Times are displayed in (UTC-05:00) Central Time (US & Canada) Change
About this paper symposium
| Panel information |
|---|
| Panel 20. Social Cognition |
| Paper #1 | |
|---|---|
| Emotion Word Production Reveals Key Dimensions Organizing Emotion Concepts in Youth and Young Adults | |
| Author information | Role |
| Chantal Alejandra Valdivia-Moreno, Princeton University, United States | Presenting author |
| Stephanie F. Sasse, Harvard University, United States | Non-presenting author |
| Hilary K. Lambert, McLean Hospital/Harvard Medical School, United States | Non-presenting author |
| Katie A. McLaughlin, Harvard University, United States | Non-presenting author |
| Leah H. Somerville, Harvard University, United States | Non-presenting author |
| Erik C. Nook, Princeton University, United States | Non-presenting author |
| Abstract | |
|
Affective scientists have long sought an understanding of how people organize emotions at the conceptual level. One leading framework, called the circumplex model, proposes that two key dimensions organize emotions: valence and arousal (Russell, 1980). Emotion fluency tasks, wherein participants generate as many emotion words as possible in 60 seconds, offer a unique opportunity to examine the underlying semantic structures that organize emotion words. Sterile as this task may seem, the process of rapidly accessing and generating emotion words mirrors the common practice of asking children to “name their emotions.” These tasks thus allow us to examine the features that shape which words most readily come to mind, and how these dimensions might vary across age. Here we provide two novel analyses of emotion fluency data. First, we deploy network analyses (Zemla & Austerweil, 2020) to visualize clusters of emotion words that reliably appear in close proximity to each other. Second, we use regression methods to test several lexical parameters (i.e., valence, word length, age of acquisition, etc.) as predictors of the order in which words are produced. We hypothesized that dimensions of the circumplex model would reliably structure word order and that youth may be especially attentive to valence (Nook et al., 2017). A cross-sectional sample of participants aged 4-25 years (N = 194) generated as many fruit words as possible in 60 seconds (to measure verbal fluency) and as many emotion words as possible in 60 seconds (to measure emotion fluency). Community Network analyses of all participants’ data are presented in the left panel of Figure 1. This sample-level network revealed 14 clusters that provide insight into semantic associations between emotion words. Early acquired “basic emotion” words clustered together (anger/fear/disgust/sadness/happiness/mad), with a bridge to more complex emotion words (upset/hurt/exhausted and grumpy/boredom). Conducting Community Network analyses within age groups reveals that the overall number of words, edges, and clusters grows across age, reflecting increases in emotion word comprehension and emotion fluency. Critically, however, a happiness/sadness cluster existed even in the youngest participants, and by adolescence, this cluster grew to include anger. We also used a mixed-effects model to test which lexical parameters were associated with word order when controlling for all other parameters (Figure 2). This analysis revealed that age of acquisition, valence, dominance, concreteness, and word length (listed in order of strength) were significant predictors of word order. Age did not moderate any associations between word order and lexical parameters (ps > .463), suggesting that the strength of each of these parameters in predicting word order does not significantly vary across age. Together, these results reveal that regardless of age, when people are prompted to generate emotion words, age of acquisition, and thus, more ‘basic’ emotion concepts, emerge as the strongest predictors of word order and semantic structure. Interestingly, these early-acquired words are first to come to mind, overpowering classic circumplex dimensions of valence and arousal. These results thus shed new light on the factors that shape emotion word accessibility across age. |
|
| Paper #2 | |
|---|---|
| Five- Through 10-Year-Olds Use Selective Attention to Learn About Emotions | |
| Author information | Role |
| Ms. Andrea Gray Stein, University of Wisconsin-Madison, United States | Presenting author |
| Seth D. Pollak, University of Wisconsin-Madison, United States | Non-presenting author |
| Abstract | |
|
