Figure 2 - uploaded by Michelle Yik
Content may be subject to copyright.
Source publication
Affect is involved in many psychological phenomena, but a descriptive structure, long sought, has been elusive. Valence and arousal are fundamental, and a key question-the focus of the present study-is the relationship between them. Valence is sometimes thought to be independent of arousal, but, in some studies (representing too few societies in th...
Contexts in source publication
Context 1
... evaluate the invariance of measures for valence and arousal across the 33 samples, we tested configural invariance (factor loadings and intercepts freely estimated across groups) and metric invariance (factor loadings constrained to be equal across groups) to ensure the meaning of the latent construct was equal across groups. For details of the procedure, please refer to the online supplemental materials 3. Figure 2 presents the final model consisting of 11 items with two correlated residuals. The model fit of the metric invariance model was compared with the configural model. ...
Context 2
... it was not possible to use the factor scores from the abovementioned final factor analytic model as input for the analyses modeling the relation between valence and arousal, as the point at which the factor scores equal zero cannot be assumed to reflect neutral valence. To circumvent this problem, using the 11 items in Figure 2, we calculated valence and arousal scores per participant by subtracting the average of the negative valence items from the average of the positive valence items, and the low arousal item from the average of the high arousal items respectively (similar approach was used by Kuppens et al., 2013Kuppens et al., , 2017). ...
Citations
... Previous studies have highlighted the challenges of capturing the full emotional spectrum in VA evaluations [16]. They have also noted the emergence of distinct patterns, including V-and U-shaped distributions, in the evaluation results [11,12]. In this study, equal weights (0.5) applied to both user and model evaluations resulted in a balanced integration of subjective user assessments and model predictions, leading to a richer emotional spectrum. ...
With the advancement of text mining and natural language processing technologies, sentiment analysis has found widespread application across various fields. However, current research often emphasizes binary or multi-class classifications, which fail to capture the full spectrum of human emotions. To address this issue, the valence-arousal (VA) model has been proposed but encounters challenges such as data imbalance and subjective labeling. This study presents a novel approach that integrates large language models with user evaluations and employs relevant augmented generation techniques to enhance data quality and consistency. In addition, the VA data is visualized to assess its utility in multidimensional sentiment analysis. Future research will focus on expanding the dataset and conducting in-depth analyses to further validate the proposed approach.
... Affective salience should not be equated with arousal. Although in some stimulus sets, normative ratings of valence appear to have a v-shaped relationship with those of arousal: i.e. arousal increases linearly with the valence distance in both the positive and negative direction (Haj-Ali et al. 2020;Kron et al. 2015), there is much variation between people, circumstances (Kuppens et al. 2013), and cultures (Yik et al. 2023), indicative of a complex relationship. For instance, individuals appear to differ in the degree to which they focus on valence or arousal in constructing their conscious affective experience (Barrett 1998). ...
... The relationship between affective salience and arousal is complex and a matter of debate (Haj-Ali et al. 2020;Kron et al. 2015). It may depend on the person, circumstances (Kuppens et al. 2013), and cultures (Yik et al. 2023). In the current results, differences in LPP amplitude are best predicted by differences in affective salience. ...
While affective salience effects have been observed consistently in the late positive potential (LPP), no event-related potential (ERP) component has consistently shown ordered valence effects. A recent study, showing images of facial attractiveness, however suggests the existence of valence-related effects at very long latencies (1000-3000 ms post stimulus). This could offer new insights into the time-course of affective neural processing. Yet, it remains unclear whether the long-latency effect was specific to facial attractiveness, or to valence in general. To corroborate the existence of a long-latency valence effect, we presented distinctly positive, neutral, and negative valenced IAPS images to a large sample of 224 participants while recording their electroenceph-alogram (EEG). Larger ERP amplitudes were elicited by both positively and negatively valenced compared to neutral stimuli (an affective salience effect) from roughly 500 until 1300 ms, followed by an ordered valence effect of larger amplitudes to negatively compared to positively valenced images from 1500 until 2500 ms. These findings corroborate the previously observed sequence of an affective salience effect followed by a long-latency valence effect. However, the polarity of this valence effect was reversed from that of the facial attractiveness study. Allostasis is discussed as potential reconciling factor. Effects in the N2 and P300 components were also found, but could not be clearly labeled as an affective salience or a valence effect. These results fit with two-stage emotion theories such as the theory of constructed emotions.
