English Emotion Research Articles

Semantic-Rich Facial Emotional Expression Recognition

The ability to perceive human facial emotions is an essential feature of various multi-modal applications, especially in the intelligent human-computer interaction (HCI) area. In recent decades, considerable efforts have been put into researching automatic facial emotion recognition (FER). However, most of the existing FER methods only focus on either basic emotions such as the seven/eight categories (e.g., <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">happiness, anger</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">surprise</i> ) or abstract dimensions ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">valence, arousal, etc.</i> ), while neglecting the fruitful nature of emotion statements. In real-world scenarios, there is definitely a larger vocabulary for describing human's inner feelings as well as their reflection on facial expressions. In this work, we propose to address the semantic richness issue in the FER problem, with an emphasis on the granularity of the emotion concepts. Particularly, we take inspiration from former psycho-linguistic research, which conducted a prototypicality rating study and chose 135 emotion names from hundreds of English emotion terms. Based on the 135 emotion categories, we investigate the corresponding facial expressions by collecting a large-scale 135-class FER image dataset and propose a consequent facial emotion recognition framework. To demonstrate the accessibility of prompting FER research to a fine-grained level, we conduct extensive evaluations on the dataset credibility and the accompanying baseline classification model. The qualitative and quantitative results prove that the problem is meaningful and our solution is effective. To the best of our knowledge, this is the first work aimed at exploiting such a large semantic space for emotion representation in the FER problem.

Manifest Quasi-Universals and Embedding Conceptual Clusters

We start this article with a discussion of the problematics involved in translating into modern English a (modern) Chinese text concerning a classical Chinese notion, namely qíng 情. Then, we suggest that it is necessary to distinguish between two levels on which a language is used. The first is the manifest level, where one finds family resemblance between, say, Chinese nù 怒 and English “anger”. The second is the generic level, where one finds qíng in classical Chinese, qínggaˇn in modern Chinese, and emotion(s) in English. We argue that the meanings of words at the generic level can only be accessed via the manifest level. It is misleading to directly identify and compare notions at the generic level (in this case, emotion(s) in English and qíng in Chinese). We call the connections at the manifest level “quasi-universals”, and we refer to the notions at the generic level as “embedding conceptual clusters”.

Dual-Language Testing of Emotional Verbal Fluency: A Closer Look at "Joy," "Sadness," "Fear," "Anger," and "Disgust".

Assessing neuropsychological functions of dual-language speakers with semantic verbal fluency should consider the impact of language proficiency. Much evidence for the proficiency effect is found in tasks targeting neutral words. The proficiency effect on emotional verbal fluency (Emo-VF) that targets emotional word retrieval, however, is far from conclusive. This study aimed to clarify the proficiency effect on Emo-VF, specifically the extent to which language proficiency impacted positive and negative word retrieval comparably. In addition, the study examined the extent to which dual-languages speakers produced duplicated items and unique, non-duplicated items in each of two languages tested. Thirty-two adult participants completed Emo-VF tasks that targeted a comprehensive set of basic emotions ("joy," "sadness," "fear," "anger," "disgust") in English and Spanish in two sessions separated by at least 72h. All participants exhibited greater proficiency in English than in Spanish according to subjective and objective measures of language proficiency. Verbal productivity was comparably lowered for all emotions in the less proficient language. Differences among categories were consistent between languages, with "joy" eliciting more words than other emotions, and "fear" yielding the highest productivity among negative emotions. Finally, dual-language speakers produced a significant number of unique items in different languages, especially for "fear." Language proficiency exerts a general effect on positive and negative word retrieval. The study extends previous work by revealing differences between "fear" and other negative categories in addition to replicating positivity biases in unbalanced dual-language speakers. Dual-language testing is valuable in capturing the richness of emotional word repertoire associated with different languages.

Wonder, beauty, ability and the natural world: The experience of wonder as a positive aesthetic emotion in Old English verse

Drawing on the recent studies on aesthetic emotions and on their recent application to the field of the Old English aesthetic emotions, this paper explores one emotion from the emotion family of AMAZEMENT in the Old English poetic corpus, attending to the type of wonder that is typically triggered by objects of beauty, excellent manufacture and by the natural world. The purpose of this paper is to understand better the poetic usage of the Old English terms for wonder as well as evidence their role in literary and everyday contexts. Through a fine-grained analysis of the above domains, this paper has shown that the wonder implicit in these texts can be triggered by perceptual or cognitive appraisals, but also by a combination of both, highlighting the complexities and particularities of the early medieval English emotion of WONDER, as well as its similarity to other emotions like the EXPERIENCE OF BEAUTY or AWE.

