Ulf Dimberg, Psychology, Uppsala University

Emotional Empathy. Causal Mechanisms and Unconscious Processes

2007-2012

The overall aim of the present project is to study causal factors behind empathic reactions. One important procedure has been to expose the participants to pictures of facial expressions of emotions while their facial electromyographic reactions, autonomic responses and their experience of the stimuli were detected. A simple exposure paradigm has been used. In order to detect automatically evoked emotional reactions we have used the "The voluntary facial action technique" which I earlier have been developed and which includes that the participants are instructed to voluntarily control their facial muscles while exposed to pictures of facial expressions. The logic behind this manipulation is that if the participants are not able to completely control their facial muscle reactions these reactions must at least partly be controlled by automatic/unconscious mechanisms.
Important questions have been to study if high empathic people are uniquely sensitive with their strong reactions to facial expression or whether the low empathic people are unique by not being able to react to facial expressions. A second question has been if the different reaction pattern has its origin in automatic/unconscious mechanisms. A third aim has been to study if facial feedback effects are related to empathic ability. Because people with Asperger diagnosis are supposed to have difficulties in interpreting facial emotional expressions those persons are included in the present project. Because the stereotypical opinion proposes that females are more empathic than are males it has also been self-evident to study sex differences in the present project.
The project work has been characterized by an intensive collection of data which are based on a great number of subjects who has attended to different conditions in different studies. It has also been necessary to produce new pictures of facial expressions with different emotional intensities.
For natural reasons it has not been possible to publish all the results. However, there are some articles published and a number of papers are in progress. Results have been presented at national and international seminars and parts of the results have also formed the basis for a PhD-thesis as well as a Master thesis.
Females in general rate themselves higher than males on empathy scales. This cannot self-evidently be interpreted as that females in reality are more empathic, because this difference can also be explained as an effect of stereotypical different gender roles in the self-reporting of empathy. This implies, however, that if only the highest scoring individuals are selected, the majority of them would be females. Therefore, to avoid this confounding problem with gender, males and females were separately selected into the respective high and low empathy groups.
One of the three important overall results is the findings on the relation between empathy and the ability to react emotionally to facial expressions. In a newly published report (Dimberg et al., 2011) we found that High in contrast to Low empathic persons react with distinctively larger corrugator activity (frowning) to angry as compared to happy faces and more zygomatic activity (smile) to happy faces. Because the Low empathic group did not differentiate at all between angry and happy faces an important follow up study was to find out which of the two groups who is unique in its reaction patterns. By selecting and comparing a High and a Low group with a group scoring between these groups we found the Between group to react weaker as compared to the High group but stronger as compared to the Low group. (Dimberg, A). The difference between High and Low groups appears even if the emotional intensity is lowered in the facial stimuli (Dimberg, B ). These differential effects are also detectable after only 0.5s of exposure which indicate that the reaction patterns are controlled by automatic mechanisms (Dimberg & Thunberg, 2012). A similar effect for the corrugator activity is also detectable when the subjects are exposed to low intensity (40%) facial stimuli in the FART paradigm (Dimberg, C).
These findings can be interpreted as that there is a simple positive relationship between Empathic ability and the tendency to react to facial expressions. The results also indicate that the reactions are controlled by automatic/unconscious processes which after only 0.5s are manifested as distinct reaction patterns in the High empathic groups.
A second type of important findings is those obtained in studies focusing on the causal role of facial muscles in the origin of emotional experience. By manipulating facial muscle activity we explored if this manipulation, by facial feedback, may influence the experience of the exposed stimuli and if this phenomenon is related to empathy. One way to test the feedback theory is for instance to instruct people to frown or to smile while they are exposed to different stimuli. In that way one could test if these muscle actions in a corresponding way influence the experience of the stimuli. To avoid that the subjects see through the true purpose of the manipulations one normally use a cover story. In one series of studies we used the FART paradigm as a cover story and accordingly instructed the participants that we detected their reaction time in different muscles. In a second type of studies the subjects were required to hold a pen/stick between their lips so that they involuntarily were forced to contract the zygomatic muscle to a smile or pull their lips down resulting in a negative expression.
In the first type of experiments we found that the participants rated pictures as more pleasant when they smiled and more unpleasant when they frowned (Dimberg & Söderkvist, 2011; Andréasson & Dimberg,2012). In the second type of experiment we found that the High- as compared to the Low-empathic group experienced humorous films as more funny when exposed to the Smile condition. During the negative condition, on the other hand, the Low group reacted quite contrary to the High group. That is, the Low group rated the films as more positive in this negative condition ( Andréasson & Dimberg, 2008). This "contrary"-effect was also found in a follow up study (Andréasson & Dimberg, 2012). A part of these studies have resulted in a PhD-thesis (Andréasson,2010).
In summary, we found that the FART-paradigm was successful when used in the study of facial feedback. By help of this paradigm we obtained strong support for the facial feedback theory. Additionally, our findings demonstrate that Low empathic people react in contrary to High empathic people as regard to facial feedback.
A third type of results is those findings on how people with Asperger diagnosis react to facial expressions. Because people with Asperger are supposed to have difficulties in interpreting facial emotional expressions and may have problems to perceive subtle cues in facial expressions we have studied how they react to less distinct, low intensity facial expressions.
In one study we compared one Asperger group with High and Low empathic groups. We found that the Asperger group in general rate different facial expressions of emotion as being less intense. (Dimberg, D). Contrary to the High empathic group the Asperger group did not differ in corrugator activity when exposed to angry and happy faces. Interestingly however, the Asperger group identically to the High empathic group differentiated with their zygomatic muscle reactions to happy and angry faces. These results show that the Asperger group did not display a reaction pattern similar to either the High or the Low empathic groups. Future research may more specifically study this interesting phenomenon.
To explore if the reaction patterns are automatically controlled we used the FART paradigm while the subjects were exposed to angry and happy faces with low emotional intensity. Contrary to a High empathic group the Asperger group did not differentiate between the low intensity facial stimuli. This was true for both the corrugator and the zygomatic muscles (Dimberg, C). These results may be interpreted as that the reactions among people with Asperger are not initially controlled by automatic mechanisms to the same degree as among high empathic people. These interesting findings may contribute to explain why people with Asperger have difficulties to perceive subtle cues in other peoples facial expression.
In general we found that there are the high empathic females that react with the strongest facial muscle reaction patterns. To decide whether this reflects a genuine effect of gender or if it is a simple effect of that females are rated as the highest on the Empathy questionnaire a future study should be performed in which one try to select one extreme group of highly empathic males which then could be compared the highest empathic females.
Among the already published articles the report on the FART paradigm is important. It shows that this paradigm is not only effective to study automatic reactions but can also be successfully used in the study of facial feedback (Dimberg & Söderkvist, 2011). A second important article is the report which shows that high empathic people are superior in their ability to rapidly react to facial stimuli (Dimberg & Thunberg, 2012). These findings indicate that spontaneously evoked empathic reactions are automatically controlled and that this can be one explanation to why empathic reactions may differ among different individuals.