... Some emotional theoretical models describe arousal and valence as orthogonal 55 . However, evidence suggests a "V" shape relationship 56 , which is consistent with the existence of low arousal, high valence ("happy" or "satisfied") music and high arousal, low valence ("angry" or "anxious") music, and also aligns with the conclusions of this study. Previous studies have primarily focused on the superficial effects of music tempo on emotional valence and arousal, often using subjective scales for assessment. ...
Music can effectively influence human emotions, with different melodies and rhythms eliciting varying emotional responses. Among these, tempo is one of the most important parameters affecting emotions. This study explores the impact of music tempo on emotional states and the associated brain functional networks. A total of 26 participants without any history of neurological or psychiatric disorders and music training took part in the experiment, using classical piano music clips at different tempi (56, 106, 156 bpm) as stimuli. The study was conducted using emotional scales and electroencephalogram (EEG) analysis. The results showed that the valence level of emotions significantly increased with music tempo, while the arousal level exhibited a “V” shape relationship. EEG analysis revealed significant changes in brainwave signals across different frequency bands under different tempi. For instance, slow tempo induced higher Theta and Alpha power in the frontal region, while fast tempo increased Beta and Gamma band power. Moreover, fast tempo enhanced the average connectivity strength in the frontal, temporal, and occipital regions, and increased phase synchrony value (PLV) between the frontal and parietal regions. However, slow tempo improves PLV between the occipital and parietal regions. The findings of this study elucidate the effects of music tempo on the brain functional networks related to emotion regulation, providing a theoretical basis for music-assisted diagnosis and treatment of mood disorders. Furthermore, these results suggest potential applications in emotion robotics, emotion-based human-computer interaction, and emotion-based intelligent control.
... Around 90% of the research reviewed is based solely on the valence vs. arousal model, making it an area that requires urgent action. Journal papers far exceed clinical trials, with the latter focused almost exclusively on depression and a few cases of Parkinson's or autism [146][147][148]. There is also no intervention aspect in attentionor stress-monitoring papers, although cognitive states are identified as necessary [149][150][151]. ...
Background/Objectives: This systematic review presents how neural and emotional networks are integrated into EEG-based emotion recognition, bridging the gap between cognitive neuroscience and practical applications. Methods: Following PRISMA, 64 studies were reviewed that outlined the latest feature extraction and classification developments using deep learning models such as CNNs and RNNs. Results: Indeed, the findings showed that the multimodal approaches were practical, especially the combinations involving EEG with physiological signals, thus improving the accuracy of classification, even surpassing 90% in some studies. Key signal processing techniques used during this process include spectral features, connectivity analysis, and frontal asymmetry detection, which helped enhance the performance of recognition. Despite these advances, challenges remain more significant in real-time EEG processing, where a trade-off between accuracy and computational efficiency limits practical implementation. High computational cost is prohibitive to the use of deep learning models in real-world applications, therefore indicating a need for the development and application of optimization techniques. Aside from this, the significant obstacles are inconsistency in labeling emotions, variation in experimental protocols, and the use of non-standardized datasets regarding the generalizability of EEG-based emotion recognition systems. Discussion: These challenges include developing adaptive, real-time processing algorithms, integrating EEG with other inputs like facial expressions and physiological sensors, and a need for standardized protocols for emotion elicitation and classification. Further, related ethical issues with respect to privacy, data security, and machine learning model biases need to be much more proclaimed to responsibly apply research on emotions to areas such as healthcare, human–computer interaction, and marketing. Conclusions: This review provides critical insight into and suggestions for further development in the field of EEG-based emotion recognition toward more robust, scalable, and ethical applications by consolidating current methodologies and identifying their key limitations.
... Research on emotion regulation has primarily focused on the valence dimension (e.g., Buhle et al., 2014). However, previous research has suggested two dimensions of emotional responses (valence and arousal; Yik et al., 2023). Further research should use both valence and arousal in emotional responses to understand the effect of positive appraisal of alternative proteins on food wanting in a more nuanced manner. ...
