University of Amsterdam
Faculty of Social – and Behaviour Sciences
2015
The Role of Emotions in the Appreciation of Art:
Measuring Blended Emotions
A literature review and research proposal Supervisors:
submitted as a requirement for a Prof. Dr. Paul De Boeck Bachelor’s degree in Psychology Mw. Dr. Mijke Rhemtulla By Sonja E.M. Vermunt
Abstract
In this paper blended emotions in the appreciation of art are explained by the Empathic Principle as formulated by Rosenberg and De Boeck (2001) and the theory of the Sublime by Edmund Burke (1757). Four studies on the emotions and the appreciation of different art forms will be discussed. The focus will be on the blended emotions involved in the Empathic Principle. The use of a facial action coding system enhanced with an algorithm to more accurately decode facial expressions as defined by (Du, Tao & Martinez, 2014) is proposed as more objective way to measure experienced, single and blended emotions
Acknowledgement
The first time Professor De Boeck told me about the interesting blend of emotions which manifests itself when people enjoy a sad or scary movie or in various art forms, he also mentioned Edmund Burke, a philosopher living in the eighteenth century. I read Burke’s work and was intrigued by the combination of beauty and fear and the difference he makes between pleasure, which is a result of beauty and delight as a result of the combination of beauty and fear or sadness. My way to understand this was: if beauty is in the eye of the beholder, the sublime should be at the rim of our thinking ( meaning that the experience is just outside of our “normal” thinking). This was years ago. It took me a long time with some pauses to finally write my bachelor thesis and I would like to thank Professor De Boeck for his patience, guidance and belief in me. I thank Mijke Rhemtulla for being the second supervisor.
Finally I would like to mention Jacques Detiger who asked me why I did not do what I wished to do: study Psychology. That was a very good question! I have greatly enjoyed my study period, even when the combination work, private life and study was complicated. I am grateful to the people dear to me (mijn dierbaren) for their love and support. And I would still like to go for my master’s degree!
"The stage is the place for violence and anguish and big tragedies.”It is wonderful if we can confine the disasters in this way. I am a happy man, but I love to suffer on stage, it is the most beautiful thing of all!" Placido Domingo
Content
Abstract ... II Acknowledgement ... III
1. Literature Review ... 2
1.1 Introduction ... 2
1.2 The Empirical Findings ... 2
1.3 Possible Interpretations ... 11 1.4 Blended Emotions ... 14 1.5 Conclusions ... 18 2. Research proposal ... 20 2.1 Introduction ... 20 2.2 Research question ... 21 2.3 Research method ... 23 2.4 Data analysis ... 24 2.5 Interpretation ... 25 References ... 26
1. Literature Review
1.1 Introduction
Previous studies show that people are capable of appreciating art that reflects negative emotions. People can highly appreciate a painting of a sad, weeping woman, watch a picture of wounded soldiers or enjoy looking at a scary movie. Burke (1757) states that pain and fear are strong emotions that heighten pleasure to delight. Pleasure is based on beauty while delight is based on a combination of beauty and, fear and or pain. This blend of emotions evokes a sublime experience, and that, people appreciate if there is enough distance mentally or physically to the experience. Rosenberg and De Boeck (2001) established the empathic principle to explain the fact that people, who relate to the negative emotions reflected in e.g. a picture, appreciate the picture even more. This blend of positive and negative emotions is an intriguing psychological phenomena and more insight in how this blending works can be obtained by improving the method to measure these blended emotions in the appreciation of art and in general.
1.2 The Empirical Findings
There have been many inquiries into the question why people appreciate art and especially how under certain circumstances negative emotions do not seem to be in the way of the appreciation. How come we go to a movie, weep and still say that was a beautiful story, or how come we go to scary movies? Seeing war photographs does not make most of us close our eyes, we can appreciate the photographs for more than the clever techniques used. This is a blend of positive and negative emotions. The art form reflects negative emotions for example pain, sadness or fear and somehow the spectator appreciates the art which is a positive emotion. Rosenberg and De Boeck ( 2001) proposed the empathic principle as complementary to the hedonistic principle ( Fechner, 1876 as mentioned in Van Bree, 2002) to explain why people can appreciate art that reflects negative events or negative emotions.
