The Effects of Inanimate Emotional Expressions on Liking and
Emotional Contagion
Name: Rosa Latour
Student number: 10526625
Supervisor: G.A. van Kleef
Date: 21-4-2019
Word count: 6033
MSc Social influence | Faculty of Social and Behavioral Sciences
Table of Contents
Abstract 3
The Effects of Inanimate Emotional Expressions on Liking and Emotional Contagion 4
Method 10 Participants 10 Materials 10 Procedure 13 Design 14 Results 15 Manipulation check 15 Reliability analyses 16 Main analyses 17 Explorative analyses 20 Discussion 22 References 27 Appendices 33
Appendix I – Interpersonal Reactivity Index: Subscale Perspective-taking 33 Appendix II – Translations of Discrete Emotions Questionnaire in Dutch 34 Appendix III – Discrete Emotions Questionnaire: Happiness subscale 35 Appendix IV – Discrete Emotions Questionnaire: Anger subscale 36
Abstract
Humans have the deep-rooted tendency to recognize and understand emotional expressions. This study examined whether this tendency, to recognize and react to emotions, also translates to inanimate objects, and thereby evokes similar interpersonal effects. To examine this issue, a questionnaire study was conducted, whereby
participants were exposed to an inanimate emotional expression. 157 participants engaged in this study and were randomly assigned to one of the following conditions: happy-, anger-, or control condition. The amount of liking and the occurrence of emotional contagion were measured. Moreover, a moderating effect of perspective-taking was explored, to see if participants with high perspective-perspective-taking ability were more prone to emotional contagion. Results showed that participants reported more liking towards a happy inanimate emotional expression, compared to anger- or control condition. However, no evidence for emotional contagion was found, regardless of taking perspective-taking into account. Future research could examine different inanimate objects.
The Effects of Inanimate Emotional Expressions on Liking and Emotional Contagion
Emotions are a thoroughly studied concept within psychology. Traditionally, emotions are examined as intrapersonal sensations and reactions, that is, as an internal state that has impact on someone’s own behavior (e.g. Van Kleef, De Dreu & Manstead, 2004). While most people comprehend what an emotion is, they often have a hard time defining it. In this study, the used definition of an ‘emotion’ is a reaction to relevant events, demonstrated by a particular expression, physiological change, experience, and/or behavioral tendencies, related to an object or cause (Van Kleef, 2016). In addition to the research about intrapersonal reactions, there is an increasing line of research that shows that emotions have a strong communicative function and that they are often caused by interpersonal situations (e.g. Parkinson, 1996), for instance, an emotional reaction evoked by another person. This development in research indicates that there is an association with an object or cause, which provides evidence for the existence of interpersonal effects of emotions. This interpersonal perspective complies with Darwin’s theory of emotion (1872; 1998), which states that emotional expressions are evolutionarily adaptive. Additionally, infants are able to recognize and understand emotional expressions of their caretakers, and are therefore capable of using emotional expressions as an informative source (Repacholi, 1998; Tronick, 1989). This
phenomenon is called social referencing (Sorce, Emde, Campos & Klinnert, 1985). In other words, social referencing is the phenomenon that infants use emotional cues to understand and evaluate ambiguous situations (Sorce et al., 1985). Another illustration that confirms this evolutionary perspective, is research indicating that apes have partly similar emotional expressions as humans, and thereby are capable of understanding primary human emotions, like joy, anger and fear (Buttelmann, Call & Tomasello, 2009;
Parr, Hopkins & De Waal, 1998). Besides, apes have the ability to use emotional
expressions of other apes as a source of information, which is beneficial for coordinating and solving their social problems (Buttelmann, Call & Tomasello, 2009; De Waal, 2009). Another study that provides evidence for emotional expressions as communicative source, is research that shows that bowlers show more emotional expressions when they are facing towards the audience, compared to facing towards the pins (Kraut & Johnston, 1979).
As has been mentioned, there is evidence for emotional expressions serving an evolutionary adaptive function. On top of this, people have the deep-rooted tendency to express emotions to other people, as mentioned in the illustration of bowlers. (Ekman & O’Sullivan, 1991; Parkinson, 2005). Moreover, research shows that the expression of emotions serve several social interpersonal functions (Van Kleef, 2016), such as
managing effective social interactions (Fischer & Manstead, 2016) or solving problems that occur with cooperation and commitment (Keltner, Haidt & Shiota, 2006). There is sufficient evidence linking positive interpersonal effects to recognizing and expressing emotions. However, research that examines these effects on inanimate emotional expressions is still missing. Inanimate objects are defined as lacking any characteristics and/or qualities of living organisms (Collins English Dictionary, 2012), that is, lifeless objects without any sign of awareness. Examining inanimate emotional expressions could complement the existing research on emotions. Moreover, it could give a clarification on how influencing unconscious behavior works and is used in the commercial sector. When similar interpersonal effects would occur when inanimate objects show an emotional expression, even though they have no awareness,
subconscious or feelings, this could have several practical applications. For instance, brands or stores could steer people to certain products, evoking liking by adding a
positive emotional expression. Liking of a product is positively associated with purchase intention (Lee, Kim, Seock & Cho, 2009; Solheim & Lawless, 1996), therefore, a possible effect could increase sales of a product. Likewise, if a phenomenon as emotional
contagion also occurs with inanimate emotional expressions, this could be applied in the commercial sector by evoking positive emotional states of consumers.