Alternative proteins have attracted increasing attention from researchers and industry. Generally, consumers exhibit reluctance toward accepting alternative proteins. However, the potential of cognitive strategies to enhance consumer acceptance of alternative proteins remains unclear. Drawing on the literature on emotion regulation, we investigated whether emotion-regulation strategies, particularly positive cognitive reappraisal, could increase positive emotions and the wanting for alternative proteins. Across two pre-registered studies, our findings revealed that positive cognitive reappraisal significantly increased the wanting for various alternative proteins, including insects, plant-based meat analogs, cultured meat, and algae compared with looking at the alternative proteins. Additionally, an increase in the wanting for alternative proteins was mediated by an increase in positive emotional responses. In other words, positive cognitive reappraisal (versus looking at alternative proteins) enhances positive emotional responses to alternative proteins, which in turn enhances wanting for alternative proteins. These findings reveal the role of cognitive strategies in enhancing consumer acceptance of alternative proteins and suggest that interventions focusing on positive cognitive reappraisal could effectively increase consumer acceptance of alternative proteins.
... As there is accumulating evidence that valence and arousal are correlated (Yik et al., 2023;Kuppens et al., 2013), we added the option to map the features to the polar coordinates (distance and angle) instead of the xand y-axes of the affect grid. ...
Subjective experience is key to understanding affective states, characterized by valence and arousal. Traditional experiments using post-stimulus summary ratings do not resemble natural behavior. Fluctuations of affective states can be explored with dynamic stimuli, such as videos. Continuous ratings can capture moment-to-moment affective experience, however the rating or the feedback can be interfering. We designed, empirically evaluated, and openly share AffectTracker, a tool to collect continuous ratings of two-dimensional affective experience (valence and arousal) during dynamic stimulation, such as 360-degree videos in immersive virtual reality. AffectTracker comprises three customizable feedback options: a simplified affect grid (Grid), an abstract pulsating variant (Flubber), and no visual feedback.Two studies with healthy adults were conducted, each at two sites (Berlin, Germany, and Torino, Italy). In Study 1 (Selection: n=51), both Grid and Flubber demonstrated high user experience and low interference in repeated 1-min 360-degree videos. Study 2 (Evaluation: n=83) confirmed these findings for Flubber with a longer (23-min), more varied immersive experience, maintaining high user experience and low interference. Continuous ratings collected with AffectTracker effectively captured valence and arousal variability. For shorter, less eventful stimuli, their correlation with post-stimulus summary ratings demonstrated the tool’s validity; for longer, more eventful stimuli, it showed the tool’s benefits of capturing additional variance. Our findings suggest that AffectTracker provides a reliable, minimally interfering method to gather moment-to-moment affective experience also in immersive environments, offering new research opportunities to link affective states and physiological dynamics.
... C. S. Kam & Meyer, 2015). Therefore, some research has aimed to capture the full dimensions by explicitly assuming either the bipolar account (Yik, 2007) or the bivariate (Affleck et al., 2001;Santos et al., 2021;Wilken & Miyamoto, 2018;Yik et al., 2023). These approaches do not resolve the issue. ...
Many psychological dimensions seem bipolar (e.g., happy–sad, optimism–pessimism, and introversion–extraversion). However, seeming opposites frequently do not act the way researchers predict real opposites would: having correlations near −1, loading on the same factor, and having relations with external variables that are equal in magnitude and opposite in sign. We argue these predictions are often incorrect because the bipolar model has been misspecified or specified too narrowly. We therefore explicitly define a general bipolar model for ideal error-free data and then extend this model to empirical data influenced by random and systematic measurement error. Our model shows the predictions above are correct only under restrictive circumstances that are unlikely to apply in practice. Moreover, if a bipolar dimension is divided into two so that researchers can test bipolarity, our model shows that the correlation between the two can be far from −1; thus, strategies based upon Pearson product-moment correlations and their factor analyses do not test if variables are opposites. Moreover, the two parts need not be mutually exclusive; thus, measures of co-occurrence do not test if variables are opposites. We offer alternative strategies for testing if variables are opposites, strategies based upon censored data analysis. Our model and findings have implications not just for testing bipolarity, but also for associated theory and measurement, and they expose potential artifacts in correlational and dimensional analyses involving any type of negative relations.
... Although numerous taxonomies of emotions exist, most tend to agree that emotions comprise of at least two dimensionspleasuredispleasure and the level of arousal (Abelson & Sermat, 1962;Gillioz et al., 2016;Russell, 1980Russell, , 2003Schlosberg, 1954). Notably, both facial expressions and emotion-laden words converge within the same conceptual space defined by dimensions of valence and arousal (Russell et al., 1989;Russell & Bullock, 1985;Yik et al., 2023). One of the most firmly established empirical laws in psychology seems to be the straightforward relationship between SCR magnitude and subjective arousal ratings of eliciting stimuli. ...