The hedonistic principle states that people will appreciate a painting when they experience positive emotions while looking at the painting or listening to a piece of music. They do not appreciate the painting or the music when they experience negative emotions. Rosenberg and De Boeck (2001) have found evidence for this hedonistic principle, but they also established a second phenomenon: the empathic principle stating that if negative emotions are reflected in a painting and the spectator experiences negative emotions him/herself, the painting is more positively appreciated. Especially when sadness and pain are reflected. This empathy can also be evoked by experiencing pity or fondness as a complementary reaction to the reflected emotions in the painting.
The hedonistic principle is investigated by measuring the experienced emotions E and relate this to the appreciation A. If E is positive, A will be positive and if E is negative, A will be negative. The reflected emotions R are not part of the equation, because it does not matter what is reflected. The empathic principle is investigated by measuring the reflected emotions
R (neg), the experienced emotions E (also negative) and relate these to the level of appreciation A (positive). If R equals E then A is positive. In this equation the negative experienced emotions and the positive appreciation show the blended emotions.
The empathic principle which includes blended emotions has been studied by Van Bree (2002), Demandt (2003), Brughmans (2004) and Goris (2007) and many others for many different art forms as a part of the role of emotions in the appreciation of art. An account of four of these studies provides insight in the broad range of application of the principle and the methodological approach used. The method used to measure the experienced emotions in the art forms according to spectators is of interest in this study since there may be a newer,
modern method to measure these emotions. In all the above mentioned studies self administered questionnaires are used.
The whole sequence of studies on the role of emotions in the appreciation of art started with three studies by Rosenberg and De Boeck (1997 and 2001) and De Boeck and Rosenberg (1997) on the emotions involved in 16 portraits Picasso made of four of his main mistresses, four of each mistress. The summary of these studies below is mostly as reported in Van Bree (2002) and Demandt (2004).
In the first two studies by Rosenberg and De Boeck (1997) and De Boeck and Rosenberg (1997) the focus was on the reflected and experienced emotions in Picasso’s portraits. A list of 19 emotions based on Tomkins’ (1962, 1963 as mentioned in Van Bree, 2002)) primary affects was used for self administered questionnaires. The emotions were clustered in three groups: 1) the positive emotions: admiration, affection, amusement, cheerfulness, pleasure, interest, relaxation, warmth, connectedness.2) the negative emotions: anger, sadness, shyness, confusion, contempt, detachment, fear, pain, repulsion. 3) the 19th emotion is surprise, a neutral emotion in this study. The participants had to report the emotions they experienced and the emotions they thought reflected in the portraits and the appreciation of each portrait. Three relations between the experienced and reflected emotions were established: a reciprocal relation and two kinds of an asymmetric relation. The reciprocal relation is the one when a participant sees a certain emotion reflected and experiences this emotion as well. The first asymmetric relation is experiencing an emotion that is not reflected in the portrait, the second asymmetric relation is not experiencing an emotion that is in fact reflected in the portrait. The variation in appreciation of the portraits was explained by the hedonistic principle. The variability in appreciation found could possibly explain by in the level of empathy evoked by
the portrait. Empathy is a reciprocal phenomena depending largely on the participant’s own emotions and experience. Rosenberg and De Boeck selected the eight portraits with the most experienced negative emotions reported and confirmed their hypothesis that reflected emotions in the portrait evoked empathy in the spectators because they experienced the same emotions. Since the test was with only eight participants a larger N experiment is necessary to really establish the empathic reciprocity.
The relation between the experienced emotions and the appreciation of the portrait was also studied by Rosenberg and De Boeck (1997) by analyzing the individual differences in the appreciation of the portraits. The outcome was that experienced emotions are good predictors of the participant‘s appreciation of the portrait. So the negative emotions reflected and experienced in a portrait evoke a positive appreciation of the portrait, blended emotions.
In a follow up study by De Boeck and Rosenberg (1997) the autobiographic emotions of Picasso and the reflected emotions in the portraits of the above mentioned four mistresses were analyzed. Picasso was able to depict his emotions very clearly, so the spectators can very well experience or the emotions reflected in the portraits. These findings are important because the reflected emotions are verified; this is what Picasso reflected, which confirms the interpretation of the experienced feelings into the empathic relation. This makes the hypothesis of the empathic principle stronger and adds to the concept of blended emotions.
