Discrete Emotion Categories Research Articles

Age differences in emotional reactivity to facets of sadness and anger

ABSTRACT Emotional reactivity, based on the discrete emotions approach (DEA), is related to opportunities or constraints across development. While prior research suggests sadness to be more adaptive in old age and anger to be more adaptive in young adulthood, there may be facets within these discrete emotion categories that further expand the DEA framework: loss-based vs. failure-based sadness and frustration-based vs. moral violation-based anger. A sample of 49 younger adults (M age = 20.00, SD = 2.26) and 51 older adults (M age = 66.00, SD = 4.94) were asked to relive and describe an emotional memory associated with facets of sadness and anger. Emotional reactivity was operationalized through self-report ratings on distinct facet categories. Results revealed a significant age difference in emotional reactivity to moral violation-based anger, with older adults being more reactive than younger adults. No other significant age differences were observed. These findings are discussed in terms of how further distinctions across emotional facets can inform a better understanding of affective experience across adulthood and old age.

"Hot" affect-related aspects in emotional information processing: The role of facial muscle responses in the direct and indirect processing of emotion categories.

The present study investigated the involvement of facial muscle responses in the indirect and direct processing of emotional facial expressions. Five discrete emotion categories were used, and we assessed both facial muscle and behavioral responses on a trial-by-trial basis. Experiment 1 tested facial muscle activation of clearly visible stimuli in an emotion categorization task. We observed emotion-specific facial muscle responses and corresponding behavioral categorization effects, providing evidence for the specificity of facial muscle activation. By contrast, under masked indirect presentation conditions in which emotional facial expressions were presented as primes in an emotion misattribution procedure, a specific pattern of emotion-congruent and cross-category behavioral misattributions was observed (in line with Rohr et al., 2015, 2018). Multilevel analyses in Study 2 suggest that an emotional reaction feeds into the behavioral decision, as indicated by differential activation of the frontalis lateralis in response to angry faces. Thus, the present study provides evidence that facial muscle responses contribute to behavioral decisions under masked indirect processing conditions. The different pattern of effects in both studies suggests that facial muscle responses index different processes, depending on the processing conditions: sensorimotor simulation in direct processing conditions and emotional reactions in masked indirect processing conditions. We discuss the implications for models that aim to account for facial muscle activity in response to emotional facial expressions. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Contextual emotion detection in images using deep learning.

Computerized sentiment detection, based on artificial intelligence and computer vision, has become essential in recent years. Thanks to developments in deep neural networks, this technology can now account for environmental, social, and cultural factors, as well as facial expressions. We aim to create more empathetic systems for various purposes, from medicine to interpreting emotional interactions on social media. To develop this technology, we combined authentic images from various databases, including EMOTIC (ADE20K, MSCOCO), EMODB_SMALL, and FRAMESDB, to train our models. We developed two sophisticated algorithms based on deep learning techniques, DCNN and VGG19. By optimizing the hyperparameters of our models, we analyze context and body language to improve our understanding of human emotions in images. We merge the 26 discrete emotional categories with the three continuous emotional dimensions to identify emotions in context. The proposed pipeline is completed by fusing our models. We adjusted the parameters to outperform previous methods in capturing various emotions in different contexts. Our study showed that the Sentiment_recognition_model and VGG19_contexte increased mAP by 42.81% and 44.12%, respectively, surpassing the results of previous studies. This groundbreaking research could significantly improve contextual emotion recognition in images. The implications of these promising results are far-reaching, extending to diverse fields such as social robotics, affective computing, human-machine interaction, and human-robot communication.

Slurring individuals

ABSTRACT This paper explores the derogatory uses of nicknames within closely-knit social settings such as villages, households, and schools. By examining ethnographic and psychological data on nicknaming practices, this paper contends that pejorative nicknames and slurs share structural and functional attributes. On the one hand, pejorative nicknames and slurs can elicit deep offence regardless of the speaker’s intentions or whether they occur within speech reports. On the other, pejorative nicknames can contribute to creating and reinforcing unjust intra-group hierarchies, hence mirroring the role of slurs within a smaller social scale. To explain these shared attributes, this paper argues (i) that both forms of verbal aggression index multiple dimensional qualities such as ‘negative valence’, ‘neutral arousal’, and ‘high dominance’ rather than discrete emotional categories such as ‘contempt’ or ‘anger’, and (ii) that the expression of high dominance in social interactions underlies their capacity to offend. Then, it translates this hypothesis into a Bayesian model of sociolinguistic variation inspired by Heather Burnett’s work on identity construction, thus integrating pragmatic reasoning into a psychologically informed framework for interpreting emotional cues. By studying both phenomena in tandem, this paper shows how understanding nicknaming dynamics within smaller speech-act communities can contribute to our comprehension of the derogatory impact of slurs in more intricate social contexts.