Publications

Andréasson, P. & Dimberg, U. (2008). Emotional empathy and facial feedback.  Journal
            of  Nonverbal Behavior. 32, 215-224.
        
Piccinelli, P. (2010). Emotionell empati och bedömning av ansiktsuttryck. 
            Psykologexamensuppsats, Institutionen för Psykologi. Uppsala Universitet.

Dimberg, U. & Söderkvist, S. (2010). The voluntary facial action technique: A method  
            to test the facial feedback hypothesis.  Journal of  Nonverbal Behavior. 35,
            17-33.

Andréasson, P. (2010). Emotional empathy, facial reactions and facial feedback.  Acta
            Universitatis Upsaliensis, Abstracst of Uppsala dissertations from the 
            Faculty of Social Sciences, 58, Uppsala.

Dimberg, U., Andréasson, P. & Thunberg, M. (2011). Empathy and Facial Reactions to
            Facial Expressions. Journal of Psychophysiology. 25, 26-31.

Dimberg, U. (2012) Ansikte mot ansikte: Omedveten emotionell kommunikation. I V. 
            Adelswärd & P.-A. Forstorp (Eds.) Människans ansikten: Emotion, interaktion
            och konst. Carlssons Bokförlag, Stockholm. Pp. 50-62.

Dimberg, U. & Thunberg, M. (2012). Empathy, emotional contagion and rapid facial    
            reactions to angry and happy facial expressions. PsyCH Journal. (in press)


Papers in progress or under editorial consideration:


Andréasson, P. & Dimberg, U. (2012). Emotional empathy, facial manipulations and 
             facial feedback.  (submitted)

Dimberg, U. (A). Empathy and facial reactions to facial expressions: A
             comparison between high, middle and low empathic groups.

Dimberg, U. (B). Empathic reactions to facial expressions with high and low
             emotional intensity.

Dimberg, U. (C). Automatic facial reactions to low intensity facial expressions
             among people with Asperger syndrom and people high or low in empathic
             ability.

Dimberg, U. (D). Empathic reactions and the experience of emotional expressions
             among people with Asperger syndrom and people high or low in empathic
             ability.


Presentations:

Dimberg, U. (2011). Face to Face: Unconscious Emotional Communication.
             Paper presented at ISCTE: Instituto Universitario de Lisboa. Department of
             social psychology: Lisabon, Portugal (November 9, 2011)

Dimberg, U: (2011). Facial expressions, empathy and emotional reactions. Department of
             Clinical Neuroscience. Karolinska Institutet. Stockholm (December 14, 2011)