Gender and Personality Trait Measures Impact Degree of Affect Change in a Hedonic Computing Paradigm

Abstract

To date, affective computing research has acknowledged individual differences with regard to detecting affect, yet little research has explored how these individual differences may determine the degree to which affective computing is successful in manipulating the affect of specific computer users. The current study used individual difference measures to predict how much an individual can be influenced by a hedonic computing paradigm: a simple trivia game. Female participants responded in a greater way to positive affective feedback about their performance than did men. Moreover, several personality traits, including neuroticism, narcissism, self-esteem, and extraversion, augmented the degree to which affect changed as a result of playing the game. The results are consistent with the gender differences hypothesis, and the authors conclude that individual differences, particularly gender and personality traits, play a large role in the potential impact of computing platforms and would be useful in personalizing the affective nature of the human-computer interaction.

Discovering gender differences in facial emotion recognition via implicit behavioral cues

“…In the no mask condition, very similar AUC scores are observed for the frontal F8 (0.55), AF3 (0.52) and T7 (0.52) electrodes. 3. EEG data over all 4s is used for this analysis.…”

“…The need to account for gender differences in interaction design and computing has spurred the evolution of the Gender HCI field [1]. Being able to recognize user demographics such as gender can benefit interactive and gaming systems in terms of a) visual and interface design [2], (b) recommending the right games and products (via ads), and (c) providing the right motivation and feedback for enhancing user satisfaction and experience [3]. Most gender recognition (GR) systems are face or voice-based, which pose privacy concerns as face or voice samples are biometrics which enable discovery of a person's identity.…”

Gender and emotion recognition with implicit user signals

“…Gender HCI [54] and A ective HCI [39] have evolved as critical HCI sub-elds due to widespread acknowledgment of the fact that computers need to appreciate and adapt to the user's gender and emotional state. e ability to identify user demographics including gender and emotion can bene t interactive and gaming systems in terms of a) visual and interface design [10,37], (b) game and product recommendation (via ads) [17,57], and (c) provision of appropriate motivation and feedback for optimizing user experience [43]. Contemporary gender recognition (GR) and emotion recognition (ER) systems primarily work with facial [18,35] or speech [25,26] cues; however, face and speech are biometrics that encode an individual's identity, and pose grave privacy concerns as they can be recorded without the user's knowledge [1].…”

“…We show how relevant gender and emotion-speci c information is captured by these low-cost devices via examination of event-related potential (ERP) and xation distribution pa erns, and also through recognition experiments. To capture gender-based di erences, we designed a facial emotion recognition experiment as men and women have been known to respond di erently to a ective information [7,14,32,43]. Our arXiv:1708.08735v1 [cs.HC] 29 Aug 2017 study performed with 28 viewers (14 male) con rms that women are superior at facial ER, mirroring prior ndings.…”

“…G Ender human-computer interaction (HCI) [1] and Affective HCI [2] have evolved as critical HCI sub-fields, as it is critical for computers to appreciate and adapt to the user's gender and emotional state. Inferring users' soft biometrics such as gender and emotion would benefit interactive and gaming systems in terms of a) visual and interface design [3], [4], (b) game and product recommendation (via ads) [5], [6], and (c) provision of appropriate motivation and feedback for optimizing user experience [7]. Gender recognition (GR) and emotion recognition (ER) systems primarily work with facial [8], [9] or speech [10], [11] cues which are biometrics encoding a person's identity.…”