Multi-modal Physiological Signal Fusion for Emotion Classification: A Multi-Head Attention Approach

In this essay, a model-level fusion technique of multi-modal physiological signals using Multi-Head Attention is studied. A framework that utilizes multi-model physiological signals for the task of emotion classification is proposed. First, the GCRNN model, which combines the Graph Convolutional Network (GCN) and the Long and Short Term Memory (LSTM), captures the unique features of electroencephalogram (EEG) signals. The spatial and temporal information that makes up impulses from the EEG can be captured precisely by such a technique. The CCRNN model, which combines the Convolutional Neural Network (CNN) integrated with the Channel-wise Attention and the LSTM, is used for peripheral physiological signals. The model can extract useful features from peripheral physiological signals and automatically learn to weigh the importance of various channels. Finally, Multi-head Attention is employed to fuse the output of the GCRNN and CCRNN methods. The Multi-head Attention can automatically learn the relevance and importance of different modal signals and weigh them accordingly. Emotion classification is implemented by adding a component of Softmax to map what the model produced to discrete emotion categories. The DEAP dataset was utilized in this study for experimental verification, and the results indicate that the method using multi-modal physiological signal fusion is substantially greater in precision than the technique using simply EEG signals. Additionally, the Multi-head Attention fusion method performs better than previous fusion techniques.

Ratings of valence, arousal, happiness, anger, fear, sadness, disgust, and surprise for 24,000 Dutch words.

Emotion is a fundamental aspect of human life and therefore is critically encoded in language. To facilitate research into the encoding of emotion in language and how emotion associations affect language processing, we present a new set of emotion norms for over 24,000 Dutch words. The emotion norms include ratings of two key dimensions of emotion: valence and arousal, as well as ratings on discrete emotion categories: happiness, anger, fear, sadness, disgust, and surprise. We show that emotional information can predict word processing, such that responses to positive words are facilitated in contrast to neutral and negative words. We also demonstrate how the ratings of emotion are related to personality characteristics. The data are available via the Open Science Framework ( https://osf.io/9htuv/ ) and serve as a valuable resource for research into emotion as well as in applied settings such as healthcare and digital communication.

Topographic representation of visually evoked emotional experiences in the human cerebral cortex

Affective neuroscience seeks to uncover the neural underpinnings of emotions that humans experience. However, it remains unclear whether an affective space underlies the discrete emotion categories in the human brain, and how it relates to the hypothesized affective dimensions. To address this question, we developed a voxel-wise encoding model to investigate the cortical organization of human emotions. Results revealed that the distributed emotion representations are constructed through a fundamental affective space. We further compared each dimension of this space to 14 hypothesized affective dimensions, and found that many affective dimensions are captured by the fundamental affective space. Our results suggest that emotional experiences are represented by broadly spatial overlapping cortical patterns and form smooth gradients across large areas of the cortex. This finding reveals the specific structure of the affective space and its relationship to hypothesized affective dimensions, while highlighting the distributed nature of emotional representations in the cortex.

Research on map emotional semantics using deep learning approach

ABSTRACT The main purpose of the research on map emotional semantics is to describe and express the emotional responses caused by people observing images through computer technology. Nowadays, map application scenarios tend to be diversified, and the increasing demand for emotional information of map users bring new challenges for cartography. However, the lack of evaluation of emotions in the traditional map drawing process makes it difficult for the resulting maps to reach emotional resonance with map users. The core of solving this problem is to quantify the emotional semantics of maps, it can help mapmakers to better understand map emotions and improve user satisfaction. This paper aims to perform the quantification of map emotional semantics by applying transfer learning methods and the efficient computational power of convolutional neural networks (CNN) to establish the correspondence between visual features and emotions. The main contributions of this paper are as follows: (1) a Map Sentiment Dataset containing five discrete emotion categories; (2) three different CNNs (VGG16, VGG19, and InceptionV3) are applied for map sentiment classification task and evaluated by accuracy performance; (3) six different parameter combinations to conduct experiments that would determine the best combination of learning rate and batch size; and (4) the analysis of visual variables that affect the sentiment of a map according to the chart and visualization results. The experimental results reveal that the proposed method has good accuracy performance (around 88%) and that the emotional semantics of maps have some general rules.