In the third study by Rosenberg and De Boeck (2001) the focus was on the relation between emotions and appreciation. The proceedings were the same as in the first study: sixteen portraits and a self administered questionnaire, only the number of participants was larger (N=27). In this study the relevance of the selected 19 emotions was established, the hedonistic principle was confirmed and the occurrence of blended emotions was manifest especially
when sadness and pain were reflected (and experienced) which confirmed the existence of the empathic principle.
One can wonder whether the empathic principle also holds for other art forms. A number of master students at the University of Leuven, Belgium did research on the role of emotions in the appreciation of arts with the research of Rosenberg and De Boeck ( 1997, 2001) and De Boeck and Rosenberg (1997) as starting point.
For this literature review four of these studies are discussed. Two studies with music as the art form, one study with war photographs and one study with lithographs depicting women in unpleasant forms by Salvador Dali. In these studies the list of the used 19 emotions is tested; is this a standard list of emotion to use? Which emotions fit which art form? The empathic principle as an explanation for the occurrence of blended emotions in the appreciation of art is challenged and the question whether the blended emotions can be explained otherwise is raised.
Van Bree (2002) wrote a thesis titled “ De Rol van Gevoelens in Esthetische Appreciatie:
een onderzoek over aria’s uit De Toverfluit van W.A. Mozart “ (translated:“ The role of Emotions in the Esthetic Appreciation, an Inquiry on Aria’s from Die Zauberflöte by W.A. Mozart”). The aim of this study is the Validation of the Empathic Principle for another art form (music). In Van Bree’s work ( N=52) the same list of 19 emotions as in the Rosenberg and De Boeck studies is used in a self administered questionnaire to report the experienced emotions, the reflected emotions and the rate of appreciation. Van Bree’s inquiry confirmed the occurrence of blended emotions leading to the empathic principle.
52 Participants indicated on a questionnaire which of the 19 emotions they thought were reflected and which they experienced in the music. For each emotion they could indicate whether the emotion was 0(not reflected or experienced), 1(slightly reflected or experienced) and 2(strongly reflected or experienced). The rate of Appreciation could be indicated on a scale from 1(not at all) to 7(totally). The positive emotions on the questionnaire were: admiration, affection, amusement, cheerfulness, pleasure, interest, relaxation, warmth, and connectedness. The negative emotions were: anger, sadness, shyness, confusion, contempt, detachment, fear, pain, repulsion. And the 19th emotion is surprise, a neutral emotion in this study. There were three conditions in this study: only music, only the text, and the music and the text.
The result of this inquiry is that the hedonistic principle as well as the empathic principle is confirmed in all three conditions. So participants liked the aria’s that reflected pleasant emotions, and appreciated aria’s that reflected unpleasant emotions which they also experienced or with which they could sympathize. This blend of emotions is consistent with the empathic principle.
Demandt (2003) wrote a thesis titled “De Rol van Gevoelens in Esthetische Appreciatie: een
onderzoek over de Goldbergvariaties van J.S. Bach” (translated: The Role of Emotions in Esthetic Appreciation, an inquiry into the Goldberg Variations by J.S. Bach. The research focused on the validation of the empathic principle for a different style of music then used in an earlier inquiry namely Bach instead of Mozart.
21 Participants indicated on a self administered questionnaire which of 20 emotions they thought reflected in the music and which they experienced with the music. For each emotion they could indicate whether the emotion was 0(not reflected or experienced), 1(slightly
reflected or experienced) and 2(strongly reflected or experienced). The rate of Appreciation could be indicated on a scale from 1(not at all) to 7(totally). ). The 20 emotions were selected by the researcher based on the list of primary affects by Tomkins (1962, 1963) as used by Rosenberg and De Boeck (1997, 2001)and adapted after a pre-study on the frequency of emotions experienced and reflected while listening to the variations by participants. The positive emotions on this questionnaire were: affection, amusement, cheerfulness, playfulness, pleasure, relaxation, warmth, connectedness and confidence. The negative emotions were: anger, confusion, restlessness, detachment, shyness, sadness, fear and pain. The emotions surprise, excitement and modesty were labeled as emotions that can be either positive or negative.
