Flash talks: Emotional responding

July 12th Room 4

Charting the Full Appraisal-Emotion Space with Parallel Process Latent Profile Analysis 

Weiqiang Qian, Craig A. Smith & Leslie D. Kirby

The associations of appraisals to the experience of individual emotions have long been studied. However, individual emotions are almost never experienced in isolation, and there are few data examining the patterns of appraisals associated with the blends of emotion typically experienced in real life. Using latent profile analysis on an aggregated dataset (N=3761) containing appraisal and emotion ratings across a wide range of emotion-eliciting situations, we extracted five appraisal profiles and five emotion profiles. These profiles represent the major patterns of appraisal and emotion, respectively, that were elicited across the sampled situations. The relations between these appraisal and emotion profiles were examined by combining the two sets of latent profile analysis with a parallel process model. A chord diagram was produced to visualize the links between the appraisal and emotion profiles, and to facilitate interpretation. The appraisal and emotion profiles were found to be systematically related to one another in ways that can be meaningfully explained by appraisal theory. For example, an appraisal profile emphasizing other-accountability was strongly linked to an emotion profile emphasizing other-directed negative emotions, and an appraisal profile emphasizing motivational congruence was associated with an emotion profile emphasizing positive emotions. Our analyses depict the ecologically valid relations among combinations of appraisals and emotions typically experienced during emotional encounters.

The Effect of Relevance Appraisal on the Emotional Response 

Larisa Olteanu & Assaf Kron

Appraisal theories assume that an emotional response is preceded by an evaluation phase that assesses the stimulus’ relevance to the perceiver’s well-being. Although the effect of relevance on the emotional response is supported by empirical findings, it has been proven difficult to control and manipulate. The current study’s primary goal is to experimentally manipulate the relevance of stimuli while ensuring that its physical properties remain unchanged. To this aim, we used stimuli with high ecological validity: participants (N=40, 26 females) were about to learn their Israeli Psychometric Entrance Test (PET) score (the equivalent of USA's SAT/ACT score - highly important for acceptance to university programs) for the first time during the experimental session. Emotional response was measured using facial electromyography (EMG), heart rate (HR), electrodermal activity (EDA), and reported experiences (self-reports). We found a substantial effect for manipulated relevance on self-reports, and HR, but not for EMG. The results provide evidence that information about a stimulus’ relevance modulates the emotional response to it.

Shared Environment Effects on Children's Emotion Recognition 

Rotem Schapira, Hillary Anger Elfenbein, Meirav Amichay-Setter, Carolyn Zahn-Waxler & Ariel Knafo-Noam

Emotion recognition (ER) is the ability to understand and recognize others’ emotions using cues such as facial expressions and tone of voice. The ability to recognize emotions is crucial for social interaction. Little is known about how this ability develops throughout childhood, and specifically the background for individual differences. The present study expanded the field by examined genetic and environmental influences on children’s ER via facial and vocal cues in 344 7-year- old (90.05 ± 3.87 months) twin children (59 MZ pairs and 113 same-sex DZ pairs), who were part of the Longitudinal Israeli Study of Twins. ER was assessed with the child version of the Diagnostic Assessment of Nonverbal Accuracy (DANVA2, Nowicki, 2010). Twin correlations were not higher for MZ twins than for DZ twins, indicating no heritability for ER in this population, for either facial and vocal cues of emotion. In contrast, correlations were positive for both types of twins and somewhat higher for DZ twins, indicating a shared environmental effect, which has been supported by a bivariate genetic analysis. The models showed no genetic (A) effect but did show shared-environment (C) and non-shared environment (E) effects on ER. This pattern was robust to controlling for twins being of the same sex and age. In addition, a bivariate genetic analysis found a shared environmental correlation between facial and vocal ER (rc=.63), indicating that the shared environmental factors contributing to vocal and facial ER overlap. The study highlights the importance of the shared environment to children’s ER.

Time to take enthusiasm seriously! 

Rijn Vogelaar, Wilco van Dijk & Eric van Dijk

Enthusiasm is a commonly used term in daily language as well as in professional settings. It plays an important role in people’s personal lives and organisations want to recruit enthusiastic employees and create enthusiasm amongst their customers. Scientific research on enthusiasm, however, is scarce and the concept is not clearly defined. The few studies that can be found on enthusiasm show, for example, that it enhances the performance of teachers, call center agents, and instructors. The current research aims at providing a richer and more specific understanding of enthusiasm and its core characteristics. To gain more insight in the phenomenon of enthusiasm, we conduct a prototype analysis, using five empirical studies that include broad samples of respondents. In Study 1, we generate the different features of enthusiasm. In Study 2, we quantify the centrality of the different features. In Study 3, we validate the prototype structure of enthusiasm by using a recall test. In Study 4, we validate the prototype structure by measuring response accuracy and speed. In Study 5, we test the ecological validity of the prototype structure by using autobiographical recall. Results of these studies will be presented and discussed in relation to the existing literature on enthusiasm. Data collection will be completed in the spring of 2019.