A Multilingual Neural Coaching Model with Enhanced Long-term Dialogue Structure

In this work we develop a fully data driven conversational agent capable of carrying out motivational coaching sessions in Spanish, French, Norwegian, and English. Unlike the majority of coaching, and in general well-being related conversational agents that can be found in the literature, ours is not designed by hand-crafted rules. Instead, we directly model the coaching strategy of professionals with end users. To this end, we gather a set of virtual coaching sessions through a Wizard of Oz platform, and apply state of the art Natural Language Processing techniques. We employ a transfer learning approach, pretraining GPT2 neural language models and fine-tuning them on our corpus. However, since these only take as input a local dialogue history, a simple fine-tuning procedure is not capable of modeling the long-term dialogue strategies that appear in coaching sessions. To alleviate this issue, we first propose to learn dialogue phase and scenario embeddings in the fine-tuning stage. These indicate to the model at which part of the dialogue it is and which kind of coaching session it is carrying out. Second, we develop a global deep learning system which controls the long-term structure of the dialogue. We also show that this global module can be used to visualize and interpret the decisions taken by the the conversational agent, and that the learnt representations are comparable to dialogue acts. Automatic and human evaluation show that our proposals serve to improve the baseline models. Finally, interaction experiments with coaching experts indicate that the system is usable and gives rise to positive emotions in Spanish, French and English, while the results in Norwegian point out that there is still work to be done in fully data driven approaches with very low resource languages.

Exploring the Motivation-Engagement Link: The Moderating Role of Positive Emotion

In self-determination theory (SDT), the role of positive emotion in the motivational process is undertheorized. Most previous research considered emotion as an outcome variable, rather than a main construct of the motivational process. There is, however, some evidence in domains outside language learning suggesting that positive emotion moderates the motivation-engagement link. The extent to which this could be applied to the domain of language learning remains unknown. Therefore, this study aimed to examine the role of positive emotion as a moderator of the motivational process of language learning. Moreover, a comparative multilingual perspective was adopted to better understand simultaneous language learning. Specifically, this study assessed whether positive emotion moderated the relationship between two types of motivation (i.e., autonomous and controlled motivation) and behavioral engagement in two foreign languages. A sample of 108 Chinese university students concurrently learning English as L2 and French as L3 participated in the survey study. Results of regression analysis suggest the differential role of positive emotion and motivation in English and French: For French learning, positive emotion served as a moderator that optimized the motivation-engagement link (i.e., positive interaction between emotion and motivation). For English learning, no moderating effect was discovered, with positive emotion being the significant and strong predictor of engagement. The findings supplied initial evidence of the potential role of positive emotion as a moderator of the motivation-engagement link, contributing to the theoretical incorporation of emotion as a more central player in the motivational process proposed by SDT. The differential role of positive emotion and motivation in learning English and languages other than English points to the value of a multilingual comparative perspective to uncovering the language-specific characteristics of the motivational processes.

What Heritage Bilinguals Tell Us about the Language of Emotion

Variation in the language experience of bilinguals has consequences for cognitive and affective processes. In the current study, we examined how bilingual experience influences the relationship between language and emotion in English among a group of Spanish–English heritage bilinguals on an emotion–memory task. Participants rated the emotionality of English taboo, negative and neutral words and then completed an unexpected recognition test. To account for language experience, data were gathered on the participants’ language dominance and proficiency. Results showed emotion–memory effects in the Spanish–English heritage bilinguals’ English (the societal language): taboo words were recognized significantly better than neutral words, while the emotionality of negative words carried over and significantly affected the recognition of preceding neutral words. Furthermore, such effects were modulated by language dominance scores with more pronounced emotion–memory effects in more English-dominant bilinguals. The findings contribute to a growing body of evidence showing that emotions are not necessarily restricted to the first acquired home language. Critically, for heritage speakers, there is often a shift in language dominance from the home language to the societal language. The present study demonstrates that the effects of emotion on memory are seen in the acquired societal language.

Semantics-Prosody Stroop Effect on English Emotion Word Processing in Chinese College Students With Trait Depression.