Demandt made a distinction between active and passive emotions because during his analysis he noted that there was an effect of positive active emotions on appreciation. Active positive emotions are: amusement, cheerfulness, playfulness, pleasure and confidence. Active negative emotions are: anger, confusion, restlessness and sadness. Passive positive emotions are: affection, relaxation and warmth. Passive negative emotions are: detachment, shyness, fear and pain. Surprise and Excitement are active emotions and modesty is passive.
Active positive emotions influence the appreciation and confirm the hedonistic principle. Active negative emotions influence the appreciation and confirm the hedonistic and empathic principle. Passive positive emotions confirm the hedonistic principle and Passive negative emotions did not confirm the empathic principle in this study. So the empathic principle is confirmed for all active emotions (negative and positive) which means that the empathic principle can also be evoked by positive emotions. For this study the fact that positive emotions can evoke the empathic principle is not further looked into. The focus is on the blend of negative emotions reflected in the music and also experienced and the positive appreciation of the art. This blend of emotions is confirmed in Demandt’s research.
Brughmans (2004) wrote a thesis titled ‘De Rol van Gevoelens in Esthetische Appreciatie:
een onderzoek over lithografieën uit ‘Dali illustre Casanova’ van Salvador Dali. “(translated: The Role of Emotions in the Esthetic Appreciation by doing an inquiry into the lithographs from “Dali illustre Casanova” by Salvador Dali). The questions in this study are: what is there to appreciate in these litho’s and how does appreciation work when bizar and shocking elements are displayed.
Picture 1. 2 litho’s from “Dali illustre Casanova” used in Brughmans ‘(2004) inquiry on the role of emotions in the appreciation of art.
In a pre-study 153 emotions were presented to participants. From these 153 the eight most mentioned negative and eight most mentioned positive emotions were chosen for the self administered questionnaire used in Brughmans’ main study. In the main study the sexe of the participants was an item because the depicted figures in the litho’s are women. 80 participants of whom 35 male and 45 female, answered two questions for each of the 14 lithographs: 1)
indicate the level of each of the 16 emotions experienced while watching the litho, and 2) indicate the level of appreciation.
There were three clusters of emotions that ended up influencing the appreciation.
1) Pain and compassion: this combination of emotions accounted for some appreciation by male participants. In the theory of Burke (1757) on the sublime, which will be addressed in paragraph 1.3 of this study, one needs distance from a situation of terror to be able to appreciate it. For men there would be enough distance from the depicted female bodies to have some appreciation. This blend of emotions confirms the empathic principle. For women there is no distance in this instance, no blending of emotions so no appreciation.
2) Interest and curiosity: this combination of emotions led to some appreciation in men and women for some of the litho’s. A very fragile blend of emotions confirming the empathic principle
3) Aversion, disgust and abhorrorence: this combination led to no appreciation by neither men nor women. The hedonistic principle is confirmed for this cluster of emotions.
Goris (2007) wrote a thesis titled “De Rol van Negatieve Gevoelens in Esthetische
Appreciatie: een onderzoek over oorlogsfoto’s van James Nachtwey” (translated: The Role of Negative Emotions in the Esthetic Appreciation by doing an inquiry into war photographs by James Nachtwey). The research questions in this study are 1) can the empathic principle explain the appreciation of war photographs? And 2) is the propensity for empathy in a person influencing the empathic principle in the appreciation of art?
A self administered questionnaire containing three questions is used for each of the 12 used photographs. First 44 participants indicated on the questionnaire which of nine emotions they
thought reflected in the photographs and to what extend (not, a little, moderately, a lot, very much). Secondly they had to indicate which emotions they experienced while watching the photographs and to what extend (not, a little, moderately, a lot, very much). And thirdly the rate of Appreciation could be indicated on a scale from -3(not at all) to +3(very much). ). The nine emotions were selected by the researcher based on a pre study. Eight most frequently indicated negative emotions were included and pity as a positive emotion was added by the researcher.
In Goris’ study the empathic principle was confirmed for the emotions pain and compassion meaning that again the blend of negative and positive emotions is present. The findings on the second research question are that when participants identify too much or too little with the negative images the appreciation is negative and, if the propensity for empathy is large, the distance between the spectator and the negative image is too small to be able to appreciate it. These results can be connected to the theory of Edmund Burke on when one can have a sublime experience and when an experience is just too sad or scary to create any positive emotion.