The role of audio-visual features in emotional and empathic responses to movies 

Katalin Balint, Yunting Huang, Mattia Savardi & Sergio Benini

Fiction films are powerful and widely used tools to elicit emotional responses in viewers. Yet, it is still unclear how the design components of movies, that is the way how the audio-visual moving image is constructed, influence viewers’ emotional responses. The present study is an interdisciplinary collaboration among scholars in media psychology and computer science to identify those formal features in audiovisual fictional narratives that have an impact on viewers’ emotional responses. In this study, fifteen film scenes from three genres were selected and analyzed for audio-visual formal features, including metrics of shot scale, color, lighting, motion dynamics, and music, assumed to be important to the emotional identity of a film scene. For the audiovisual analysis we employed computer algorithms to examine these features on a frame-by-frame basis. With the same set of film 15 film scenes a 3 (genres; action vs drama vs romance) x 5 (film) within-subject experiment was conducted (N = 75). Arousal and valence responses were continuously recorded by FaceReader while participants were watching the 15 film scenes. After each scene, participants filled in a short state empathy scale. Data is analysed (in progress) by cross-correlational analyses to test whether the time-series pattern of emotional valence and arousal in viewers is associated with the pattern of audio-visual formal features. Furthermore, in a regression analysis we investigated if the formal features can predict empathy responses. Findings will generate new knowledge for a theoretical model on the interaction of formal and content features in media-elicited emotional responses.

Fluctuations in affect represent integration of information based on stimulus relevance and prior affect 

Erkin Asutay & Daniel Västfjäll

Understanding how affect fluctuates in the face of ongoing sensory information is critical in understanding the role of affect in human behavior. Here, using a novel paradigm, we show that experienced affect at a given time represents affective integration of sensory input and previously experienced affect, and this integration is influenced by stimulus relevance. Participants (N=89) sequentially viewed four images in each trial (4sec/image rate) and subsequently reported their momentary affective experiences on valence and arousal features. We then formulated predictive models of valence and arousal based on the normative image ratings in each trial and prior affective experience (reported in the previous trial). To manipulate stimulus relevance, we incorporated an attentional task (identifying a visual shape presented on top of the images) into this basic paradigm, which rendered a subset of the images as task-irrelevant. Critically, the results revealed a weighted-averaging mechanism for affective integration that assigns significantly higher weights to relevant (B=.21±.015) compared to irrelevant images (B=.17±.014). Importantly, affective averaging of stimuli and prior affect accounted for distinct contributions to currently experienced affect. The results provide support for the formulation that currently experienced affect carries information about both recent sensory information and prior affective experiences. In addition, during this dynamical process goal-relevant information receives higher weights compared to irrelevant information.

Data-driven approach reveals universal patterns in colour-emotion associations across 30 nations 

Domicele Jonauskaite, Nele Dael, Ahmad Abu-Akel, Daniel Oberfeld, Jörg Wicker, Lausanne International Colour Survey & Christine Mohr

While symbolic meanings of colour might be the making of cultural customs (e.g., white vs. red worn at weddings in Western world vs. China/Japan), little is known whether emotion associations with colour are also culture-specific, or rather universal. We performed a comprehensive, systematic survey on conceptual colour-affect associations in 30 countries (N = 4,598; males = 1,114; mean age = 35.63 y.), completed in individuals’ respective native languages (https://www2.unil.ch/onlinepsylab/colour/main.php). Participants associated 12 colour terms with one, several, or none of 20 emotion concepts presented on the Geneva Emotion Wheel, organised along the axes of valence and power. Most colours were associated with several emotions, usually of similar valence and/or power. There were only a few associations with discrete emotions (e.g., RED and love, RED and anger, BROWN and disgust). To test for cross-cultural effects, we computed association matrices on the likelihood of associating 12 colour terms with 20 emotions for each country. Colour-emotion associations of individual countries were relatively close to the colour-emotion associations of the “global” matrix (average likelihood); similarity was the highest for Spain (94.1%) followed by 14 additional countries with a similarity level of ≥ 85%. High similarity was reported on other measures too. Nevertheless, a machine-learning algorithm could predict participants’ country of origin with accuracies above the chance level. The confusions mainly occurred between countries that were neighbours or used the same language. Our results suggest there is a high inter-country agreement on the affective connotations of colour terms with some cultural specificities.

Predicting moral judgments using facial muscle activity: United Kingdom, New Zealand, and Hong Kong 

Pete Cannon, Emma Buchtel, Alexander Bidwell & Leo Ng

Moral judgments have been demonstrated to cluster into five concerns or foundations (Graham, Haidt, & Nosek, 2009). Previous research using facial electromyography found that in a UK sample muscle activity while listening to and thinking about 90 statements split into good and bad behaviours across the five moral foundations predicted later subjective moral judgments of these same behaviours (Cannon, Schnall, & White, 2011). There was no single facial muscle’s activity that predicted judgments across all foundations, but contemplating statements that involved harm increased activity of corrugator supercilii (knitting the brow to frown), unfairness increased activity of the levator labii muscles (wrinkling the nose), and purity violations resulted in a full face disgust expression (corrugator supercilii and levator labii). We present a direct replication with data collected in New Zealand (N=30) and Hong Kong (N=40) along with a re-analysis of the original data (N=37). Analyses using Bayesian Mixed Models supports the original study’s finding that facial muscle activity can predict subjective judgments, but the new data and re-analysis finds that the muscles that predict negative judgments vary by culture.