Research shows that people have the tendency to project human-like attributes to geometric figures (Heider & Simmel, 1944). In the study of Heider and Simmel, almost all participants describe the movements of the geometric shapes as behaviors of animate beings. This phenomenon is called anthropomorphism and refers to the tendency to attribute humanlike characteristics, intentions, emotions or motivations to inanimate objects (Epley, Waytz & Cacioppo, 2007). Anthropomorphism is often considered an automatic process (Guthrie, 1993). However, recent studies explain that there are individual differences in people in why and when they are likely to anthropomorphize (Epley et al., 2007). Knowledge about human behavior and characteristics, lacking social contact to other humans and motivation to understand the behavior of other’s are identified as factors which stimulate anthropomorphism. These findings indicate that humans could project human-like characteristics on an inanimate object, when the right circumstances are present.
The aim of this study is to determine whether inanimate emotional expressions serve similar effects as human emotional expressions. Considering the existence of specific brain regions to process different emotional expressions (e.g. Batty & Taylor, 2003), it can be assumed that humans have a hardwired tendency to recognize
emotional expressions. The processing of emotional expressions could be ingrained to such an extent, that it also translates to inanimate emotional expressions, for instance,
more experienced liking towards objects with an emotional expression, compared to neutral objects, or the occurrence of emotional contagion after being exposed to inanimate emotional expressions.
First, this study will examine whether there is a positive association between liking and inanimate emotional expressions. Liking will be used interchangeably with having a positive attitude (Greenwald, Brock & Ostrom, 2013). Human emotional expressions have several effects on liking, for instance, people tend to like others more when they express happiness (Fischer & Manstead, 2016; Van Kleef, De Dreu, &
Manstead, 2010). Moreover, an increased liking is associated with eliciting cooperative behavior in the opponent (Van Kleef, 2006; Van Kleef, De Dreu & Manstead, 2004) and an increase in willingness to make concessions (Baron, 1990). On the other hand,
negative emotions, such as anger, elicit negative affective reactions in the opponent, and thereby less liking of the opponent (Van Kleef et al., 2009). This study will examine if these findings, more liking with expressions of happiness and less liking with
expressions of anger, are similar when examining inanimate emotional expressions.
Second, if it is assumed that recognizing emotional expressions is hard-wired in the human brain, it is plausible that inanimate emotional expressions, just like animate emotions, are contagious (e.g. Jung, 1968). Emotional contagion is defined as the (automatic) inclination to synchronize and mimic another person’s facial and posture expressions and therefore establish emotional convergence (Hatfield, Cacioppo, Rapson, 1993). More specifically, this process could occur without any awareness of adopting the emotion. A known application of emotional contagion can be found in television shows, the canned laughter when a joke is being made (Niedenthal & Ric, 2017), which even appears to work (Bush, Barr, McHuho & Lanzetta, 1989). In the laboratory, research has
shown that participants in a negotiation adopted as well positive (happiness) as negative (anger) affective states of the confederate (Barsade, 2002). Furthermore, research suggest that, in groups, negative emotions, such as anger, may be more
contagious than positive emotions, such as happiness (Spoor & Kelly, 2004). In contrast, other studies suggest that positive emotions are more strongly associated with
emotional contagion, compared to negative emotions (e.g. Bhullar, 2012). Translated to inanimate emotional expressions, it is interesting to examine if positive or negative emotions are more contagious.
There are individual differences in the degree people are prone to emotional contagion, meaning not all humans are equally receptive of other people’s emotions (Niedenthal & Ric, 2017). With this in mind, the susceptibility of taking over emotions could depend on their ability to switch between perspectives. Over time, people develop the ability to switch between perspectives (Urberg & Docherty, 1976), where they could implement non-egocentric behavior (see Piaget’s Stage theory, 1970). This perspective-taking ability facilitates the anticipating of other people’s reactions and could therefore increase rewarding interpersonal relationships (Davis, 1983). When people have a higher perspective-taking ability, it could be assumed that they have higher
susceptibility for emotional contagion. Put differently, when someone could easily switch between perspectives, they are better in empathizing someone’s emotions and emotional contagion. The question remains, if these findings are applicable to inanimate objects as well. Research has shown that humans are capable of empathizing with inanimate objects, under the condition that the inanimate object have some humanlike characteristics, but are not excessively humanlike (Misselhorn, 2009). Accordingly, there could be perspective-taking with the inanimate objects in this study, since they express humanlike emotions, but no human bodily characteristics. It is expected that liking is not
dependent of the perspective-taking ability, because experienced liking is not associated with the ability to switch to the other person’s perspective.
In conclusion, a lot of research has been done about the intra- and interpersonal effects of emotions. For instance, people are liked more by others when they express positive emotions, compared to negative emotions. Furthermore, emotions could be contagious. However, there is no evidence whether these effects also apply to inanimate emotional expressions. Research shows that people have a hard-wired tendency to recognize emotional expressions, therefore it could be possible that humans
unknowingly assign similar effects to inanimate objects. This study will test the following hypotheses: (1) There is more liking of the inanimate object in the happy condition, compared to the anger condition and the control condition. (2a) There is no difference in mood prior to exposure to the stimulus; (2b) Participants in the happy-condition will experience more happiness at the second measurement of emotional contagion, compared to anger- and control condition; (2c) Participants in the anger condition will experience more anger at the second measurement of emotional contagion, compared to happy- and control condition. (3) Respondents with high perspective-taking ability tend to show more emotional contagion, compared to low perspective-taking ability. This study focuses on the effects of inanimate emotional expressions, whereby the spill-over effects of liking and emotional contagion are examined.
Method
Participants
A total of 157 participants1 (45 men, 111 women) participated in this study, aged
between 18 and 67 years (M = 23.96, SD = 9.55). The majority of the participants (61.8%) described themselves as Dutch; subsequently another European country (29.3%); North-America (1,3%); Asia (1,3%) or other (5.7%). Undergraduate
psychology students who participated (n = 84, 53.5%) received compensation in the form of research credits for their participation. The following exclusion criteria were applied. Firstly, participants who reported to have insufficient understanding of the English language were excluded from the study, since the study was conducted in English. Secondly, participants who failed the manipulation check were excluded from the study, as they did not recognize the assigned emotional expression.
Materials
Questionnaire. The program used to conduct the online questionnaire is
Qualtrics Lab (© 2019, Qualtrics®). Access to the questionnaire was granted through an
online link, which enabled participants to fill in the questionnaire at any time or location.
Stimuli. Photoshop software was used to modify inanimate emotional
expressions on a watering can, giving it either a happy emotional expression, an angry
1
This study was part of a joint data collection. The total data collection (n = 420) consisted of eight conditions, whereof three conditions are included in this study. The “stuffed-animal study” (containing four conditions) and the sad condition are not included in this study. Furthermore, only the subscale ‘perspective-taking’ of the Interpersonal Reactivity Index is included, the other subscales (fantasy; emphatic concern; personal distress) are not discussed in this study.
emotional expression, or no expression (control condition). Research on facial expressions (e.g. Ekman & Friesen, 2003) was used to simulate the appropriate expressions. The use of a watering can as inanimate object was chosen because it is a familiar object and moreover, an object that is presumably not strongly positively or negatively evaluated. Prior to the study, a short pilot was conducted to test if the modified emotional expressions were recognizable. The emotional expressions were well recognized and are presented in Figure 1a-c.
Figure 1-a. Happy emotional expression. Figure 1-b. Angry emotional expression,
Figure 1-c. Control condition.
Attitude Semantic Scale. Liking was measured by conducting the Attitude Semantic Scale (ASS). Similar studies using the ASS (Spears & Singh, 2004) reported a high reliability of α = .95. Five opposite adjectives were evaluated on a 7-point bipolar scale, ranging from unpleasant-pleasant; unfavorable-favorable; unappealing-appealing; bad-good; unlikeable-likeable (Appendix V) . Two out of five items were reverse-coded,
to control for response tendencies. A higher score on the ASS implied a more positive attitude towards the inanimate object, indicating higher liking (Greenwald, Brock & Ostrom, 2013).
Discrete Emotions Questionnaire. Emotional contagion, that is, if the
participant unconsciously comes to experience the emotion that was expressed by the inanimate object, was measured by a shortened version of the Discrete Emotions Questionnaire (DEQ; Harmon-Jones, Bastian, Harmon-Jones, 2016). The original DEQ consists of 32 items, measuring eight distinct emotions (anger, happiness, sadness, disgust, anxiety, fear, relaxation, and desire), and has a satisfactory internal reliability of 0.85-0.97 (Harmon-Jones et al., 2016). For this study, only the anger (α = .95) and
happiness (α = .96) subscales are used, as shown in Appendices III and IV. Hence, 8 items were administered (four for happiness and four for anger), using 7-point Likert scales ranging from 1 (= not at all) to 7 (= an extreme amount), to measure participants' current emotional state. Dutch translations were provided in parentheses (Appendix II) to enable better understanding for Dutch participants. The DEQ was administered twice; before exposure to the inanimate object (to establish a baseline) and after exposure to the inanimate object. This baseline analysis was conducted to make sure that potential effects were due to the exposure to the stimulus and not present prior to the exposure. When participants in the happy condition did not report more happiness on the first measurement (compared to the anger- and control condition), but did report
significantly more happiness on the second measurement, the occurrence of emotional contagion is inferred. Likewise, when participants in the anger condition did not report more anger on the first measurement (compared to the happy- and control condition), but reported significantly more anger on the second measurement, emotional contagion is inferred.
Interpersonal Reactivity Index. Perspective-taking was measured by a subscale of the Interpersonal Reactivity Index (IRI; Davis, 1980), which has a satisfactory internal reliability of 0.75-0.78. Perspective-taking refers to the ability to adopt another one’s point of view in daily life, that is, the ability to shift between perspectives (Davis, 1980; 1983). An example item includes “I sometimes try to understand my friends better by imagining how things look from their perspective”, the complete scale is presented in Appendix I. Perspective-taking is measured by a 7-item subscale, administered on 5-point Likert scales ranging from 1 (= does not describe me at all) to 5 (= describes me very well). A higher score on the subscale implied a higher perspective-taking ability. Procedure
The study was approved by the Ethics Committee of the University of Amsterdam (UvA). Participants accessed the questionnaire online, either by using a link or through the UvA Lab website. After a short introductory information brochure, the participants were asked to agree to the informed consent form. All participants were then randomly assigned to one of the three conditions: happy condition (n = 52), anger condition(n = 52), or control condition (n = 53). To measure the current emotional state of the
participants, the DEQ was administered. Afterwards, the perspective-taking ability was measured by the IRI. Then, the participants were presented with either a happy, angry, or neutral watering can image, whereby a timer of five seconds was set on this page to prevent the participants from clicking further without observing the image. The ASS was conducted to measure the amount of liking towards the object, followed by the second measurement of DEQ, to see if there is a difference in emotional state after being exposed to the image. Next, a manipulation check was administered, asking the participants which emotion they recognized on the watering can. To conclude,
and occupation, followed by a question about their mastery of English and a voluntary option to fill in their e-mail. A monetary reward (thirty euros) was raffled off among the participants who filled in their e-mail. Participants were thanked for their participation and their responses were recorded.
Design
As previously stated, three experimental conditions were included in this study. Liking and emotional contagion were included as dependent variables. An additional dependent variable was perspective-taking, as a possible moderator on emotional
contagion. The relation between the inanimate emotional expressions and liking was not expected to be moderated by perspective-taking, as liking is expected to occur without taking one’s perspective in a situation. First, a one-way ANOVA on liking was conducted, whereby it was expected that participants in the happy condition reported higher ASS scores, compared to the participants in the anger- or control condition. Second, a one-way ANOVA was conducted to analyze emotional contagion, whereby the anger-subscale and happiness-subscale of DEQ were analyzed separately. Additionally, variance
analyses were conducted for the first and second measurement of DEQ. On the first measurement of DEQ, no significant differences between conditions were expected. However, on the second measurement, differences between conditions were expected: participants in the happy condition were expected to report higher on the happy subscale; participants in the anger condition were expected to report higher on the anger subscale. Finally, a moderation analysis was conducted to examine whether perspective-taking moderated the effect of inanimate emotional expression on emotional contagion. Expected was that perspective-taking would interact with the inanimate emotional expression on emotional contagion, whereby more emotional contagion would occur when the perspective-taking ability of participants was higher.
Results
The analyses were performed using Statistical Package for the Social Science (SPSS; version 24), whereby p-values lower than 0.05 were considered statistically significant. For comparisons of equality of gender distribution across conditions, a chi-square test was conducted, Χ2 (4) = 11.63, p = 0.02, which indicated gender was not
equally distributed over conditions. Therefore, gender was included as covariate in the main analysis, which indicated that gender was not a significant predictor. In the following analyses, gender was not included, since it was no significant predictor. To compare the mean age between conditions a one-way ANOVA was conducted, F (2, 154) = 0.45, p = .64, which suggested that age was equally distributed over conditions.
Manipulation check
Participants had to indicate on 7-point Likert scales whether they identified the inanimate emotional expression as happy, angry, or neutral. An analysis of variance on this manipulation check was conducted, to verify if the manipulation was successful. Descriptive statistics of the different reported emotions per condition are shown in Table 1.
Table 1
Means and Standard Deviations of the reported Emotions on the Manipulation Check Reported happiness Reported anger Reported as neutral
M SD M SD M SD
Happy condition 6.44 0.75 1.13 0.49 2.31 1.38 Anger condition 1.27 0.63 6.06 1.39 1.85 1.23 Control condition 3.49 1.73 1.83 1.28 5.40 1.72
The results indicated that participants in the happy condition identified the stimulus significantly more as happy, compared to anger- and control condition, F (2,154) = 264, p < .001. Likewise, participants in the anger condition identified the stimulus significantly more as angry, compared to happy- and control condition, F (2, 154) = 290.05, p < .001. Moreover, participants in the control condition identified the stimulus significantly more as neutral, compared to happy- and anger condition, F (2, 154) = 91.96, p < .001. These results were in line with expectations, and suggested that the manipulation was successful.
Reliability analyses
Prior to the main analyses, reliability analyses were conducted over each of the used scales. All the scales indicated satisfactory reliabilities, as shown in Table 2, accompanied by overall mean scores on the scales.
Table 2
Overall Means, Standard Deviations and Calculated Internal Reliability (Cronbach’s α) of the used Scales
Name questionnaire Subscale Mean SD Cronbach’s α
Attitude Semantic Scale - 4.13 1.26 .90
DEQ 1st measurement Happiness 3.90 1.25 .90
DEQ 1st measurement Anger 1.52 0.95 .85
DEQ 2nd measurement Happiness 3.73 1.32 .92
DEQ 2nd measurement Anger 1.48 0.96 .87
IRI Perspective-taking 3.69 0.57 .75
Main analyses
Based on aforementioned exclusion criteria, no participants were excluded from the study. The data of all participants (n = 157) was included for further analyses. The participants were randomly assigned to the happy condition (n = 52), anger condition (n = 52 ), or control condition (n = 53).
Liking. Prior to the analysis, the corresponding assumptions were verified2.
Results of a one-way ANOVA indicated a significant effect of condition on liking, F(2, 132.22) = 17.63, p < .001, ηp2 = .19. As expected, contrasts indicated that participants in
the happy-condition (M = 4.72, SD = 1.40) reported significantly more liking towards the inanimate object, compared to the participants in the anger condition (M = 3.41, SD = 1.12) and control condition (M = 4.25 , SD = 0.83), t(76.86) = 4.13, p < .001, as shown in Figure 2.
Figure 2. The mean scores of Attitude Semantic Scale, sorted by Condition.
2 Only the assumption of homogeneity of variance was violated, F (2,154) = 7.13, p = .001.
Therefore, alternative F statistics and corresponding degrees of freedom (Brown-Forsythe test) were reported.
Emotional Contagion. Prior to the analysis, the corresponding assumptions were verified. The mean scores and standard deviations of DEQ on the first and the second measurement were calculated for each condition, as shown in Table 3a-b.
Table 3a-b
Means and Standard Deviations of Happy scale on first and second measurement of DEQ (left) and Means and Standard Deviations of Anger scale on first and second measurement of DEQ (right). Condition Happy-scale Mean DEQ 1 SD DEQ 1 Mean DEQ 2 SD DEQ 2 Condition Anger-scale Mean DEQ 1 SD DEQ 1 Mean DEQ 2 SD DEQ 2 Happy condition 3.74 1.35 3.80 1.34 Happy condition 1.59 0.99 1.46 0.91 Anger condition 4.01 1.24 3.64 1.38 Anger condition 1.47 0.79 1.51 0.93
Control 3.95 1.18 3.75 1.27 Control 1.50 1.05 1.47 1.06
Note. DEQ 1 = first measurement; DEQ 2 = second measurement.
Several analyses were conducted to analyze emotional contagion in the happy and anger condition. The first measurement of DEQ was analyzed by conducting a one-way ANOVA. As expected, results showed that there were no significant differences of happiness on the first measurement, F(2, 154) = 0.68, p = .51, ηp2 = .009. Likewise, there
were no significant differences of anger on the first measurement, F(2,154) = 0.22, p = .81, ηp2 = .003. This indicated that participants did not significantly differ in emotional
state, prior to exposure to the stimulus. Surprisingly, at the second measurement, there were no significant differences either; participants in the happy condition did not report significantly more happiness after being exposed to the stimulus, F(2,154) = 0.21, p = .81, ηp2 = .003. Also, there were no significant differences in the anger condition at the
are no considerable differences of happiness or anger between conditions, which was not in line with the hypotheses.
Figure 3. The mean scores of DEQ on the second measurement; separated lines for Happiness- and Anger subscale.
Perspective-taking. A moderation analysis (factorial ANOVA) was conducted to determine if there is a moderating effect of perspective-taking on emotional contagion. Results showed that there was no significant main effect of condition on happiness, F (2,154) = 1.52, p = .22, ηp2 = .020, which implied that condition was not a significant
predictor. Also, there was no significant main effect of perspective-taking on happiness, F (1,154) = 1.91, p = .17, ηp2 = .012. Opposed to the hypothesis, there was no significant
interaction effect between perspective-taking and emotion condition on happiness, F(2,154) = 1.54, p = .22, ηp2 = .020.
Moreover, results indicated that there was no significant main effect of condition on anger, F (2,154) = 1.49, p = .23, ηp2 = .019, which implied that condition is not a
significant predictor. Also, no significant main effect of perspective-taking on anger subscale was found, F (1,154) = 0.21, p = .65, ηp2 = .001. Opposed to the hypothesis, there
was no significant interaction effect between perspective-taking and emotion condition on anger, F (2,154) = 1.58, p = .21, ηp2 = .020.
Explorative analyses
In addition to the directional hypotheses tests reported above, exploratory analyses were conducted. Firstly, a moderation analysis was conducted with age as a moderating variable on the second measurement of DEQ. Although no moderation effect was found, a marginally significant3 main effect was found on happiness, b = 0.02, t
(153) = 1.90, p = 0.059, which implied that, on average, when age increases, happiness also increases.
Secondly, the dataset was split into two groups: the student-group (n = 84), who received research credits for their participation, and the nonstudent-group (n = 73) who did not receive any compensatory reward. In the student-group, the majority of the participants was female (n = 69, male = 14) and had a mean age of 20.64 (SD = 3.46). Furthermore, the majority of participants originated from another European country (51.2 %, n = 43), subsequently the Netherlands (36,9%, n = 31). In the nonstudent-group, a slight majority was female (n = 42, male = 31) with a mean age of 27.78 (SD = 12.44). The vast majority (90.4%) of the nonstudent-group originated from the
Netherlands. Given the considerable demographic differences between the student- and nonstudent-group, additional moderation analyses were conducted over the separate groups. For the student-group, no significant interaction effect of perspective-taking was found on happiness, F (2,81) = 0.38, p = .68, ηp2 = .010. Furthermore, in the
3 P-values above 0.05 but below the level of 0.10 are seen as marginally significant, or close to a
group, no significant interaction effect was found on anger, F (2,81) = 0.20, p = .82, ηp2 =
.005.
However, in the nonstudent-group, a significant main effect of perspective-taking on happiness was found, F (1,70) = 5.21, p = .026, ηp2 = .072, whereby a small positive
coefficient (r = .07) suggested that when perspective-taking increases, happiness increases. Moreover, a marginally significant main effect of condition on anger was found, F (2,70) = 2.69, p = .075, ηp2 = .074, More importantly, a marginally significant
interaction effect between perspective-taking and emotion condition on anger was found, F (2,70) = 2.88, p = .063, ηp2 = .079. To further examine this interaction, a
scatterplot is added, as shown in Figure 4. The scatterplot showed a negative effect (r = -.26) of perspective-taking on anger in the happy condition and a positive effect (r = .30) of perspective-taking on anger in the control condition. Surprisingly, there seems to be no effect (r = -.007) of perspective-taking on anger in the anger condition. These findings implied that participants in the happy condition experienced less anger as they reported higher perspective-taking ability. Also, participants in the control condition reported more anger as they reported higher perspective-taking ability.
Figure 4. Interaction between Perspective-taking and Condition on Anger subscale on the Second measurement.
Discussion
This study examined the effects of inanimate emotional expressions, in particular the effects of liking and emotional contagion. A possible moderating effect of
perspective-taking on emotional contagion was also explored. Hypothesized was that an inanimate object is liked more when it expresses happiness, compared to expressions of anger or no expressed emotion. This hypothesis was supported: participants reported more liking towards a watering can with a happy emotional expression, compared to an angry emotional expression and a neutral watering can. There were two measurements of emotional contagion in this study. It was hypothesized that participants did not differ in emotional state prior to exposure to the inanimate object. This was indeed the case, indicating that participants did not differ in emotional state over the different
conditions. However, analyses on emotional contagion on the second measurement indicated that there were no differences between conditions either, which was not in
line with the hypothesis. There were no significant differences on either the happiness or the anger subscale. This suggests that the exposure to either a happy, anger, or neutral object did not affect the emotional state of the participants. Moreover, adding perspective-taking as a moderating variable did not produce a significant effect on either the happy or anger subscale. These findings were not in line with the hypotheses.
The most noteworthy finding of this study is the higher amount of liking towards an inanimate object with a happy emotional expression, compared to angry or neutral object. This finding could be beneficial to apply in the commercial sector, for instance in marketing or in department stores. When marketeers use positive emotional
expressions on their products in, for instance, an advertisement, the object is presumably liked more, which could in turn lead to a positive association with the brand. Research has shown that high amount of liking increases the mood and therefore increases the willingness to spend money. Therefore, by adding a positive emotional expression on a certain product, it could lead to a higher willingness to spend money, which in turn could lead to a higher turnover for the brand. Moreover, as previously mentioned, experienced liking towards a product is associated with purchase tendency (Lee, Kim, Seock & Cho, 2009). This could be applicated in marketing as well, but also in for instance, supermarkets. A supermarket chain could steer people into buying certain products, by adding a positive emotional expression. All in all, the consumer could be inclined to prefer a certain brand or product over other brands or products in future purchases.
There was no evidence in this study for the occurrence of emotional contagion of inanimate objects. A possible explanation for why this effect was not found, is that a large amount of the participants (53.5 %) were undergraduate Psychology students.
Psychology students are familiar with theories of emotion and the phenomenon of emotional contagion. It could be that they were aware of the manipulation this study tried to implement, which is why the manipulation did not work. The number of questions that were asked about the inanimate object could excite excessive thinking about which manipulation this study tried to implement. The conscious thinking about the object, could lead to the realization of being influenced. Dual-process models state two possible routes: the unconscious, implicit route is unreflective and effortless and the conscious route is explicit, intentional and effortful (Erisen, Lodge & Taber, 2014). This could also be explained by the EASI theory of Emotion (Van Kleef, De Dreu, Manstead, 2010), which explains the distinction between affective and inference processes. Inferential processes are more deliberate and affective processes are relatively more automatic. According to the EASI theory, a phenomenon as emotional contagion occurs more prevalent when someone uses affective processes, compared to inferential
processes, since inferential reactions are often less driven by someone’s own emotional state. This is not to say that, when someone uses the inferential route, emotional
contagion could not occur at all. However, emotional contagion is less prevalent when someone uses the inferential route and it is likely to be overruled by strategic inferences (Van Kleef et al., 2010). This presumably resulted in no effect of emotional contagion.
A second explanation of why emotional contagion did not occur is that the time between exposure to the inanimate object and the second measurement could be too long. After exposure to the stimulus, the participant had to answer three questionnaires, before the (second) measurement of emotional contagion. It could be that emotional contagion occurred at first, but that it faded away while answering the other
questionnaires. A third explanation is the used stimuli. The underlying idea is that people are susceptible to emotions of their surroundings and therefore emotional
contagion occurs. It could be that the used inanimate emotional expressions were not ‘contagious’ enough, since it is a flat 2D picture. In future research, emotional contagion could be measured by using a moving stimulus (e.g. short movie clip), instead of just a picture.
As mentioned earlier, there are stable individual differences in whether people are likely to project human-like characteristics to inanimate objects (Epley, Waytz & Cacioppo, 2007). Different factors determine whether it is more or less likely that anthropomorphism occurs. It could be that in this study, the motivation to understand other’s behavior was not high enough to stimulate anthropomorphism. Therefore, the likelihood of emotional contagion decreases. In future research, an additional factor could be added, which increases the motivation to anthropomorphize, for example, ask participants to imagine a story that the inanimate object went through.
Opposed to the hypothesis, no interaction effect of perspective-taking was found. An explanation for this, could be that the use of the IRI to measure perspective-taking ability could be misplaced. Other studies that examined the behavioral measures of perspective-taking, for instance the ‘imitation task’ (Newey, Koldewyn, Ramsey, 2019), showed a poor correlation with the IRI. It could be that the self-reported perspective-taking ability and behavioral perspective-perspective-taking ability tap into different processes. A high perspective-taking ability on the IRI could possibly not generalize to a high perspective-taking ability in behavioral settings. Although no effect of emotional contagion was found on the entire data-set, when the data-set was split into two
separate groups (students vs. non-students) a marginally significant interaction effect of perspective-taking was found on the anger subscale of emotional contagion. More specifically, this effect was only found in the nonstudent-group. Surprisingly, in the
found interaction effect, participants in the control group reported more anger when they had a higher perspective-taking ability. This was not hypothesized, since the control group was exposed to a neutral watering can, no emotional contagion was expected to occur. An explanation for this finding could be that participants were
irritated because they had to look at a neutral watering can for some time and then were asked to answer several questions about it. Irritation could lead to higher experienced anger on the subscale.
Several limitations and suggestions for future research are mentioned above. Another interesting matter to examine in further research, is the use of different inanimate objects. In this study, only a watering can is examined and presented to participants. It could be interesting to examine a variation of different objects in a follow-up study, where a distinction could be made between objects that are manmade and natural objects. Another suggestion for future research could be conducting a facial EMG to measure contagion, instead of a self-report scale. When participants are looking at a smiling watering can in the laboratory, it could be that their face muscles tell more than a self-report measure. Overall, there are sufficient notions that are worth looking into for possible follow-up studies.
All in all, this was a pioneer study in exploring if several functions of human emotions also translate to inanimate emotional expressions. Several aspects of the design could be improved in future research. For now, we could conclude that it is worth further examining if positive inanimate emotions will actually evoke more liking in the field, for instance, conducting a field study with happy and angry inanimate objects.
References
Baron, R. A. (1990). Environmentally Induced Positive Affect: Its Impact on Self-Efficacy, Task Performance, Negotiation, and Conflict 1. Journal of Applied Social
Psychology, 20(5), 368-384.
Barsade, S. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47, 644-675.
Batty, M., & Taylor, M. J. (2003). Early processing of the six basic facial emotional expressions. Cognitive Brain Research, 17(3), 613–620.
Bhullar, N. (2012). Relationship between mood and susceptibility to emotional contagion: is positive mood more contagious?. North American Journal of Psychology, 14(3).
Boyle, G. J. (1984). Reliability and validity of Izard’s differential emotions scale. Personality and Individual Differences, 5(6), 747–750.
Buck, R., Losow, J. I., Murphy, M. M., & Costanzo, P. (1992). Social facilitation an
inhibition of emotional expression and communication. Journal of Personality and Social Psychology, 63(6), 962.
Bush, L. K., Barr, C. L., McHugo, G. J., & Lanzetta, J. T. (1989). The effects of facial control and facial mimicry on subjective reactions to comedy routines. Motivation and emotion, 13(1), 31-52.
Buttelmann, D., Call, J., & Tomasello, M. (2009). Do great apes use emotional expressions to infer desires? Developmental Science, 12, 688–698.
Collins English Dictionary. (2012). British Dictionary definitions for inanimate. Retrieved January 15, 2019, from
https://www.dictionary.com/browse/inanimate.
Darwin, C. (1872). The expression of emotion in animals and man. London: Methuen.(1877), A biographical sketch of an infant. Mind, 2, 285-294
Darwin, C., & Prodger, P. (1998). The expression of the emotions in man and animals. Oxford University Press, USA.
Davis, M. H. (1980). A multidimensional approach to individual differences in empathy.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of personality and social psychology, 44(1), 113.
de Waal, F. B. M. (2009). The age of empathy: Nature’s lessons for a kinder society. New York: Harmony Books.
Ekman, P., & Friesen, W. V. (2003). Unmasking the face. A guide to recognizing emotions from facial clues. Los Altos, CA: Malor Books.
Ekman, P., & O’Sullivan, M. (1991). Who can catch a liar? American Psychologist, 46, 913– 920.
Elfenbein, H. A. (2014). The many faces of emotional contagion: An affective process theory of affective linkage. Organizational Psychology Review, 4, 326–362.
Epley, N., Waytz, A., & Cacioppo, J. T. (2007). On seeing human: a three-factor theory of anthropomorphism. Psychological review, 114(4), 864.
Erisen, C., Lodge, M., & Taber, C. S. (2014). Affective contagion in effortful political thinking. Political Psychology, 35(2), 187-206.
Fischer, A. H., & Manstead, A. S. R. (2016). Social functions of emotion and emotion regulation. In M. Lewis, J. Haviland, & L. Feldman Barrett (Eds.), Handbook of emotion (4th ed.). New York: Guilford.
Greenwald, A. G., Brock, T. C., & Ostrom, T. M. (Eds.). (2013). Psychological foundations of attitudes. Academic Press.
Gross, J. J., & Levenson, R. W. (1995). Emotion elicitation using films. Cognition & emotion, 9(1), 87-108.
Guthrie, S. E. (1993). Faces in the clouds: A new theory of religion. New York: Oxford University Press.
Harmon-Jones, C., Bastian, B., & Harmon-Jones, E. (2016). The Discrete Emotions Questionnaire: A New Tool for Measuring State Self-Reported Emotions. PLOS ONE, 11(8).
Hatfield, E., Cacioppo, J. T., & Rapson, R. L. (1993). Emotional contagion. Current directions in psychological science, 2(3), 96-100.
Heider, F., & Simmel, M. (1944). An experimental study of apparent behavior. The American journal of psychology, 57(2), 243-259.
Joiner, T. E. (1994). Contagious depression: Existence, specificity to depressed
symptoms, and the role of reassurance seeking. Journal of Personality and Social Psychology, 67, 287-296.
Jung, C. G. (1968). Lecture five: Analytical psychology: Its theory and practice(151–160). New York: Random House.
Kelly, J. R., Iannone, N. E., & McCarty, M. K. (2015). Emotional contagion of anger is automatic: An evolutionary explanation. British Journal of Social
Psychology, 55(1), 182–191.
Keltner, D., Haidt, J., & Shiota, M. N. (2006). Social functionalism and the evolution of emotions. In M. Schaller, J. A. Simpson & D. T. Kenrick (Eds.), Evolution and social psychology (pp. 115–142). Madison, CT: Psychosocial Press.
Kraut, R. E., & Johnston, R. E. (1979). Social and emotional messages of smiling: An ethological approach. Journal of Personality and Social Psychology, 37, 1539– 1553.)
Lee, Y., Kim, S., Seock, Y. K., & Cho, Y. (2009). Tourists' attitudes towards textiles and apparel-related cultural products: A cross-cultural marketing study. Tourism Management, 30(5), 724-732.
Misselhorn, C. (2009). Empathy with Inanimate Objects and the Uncanny Valley. Minds and Machines, 19(3), 345–359.
Newey, R., Koldewyn, K., & Ramsey, R. (2019). The influence of prosocial priming on visual perspective taking and automatic imitation. PloS one, 14(1), e0198867.
Niedenthal, P. M., & Ric, F. (2017). Psychology of emotion. Psychology Press.
Parkinson, B. (1996). Emotions are social. British journal of psychology, 87(4), 663-683.
Parkinson, B. (2005). Do facial movements express emotions or communicate motives? Personality and Social Psychology Review, 9, 278–311.
Parr, L. A., Hopkins, W. D., & de Waal, F. B. M. (1998). The perception of facial
expressions in chimpanzees, Pan Troglodytes. Evolution of Communication, 2, 1– 23.
Piaget, J. (1970). Piaget's theory.
Repacholi, B. M. (1998). Infants’ use of attentional cues to identify the referent of another person’s emotional expression. Developmental Psychology, 34, 1017– 1025.
Solheim, R., & Lawless, H. T. (1996). Consumer purchase probability affected by attitude towards low-fat foods, liking, private body consciousness and information on fat and price. Food Quality and Preference, 7(2), 137-143.
Sorce, J. F., Emde, R. N., Campos, J., & Klinnert, M. D. (1985). Maternal emotional signaling: Its effect on the visual cliff behavior of 1 year olds. Developmental Psychology, 21, 195–200.
Spears, N., & Singh, S. N. (2004). Measuring Attitude toward the Brand and Purchase Intentions. Journal of Current Issues & Research in Advertising, 26(2), 53–66. Spoor, J. R., & Kelly, J. R. (2004). The evolutionary significance of affect in groups:
Communication and group bonding. Group Processes and Intergroup Relations, 7, 401–415.
Stel, M., Van Baaren, R. B., & Vonk, R. (2008). Effects of mimicking: Acting prosocially by being emotionally moved. European Journal of Social Psychology, 38(6), 965-976.
Tronick, E. Z. (1989). Emotions and emotional communication in infants. American Psychologist, 44, 112–119.
Urberg, K. A., & Docherty, E. M. (1976). Development of role-taking skills in young children. Developmental Psychology, 12(3), 198.
Van Kleef, G. A. (2016). The interpersonal dynamics of emotion: Toward an integrative theory of emotions as social information. Cambridge, UK: Cambridge University Press.
Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2004). The interpersonal effects of emotions in negotiations: A motivated information processing approach. Journal of Personality and Social Psychology, 87, 510-528.
Van Kleef, G. A., De Dreu, C. K. W., & Manstead, A. S. R. (2010). An interpersonal approach to emotion in social decision making: The emotions as social information model. Advances in Experimental Social Psychology, 42, 45–96.
Van Kleef, G. A., De Dreu, C. K., & Manstead, A. S. (2004). The interpersonal effects of anger and happiness in negotiations. Journal of personality and social psychology, 86(1), 57.
Van Kleef, G. A., Homan, A. C., Beersma, B., Van Knippenberg, D., Van Knippenberg, B., & Damen, F. (2009). Searing sentiment or cold calculation? The effects of leader emotional displays on team performance depend on follower epistemic motivation. Academy of Management Journal, 52, 562-580.
Appendices
Appendix I – Interpersonal Reactivity Index: Subscale Perspective-taking The following statements inquire about your thoughts and feelings in a variety of situations. For each item, indicate how well it describes you on a 5-point scale, from 1=does not describe me at all, to 5=describes me very well. Read each item carefully before responding. Please answer as honestly as you can. Thank you.
1 2 3 4 5
Does not describe me at all
Describes me very well
1. I sometimes find it difficult to see things from the "other guy's" point of view. * 2. I try to look at everybody's side of a disagreement before I make a decision. 3. I sometimes try to understand my friends better by imagining how things look
from their perspective.
4. If I'm sure I'm right about something, I don't waste much time listening to other people's arguments. *
5. I believe that there are two sides to every question and try to look at them both. 6. When I'm upset at someone, I usually try to "put myself in his shoes" for a while. 7. Before criticizing somebody, I try to imagine how I would feel if I were in their
place.
Appendix II – Translations of Discrete Emotions Questionnaire in Dutch
Original DEQ Dutch translation
Anger Boos Satisfaction Tevreden Rage Woedend Happiness Gelukkig Mad Kwaad Enjoyment Blij Pissed off Geërgerd
Appendix III – Discrete Emotions Questionnaire: Happiness subscale
To what extent do each of the following words describe how you are feeling at the moment? Please indicate your response using the scale provided.
Satisfaction
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Happiness
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Enjoyment
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Liking
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Appendix IV – Discrete Emotions Questionnaire: Anger subscale
To what extent do each of the following words describe how you are feeling at the moment? Please indicate your response using the scale provided.
Anger
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Rage
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Mad
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Pissed off
1 2 3 4 5 6 7
Not at all Slightly Somewhat Moderately Quite a bit Very much An extreme amount
Appendix V – Attitude Sematic Scale