1.3 Possible Interpretations
Although the hedonistic and empathic principles can explain the positive or negative appreciation of art forms, the reason why human beings do sympathize with a sad picture or why we are even able to stand in a zoo and fearfully exclaim how beautiful that angry tiger is and not run for our lives, is not explained yet. The Theory of the Sublime by Edmund Burke (1757) supplements our thinking on the appreciation of fearful and sad things with what he calls a sublime experience. The sublime describes experiencing danger or pain from a certain distance and being able to greatly enjoy it; because of the distance we are emotionally
strongly moved by the fear or pain but not contained by the experience. With a sublime experience comes relief because the terribleness enclosed in the tiger or the deep sadness of a grieving widow depicted in a painting is not the subject’s own fear or pain. This relief based on a pain or danger experience, this blend of negative and positive emotions, is called delight. Burke speaks of delight as the strongest form of pleasure. Pleasure is an emotion born from beauty, it caresses our senses. Delight is an emotion born out of the sublime after an experience that shocks our senses. Our senses will be/ can be shocked by
a) dimensions in objects and in sounds, everything that is larger than us and loud, great sounds
b) Darkness, literally and figuratively: everything that is dark, not transparent, veiled, unknown, and infinite.
c)distance and proximity: when we are watching something terrible from a distance it can be sublime, when something happens to somebody else it can be sublime, when it is too close by it is just scary and it cannot be sublime, there is no delight , we just have a horrible experience.
d) Power: the sublime is based on powerlessness of the subject and power of the object or happening. The sublime happens, there is no control over the instance but then there is delight because we are not the victim.
The results in Goris’ research (2007), two and a half centuries and a lot of knowledge substantiated by measurements in research later then Burke (1757), provides evidence for Burke’s thoughts on distance and proximity as being important in the ability to have a blend of positive and negative emotions when confronted with a negative situation (actually or in an art form). As mentioned in the previous paragraph the propensity of empathy influences the distance one has to the war photographs and this influences the appreciation. If people
identify a lot, or scarcely identify with a negative image the appreciation is negative. The propensity of empathy and the identification say something about how close or connected one feels towards in this case an image, and that has to do with distance and proximity.
As far as emotions are involved Burke mentions four emotions as consequences/ effects of the sublime: 1) astonishment/ being overwhelmed is the most powerful effect of the sublime according to Burke. 2) admiration, 3)reverence and 4) respect.
Burke states that the shocking of senses as a negative confrontation is influenced by the individual’s experiences in life resulting in a subjective labeling of the situation. If the situation is one the subject can handle one of the four emotions is the effect which is positive and the experience is sublime. If the situation is too scary or heartbreaking, the subject will not experience delight but will just be scared or sad.
The idea that negative emotions have a greater impact than positive emotions is also noted by Ben-Zeév (2010) who states that the value of negative emotions is greater than the value of positive emotions because of the functionality of these emotions. Negative emotions occur when a goal is blocked and another way needs to be found to achieve the goal or the situation needs to be accepted , while positive emotions are usually experienced when a goal is achieved. Accordingly, negative emotions require more cognitive resources to deal with the given situation according to Ben- Zeév.
The emphatic principle explains how it is possible to appreciate a picture of Picasso’s sad mistress. The emotions of the mistress are negative but the spectator can relate to the sad feelings for all of us have been hurt in love. So the negative emotion reflected in the picture resonates with the spectators own feelings, resulting in appreciation based on empathy. There seems to be closeness and a recognition in the encounter between art and spectator.
The sublime, to the contrary, is based on the distance between the reflection and the spectator. The spectator can appreciate the picture or other artform because the pain or fear reflected is not happening to him/her. The spectator can distance himself from the experience and be happy it is not happening to her/him. And only because of the distance will he feel delight and will he be able to appreciate the art which is confirmed in the study of Brughmans (2004) and Goris (2007).
The empathic principle and the sublime are complementary explanations for blended emotions leading to the appreciation of art. Even though the conclusion of the studies is that there is a blending of emotions possible in the appreciation of art, the measuring of the emotions was done by lists of single emotions. In all the studies a self administered questionnaire was used which included a list of ranging from 8 to sometimes 25 single emotions. Earlier studies have been done on the idea of blended emotions and how to detect these. In the next paragraph studies on detecting blended emotions are discussed.