Temporal relative transformer encoding cooperating with channel attention for EEG emotion analysis

Electroencephalogram (EEG)-based emotion computing has become a hot topic of brain-computer fusion. EEG signals have inherent temporal and spatial characteristics. However, existing studies did not fully consider the two properties. In addition, the position encoding mechanism in the vanilla transformer cannot effectively encode the continuous temporal character of the emotion. A temporal relative (TR) encoding mechanism is proposed to encode the temporal EEG signals for constructing the temporality self-attention in the transformer. To explore the contribution of each EEG channel corresponding to the electrode on the cerebral cortex to emotion analysis, a channel-attention (CA) mechanism is presented. The temporality self-attention mechanism cooperates with the channel-attention mechanism to utilize the temporal and spatial information of EEG signals simultaneously by preprocessing. Exhaustive experiments are conducted on the DEAP dataset, including the binary classification on valence, arousal, dominance, and liking. Furthermore, the discrete emotion category classification task is also conducted by mapping the dimensional annotations of DEAP into discrete emotion categories (5-class). Experimental results demonstrate that our model outperforms the advanced methods for all classification tasks.

Intensity, frequency, and differentiation of discrete emotion categories in daily life and their associations with depression, worry, and rumination.

Previous research examining links between psychopathology and emotions in daily life have largely focused on disorder-specific patterns of mean-level positive and negative affect. In this study, we examined three transdiagnostic dimensions of psychopathology, namely depression, worry, and rumination, and their association with the intensity, frequency, and differentiation of discrete emotion categories endorsed in daily life. Three independent samples of young adults (total n = 614) completed baseline measures assessing depression, worry, and rumination followed by a 1-week daily diary procedure. Daily diaries assessed the following emotion categories: "tranquility," "contentment," "happiness," "excitement," "pride," "sadness," "boredom," "guilt," "fear," and "anger." Intensity, frequency, and differentiation were all calculated from daily emotion ratings. Psychopathology was associated with intensity and frequency of pleasant emotion, with intensity showing stronger associations. Intensity of contentment and happiness were most strongly associated with all three forms of psychopathology, whereas pride had the weakest associations. Psychopathology was associated with intensity and frequency of unpleasant emotions, with frequency tending to show stronger associations. Frequency of sadness and boredom were most strongly associated with depression, whereas the frequency and intensity of fear were similarly associated with depression and worry. Although psychopathology was associated with pleasant and unpleasant emotion differentiation, to our surprise, effect sizes were negligible to small. Clinical implications and recommendations for future research are discussed. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Music and Lyric Characteristics of Popular Dutch Funeral Songs.

This study compared the characteristics of 150 songs (Dutch lyrics, N = 47, English lyrics, N = 103), popular at Dutch funerals, to an equal number of non-funeral songs. The variables explored included those linked with the music (valence, energy, danceability, acousticness, key, and tempo); and lyrics, namely: linguistics-related (first-person singular/plural, second-person pronouns; past, present, future tense; expressed emotion (positive, negative words, and the discrete emotional categories anger, anxiety, sadness); and category words (those relating to family, friends, death, religion). Funeral music was lower in valence, energy, and danceability and higher in acousticness than non-funeral music. Furthermore, English funeral music lyrics contained more second-person pronouns and were more future-focused than comparison songs. Funeral lyrics were not particularly negative, but English texts contained more words relating to sadness. In conclusion, funeral music differs in severable notable respects from general popular songs that may reflect the special purpose of this music.