This study explored the performance of Chinese college students with different severity of trait depression to process English emotional speech under a complete semantics–prosody Stroop effect paradigm in quiet and noisy conditions. A total of 24 college students with high-trait depression and 24 students with low-trait depression participated in this study. They were required to selectively attend to either the prosodic emotion (happy, sad) or semantic valence (positive and negative) of the English words they heard and then respond quickly. Both prosody task and semantic task were performed in quiet and noisy listening conditions. Results showed that the high-trait group reacted slower than the low-trait group in the prosody task due to their bluntness and insensitivity toward emotional processing. Besides, both groups reacted faster under the consistent situation, showing a clear congruency-induced facilitation effect and the wide existence of the Stroop effect in both tasks. Only the Stroop effect played a bigger role during emotional prosody identification in quiet condition, and the noise eliminated such an effect. For the sake of experimental design, both groups spent less time on the prosody task than the semantic task regardless of consistency in all listening conditions, indicating the friendliness of basic emotion identification and the difficulty for second language learners in face of semantic judgment. These findings suggest the unneglectable effects of college students’ mood conditions and noise outside on emotion word processing.

Musical colossal drama project-based learning to boost students’ engagement in English language learning

Various factors influence students to be active in learning English. One of them is role-playing and drama in learning songs. The present study aims to showcase EFL students’ emotions in learning English as a foreign language by adopting musical colossal drama project-based learning Nested in a qualitative case study, the eleventh-grade students of a senior secondary school in Purbalingga, Central Java, Indonesia were recruited to participate. Data were garnered through semi-structured interviews and classroom observation. The data were analyzed by employing data reduction, data display, and conclusion verification. Findings of the study demonstrate that the senior high school EFL students showed positive emotions when they engaged in learning English, particularly adopting songs, in the frame of musical colossal drama project-based learning. Another finding reveals that musical colossal drama project-based learning boosted the EFL students’ 21st century skills, i.e., critical thinking, creative thinking, communicating, and collaborating. This study promotes some empirical evidence for how musical colossal drama is organized to foster students’ involvement in a project-based English language learning.

The role of mixed emotions in language learning and teaching: A positive psychology teacher intervention

This study adopted a quasi experimental design to examine the impact of a pedagogical positive psychology (PP) intervention on how the integration of positive and negative emotions in English as a foreign language (EFL) learning and teaching affects learner language motivation as a positive affective variable, anxiety as a negative emotion, and EFL achievement as a learning outcome. To address this question empirically, the study implemented three emotion-regulation quasi experiments employed by 6 EFL teachers with 209 EFL learners divided into three groups for 12 weeks: in group 1, learners were exposed extensively to anxiety-regulating strategies, those in group 2 were thoroughly exposed to motivation-promoting strategies, and those in group 3 were exposed to combined anxiety-regulating and motivational strategies together. The influence of the treatment was assessed using questionnaires, classroom observations, and achievement tests. ANOVAs and MANOVAs were employed to capture treatment effects. The findings show that integrating positive and negative emotions in language learning resulted in the largest positive change in learner's motivation, anxiety, and language achievement. These findings support the assumptions of the Broaden-Build Theory in that positive emotions not only broaden people's vision and build strengths but further help undo the lingering effects of negative emotions.

Pengenalan Emosi Pembicara Menggunakan Convolutional Neural Networks

Recognition of the speaker's emotions is an important but challenging component of Human-Computer Interaction (HCI). The need for the recognition of the speaker's emotions is also increasing related to the need for digitizing the company's operational processes related to the implementation of industry 4.0. The use of Deep Learning methods is currently increasing, especially for processing unstructured data such as data from voice signals. This study tries to apply the Deep Learning method to classify the speaker's emotions using an open dataset from SAVEE which contains seven classes of voice emotions in English. The dataset will be trained using the CNN model. The final accuracy of the model is 88% on the training data and 52% on the test data, which means the model is overfitting. This is due to the imbalance of emotion classes in the dataset, which makes the model tend to predict classes with more labels. In addition, the lack of heterogeneity of the dataset makes the character of the emotion class more different from the others so that it can reduce the bias in the model so as not to overfit the model. Further development of this research can be done, such as over-sampling the existing dataset by adding other data sources, then performing data augmentation to get the data character of each emotion class and setting hyperparameter values ​​to get better accuracy values.&#x0D;

EmCat-Eng: A catalogue of 1,759 basic emotion terms in English

Abstract This study investigates the lexicalization patterns of six basic constructs of emotion in English: anger, disgust, fear, joy, sadness, and surprise. These words, along with all their synonyms in noun, verb, and adjective forms were recorded and supplied with corpus frequency data. The resulting catalogue of basic emotion terms in English was analyzed. The categories of words denoting different emotions were quantified in order to determine their relative cultural significance. Word frequency patterns were analyzed in order to determine any manifestations of display rules. The results indicate that in English all emotions are preferentially lexicalized as adjectives. Negative emotions are preferentially expressed as verbs, and positive emotions – as nouns. English boasts more words for negative than positive emotions, confirming the presence of the negative differentiation effect. At the same time, the less numerous words for positive emotions were found to be more frequently used, confirming the Pollyanna effect. The study revealed the central role of fear in the English-speaking world. Uniquely, fear was found to conceptually and semantically overlap with all other basic emotions regardless of their valence; the mean frequency of all the words denoting fear made it the second most frequent overtly, verbally communicated emotion in English – after joy.