1.4 Blended Emotions
Blended emotions are an intriguing part of the role of emotions in the appreciation of art because the idea of weeping because of the sad story played in a movie and still enjoying the movie sounds as a contradiction and yet we deal with this blend of emotions all the time, We even go looking for this experience by visiting expositions of war photographs or listening or even buying songs about the pain of a lost love. The empathic principle and the theory of the sublime provide an explanation for this phenomenon. The study of Rosenberg and De Boeck (2001) shows that experienced emotions E (which can be negative) are good predictors of the appreciation A (which is positive) of a painting when pain and sadness are involved. So knowing which emotions are experienced can provide information on how people will react,
maybe predict whether people would want to buy a painting or a song or explain why some art is more appreciated than other art or some art is more appreciated by some people with certain experiences. Being able to measure the experienced emotions might provide important information into the cognitive and affective functioning of humans.
Self administered questionnaires including lists of emotions have been used by Rosenberg and De Boeck as well as in the inquiries by Van Bree (2002), Demandt (2003), Brughmans (2004) and Goris (2007) on The Role of Emotions in Esthetic Appreciating. The lists of emotions in these questionnaires are single emotions such as sadness, happiness. In order to inquire whether it is possible to more objectively and accurately measure the emotions experienced by spectators the use of a computer program is proposed .This program is based on a facial action coding system enhanced by Du, Tao and Martinez ( 2014) by adding a pair wise landmark algorithm. The term blended emotions is used by Du et al. (2014) for a blend of two emotions.
Du et al. (2014) started out by using the facial action coding system defined by Ekman and Friesen (1978 as mentioned in Du et al. (2014) that contains representations for the muscle activation of a facial expression. Ekman is known for establishing the six basic emotions: happiness, surprise, anger, fear, disgust, sadness, all single emotions. The individual muscle or group of muscles used in the production of a facial expression is called an Action Unit (AU) with a number, e.g. AU4. AU4 defines the lowering of the eyebrows as a result of the contraction of two muscles. AU4 is seen in the expression of sadness, anger and fear. Distinguishing one facial expression from another is done by measuring the strength of the muscle activity and the combination of AU’s. In blended emotions AU‘s that are distinctly for happiness and AU’s distinctly used for surprise are both activated and can be measured. Du et al. (2014) have defined 22 distinct emotion categories, of which 15 are blended one is
neutral and six are the basic single emotions In picture two is depicted how the AU’s are combined into a blended emotion.
Picture 2 copied from Du et al. (2014) The combination of AU’s in a happily surprised face
Precision in the detection of all muscle movement is of the utmost importance in blended emotions as there is a main emotion and a secondary emotion. The muscle groups are contracted according to the first or secondary place. The study of De Bonis, De Boeck, Perez-Diaz and Nahas (1999) testing their theory that the human face when expressing emotions is like a building- block -construct in which some emotions need more blocks than others also emphasizes the fact that different parts of the face can be involved in expressing different emotions The measuring or identification of which parts of the face move when is very important in labeling blended emotions, so Du et al.(2014) developed an algorithm involving 94 fiducial (landmark) points for face-muscle movement. that can more accurately analyze the subtle muscle movements. A second improvement on analyzing AU’s by a computer is the pair wise positioning of each landmark. This pair wise positioning strengthens the AU recognition because of a joint probability of e.g. the left and right mouth corner constraining the location of the mouth corners. This finding has been acknowledged in a study by Meulders, De Boeck and Van Mechelen (2004) as they studied the importance of the spatial
relationship between parts of the face. They state that this relationship is an additional source of information when interpreting emotions in facial expressions.
Picture 3 copied from Du et al. (2014): the 94 fiducial points in faces are shown.
In picture 3 the spreading of the 94 fiducial points in two different faces are shown. These fiducial points make the detection AU’s more precise so the recognition of blended emotions, deciding which emotion is the main and which the secondary emotion which is important for naming the blend, is possible. The set of muscles in the face used to express the blend of happy and surprise resulting in a happily surprised face is consistently and distinctively different than the set of face muscles used for a angrily surprised face.
With this improved, more precise computer program it would be interesting to measure the blended emotions involved in the appreciation of art when the hedonistic principle does not explain the appreciation or non-appreciation.