Fear and cultural background drive sexual prejudice in France – a sentiment analysis approach

Abstract Sexual prejudice and its negative consequences remain major issues in Western societies, and numerous studies have tried to pinpoint its sociocultural underpinnings. However, most research has operationalized sexual prejudice via self-report measures or via implicit association tests (IATs), although it surfaces in language use and can be traced in spontaneous speech. Here, we report results from an experimental study investigating sexual prejudice in a corpus of spontaneous speech samples. Specifically, we tested in a context-sensitive sentiment analysis approach which attitudes (negative vs. positive) and emotions (joy, sadness, anger, fear, disgust) were voiced by the participants in response to picture prompts displaying homosexual couples. We also considered the sociocultural basis of prejudicial attitudes, in particular the effects of the participants’ cultural background (France vs. Maghreb), age and gender. We find strong effects of cultural background and gender both on the frequency of negative vs. positive attitudes expressed, and on discrete emotion categories, namely that male Maghrebian participants were more negative and conveyed more fear. The results are discussed in the context of current diversity approaches in France and their implications for potential prejudice regulation strategies. We further discuss in how far our context-sensitive sentiment analysis approach advances research on sexual prejudice.

Development of an affective database made of interactive virtual environments

Despite the great potential of Virtual Reality (VR) to arouse emotions, there are no VR affective databases available as it happens for pictures, videos, and sounds. In this paper, we describe the validation of ten affective interactive Virtual Environments (VEs) designed to be used in Virtual Reality. These environments are related to five emotions. The testing phase included using two different experimental setups to deliver the overall experience. The setup did not include any immersive VR technology, because of the ongoing COVID-19 pandemic, but the VEs were designed to run on stereoscopic visual displays. We collected measures related to the participants’ emotional experience based on six discrete emotional categories plus neutrality and we included an assessment of the sense of presence related to the different experiences. The results showed how the scenarios can be differentiated according to the emotion aroused. Finally, the comparison between the two experimental setups demonstrated high reliability of the experience and strong adaptability of the scenarios to different contexts of use.

Supplemental Material for Intensity, Frequency, and Differentiation of Discrete Emotion Categories in Daily Life and Their Associations With Depression, Worry, and Rumination

Supplemental Material for Intensity, Frequency, and Differentiation of Discrete Emotion Categories in Daily Life and Their Associations With Depression, Worry, and Rumination

Real Time Facial Gesture Recognition using Deep Learning

The human faces have vibrant frequency of characteristics, which makes it difficult to analyze the facial expression. Automated real time emotions recognition with the help of facial expressions is a work in computer vision. This environment is an important and interesting tool between the humans and computers. In this investigation an environment is created which is capable of analyzing the person’s emotions using the real time facial gestures with the help of Deep Neural Network. It can detect the facial expression from any image either real or animated after facial extraction (muscle position, eye expression and lips position). This system is setup to classify images of human faces into seven discrete emotion categories using Convolutional Neural Networks (CNNs). This type of environment is important for social interaction.

FANCat: French affective norms for ten emotional categories.

The present study develops key research for French word norms that combines the predominant theories of dimensional and discrete (or categorical) emotions. As a result, we provide the database FANCat, affective norms for a set of 1031 French words on ten discrete emotion categories: fear, anger, disgust, sadness, anxiety, awe, excitement, contentment, amusement, and serenity. FANCat complements a previous word set, FAN, which provides only the dimensional norms, valence, and arousal (Monnier & Syssau, 2014). Herein, we introduce five discrete positive emotions in efforts to differentiate positive emotions at higher resolution and specificity. Although ten emotional categories were considered in FANCat norms, results showed a high degree of inter-rater reliability and a good external validity. Then, distributional analyses of words into the ten emotion categories revealed that positive words evoked mainly the emotions awe, contentment, and amusement, and principally evoked either one positive emotion only ("pure" words) or two (mixed words). This study contributes to a deeper understanding of the relationship between language, and negative and positive emotions. It is also currently the only norms database in French that analyses ten discrete emotions as well as including valence and arousal. FANCat is available at https://www.researchgate.net/publication/338622765_FANCat_database .