To figure out how someone is feeling, we can use a multitude of signals, from what the individual is doing with their face to our knowledge of the surrounding context. One open question is how our use of these cues to learn and reason about emotions may change across development. We investigated developmental differences in (1) the primacy of facial vs. contextual cues and (2) the use of selective vs. diffused attention strategies when learning about emotions. This is the first study we know of to investigate these two phenomena simultaneously and in the context of new emotion learning. In our learning task, five- through ten-year-olds classified the feelings of an alien (“Wiggle”) into one of two novel categories (feeling “binty” or feeling “daxy”). Stimuli consisted of combinations of two facial and three contextual cues: Wiggle’s eyes, Wiggle’s mouth, the social context, the setting, and the activity (see Figure 1). Categories were structured such that one cue was perfectly diagnostic of category membership, while the others were moderately predictive. Accordingly, participants could learn to classify the stimuli either via a diffused attention strategy (attending to all cues), or a selective attention strategy (attending to the perfectly diagnostic cue). Which cue was perfectly diagnostic, as well as which group of stimuli corresponded to which novel emotion, was randomized and counterbalanced within age bins. The final sample to be presented at the meeting will include 180 children. To date, 160 children completed 64 learning trials in which they classified Wiggle’s emotion and received corrective feedback. They then completed 8 ambiguous trials, consisting of stimuli in which the perfectly diagnostic cue matched one category (consistent with a selective attention strategy), but the majority of cues matched the other (consistent with a diffused attention strategy). Preliminary results are as follows. Overall, participants demonstrated learning, averaging 79% accuracy on the final training block (vs. chance: t(159) = 15.948, p < .001). A beta regression model regressing accuracy in the final training block on participant age, perfectly diagnostic cue, and their interaction suggests developmental stability in the relative primacy of facial vs. contextual cues (b = -.041, z = -.385, p = .70; see Figure 2). Similarly, a beta regression model predicting strategy use on the ambiguous trials suggests there are not significant developmental differences in children’s use of diffused versus selective attention strategies (b = .093, z = 1.547, p = .12). Children across the age range studied tended to use selective attention; the average participant was predicted to use a selective attention strategy on 72% of trials. These preliminary results suggest that by age five, we may form relatively stable expectations regarding what cues are informative about others’ emotions. Similarly, our tendency to deploy selective attention when learning and reasoning about others’ emotions may emerge early in development and remain relatively stable into late childhood. We will discuss how these results fit into the bigger developmental picture when it comes to both the development of attention and the development of emotion reasoning. |
|
| Paper #3 | |
|---|---|
| Four- to 10-Year-Olds’ Self-Reports of Their Worry: Comparison to Parent Perspectives | |
| Author information | Role |
| Maritza Miramontes, University of California, Davis, United States | Presenting author |
| Hannah J. Kramer, University of Wisconsin-Madison, United States | Non-presenting author |
| Karen H. Lara, Southwestern University, United States | Non-presenting author |
| Kristin Hansen Lagattuta, University of California, Davis, United States | Non-presenting author |
| Abstract | |
|
Past research has relied on caregiver questionnaires to assess young children’s typical emotions (Bilancia & Rescorla, 2010), with child self-reported perspectives not incorporated until children reach about 8 to 10 years of age (Silverman & Ollendick, 2005). Several factors motivate this exclusion of young children’s self-views, including the underlying assumption that evaluating children’s emotions should be easy for parents. The validity of parent reports of children’s emotions has rarely been questioned, outside from concerns that parents with clinical diagnoses may provide biased reports (Affrunti & Woodruff-Borden, 2015). A further guiding assumption, influenced by the legacy of Piaget, is that pre-operational children are too illogical to sensibly and reliably answer such reflective questions. As an exception, Lagattuta et al. (2012) found that 4- to 10-year-old children can reliably report their worry, that there was no agreement between mother and child perspectives, and that mothers perceived their children to worry less frequently than what their children self-reported (suggesting a parental positivity bias). Here, we aimed to replicate their findings six years later with a new, non-clinical sample. More specifically, the current work addressed: 1)Can 4- to 10-year-olds reliably self-report their worries? 2)Is there agreement between child and mother perspectives? 3)Is there a directional difference between how children and mothers evaluate children’s emotions? These questions are of theoretical and empirical importance as researchers typically exclude child self-reported emotions in children under 8 years of age and parents are essential gatekeepers for child mental health services. Four- to 10-year-olds (N=149) reported the frequency that they experienced emotions and thoughts reflective of worry (e.g., “I think about bad or scary things”) using a 5-point pictorial scale ranging from 0 (none of the time) to 4 (all of the time; training and scale from Lagattuta et al., 2012). Mothers predicted how their child would respond to the same items (e.g., “My child thinks about bad or scary things”). All reports were scored for average rating. Results showed high internal consistency in children’s and mothers’ responses, across three child age groups (.81<α<.87; Table 