1.5 Conclusions
1. The empathic principle is validated for different art forms, although in some art forms the effect is very fragile (Brughmans, 2004, the litho’s by Dali).
The empathic principle as defined by Rosenberg and De Boeck (2001) has been confirmed in various studies by master students Psychology at the University of Leuven. The main conclusion is that the empathic principle works in different art forms. The blending of negative and positive emotions has been found portraits (Rosenberg and De Boeck, 2001), pictures (Goris, 2007), and in different music styles (Van Bree, 2002 and Demandt, 2003).
2. The theory of the Sublime by Edmund Burke provides an addition to the empathic principle.
As the empathic principle explains, the reflected emotions in an art form can be negative, the emotions experienced by the spectator can be negative but the appreciation for the art form is positive. The hedonistic principle is about liking an art form that gives you a pleasant emotion, or disliking it when it evokes negative emotions. The sublime explains that one can appreciate negative reflected emotions while not experiencing these emotions oneself. This is because one can understand the impact of the negative emotion, fearing it from a safe distance and being very happy it is not one’s own emotion and experience. The result of such an experience brings delight, a stronger form of pleasure according to Burke (1757).
3. Self description has limitations.
Even though emotions are subjective, the way we try to measure them should be as objectively as possible. In the self administered questionnaires the participants self reported their emotions and reading through the list of emotions presented, can influence the choice of emotions they indicate. It is also possible that the choice of the emotion on the list is
influenced by the linguistic interpretation of the participant. This indicates that self description has its limitations.
4. No blended emotions were included in the self administered questionnaires.
In the inquiries on the role of emotions in esthetic appreciation, the emotions provided on the questionnaires were single emotions. There were no blended emotions provided to report on, although as measured in different studies discussed in the literature review this blend of emotions does occur and the theory of the sublime and the empathic principle explain the occurrence of mixed emotions especially a blend of pain and/or danger with emotions of positive appreciation.
5. A facial recognition technique for blended emotions may help.
The development of a computer program that can analyze blended as well as single emotions in facial expressions provides an opportunity to use a more objective way to measure the emotions participants experience while watching or listening to different forms of art.
As mentioned above self description has its limitations and the lists of emotions used in the studies on the role of emotions in the appreciation of art consisted only of single emotions.
Du, Tao and Martinez (2014) described the use of a more refined facial action coded system that can recognize single as well as blended emotions. The addition of an algorithm for 94 fiducial landmarks and the pair wise detection mode made the facial coding system as defined by Du et al., (2014) a more accurate instrument to use for the analysis of the facial perception of emotions. It would be interesting to use this program in the replication of the studies described in the literature review to see whether the same results and validations would be found. In the research proposal following this literature review the possibility to use this specific facial coding system is described.
2. Research proposal
2.1 Introduction
There have been many inquiries on the question why people appreciate art and especially how under certain circumstances negative emotions do not seem to be in the way of the appreciation. How come we go to a movie, weep and still say that was a beautiful story, or how come we go to scary movies? Seeing war photographs does not make most of us close our eyes, we can positively appreciate the photographs for more than the clever techniques used. Rosenberg and De Boeck (1997, 2001) proposed the empathic principle as complementary to the hedonistic principle to explain why people can appreciate art that reflects negative events or negative emotions.
The hedonistic principle (Fechner, 1876, as mentioned in Van Bree (2002)) states that people appreciate art when they experience positive emotions while looking at it or listening to it. People do not appreciate art if they experience negative emotions. Rosenberg and De Boeck (2001) have found evidence for this hedonistic principle, but they also established a second phenomenon which they interpreted as an empathic principle. They found that if negative emotions are reflected in a Picasso painting and the spectator experiences negative emotions him/herself as an spectator, the painting is even more appreciated. This is especially true when sadness and pain are reflected in the painting. These initial results have led to further studies on other kinds of art and have been replicated to some extent in master theses by Van Bree (2002), Demandt (2003), Brughmans (2004) and Goris (2007) under the supervision of Paul De Boeck.