Speech emotion recognition using fusion of three multi-task learning-based classifiers: HSF-DNN, MS-CNN and LLD-RNN

Speech emotion recognition plays an increasingly important role in emotional computing and is still a challenging task due to its complexity. In this study, we developed a framework integrating three distinctive classifiers: a deep neural network (DNN), a convolution neural network (CNN), and a recurrent neural network (RNN). The framework was used for categorical recognition of four discrete emotions (i.e., angry, happy, neutral and sad). Frame-level low-level descriptors (LLDs), segment-level mel-spectrograms (MS), and utterance-level outputs of high-level statistical functions (HSFs) on LLDs were passed to RNN, CNN, and DNN, separately. Three individual models of LLD-RNN, MS-CNN, and HSF-DNN were obtained. In the models of MS-CNN and LLD-RNN, the attention mechanism based weighted-pooling method was utilized to aggregate the CNN and RNN outputs. To effectively utilize the interdependencies between the two approaches of emotion description (discrete emotion categories and continuous emotion attributes), a multi-task learning strategy was implemented in these three models to acquire generalized features by simultaneously operating classification of discrete categories and regression of continuous attributes. Finally, a confidence-based fusion strategy was developed to integrate the power of different classifiers in recognizing different emotional states. Three experiments on emotion recognition based on the IEMOCAP corpus were conducted. Our experimental results show that the weighted pooling method based on attention mechanism endowed the neural networks with the capability to focus on emotionally salient parts. The generalized features learned in the multi-task learning helped the neural networks to achieve higher accuracies in the tasks of emotion classification. Furthermore, our proposed fusion system achieved weighted accuracy of 57.1% and unweighted accuracy of 58.3%, which were significantly higher than those of each individual classifier. The effectiveness of the proposed approach based on classifier fusion was thus validated.

Comment: A role of Language in Infant Emotion Concept Acquisition

Ruba and Repacholi (2020) review an important debate in the emotion development literature: whether infants can perceive and understand facial configurations as instances of discrete emotion categories. Consistent with a psychological constructionist account (Lindquist & Gendron, 2013; Shablack & Lindquist, 2019), they conclude that infants can perceive valence on faces, but argue the evidence is far from clear that infants perceive and understand discrete emotions. Ruba and Repacholi outline a novel developmental trajectory of emotion perception and understanding in which early emotion concept learning may be language-independent. In this comment, we argue that language may play a role in emotion concept acquisition even prior to children’s ability to produce emotion labels. We look forward to future research addressing this hypothesis.

Simultaneous Prediction of Valence/Arousal and Emotions on AffectNet, Aff-Wild and AFEW-VA

Abstract We address the problem of emotional state detection from facial expressions. Our proposed approach simultaneously detects faces and predicts both discrete emotion categories and continuous valence/arousal values from raw input images. We train and evaluate our approach on 3 different datasets, compare our approach to other state-of-the-art approaches and perform a cross-database evaluation. In this way, we found, that our approach generalizes well and is suitable for real-time applications.

Influence of Acculturation on Emotional Learning and Memory in Spanish and English Speakers

Abstract Objective The Emotional Verbal Learning Test-Spanish (EVLT-S) is a novel list learning test similar in structure to traditional verbal memory tests but contains words from 4 discrete emotional categories. Influence of acculturation on emotional learning was examined on EVLT-S learning trials 1 to 5. Participants and Method Participants included 50 bilingual healthy individuals in the Spanish-dominant group (Age M = 20.3years; 72% female) and 27 in the English-dominant group (Age M = 23.4years, 51.9% female). They were administered a battery of tests including the EVLT, a non-emotional verbal list learning test (LLT), Vocabulary (VC), Digit Span (DS), and The Short Acculturation Scale for Hispanics (SASH) in their dominant language. To examine learning curves a mixed model ANOVA was conducted with test (EVLT-S/EVLT, LLT-S/LLT) and trial as within-subjects variables and group (Spanish, English) as a between-subjects variable. Results There were significant main effects (p < .005) for test, trial, and group. The Spanish-dominant group had the lowest overall performance on the EVLT-S (main effect for group). Post hoc comparisons of the Spanish-dominant group’s LLT-S and EVLT-S performance indicated significant differences at each trail (p < .01), with the EVLT-S scores being lower than the LLT-S scores. The ANOVA was repeated with VC, DS, and SASH included as covariates. When controlling for these variables, the main effect for group was no longer significant. A significant trial by acculturation interaction effect emerged (p<.05), with the English-dominant group generally performing better than the Spanish-dominant group. Conclusions Results suggest that Spanish and English language group differences were largely accounted for by word knowledge (VC), working memory (DS), and acculturation differences (SASH) between the groups. The role of these variables is discussed further in the context of acculturation.