Context, positioning and goals
One of the Marketing Science Institute’s priorities for 2018-20 is to better understand the benefits of Customer-Technology Interfaces (CTIs), more commonly known as Human-Machine Interfaces (HMIs). Consumers are faced with a wide range of interfaces when interacting with companies, which significantly affects their experience. Companies are increasingly interested in the user experience as a differentiating factor on competitive markets and as a source of value (Lee, 2013). For example, GfK – one of the world’s leading research institutes – has developed a specific methodology that allows its clients to improve the users’ experience of their existing or newly developed offerings (Wildner et al., 2015).
The user experience is the final step in the process of adopting new technologies (Rogers, 2003). The term “user experience” first appeared in Norman’s “The Design of Everyday Things” in 1988, but this field of research has only recently been studied in marketing. The user experience is defined as the outcome of the characteristics of the interface (e.g. aesthetics and usability), of the user and of the context of the interaction (Hassenzahl & Tractinsky, 2006). It has become essential to study this issue, at a time when a growing number of new products are poorly or insufficiently used, due to problems of usability (Dupré et al., 2018).
The design of HMIs requires a detailed understanding of the attitudinal, behavioural and emotional reactions that come into play during their use (Dupré et al., 2018). Research on human-machine interactions specifically places the use of interfaces and the user’s perspective at the heart of its concerns, as shown by the virtually systematic use of a user-centred design approach in the research activities. This research project is positioned at the intersection of computer science, consumer behaviour and psychology. It contributes to the development of human-machine interfaces by taking account of users’ perceptions and emotions, in order to improve companies’ value propositions by promoting the adoption of digital interfaces. The EmotX project studies the influence of an interface’s characteristics (i.e. its aesthetics and usability) and people’s emotions on the user experience.
The literature on information systems, marketing and human-machine interaction has long focused on the instrumental characteristics of the user experience. In particular, research has examined ease of use (or usability) and its impact on the adoption of technology (e.g. Venkatesh, Morris & Gordon, 2003). More recent studies have emphasised the importance of considering other non-functional features (Tuch et al., 2012), including aesthetics, which is a crucial criterion (Mishra et al., 2015). In human-machine interaction, user emotions appear to be important to the development and evaluation of interfaces (Hook, 2012). When their role in the user experience is analysed, two approaches become apparent (Hassenzahl, 2006). The first highlights the importance of emotions as consequences of the use of a product. The second consists in studying emotions as antecedents to the use and evaluation of a product.
The originality of the EmotX project stems from the use of an integrative approach, bringing together researchers from different disciplines: marketing (from the Centre for Applied Research and Studies in Management – CERAG and the Research Laboratory for In Situ Action Design – COACTIS), computer science (from the Grenoble Computer Science Laboratory – LIG), and psychology (from the Inter-University Psychology Laboratory – LIP/PC2S), revolving around the study of emotions, which will be measured and collected (i.e. measured using questionnaires, but also collected by using software such as Face Reader to interpret facial expressions). One of the goals of this project is to analyse the role of emotions in the evaluation and adoption of an interface. To this end, EmotX is based on two case studies which are complementary in terms of product benefits. The first case concerns a hedonic interface (related to a “pleasure-related” purchase, e.g. linked to tourism), and the second case concerns a utilitarian interface (connected with a search for information, e.g. related to natural risks).
EmotX combines both of Hassenzahl’s (2006) above-mentioned approaches to studying emotions within the user experience. From the perspective of human-machine interaction, the research question amounts to studying the role of emotions as characteristic responses at different levels (high or low) of aesthetics and usability in interface design (T1). From the marketing perspective, the research question is about understanding the central role of emotions in the adoption of an interface (T2). Figure 1 summarises the two approaches to this research project.
Florence Jeannot, Head of the “Emotions as explanatory mechanisms for interface adoption” division
The PAD (Pleasure, Arousal and Dominance) model, devised by Mehrabian and Russell (1974) within the widely recognized Stimulus-Organization-Response (SOR) model, was first used to examine the mediating (or explanatory) effect of emotions on a psychological environment and its stimuli, on the one hand, and the behavioural responses, on the other. We adhere to this theoretical approach by supplementing this model with the approaches of Wood and Moreau (2006) and Jeannot and Jolibert (2013). This enables us to simultaneously consider both emotional and cognitive dimensions when using an interface. Understanding the relationship between these dimensions, which are often studied separately, is a contribution to the literature. In addition, we enrich the HMI literature by examining the acceptance of the interface beyond its intended uses, while also considering loyalty-related aspects (e.g. recommendations). This work will focus on a hedonic interface for which positive emotions (e.g. joy) will be considered. Consequently, the first goal of task T2 is to analyse the influence of usability and aesthetics on positive emotions and their role in the acceptance of an interface.
To better understand the role of negative emotions, we will extend the study of their mediating role in the case of a utilitarian interface. As mentioned above, under conditions of uncertainty, discrepancies between reported and actual behaviours are likely to become apparent. Moreover, it is known that when making decisions in a context involving losses (e.g. a natural event), individuals have a tendency to take risks (Kahneman, 2003). That is why we use another application case, related to a context of greater uncertainty and a utilitarian purpose. This study will be an extension of doctoral research initiated as part of the RISK project (CDP Labex 2018). More specifically, the second goal of the T2 task is to better understand the role of negative emotions in decision-making under conditions of uncertainty (e.g. an interface related to natural risk management).
The method used to conduct the research is based on the THEDRE method (Mandran and Dupuy-Chessa, 2017), underpinned by the continuous improvement approach and advocating the use of several data production methods to clearly identify the problem and obtain solid elements to validate the research. We will firstly use a quantitative method to test the proposed statistical models. There will be two collections per panel for each interface developed (i.e. a sample of 200 respondents aged 18 and over for each type of interface), as well as a laboratory experiment designed to calibrate the measurement of the emotions collected. The independent variables will be manipulated according to the aim of the study, which will lead to the development of several interface conditions (e.g. low v. high aesthetics). The dependent variables measured on scales taken from the literature will be emotions (i.e. self-reported) and interface adoption indicators (i.e. intention to use and recommend), as well as the collected emotions. We will also use a qualitative method to understand users’ behaviour and reactions when using the application by conducting individual interviews with about 20 users before and after the laboratory test.