Facial expressions elicit multiplexed perceptions of emotion categories and dimensions

SummaryHuman facial expressions are complex, multi-component signals that can communicate rich information about emotions,1, 2, 3, 4, 5 including specific categories, such as “anger,” and broader dimensions, such as “negative valence, high arousal.”6, 7, 8 An enduring question is how this complex signaling is achieved. Communication theory predicts that multi-component signals could transmit each type of emotion information—i.e., specific categories and broader dimensions—via the same or different facial signal components, with implications for elucidating the system and ontology of facial expression communication.9 We addressed this question using a communication-systems-based method that agnostically generates facial expressions and uses the receiver’s perceptions to model the specific facial signal components that represent emotion category and dimensional information to them.10, 11, 12 First, we derived the facial expressions that elicit the perception of emotion categories (i.e., the six classic emotions13 plus 19 complex emotions3) and dimensions (i.e., valence and arousal) separately, in 60 individual participants. Comparison of these facial signals showed that they share subsets of components, suggesting that specific latent signals jointly represent—i.e., multiplex—categorical and dimensional information. Further examination revealed these specific latent signals and the joint information they represent. Our results—based on white Western participants, same-ethnicity face stimuli, and commonly used English emotion terms—show that facial expressions can jointly represent specific emotion categories and broad dimensions to perceivers via multiplexed facial signal components. Our results provide insights into the ontology and system of facial expression communication and a new information-theoretic framework that can characterize its complexities.

With Emotions, We Learn: Understanding Emotions in Learning English as A Second Language (ESL)

The English proficiency level of ESL learners will be affected by enrolling emotions in English language classes. The aim of this research is to analyse the dominant emotions in learning English as a Second Language (ESL) through reading and writing skills by focusing on three positive emotions (enjoyment, hope and pride) and three negative emotions (anxiety, shame and boredom). Thus, a quantitative research design was carried out by involving (n=120) pre-services teachers from a public university in Malaysia. The results revealed that the dominant positive emotions in learning English reading skill was pride while the dominant positive emotions in learning English writing skill was hope. Meanwhile, both skills shared the same emotion of anxiety as the dominant negative emotions. The findings of this study strongly supported the involvement of emotions in the ESL settings. Hence, ESL educators are encouraged to design and formulate an engaging learning environment that ignites more activating positive emotions than negative emotions in the educational settings which would result in better enhancement mainly in the context of ESL learning.

Evolution of emotion semantics

Humans possess the unique ability to communicate emotions through language. Although concepts like anger or awe are abstract, there is a shared consensus about what these English emotion words mean. This consensus may give the impression that their meaning is static, but we propose this is not the case. We cannot travel back to earlier periods to study emotion concepts directly, but we can examine text corpora, which have partially preserved the meaning of emotion words. Using natural language processing of historical text, we found evidence for semantic change in emotion words over the past century and that varying rates of change were predicted in part by an emotion concept's prototypicality—how representative it is of the broader category of “emotion”. Prototypicality negatively correlated with historical rates of emotion semantic change obtained from text-based word embeddings, beyond more established variables including usage frequency in English and a second comparison language, French. This effect for prototypicality did not consistently extend to the semantic category of birds, suggesting its relevance for predicting semantic change may be category-dependent. Our results suggest emotion semantics are evolving over time, with prototypical emotion words remaining semantically stable, while other emotion words evolve more freely.

