RESUMEN

Emotional facial expressions can inform researchers about an individual's emotional state. Recent technological advances open up new avenues to automatic Facial Expression Recognition (FER). Based on machine learning, such technology can tremendously increase the amount of processed data. FER is now easily accessible and has been validated for the classification of standardized prototypical facial expressions. However, applicability to more naturalistic facial expressions still remains uncertain. Hence, we test and compare performance of three different FER systems (Azure Face API, Microsoft; Face++, Megvii Technology; FaceReader, Noldus Information Technology) with human emotion recognition (A) for standardized posed facial expressions (from prototypical inventories) and (B) for non-standardized acted facial expressions (extracted from emotional movie scenes). For the standardized images, all three systems classify basic emotions accurately (FaceReader is most accurate) and they are mostly on par with human raters. For the non-standardized stimuli, performance drops remarkably for all three systems, but Azure still performs similarly to humans. In addition, all systems and humans alike tend to misclassify some of the non-standardized emotional facial expressions as neutral. In sum, emotion recognition by automated facial expression recognition can be an attractive alternative to human emotion recognition for standardized and non-standardized emotional facial expressions. However, we also found limitations in accuracy for specific facial expressions; clearly there is need for thorough empirical evaluation to guide future developments in computer vision of emotional facial expressions.

RESUMEN

Assessing design solutions via domain-specific emotions has been widely concerned and explored in the field of affective design. However, the examination and accommodation of individual differences have not been settled sufficiently in the literature. To address this research gap, this paper proposes a descriptive approach to draw calibrated collective emotion patterns in survey-based affective design assessment. A 'Repertory Grid Interview linked with Rate-All-That-Apply' (RGI/RATA) procedure is firstly conducted to elicit and code the individual's personal emotional descriptions into mid-level Emotion Words (EWs) and to gather emotion data grids with each grid quantified by an individual's own EWs. The obtained individualised emotion data grids are then subjected to Multiple Factor Analysis (MFA) to extract collective emotional space, thus to enable conceptualising collective emotional dimensions and measuring calibrated collective responses. A case study demonstrating the implementation process for a simple project of appearance design assessment is also presented. Practitioner Summary: The proposed methodology may help a design team to investigate the shared patterns of domain-specific emotions through a single assessment survey. With the provided post hoc analysis tools, designers may also evaluate multi-level individual differences (e.g. regarding user groups or even intra-individual) quantitatively and at a low cost. Abbreviations: EWs: emotion words; HF/E: human factors and ergonomics; IEA: International Ergonomics Association; MFA: multiple factor analysis; PCT: personal construct theory; PCA: principal component analysis; RGI/RATA: repertory grid interview linked with rate-all-that-apply; RGI: repertory grid interview; RATA: rate-all-that-apply; SD: standard deviation; USB: universal serial bus.

Asunto(s)

RESUMEN

The number of studies investigating music processing in the human brain continues to increase, with a large proportion of them focussing on the correlates of so-called musical emotions. The current Review highlights the recent development whereby such studies are no longer concerned only with basic emotions such as happiness and sadness but also with so-called music-specific or "aesthetic" ones such as nostalgia and wonder. It also highlights how mechanisms such as expectancy and empathy, which are seen as inducing musical emotions, are enjoying ever-increasing investigation and substantiation with physiological and neuroimaging methods. It is proposed that a combination of these approaches, namely, investigation of the precise mechanisms through which so-called music-specific or aesthetic emotions may arise, will provide the most important advances for our understanding of the unique nature of musical experience.

Asunto(s)