Causing emotion

Cognitive appraisal theory describes the way in which appraisals, or

perceptions, of events i a i di idual s e i o e t cause emotional responses (Moors, 2013; Roseman, 2011; Scherer, 2009). These appraisals typically feed forward to drive changes in other emotion components, which shape its adaptive response (Figure 15). That is, they determine for a large part the changes that an emotion brings about in the way people think and act. According to this theory, appraisals that indicate goal-conduciveness and goal-obstruction, differentiate positive from negative emotions. Goal- conduciveness and goal-obstruction refer to the way in which an event i flue es the p og ess to a d attai i g the i di idual s goals. That is, if the event implies that the current situation can lead to or led to attaining the

137

i di idual s goals (e.g. good performance when the goal is to perform well), positive emotion is elicited, but when it implies the reverse (e.g. bad performance when the goal is to perform well), negative emotion is elicited. Other appraisals (e.g. of cause, coping potential, and norm violation) further differentiate the type emotion that unfolds (e.g. the difference between the positive emotions of joy and pride). See (Moors, 2013; Roseman, 2011; Scherer, 2009) for overviews.

Figure 15 Schematic of the function of cognitive appraisal processes in emotion causation. An event in the environment causes emotion (e.g. a positive emotion), via cognitive appraisal processes (e.g.an appraisal of an event that is perceived as goal-conducive, such

as good performance), by feeding forward to drive changes (green arrows) in the other emotion components.

There are, however, two conditions that need to be taken into account to enable these appraisals to lead to a sufficiently strong emotional response to impact the link between emotion and creativity during idea generation. We believe that both these two factors need to be taken into account when designing our interactive system.

138

First, certain interactions between appraisals can be conditional for an emotion to emerge (Brans & Verduyn, 2014; Sonnemans & Frijda, 1994). In addition to the influence of appraised goal-conduciveness or - obstructiveness on positive or negative emotion, the appraised goal-

relevance of an event, i.e. the evaluation of how strongly the event affects

the i di idual s u e t goals, ode ates the degree to which other appraisal processes can cause an emotion (Kreibig et al., 2012; Nyer, 1997). For instance, when primed with achievement goals, performance feedback that is positive (success) and negative (failure) can elicit positive and negative emotions, but only when people appraise the performance feedback to be sufficiently relevant to their current goals (Kreibig et al., 2012). This suggests that an event should be perceived as both goal- relevant and goal-conducive to enable it to cause emotion. That is, without any goal-relevance an event is unlikely to cause an emotion that brings about noticeable change in the way people think and act.

Second, feedback connections among the emotion components (Figure 1), can create a temporary disposition to have the same emotion that was initially caused when they were first manipulated (Lewis, 2005; Scherer, 2009; Siemer, 2005). Thus, appraising an event in a particular way increases the likelihood that subsequent events will be appraised in a similar manner, because the changes that cognitive appraisal processes bring about in the other emotion components not only feed forward, but also feed back into these same emotion components (Siemer, 2005). It follows that when appraisals of a certain kind happen more closely together, this facilitates the emergence of the associated emotional response (Roseman, 2011). For instance, if there are only a few goal-conducive events over a period of time, one might feel slightly positive, but when something obstructive

139

happens, o e s emotional state might be prone to change. However, if the rate of goal-conducive events increases, positive emotion will emerge in a way that is more intense, and less prone to negative influences (Lewis, 2005; Roseman, 2011). Therefore, a certain rate of goal-conducive events is likely also to be necessary to cause a sufficiently strong emotional response for our approach to be effective.

It could be argued that interactive systems that are designed to hack into the cognitive appraisal processes that form part of emotions, exist implicitly in many different types of technologies. That is, it is unlikely that the use of emotion induction techniques from psychology (section 2.3.1.1), affective mirrors (section 2.3.1.3), and ways of mimicking social interactions (section 2.3.1.4) do not rely on some form of appraisal process to enable emotions to be caused. We believe, however, that our approach is closer to technologies that target reward and punishment, such as gaming technologies, which are conceptually closely related to the processes underlying appraisal theory (Koster, 2013). For instance, a popular approach to designing games is to set a minimal amount of points that need to be scored, as the goal that needs to be met to advance in the game. The accumulation of points scored throughout the game informs the appraisal of the progress of the player towards his or her goals, and causes positive emotion accordingly (Järvinen, 2007). Similar approaches have been taken outside the context of games, such as in the design of positive technologies (Calvo & Peters, 2014), persuasion, and generally technologies that aim to change behaviour (Eslambolchilar & Rogers, 2013). This indicates that interactive systems can be used to hack into the cognitive appraisal processes that form part of emotions. However, technologies that explicitly target appraisal processes with the goal to cause emotion, are relatively

140

rare such as (cf. van Reekum et al., 2004). No interactive systems currently exist that explicitly attempt to cause emotion, rather than induce emotion in a more indirect manner (section 2.3.3.1), to influence the emotion-creativity link.

In this chapter we develop such a technology, by explicitly enabling an interactive system to manipulate the cognitive appraisal processes that cause and differentiate positive and negative emotions during idea generation.