Analyzing User Digital Emotions from a Holy versus Non-Pilgrimage City in Saudi Arabia on Twitter Platform

Humans are the product of what society and their environment conditions them into being. People living in metropolitan cities have a very fast-paced life and are constantly exposed to different situations. A social media platform enables individuals to express their emotions and sentiments and thus acts as a reservoir for the digital emotion footprints of its users. This study proposes that the user data available on Twitter has the potential to showcase the contrasting emotions of people residing in a pilgrimage city versus those residing in other, non-pilgrimage areas. We collected the Arabic geolocated tweets of users living in Mecca (holy city) and Riyadh (non-pilgrimage city). The user emotions were classified on the basis of Plutchik’s eight basic emotion categories, Fear, Anger, Sadness, Joy, Surprise, Disgust, Trust, and Anticipation. A new bilingual dictionary, AEELex (Arabic English Emotion Lexicon), was designed to determine emotions derived from user tweets. AEELex has been validated on commonly known and popular lexicons. An emotion analysis revealed that people living in Mecca had more positivity than those residing in Riyadh. Anticipation was the emotion that was dominant or most expressed in both places. However, a larger proportion of users living in Mecca fell under this category. The proposed analysis was an initial attempt toward studying the emotional and behavioral differences between users living in different cities of Saudi Arabia. This study has several other important applications. First, the emotion-based study could contribute to the development of a machine learning-based model for predicting depression in netizens. Second, behavioral appearances mined from the text could benefit efforts to identify the regional location of a particular user.

Engaging with customer’s emotions

AbstractInformational-persuasive discourse may be encoded in promotional strategies through which a given product is described in a positive way to persuade potential customers. For this, evaluation may appeal to reason or may tickle emotions (Cook, 2001). This study compares the way in which advertising texts for herbal tea engage with customers’ emotions in English and in Spanish. We examined the strategies of ‘enjoying the experience’ and ‘aesthetic appeal’ from an Appraisal Theory approach (Martin and White, 2005). We categorised these according to the attitude sub-systems of ‘affect’, ‘appreciation’, and ‘judgement’, determined how explicit the evaluation was, and identified gradable resources. Results show that English texts display more ‘affect’-like resources that can awaken a desire in the customer. By contrast, in the Spanish sample ‘appreciation’ resources that evaluate the composition of the product play a greater role. ‘Enjoying the experience’ seems to engage with the customers’ emotions more overtly than ‘aesthetic appeal’.

Social psychological accounts of peer emotion transfer in EFL classrooms: A doubly latent multilevel analysis

Based on theories of emotion contagion and social appraisal in interpersonal affect transfer and the control-value theory of achievement emotions, the present study examined associations between students’ perceptions of peer emotions and their own self-perceived emotions in English as a foreign language (EFL) classrooms. Data were collected from 103 seventh to ninth grade classrooms ( n = 3,643) using self-report questionnaires. Doubly latent multilevel structural equation modeling showed that perceived peer enjoyment, anxiety and boredom, and students’ corresponding emotions for language learning, were positively related and that the effects of perceived peer enjoyment, anxiety and boredom on corresponding student emotions were mediated by control and value appraisals at the individual level. At the class level, however, the mediation effects were only significant for control appraisal as a mediator of effects on anxiety and value appraisal as a mediator of effects on boredom. Effects were robust across grade level, gender and previous language achievement. The discussion centers on the practical implications of peer emotion interactions for promoting foreign language development in classroom instruction.

Keywords, Structures of Feeling, and the Novel

Keywords, Structures of Feeling, and the Novel

Changing the Type of Predicate as a Means of Cognitive Adaptation in Translation

The linguistic worldview is a reflection of the national cognitive worldview. ‘Worldview’ is often defined as a way of perceiving the surrounding reality, yet the way people perceive their personal inner world also reflects their national self-identification. It is difficult to compare how people of different nations experience emotions and perceive such experiences because these processes are not available for direct observation and objective assessment. The most complete representation of the way a person experiences a particular emotion can be found in fiction. Contrastive analysis of how this process is reflected in different languages can be based on a comparison of a literary text with its translation into another language, since, in this case, both texts present the same character in the same situations that cause certain emotions. To exclude the influence of the translator’s personality, in our analysis we have used three different translations of selected passages from Dostoevsky’s The Idiot. A quantitative analysis of the means employed by the translators shows that representation of emotions in English does indeed reflect the way of perceiving the world that is typical of the national linguistic worldview as a whole. In all the three English texts, state predicates prevail over ac-tion predicates, and predicatively used adjectival words prevail over those used attributively. It means that emotional states are mostly perceived by English speakers as something that, while not permanent or inherent to a person, is, at the same time, static: less a process than a result of that process. In contrast, native speakers of Russian perceive emotional states as actions, and the Russian text reveals no inclination toward perceiving emotional states as personal characteristics, whether temporary or permanent. All these regularities are statistical and not absolute, which means that they reflect usage and not the linguistic norm, and, thus, the change of predicates in translation should be regarded as a way of cognitive adaptation rather than a structural transformation.