... The utilization of valence and arousal ratings for Estonian emotion words (Russell et al., 1989) was precluded in this study due to the absence of surprise, disgust, and neutral categories. However, the valence-arousal relationship appears to hold true cross-culturally, as supported by evidence including Estonian data (Yik et al., 2023), suggesting that the potential error caused using other language data is expected to be relatively small. ...
... Valence and arousal ratings for Actions, Clothing and Landscapes subsets and arousal in emotional stimuli was found in recent studies (see Yik et al., 2023). Moreover, this pattern between valence and arousal evaluation was also found in previous studies, which were conducted in the same cultural background (e.g., Dal Fabbro et al., 2021), or used bipolar semantic scales to rate valence and arousal (e.g., Haberkamp et al., 2017;Ma et al., 2015;Marchewka et al., 2014), in contrast to the Self-Assessment Manikin (SAM) scale. ...
... Emotion induction is a fundamental methodological issue (for a review, see Joseph et al., 2020) and pictorial databases supply a great number of images validated as effective emotional inductors (Barrett & Kensinger, 2010;Colden et al., 2008;Ebner et al., 2010;Kim et al., 2018;Marchewka et al., 2014). As SocialPICS could induce emotions, and negative images produce stronger reactions than positive ones (i.e., negativity bias, Yik et al., 2023), we suggest that future research using this database should evaluate participants' mood state at the beginning and at the end of the experimental session. ...
The present study developed and validated an image database, SocialPICS, with information of socioeconomic status (SES) and affective space (i.e., valence and arousal). Images were selected from public-domain websites and a pilot study pre-selected the images to compose the database. A sample of 132 participants (mean age = 24.3 ± 9.4; 61% women, 36% men, and 3% other) responded a questionnaire and rated the SES, the valence, and the arousal of the images displayed. Results showed that the instrument is validated for the Brazilian context. The image dataset covers a broad range of SES levels in a continuum ranging from lower- to higher-status images and provides subsets of images in a categorical classification (high, medium, and low). In addition, the affective space analysis showed that SES image ratings are positively associated to valence, and negatively associated to arousal. It is likely that SocialPICS portraits social differences regarding power, prestige, and control of resources that SES status communicates. As a theoretical outcome, we argue that SES images are emotional images. SocialPICS is a novel, high-quality resolution, standardized database of 429 SES static images composed of 136 human action images, 157 clothing images, and 136 landscape images. These three subsets comprise stimuli usually employed by researchers in social cognition and neuroscience. Its use should simplify and favor original and replication studies with a higher level of standardization and control over visual SES-content stimuli. Information on physical proprieties is provided for each image. Download SocialPICS: https://osf.io/3t9r2/.
... Thorough screenings are required of culturally situated discretized emotions relevant to moral dilemmas (Michelini et al., 2019). Indeed, condensing complex emotions and emotion-laden stimuli into basic affective features such as valence and arousal risks missing cultural differences key for comparing samples (Ferré et al., 2022;Lim, 2016;Schiller et al., 2023;Yik et al., 2023). Furthermore, harmonized parameters are needed to constrain assumptions on emotional state types (McFarlane & Perez, 2020), and normative data to define emotion valence baselines between groups (McFarlane & Perez, 2020). ...
Socio-cognitive research on bilinguals points to a moral foreign-language effect (MFLE), with more utilitarian choices (e.g., sacrificing someone to save more people) for moral dilemmas presented in the second language (L2) relative to the first language. Yet, inconsistent results highlight the influence of subject-level variables, including a critical underexplored factor: L2 proficiency (L2p). Here we provide a systematic review of 57 bilingualism studies on moral dilemmas, showing that L2p rarely modulates responses to impersonal dilemmas, but it does impact personal dilemmas (with MFLEs proving consistent at intermediate L2p levels but unsystematic at high L2p levels). We propose an empirico-theoretical framework to conceptualize such patterns, highlighting the impact of L2p on four affective mediating factors: mental imagery, inhibitory control, prosocial behavior and numerical processing. Finally, we outline core challenges for the field. These insights open new avenues at the crossing of bilingualism and social cognition research.
