstraction.

Figure 1.

Development of emotion comprehension and abstraction for each emotion.

Control analyses.

Results

Interrelations Between Dependent Measures

Table 1.

Age-related Change in Emotion Comprehension and Abstraction

Table 2.

Emotion comprehension scores.

Coder abstractness scores.

Linguistic abstractness scores.

Figure 2.

General definition use.

Figure 3.

Synonym use.

Example situation use.

Physiological marker use.

Controlling for word count.

Control analyses investigating trials on which participants only provided synonyms.

Age-related Change for Each Emotion

Figure 4.

Discussion

An Interview-Based Method for Measuring Emotion Comprehension

Emotion Abstraction: Implications for Affective, Developmental, and Cognitive Theory

Toward an Automated Linguistic Measure of Abstraction

Limitations and Future Directions

Conclusion

Supplementary Material

Acknowledgments

Footnotes

REFERENCES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases