The Influence of Emotional Change on
Person Perception
A STUDY TO THE EFFECT OF FACIAL EXPRESSIONS
CHANGE AND GENDER ON
JUDGEMENTS OF
DOMINANCE AND AFFILIATION
Name: L. Pijnacker
Student number: 10009248
University: University of Amsterdam Master: Psychology, Social Psychology Program: Master these
Supervisor: X. Fang
Abstract
Previous research on emotional expressions and person perception mainly focussed on single static emotions. However, people often shift from one emotional state to another. This study aims to identify the effect of emotional change on dominance and affiliation perception, and also investigates the mediation role of the gender of expressers. Here, we focussed on the emotional transitions fear-to-anger, disgust-to-anger, fear-to-disgust, to-disgust, anger-to-fear and disgust-anger-to-fear. Findings suggest that the transition of emotional expression to anger leads to the highest perceived dominance, followed by a transition to disgust. While a transition to fear leads to the lowest dominance. Furthermore, we found that the transition of emotional expression to anger leads to a lower perceived affiliation, and a transition to fear leads to the highest affiliation. The emotional change anger-to-disgust led to the lowest affiliation. Moreover, the affiliation perception for disgust was mediated by the first emotion (anchor emotion) in the emotional change. At last, female expressers had an overall higher dominance score than male expressers. This study provides novel insights into the effect of emotional change on person perception.
Keywords: person perception, trait inference, dominance, affiliation, dynamic emotions, emotional change
The Influence of Facial Expressions Change and Gender on Judgments of Dominance and Affiliation
Emotions are functional. According to the Emotion as Social Information (EASI) theory, emotional expressions influence the perception of cognitions, feelings, attitudes and behaviours about others (Van Kleef, De Dreu, & Manstead, 2010). Researchers found, in line with this theory, that emotional expressions could help people to form a personality impression (Hess, Blairy, & Kleck, 200; Montepare & Dobish, 2003). Previous research has shown that most trait inferences are made on two dimensions: dominance and affiliation (Wiggins, Phillips, & Trapnell, 1988; Todorov, Said, Engell, & Oosterhof, 2008). Much research has been done on the effect of emotional expressions on these personality traits. Unfortunately, most studies had a lack of ecological validity. This has to do with the fact that most previous research has been limited to the use of static emotional expressions. Actually, people in real life do not stick to one emotion but rather transfer from one emotional state to another, which is called emotional change. The present study expanded the existing literatures by examining the effects of emotional change on the perception of dominance and affiliation, and also investigated the mediation role of the gender of expressers. Before introducing hypothesis, this paper will first present important findings about person perception and the relation between emotions and gender with dominance and affiliation.
Person perception; dominance and affiliation
So, the interest of present research refers to the effect of dynamic facial expressions on person perception. Facial appearance is important to make a first impression of a person. Thereby, emotional expressions serve as an important cue to convey information about the personality of the expresser. Several researchers found evidence for this assumption (Knutson; 1996, Hess et al., 2000; Montepare & Dobish, 2003). These impression formations have important implications for the understanding of social interactions. Eliot Asch (1946) found
that the first impressions of someone’s personality traits are very dominant and has a great influence, in comparison with subsequently impressions. Actually, it is hard to correct your first impression. This is also known as the primacy-effect. For example, the first impressions of the personality of your new colleague will have a great impact on your future social interactions with him. If you think the person is warmth and friendly, you will like to cooperate with each other. On the other hand, if you think the person is dominant and rough, you do not want to cooperate with each other.
So everybody made up a first impression when meeting someone new. Wiggins and his colleagues have shown an interesting phenomenon regarding this interpersonal perception. They found that people evaluate others on two orthogonal dimensions, affiliation and dominance, which they called the circle of ‘interpersonal circumplex’ (see figure 1: Wiggins, 1979; Wiggins, Phillips & Trapnell 1988). This model of interpersonal perception is similar to the face evaluation model of Todorov, Said, Engell, and Oosterhof (2008). They found that faces are evaluated on two fundamental dimensions, trustworthiness and dominance, which are comparable with affiliation and dominance. Luxen (2005) refers to affiliation and dominance as ‘the ink with which human interaction is written’.
So, people evaluate faces on dominance and affiliation, but what does these traits tells us? Well, dominance refers to the extent in which a person is perceived as powerful, whereas affiliation refers to the extent in which a person is perceived as warmth and friendly (Wiggins, 1979). In present study we are interested in the relation between emotional expressions, specific in emotional changes, and trait inferences like dominance and affiliation. There is overwhelming evidence corroborating the notion that facial expressions are associated with a certain level of dominance and affiliation.
Anger, disgust and fear in relation to dominance and affiliation
Three emotions commonly related to dominance and affiliation are anger, disgust and fear. Angry and disgusted expressions are rated as high in dominance and low in affiliation (Hess, Blairy, & Kleck, 2000; Algoe, Buswell, & De Lamater, 2000; Knutson, 1996; Montepare & Dobish, 2003). Fearful expression is rated low in dominance. However, the ratings of affiliation of fearful expression are not consistent across research (Algoe, Buswell, & De Lamater, 2000; Hess, Blairy, & Kleck, 2000; Montepare & Dobish, 2003). When compared to anger or disgust, research agreed that affiliation ratings are higher for fearful expressions. Explanation for most of the findings derive from the different face gestures of emotional expressions, specific the mouth and brow configurations (Keating, 1985; Keating et al., 1981; Ekman & Davidson, 1994). Faces with thick and low eyebrows, expressing anger and disgust, are likely to be judged high on dominance. Faces with thin and high eyebrows, expressing fear, are likely to be judged low on dominance. Faces with an open, inverted mouth, like anger, disgust and fear, are likely to be judged low on affiliation.
However, the interest of present study refers to the effect of emotional changes rather than the effect of single emotions. Thereby, present study’s hypotheses focus on the target emotion of the emotional change. The target emotion infer information about the ‘emotional end-state’ of the expresser and therefore we belief that the target emotion will have a greater impact on the person perception than the anchor emotion. We are interested in the effect of the target emotion on judgements of dominance and affiliation. Besides looking between emotional changes with different target emotions, it is also interesting to look within emotional changes with the same targets. Therefore, we can explore whether there is a role of the context of the emotional change, where context refers to the anchor emotion. Because we do not expect any role of the context (because of the overwhelming impact of the target emotion), we hypothesize that there will be no differences within emotional changes with the
same targets. Despite that, it is still interesting to explore, because it is yet an unexplored field.
Based on previous findings and the mouth- and face gestures, we hypothesize that disgust-to-anger and fear-to-anger will be judged as high in dominance and low in affiliation. Furthermore, we hypothesize that emotional changes with target emotion anger (either disgust-to-anger or fear-to-anger ) will be judged as higher in dominance and lower in affiliation compared with emotional changes with fear as the target emotion (either anger-to-fear or disgust-to-anger-to-fear) (hypothesize 1a).
Second, we hypothesize that anger-to-disgust and fear-to-disgust will be judged as high in dominance and low in affiliation. Furthermore, we hypothesize that the emotional changes with target emotion disgust (either anger-to-disgust or fear-to-disgust) will be judged as higher in dominance and lower in affiliation compared with emotional changes with fear as the target emotion (either anger-to-fear or disgust-to-fear) (hypothesize 1b).
At last, we hypothesize that anger-to-fear and disgust-to-fear will both judged low in dominance and high in affiliation. Furthermore, we hypothesize that the emotional changes with target emotion fear (either anger-to-fear and disgust-to-fear) will be judged as lower in dominance and higher in affiliation compared to the other emotional changes (hypothesize 1c).
Gender differences in dominance and affiliation
Besides facial expressions, gender is another factor that is often discussed in dominance and affiliation research. Women are perceived as more affiliative than men (Algoe, Buswell, & De Lamater, 2000; Diekman & Eagly, 2000), whereas men are perceived as more dominant than women (Alexander & Wood, 2000). Regarding emotion, angry men are perceived as more dominant than angry women (Hess, Blairy, & Kleck, 2000). This gender differences in dominance and affiliation perception may be modulated by the
perceived appropriateness of the emotions. Anger and disgust are perceived as inappropriate emotions to experience and express for females, however, are perceived as appropriate emotions for males (Plant, Hyde, Keltner & Devine, 2000). In contrast, fear is often perceived as an inappropriate emotion for males to experience and express, and an appropriate emotion for females (Plant, Hyde, Keltner & Devine, 2000).
It seems that this gender stereotypes for the appropriateness to express and experience an emotion are ground covered. Moreover, women and men are not only perceived as more dominance or affiliative, but they actually show behaviour regarding these stereotypes. Fluxen (2005) observed men and women in social interactions and found that men showed more dominance, like head shaking and using closed questions. Whereas, women showed more affiliation, like higher frequency of laughing and posing open questions. An explanation for this gender stereotype behaviour derives from an ecological vision who argue that men are more concerned to acquiring and showing status to be reproductively successful and therefore show more dominance (Moskowitz, 1993; Pratto, 1996). In contrast, women are less concerned to show status, but are more concerned to emphasis on relational aspects and therefore show more affiliation (Buss & Malamuth, 1996; Taylor et al. 2000). However, this knowledge is very useful for understanding gender stereotypes regarding emotions.
In sum, previous research suggests that negative emotions are more appropriate to express and experience for males than for females, except for fear. However, we expect a contrast effect on dominance and affiliation for inappropriate facial expressions. In other words, we predict that when an emotional expressing is inappropriate for the gender, the judgements of dominance and affiliation will be conversely to the gender stereotypes of dominance and affiliation perception. So for example, females would be perceived as high in dominance and low in affiliation when expressing an incongruent stereotype emotion. Well, the emotions used in present study are all negative, so the majority of the facial expressions
will be perceived as inappropriate for females. Therefore, we hypothesize that females will be perceived as higher in dominance and lower in affiliation than males, for all emotional changes (hypothesis 2).
Methods Design
The present study will employ a 6 (Emotional Change: Anger-to-Disgust, Disgust-to-Anger, Anger-to-Fear, Fear-to-Disgust-to-Anger, Disgust-to-Fear, Fear-to-Disgust) x 2 (Expresser Gender: Male, Female) within-subject design, with ratings on dominance and affiliation as dependent variables. This means that participants will be assigned to all conditions. In total there will be 12 conditions, of which six conditions contains a different combination of emotional changes, displayed by either a female or male.
Materials
Facial Stimuli. Facial expressions of anger, disgust and fear were chosen as the
emotional stimuli in the present research. Twelve Dutch actors (six male and six female) were selected from the Radboud Faces Database (Langner, Dotsch, Bijlstra, Wigboldus, Hawk, & van Knippenberg, 2010). In the validation study by Langner et al. (2010), the mean agreement on emotion categorization was 95.83 (SD = 2.56) for anger, 87.83 (SD = 4.54) for disgust and 87.83 (SD = 6.11) for fear. Using a digital morphing program (Fantamorph;
http://www.fantamorph.com/), 26-frame morphs in which three emotion pairs with two directions (anger-to-disgust, disgust-to-anger, anger-to-fear, fear-to-anger, disgust-to-fear, fear-to-disgust) for the same individual’s face will be produced. In total, 72 morphs (6 actors × 2 genders × 3 emotion pairs × 2 directions) were included.
Measurement of Dominance/Affiliation. The Revised Interpersonal Adjective Scale
(IAS-R) measures the interpersonal dimensions of dominance and affiliation, containing 32 trait adjectives items (Wiggins, Trapnell, & Phillips, 1988). The items are rated on a
seven-point Likert scale ranging from extremely inaccurate (1) to extremely accurate (7). To simplify the measure, the short version with four adjectives representing the dominance items (dominant, self-assured, assertive, self-confident) and four adjectives representing the affiliation items (gentle, agreeable, tender, sympathetic) will be employed. This results in a total of eight items, adapted from the IAS-R, to measure dominance and affiliation.
The formula to compute the dominance and affiliation score is adapted from Wiggins et al. (1988). Because we only use four trait adjectives for both dominance items (Cronbach’s α = .77) and affiliation items (Cronbach’s α = .87), we will add up the scores of the dominance- and affiliation items and divide it by the amount of items (Dominance = assured-dominance (PA) = (dominant, self-assured, assertive, self-confident)/4; Affiliation = warm-agreeable (LE) = (gentle, warm-agreeable, tender, sympathetic)/4). The final scores of dominance and affiliation could range from 1 to 7 with higher scores representing higher dominance and affiliation.
Participants
99 participants were (83 female; 20.47±2.93 years) recruited from the subject’s pool of University of Amsterdam. They received 0.5 research point for participating the study.
Procedure
Each trial will start with a fixation cross, displayed in the center of the screen for 500 ms, followed by a movie clip displaying emotional changes lasting 2000 ms. Immediately following the last frame of the clip, a grey screen with two items measuring either dominance or affiliation (IAS-R; Wiggins, Trapnell, & Phillips, 1988) will appear. Participants will be asked to rate on seven-point Likert scales ranging from extremely inaccurate (1) to extremely accurate (7). The order of the items will be the same within each participant, but will be counterbalanced between each participant. The questions will appear simultaneously on screen. Participants will complete 4 practice trials, followed by four blocks of 12 trials each
Comment [XF1]: In the research
proposal, we use future tense to describe the design and experiments. However, in the final draft, we have done the experiments and we should use past tense
(each emotional change will be displayed by a unique actor). To complete all eight questions to measure dominance and affiliation, each participant will receive four blocks. Within each block, all the stimuli will be presented in a random order. Present study is combined with another study who has equally methods but with a different dependent variable. The order of the studies will be counterbalanced between each participants. Participants fulfilled an unrelated task between the two studies to be sure that the studies do not influence each other. While playing the game, participants will forget the photographs they saw in the experiment of the first study. The Snake Game is used for this task and is well known under people. As a result, three components in total will be used for the present study (study 1, unrelated task, study 2). Each component took approximately 10 minutes, which results in a total duration of 30 minutes.
Results
Because hypotheses 1a, 1b and 1c assume that there is a main effect of emotional effect on dominance and affiliation, we first explored this effect by conducting an one-way repeated measure ANOVA for both dominance and affiliation. Since the data violated the assumption of sphericity, Greenhouse-Geisser corrections were employed for all repeated measures analyses reported below. The test revealed that there was a significant main effect of emotional change on ratings of dominance, F(4,396) = 23,57, p < .001 (Table 1). The partial eta squared
(ɳ2
𝑝𝑝= .19) was of large size. There was also a significant main effect of emotional change on ratings of affiliation, F(4,335) = 30,84, p < .001 (Table 1). The partial eta squared (ɳ2
𝑝𝑝 = .46) was of large size. We used the post hoc test with a Bonferroni correction to test which means differs from each other. The results are discussed in order of the hypotheses of present study.
Table 1
Analysis of Variance on Dominance and Affiliation Ratings as a Function of Emotional Change and Gender of the Expresser
Dominance Affiliation
df / df error F p partial eta² F p partial eta² Emotional Change 4/396 23.57 .001* .19 81.68 .001* .46 Gender of expresser 1/98 .27.34 .001* .22 .244 .623 .002 Note. Results are reported at a significant value p < .05, *p < .001
Hypotheses 1a Testing
We hypothesized that disgust-to-anger and fear-anger will be judged as high in dominance and low in affiliation. Furthermore, we hypothesized that emotional changes with anger as the target emotion (either disgust-to-anger or fear-to-anger) would be judged as higher in dominance and lower in affiliation than emotional changes with fear as the target emotion (either anger-to-fear or disgust-to-fear). We conducted an one-way repeated measure ANOVA and used the post hoc test.
For dominance, the post hoc test revealed that there were no significant differences between the means of disgust-to-anger and fear-to-anger on dominance, as expected. However, the test revealed that there were significant differences between the means of fear/disgust-to-anger and anger/disgust-to-fear (Mdiff = .676 ; p < .001, Mdiff = .720 ; p < .001,
Mdiff = .557 ; p < .001 and Mdiff = .602 ; p < .001). With higher dominance ratings for
fear/disgust-to-anger (Table 2). This was as expected. In contrast to expectations, there were significant differences between the means of fear-to-anger and anger/fear-to-disgust (Mdiff
(Table 2). Another remarkable and unexpected finding was the significant difference between the means of disgust-to-anger and anger-to-disgust (Mdiff = .242 ; p = .026), with slightly
higher dominance score for disgust-to-anger. There were no significant differences between disgust-to-anger and fear-to-disgust.
For affiliation, the post hoc test revealed that there were no significant differences between the means of disgust-to-anger and fear-to-anger on affiliation, as expected. Furthermore, the test revealed that there were significant differences between the means of disgust/fear-to-anger and anger/disgust-to-fear (Mdiff = -.990 ; p < .001, Mdiff = -.677 ; p < .001,
Mdiff = -1.057 ; p < .001 and Mdiff = -.744 ; p < .001). With lower affiliation ratings for
disgust/fear-to-anger (Table 2), just as expected. But again, there is an unexpected significant differences too between the means of disgust/fear-to-anger and anger-to-disgust (Mean difference = .33 ; p < .001 and mean difference= .400 ; p < .001). With higher ratings on affiliation for disgust/fear-to-anger (Table 2).
Table 2
Mean and Standard Deviation of the average dominance and affiliation score per condition (emotional change and gender of expresser)
Condition Mean (SD) Dominance Mean (SD) Affiliation Disgust-to-Anger* 4.08 (.80) 3.08 (.97) Fear-to-Anger* 4.20 (.83) 3.15 (.98) Anger-to-Disgust* 3.84 (.90) 2.74 (.85) Fear-to-Disgust* 3.87 (.81) 3.02 (.90) Disgust-to-Fear* 3.48 (.79) 3.82 (.88) Anger-to-Fear* 3.52 (.83) 4.14 (1.01) Female Expresser* 3.95 (.67) 3.34 (.70) Male Expresser* 3.62 (.65) 3.31 (.72)
*N=99
Hypothesis 1b Testing
We hypothesized that anger-to-disgust and fear-to-disgust will be judged as high in dominance and low in affiliation. Furthermore, we hypothesized that the emotional changes with disgust as target emotion (either anger-to-disgust or fear-to-disgust) will be judged as higher in dominance and lower in affiliation than emotional changes with fear as the target emotion (either anger-to-fear or disgust-to-fear). Once again, an one way repeated measure ANOVA was conducted to test the hypothesis.
For dominance, the post hoc test revealed that there were no significant differences between the means of anger-to-disgust and fear-to-disgust, which was expected. From the post hoc tests discussed above we already know that anger-to-disgust leads to lower dominance than disgust/fear-to-anger (Mdiff = -.3242 ; p = .026 and Mdiff = -.360 ; p = .001).
We also already know that fear-to-disgust leads to lower dominance than fear-to-anger (Mdiff
= -.332 ; p = .001), but that there were no differences between fear-to-disgust and disgust-to-anger on dominance. These differences between target disgust and target disgust-to-anger were all unexpected. Furthermore, the test revealed that there were significant differences between the means of anger/fear-to-disgust and anger/disgust-to-fear (Mdiff = .316 ; p = .011, Mdiff = .360 ;
p = .001, Mdiff = .343 ; p = .002 and Mdiff = .388 ; p < .001). These differences between target
disgust and target fear were as expected.
For affiliation, the post hoc test revealed that there were significant differences between the means of anger-to-disgust and fear-to-disgust (Mdiff = -.275 ; p = .001), which is
not in line with the hypothesis. With lower affiliation ratings for anger-to-disgust (Table 3). In fact, the data suggest that anger-to-disgust had the lowest affiliation ratings when comparing with disgust-to-anger, fear-to-anger, disgust-to-fear and anger-to-fear (Mdiff =
-.333 ; p < .001, Mdiff = -.400 ; p < .001, Mdiff = -1.077 ; p < .001 and Mdiff = -1.390 ; p < .001).
These results are as expected, except for the finding that anger-to-disgust leads to lower affiliation ratings too compared with disgust/fear-to-anger. Furthermore, the post hoc test revealed that there were significant differences when comparing fear-to-disgust with anger/disgust-to-fear (Mdiff = -1.115 ; p < .001 and Mdiff = -.802 ; p < .001). With lower ratings
for fear-to-disgust (Table 2), as expected. There were no differences found between the means of fear-to-disgust and disgust/fear-to-anger, this was as expected.
Hypothesis 1c Testing
We hypothesized that anger-to-fear and disgust-to-fear will be judged as low in dominance and high affiliation. Furthermore, we hypothesized that anger/disgust-to-fear will be judged as lower in dominance and higher in affiliation compared to the other emotional changes. Once again, an one way repeated measure ANOVA was conducted to test the hypothesis.
For dominance, the post hoc test revealed that there were no significant differences between the means of anger-to-fear and disgust-to-fear. This was as expected. From the post hoc tests discussed above we already know that anger/disgust-to-fear leads to lower dominance than disgust/fear-to-anger (Mdiff = -.557 ; p < .001, Mdiff = -.676; p < .001, Mdiff =
-.602 ; p < .001 and Mdiff = -.702; p < .001). And that anger/disgust-to-fear leads to lower
dominance than anger/fear-to-disgust (Mdiff = -.316 ; p = .011, Mdiff = -.343; p = .002, Mdiff =
-.360 ; p = .001 and Mdiff = -.388; p < .001). All these results are as expected and support
hypothesis 1c.
For affiliation, the post hoc test revealed that there were significant differences between the means of anger-to-fear and disgust-to-fear (Mdiff = .313 ; p < .001). With higher
affiliation ratings for anger-to-fear (Table 2), which was not expected. Furthermore, we already know from the post hoc tests discussed above that there were significant differences
between the means of anger/disgust-to-fear when comparing with anger-to-disgust, fear-to-disgust, disgust-to-anger and fear-to-anger (Mdiff = 1.390 ; p < .001, Mdiff = 1.115 ; p < .001,
Mdiff = 1.057 ; p < .001 and Mdiff = .990 ; p < .001). With higher ratings of affiliation for
anger/disgust-to-fear (Table 2), which was as expected.
In sum, the first hypotheses are partly supported by the data. The data suggests that the expectations regarding dominance- and affiliation scores between target anger and target fear were supported. The relations with target disgust were slightly different than excepted. Whereas, we expected that dominance- and affiliation scores for target anger and target disgust would be equal, but the data showed us that this was not the case.
Hypothesis 2 Testing
We hypothesized that, for all emotional changes, females will be perceived as higher in dominance and lower in affiliation than males. To explore whether there is a main effect of gender of expresser on dominance and affiliation, we conducted an one-way repeated measures ANOVA for both dominance and affiliation. The assumption of sphericity is met, because the repeated measure variable gender of expresser has only two levels. For dominance, the test revealed that there was a significant main effect of gender of expresser on ratings of dominance, F(1,98) = 27, 24, p < .001 (Table 2). The partial eta squared (ɳ2
𝑝𝑝 = .22) was of large size. The post hoc test with a Bonferonni correction revealed that there were significant differences between dominance ratings of male expressers and female expressers (Mdiff = -.232 ; p < .001). With higher overall dominance ratings for female expressers (Table
2), as expected. Furthermore, for affiliation, the one-way repeated measure ANOVA revealed that there was no significant main effect of gender of expresser on affiliation. This means that the overall affiliation score for females and males did not differ from each other.
Present study lend support to previous findings who argued that expressing anger leads to high perceived dominance and low perceived affiliation when comparing with fear. Moreover, expressing fear leads to low perceived dominance and high perceived affiliation when comparing with anger or disgust. These results support the hypothesis that fear leads to the (relatively) highest perceived affiliation and lowest perceived dominance. However, the results do not lend support for the hypothesis that anger and disgust would be perceived as equal in affiliation and dominance. Whereas, expressing disgust leads to lower dominance and lower affiliation when comparing with anger. Another remarkable finding regarding disgust is that the affiliation judgements are mediated by the anchor emotion. Anger-to-disgust is perceived as lower in affiliation than fear-to-Anger-to-disgust. This result do not support present study’s hypothesis that emotional changes with target disgust would lead to the same affiliation perception. In contrast, the dominance perception of disgust was independent of the anchor emotion. However, anger and fear as target emotion determined the perceived dominance and affiliation. In other words, the dominance and affiliation judgements for anger and fear were both independent of the anchor emotion. These results of anger and fear do support the hypotheses that there will be no differences in dominance within emotional changes with the same target emotion. At last, present study showed, in contrast with previous research, that females were perceived as higher in dominance than males. The hypothesis that females would be perceived as lower in affiliation than males was not supported. Furthermore, we have to keep in mind that the results in function of perceived dominance and affiliation are all relative.
Emotional Change and Dominance/Affiliation Perception
The present study provides convincing evidence that the perceived affiliation for disgust is dependent of the context or in other words dependent of the anchor emotion. At first, this may be due to the strength of association between affiliation and disgust. Anger and
fear are two emotions strongly associated with low and high affiliation, respectively. Whereas, disgust is less often expressed than anger and fear do, so maybe people are not that familiar with disgust. Therefore, we do not have such strength associations between disgust and affiliation as we have for anger and fear. As a result, people rely on available contextual information to make a judgement when disgust is expressed. The available information in present study was the anchor emotion. Second, it might be to do with the fact that disgust is known as an object- or food-directed emotion (Izard, 1991; Haidt, 2003), rather than a person-directed emotion. However, in present study, the emotion was directed to the participant. As a result, people do not know what to think of this person-directed disgust and rely on other available information to judge about dominance and affiliation. Again, this available information was the anchor emotion. And as we see for disgust, the affiliation judgements were in line with the associated affiliation of the anchor emotion. The point of all what is discussed above is that it is reasonable to think that the findings for disgust and affiliation are a result of the ‘unfamiliarity’ with disgust.
Another remarkable finding regarding disgust that needs to be discussed is that expressing disgust leads to particularly low affiliation judgements, lower than previous research showed. It might be that this is due to the feeling that is caused when someone is displaying disgust directed to you. Disgust is moral emotion and contains meanings of what is wrong/unaccepted and what is right/accepted (Haidt, 2003). It seems that disgust is an emotion associated with social rejection. When someone express disgust to you, it seems reasonable that you get a feeling of rejection. As a result, person-direct disgust will be judged as low in affiliation.
For dominance, it seems that disgust is perceived as somewhat less dominant than anger. This might be due to the associated action tendencies for disgust and anger. Disgust action tendencies are associated with avoidance, you want to get away or have nothing to do
with the object. Whereas anger is associated with approach and attack (Haidt, 2003). So, taking this knowledge into account, it seems that disgust is a more passive emotion, whereas anger is a more active emotion. Moreover, disgust is in contrast with anger associated with decreases in hart rate, which could be interpreted as disgust as a passive instead of active emotion (Haidt, 2003). As a result, disgust is perceived as lower in dominance than anger. Furthermore, disgust is just like fear an inhibitory or avoidance emotion, whereas anger is an approach emotion. Previous research showed that inhibitory or avoidance emotions are perceived as lower in dominance than approach emotions (Montepare & Dobish, 2003; Carney, Hall & LeBeau, 2005; Hareli & Shomrat, 2009). This can also explain why anger is perceived as higher in dominance than disgust. Based on findings of present study, we recommend future research to explore disgust, specific person-direct disgust, more clear. The central question that can be kept in mind is ‘How is person-directed disgust related to perceived dominance and affiliation’?
Gender and Dominance/Affiliation Perception
The last thing we will discuss refers to the role of gender. Present study did not found any differences between the judgements of female and male expressers on affiliation. As mentioned earlier, we expected contrast effects on affiliation when the facial expression was perceived as inappropriate. However, no lower affiliation ratings were found for females compared with males. A convenient explanation can be that the powerful impact of emotional change on affiliation undermined the potential contrast effect on affiliation for gender inappropriate expressions. The results in present study showed that the effect size of the main effect of emotional change on affiliation was almost twice as big then the main effect of emotional change on dominance. Taking this into account, it seems reasonable to think that this can be an explanation for the lack of the role of gender of expresser on affiliation.
In contrast with previous research, present study find support for the hypothesis that females are perceived as higher in dominance than males. Earlier, we based this on the potential contrast effect when displaying gender inappropriate emotions. Another suitable explanation can be that the three negative emotions used in present study were more salient for female expressers than for male expressers. As a result, females were judged as higher in dominance. Recommendation for future research refers to the effect of gender inappropriate emotional expressions on person perception.
Limitations of present study
It is clear that there are some remarkable and interesting findings from present study that can be discussed. However, there are also some limitations of the study that have to be noticed. First of all, female participants were over-represented and only 18% of the participants were male. Previous studies have shown that women and men process emotions differently. Spalek et al. (2015) found that women evaluate emotional stimuli as more arousing than men, whereas Thayer & Johnsen (2000) found that women are better in recognizing negative emotions than men. These different ways of processing emotions could lead to different judgements of dominance and affiliation. So, we have to keep in mind that the findings of present study mainly depend on female observers and cannot be generalized to the whole population. Recommendation for future research is to balance males- and females percentages in the participant’s group. Besides generalisation of the results, this makes it also possible to compare judgements of male and female observers.
In present study we are interested in impression formation. One thing known about this phenomenon is the implicit personality theory. This theory describes the specific patterns and biases that people use when they form an impression of an unfamiliar person. One factor that affects the impression we make is mood, this is also known as the mood-congruent judgement. This means that in general, happy people will judge more positive and sad people
will judge more negative. Several studies have shown that a happy mood leads to more favorable impressions and judgments than sad moods (Forgas & Bower, 1987; Mayer 1992; Forgas, 2011). Taking this knowledge into account, it seems valuable to control the mood of participants. Limitation of present study is the absent of such a mood control. Future research needs to take the influence of mood in consideration and control for mood when it is possible. In this way, potential effects of mood can be controlled.
The lack of available situational context in which the emotions were expressed. In real-life, people often do have situational information. It is reasonable that situational information might lead to different dominance and affiliation judgements. This limitation reduces the ecological validity of present study. However, the use of dynamic stimuli instead of static stimuli and the present of emotional changes instead of single emotions maximize the ecological validity of present study.
Implications of Present Study
Present study lends support for the EASI-theory and earlier research who argued that emotions can serve as an informational cue. We supported the notion that emotional expressions serve as an important cue to convey information about the personality of the expresser. In other words, facial expressions influence the person perception and impression formation. This knowledge plays an important role in understanding how people judge others and how they form impressions. Consequently, the impressions about another’s person dominance and affiliation can tell us something about the meaning of the social interaction. For example, it can tell you something about the relationship like low versus high status and about the attitude of the expresser. As mention earlier, this first impressions can also influence your future interactions with the person. Especially in business life this can be useful information. If you are aware of what kind of information others convey when expressing a certain emotion, you can use this knowledge in favour of your goals.
Comment [ef2]: This point is not so
Unfortunately, present study did not look to actions tendencies associated with the judgements that participants made. It is still interesting for further research to look at this more closely to get a more clear understanding about the implications for social interactions.
However, present study showed also a more technical implication. Moreover, present study showed that there is not always a negative correlation between dominance and affiliation. Whereas, it appears that disgust can be perceived as relatively low in dominance and low in affiliation. The last and most important implication of present study refers to findings of emotional change. Actually, emotional change was our main interest, because of the lack of ecological validity in researchers who used single static emotions. Our aim was to explore the effect of this phenomenon, emotional change, on person perception. The results showed that it is very useful and interesting to look at emotional changes in relation to dominance and affiliation. Present study gave clear evidence that the target emotion is of great importance in dominance and affiliation judgements. Moreover, this study showed that the influence of the target emotion disgust differs from target emotion anger and fear. People use information that they can convey from the anchor emotion when judging disgust expressions on affiliation. In the discussion we made the notion that his might be due to the unfamiliarity with person directed disgust in relation to dominance and affiliation. Is seems to be useful and interesting for future research to look at this person-directed disgust more closely. Our study proved that the effect of emotional changes with disgust on person perception is different from emotional changes with anger and fear and that disgust is a less dominant and less affiliative emotion than we thought. Further research is needed in studying the effect of other emotions, like happiness and sadness, in emotional changes to give rise to the knowledge of the phenomena emotional change. This new field in the social psychology literature can give us a more clearly/enhanced/wider understanding of emotions and person perception.
Conclusions
Present studies aim was to explore the effect of emotional change on the perception of dominance and affiliation and to investigate the mediation role of the gender of expressers. The results showed that emotional changes have a great impact on this person perception. The role of gender of the expresser is only found for dominance, not for affiliation.
When looking within the emotional changes with the same target emotion, it appears that the dominance perception do not differ when either anger, disgust or fear is the target emotion. In other words, it seems that the anchor emotion in the emotional change do not have a particularly influence on judgements, whereas the target emotion have. However, for affiliation, it appears that there were differences within the emotional changes with target emotion disgust. Where anger-to-disgust leads to lower perceived affiliation than fear-to-disgust. In other words, it seems that the perception of affiliation for disgust is mediated by the anchor emotion. Moreover, there were no differences found within the emotional changes with target emotions anger and fear.
Moreover, it appears that there are differences in dominance and affiliation perception when looking between emotional changes with different target emotions. In other words, anger, fear and disgust caused different dominance and affiliation perceptions. Moreover, for dominance, an emotional change with target anger or disgust leads to relatively higher perceived dominance than target fear and vice versa. Another remarkable result was that the emotional changes with target anger caused relatively higher dominance perception than the emotional changes with target disgust. In other words, anger leads in general to higher perceived dominance than disgust do. Furthermore, for affiliation, it appears that emotional changes with target fear caused the highest affiliation perceptions. The emotional change anger-to-disgust caused the lowest affiliation perception. Emotional changes with target anger and fear-to-disgust caused lower affiliation perceptions than target fear. On basis of the
evidence currently available, it seems fair to suggest that disgust is perceived as lower in affiliation than anger. At last, present study showed that dominance perceptions were higher for female expressers than for male expressers.
In conclusion, the target emotion in emotional changes have a powerful influence on dominance and affiliation judgements, except for disgust and affiliation judgements. These judgements for disgust are mediated by the anchor emotion. Despite the interesting findings of present study, future research is still needed to get a full understanding of this phenomenon.
References
Asch S. E. ( 1946). Forming impressions of personality. Journal of Abnormal and Social
Psychology, 41, 258-290.
Alexander, M. G., & Wood, W. (2000). Women, men and positive emotions: A social role interpretation. Gender and emotion: Social psychological perspectives,189–211. Cambridge University Press.
Algoe, S. B., Buswell, B. N., & De Lameter, J. D. (2000). Gender and job status as contextual cues for the interpretation of facial expression of emotion. Sex Roles, 42, 183- 208.
Carney, D. R., Hall, J. A., & LeBeau, L. S. (2005). Beliefs about the nonverbal expression of social power. Journal of Nonverbal Behavior, 29, 105-123.
Diekman, A. B., & Eagly, A. H. (2000). Stereotypes as dynamic constructs: Women and men of the past, present and future. Personality and Social Psychology Bulletin, 26, 1171– 1188.
Dovidio (Eds.), Power, dominance, and nonverbal behaviour (pp. 89–108). New York:
Ekman, P., & Davidson, R. J. (1994). The nature of emotion. New York: Oxford University Press.
Forgas, J. P., & Bower, G. H. (1987). Mood effects on person-perception judgments. Journal of Personality and Social Psychology, 53, 53-60.
Forgas, J.P. (2011). Can negative affect eliminate the power of first impressions? Affective influences on primacy and recency effects in impression formation. Journal of Experimental Psychology, 47, 425-429.
Haidt, J. (2003). The moral emotions. Handbook of affective sciences. Oxford University Press, 852-870.
Hareli, S., & Shomrat, N. (2009). Emotional versus neutral expressions and perceptions of social dominance and submissiveness. Emotion, 9, 378-384.
Heider, F. (1958). The Psychology of Interpersonal Relations. New York: Wiley.
Hess, U., Blairy, S., & Kleck, R. E. (2000). The influence of facial emotion displays, gender, and ethnicity on judgments of dominance and affiliation. Journal of Nonverbal Behavior, 24, 265-283.
Izard CE. 1991. The Psychology of Emotion. New York: Plenum
Keating, C. (1985). Human dominance signals: The primate in us. Journal of Nonverbal Behavior, 20, 165–182.
Knutson, B. (1996). Facial expressions of emotion influence interpersonal trait inferences. Langner, O., Dotsch, R., Bijlstra, G., Wigboldus, D. H. J., Hawk, S. T., & Van Knippenberg,
A. (2010). Presentation and validation of the Radboud Faces Database. Cognition & Emotion, 24, 1377-1388.
Mayer, J. D., Gaschke, Y. N., Braverman, D. L., & Evans, T. W. (1992). Mood-congruent judgment is a general effect. Journal of Personality and Social Psychology,
63, 119-132.
Montepare, J. M., & Dobish, H. (2003). The contribution of emotion perceptions and their overgeneralizations to trait impressions. Journal of Nonverbal Behavior, 27,
Plant, E. A., Hyde, J. S., Keltner, D., & Devine, P. G. (2000). The gender stereotyping of emotions. Psychology of Women Quarterly, 24, 81-92. Springer-Verlag.
Spalek, K., Fastenrath, M., Ackermann, S., Auschra, B., Coynel, D., Frey, J., Gschwind, L., Harmann, F., Maarel, N., Papassotiropoulos, A., Quervain, D., & Milnik, A. (2015). Sex-Dependent Dissociation between Emotional Appraisal and Memory: A
Large-Scale Behavioral and fMRI Study. The Journal of Neuroscience, 35, 920-935. Thayer, J. F., & Johnsen, B. H. (2000). Sex differences in judgement of facial affect: a
multivariate analysis of recognition errors. Scandinavian Journal of Psychology, 41, 243-246.
Todorov, A., Said, C. P., Engell, A. D., & Oosterhof, N. N. (2008). Understanding evaluation of faces on social dimensions. Trends in Cognitive Sciences, 12, 455-460.
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.
Wiggins, J. S., Trapnell, P., & Phillips, N. (1988). Psychometric and geometric characteristics of the Revised Interpersonal Adjective Scales (IAS-R). Multivariate Behavioral Research, 23, 517-530.