1). Overall, there was a non-significant bivariate (r=.11, p=.184) and partial correlation (controlling for child age; r=.12, p=.149) between how mothers and children perceived children’s worry. When separated into child age groups, parents and their 8- to 10-year-olds shared moderate agreement in perspectives of the child’s worry (r=.55, p<.001); however, mothers and their 4- to 7-year-olds held unrelated views (ps>.359). A 3 (Age: 4/5 years, 6/7 years, 8/10 years) x 2 (Reporter: child, parent) ANOVA on worry scores indicated a near-significant main effect of Reporter, F=3.82, p=.053, ηp2=.025 (Figure 1). In sum, findings partially replicated Lagattuta et al. (2012), with signs of increasing parental awareness that children experience worry. We confirmed high internal consistency in 4- to 10-year-olds’ responses, moderate parent-child agreement about child worry at 8 to 10 years (versus no agreement at any age group), and a trending versus significant parental positivity bias. These data support the inclusion of young children’s self-reported emotions in developmental research and clinical practice. |
|
| Paper #4 | |
|---|---|
| Finding Happiness: How Goals Shape Children's Perceptions Beyond Simple Preferences | |
| Author information | Role |
| Lingyan Hu, M.Ed., University of Pennsylvania, United States | Presenting author |
| Douglas A. Frye, University of Pennsylvania, United States | Non-presenting author |
| Fan Yang, The University of Chicago, United States | Non-presenting author |
| Abstract | |
|
Philosophers and psychologists have long wrestled with why it can be so hard to achieve happiness. It seems simple enough: just pursue what we enjoy and avoid what we do not. Yet, the reality is far more complex. While we have simple preferences, we also set goals that often require sacrificing immediate pleasures for more meaningful rewards. Even children may frequently encounter conflicts between goals and preferences, such as choosing between watching cartoons and pursuing important learning goals. How do children perceive the role of goals in shaping our sense of happiness, beyond merely fulfilling our likes and dislikes? We conducted a series of studies to explore this question. In Study 1, 4-5-year-old (N = 30) and 7-8-year-old (N = 30) children listened to stories about individuals whose simple preferences (like or dislike doing something) conflicted or aligned with more important goals. A linear mixed effects model revealed an interaction between Preference Valence (like or dislike) and Conflict Levels (Simple-Preference or Conflict) (F(1, 232) = 80.475, p < .001), Figure 1. Children attributed greater happiness to characters who performed a disliked activity (e.g., doing homework) to achieve an important goal (e.g., getting high testing scores), compared to similar actions that did not conflict with a goal (t(174) = 5.351, p < .001). Conversely, children attributed less happiness to characters who did something they liked (e.g., watching TV) at the expense of a goal (e.g., getting lower scores), compared to similar actions not conflicting with a goal (t(174) = -7.33, p < .001), Figure 1. This pattern was consistent across ages. The findings of Study 1 raised the question of whether children’s attributions of happiness were also sensitive to conflicts between two simple preferences, not just conflicts between preference and important goals. Study 2 introduced a new condition where characters had to choose between two competing preferences. Conceptually replicating and extending results from Study 1, across age groups (N = 126), children attributed highest happiness to characters who fulfill simple preference (M = 3.64), lower happiness to characters fulfilling simple preference that conflicts with another preference (M = 2.98), and lowest happiness to those fulfilling simple preference that conflicts with goals (M = 2.15), Figure 2. Moreover, Study 2 asked children to choose between a gift offering long-term benefits versus one providing immediate satisfaction. The less happiness a child attributed to the preference-goal conflict story, the more likely they were to choose a gift with long-term benefits, β = -0.08, p = .03. Study 3 (ongoing, target N = 120) examines how children view different beliefs about happiness, such as happiness resulting from fulfilling preferences versus achieving goals. Children predict each character’s future decisions and achievements based on the character’s perspective on happiness and express their own preferences for each character and their corresponding beliefs. Our studies suggest that children recognize that achieving important goals influences happiness beyond the satisfaction of immediate desires. This recognition also predicted children’s own choice between long-term benefits over immediate satisfaction. These findings illuminate the nature of happiness in the eyes of children, with significant implications for promoting their motivation and achievement. |
|
⇦ Back to session
Emotion-Wise: New Insights into the Multifaceted Nature of How Children Conceptualize, Learn, and Perceive Emotions
Submission Type
Paper Symposium
Description
| Session Title | Emotion-Wise: New Insights into the Multifaceted Nature of How Children Conceptualize, Learn, and Perceive Emotions |