Although empathy can explain the phenomenon of enjoying negative emotions in art, Brughmans (2004) offers an alternative explanation in terms of the sublime. In Brughmans’study lithographs by Salvador Dali depicting women in very unpleasant and even
disgusting ways were used. Gender was taken into account in the appreciation of this art as women even less appreciated the litho’s. The theory of the sublime (Burke , 1757) provides an explanation by addressing the distance factor. As the figures depicted are women, the female participants have no distance to the disgusting litho’s and can not in any way appreciate the litho or resonate the reflected emotions. Men on the other hand have somewhat more distance and can manage some level of empathy. The reason why human beings are positively intrigued by dramatic events such as in some movies when they seem to enjoy sad scenes or such as when they see an angry tiger in the zoo while they exclaim how beautiful it is instead of running away, can be found in the theory of the sublime by Edmund Burke (1757). The sublime refers to experiencing danger or pain from a certain distance, the experience of an emotion while it does not affect one’s own condition. Because of negative events people can be strongly moved emotionally while but not being completely overthrown by the experience due to the distance between the event and themselves. With a sublime experience comes relief because the terribleness enclosed in the angry tiger in the zoo or the desperate sadness of a widow depicted in a painting does not really affect the life of the spectator. This relief based on a pain or danger experience is called delight. Burke speaks of delight as the strongest form of pleasure. According to Burke, pleasure is an emotion born from beauty, it caresses our senses. Delight is an emotion born out of the sublime after an experience that shocks our senses. Both these interpretations, the empathic principle and an experience of the sublime imply a combination of negative and positive and therefore blended emotions.
2.2 Research question
The main research question that can be inferred from the previous is whether pieces of art that express something negative would lead spectators to experience blended emotions (blends of positive and negative emotions) and if these can be objectively be measured. None of the
described studies tried to measure the blended emotions even though there is theoretical evidence for these blended emotions provided by the empathic principle and the theory of the sublime. In terms of methodology the blended emotions are also a fact since they can be identified through the facial action coding system by Du et al. (2014). The methodology that was used thus far to measure emotions in the appreciation of arts is based on self -report: self administered questionnaires to indicate the experienced and reflected emotions. There is a chance that participants would not admit enjoyment as a spectator of negative events. Because of this and other limitations of self-description , we have looked for alternative methodologies. One such alternative methodology would be the registration of facial expressions as studied by Ekman (1978 as mentioned in Du et al.(2014).
However, for our specific purpose we are interested in blended emotions and the identification of such emotions. This is why the facial action coding system by Du, Tao, and Martinez (2014) may be useful. These authors were able to identify 22 emotions with their advanced facial action coding system, 15 blended emotions, one neutral face and six single emotions. The six single emotions are the basic emotions identified by Ekman (1976 as mentioned in Du et al.(2014).When such blended emotions can be identified in the spectators, a second research question is whether they are related to the appreciation of the pieces of art. Both research questions will be studied from a between-subjects and a within-subjects perspective using mixed modeling, to capture:
- global individual differences (global differences in experiencing blended emotions, global differences in appreciation of art);
- differences between pieces of art regarding blended emotions and appreciation of art; - individual differences in how the experience of blended emotions across pieces of art is related to appreciation of art.
Gender is specifically important when the depicted person in the painting is of the same gender as the spectator as became clear in the study by Brughmans (2004). One can expect less blending with a positive component when the distance is smaller (same gender).
2.3 Research method
Six paintings by Picasso and six paintings by Dali will be presented individually to subjects through projection on a screen. The paintings will be chosen on the basis of research by Rosenberg and De Boeck (1997) for Picasso and Brughmans (2004) for Dali. Although far from the ideal of an experimental study, because the selection of the stimuli is on gender and negativity ( all these paintings depict women and they all represent something negative) this is a first step. This experiment will be complementing the previous studies and it is of interest to further explore the theory of the sublime on distance with regard to women and appreciation of the litho’s. In follow up studies art depicting men in a disgusting way could be used to see if the same results come up. So while the gender issue is an experimental limitation, it gives the study more ecological validity for the world of art and it is a better follow-up of the earlier studies. In the pilot stage of the study only five male subjects and five female subjects will be included. If a sufficient amount of blends of positive and negative emotions are observed, the full study will be implemented with 20 male subjects and 20 female subjects. If almost no blends are observed, then the study will be continued with another pilot but with different kinds of art. While looking at the paintings, the facial expression of the subjects will be registered and later on subjected to the coding system. For each painting, the subjects will be asked to rate their appreciation of the painting.
Follow-up studies can be planned with a more complete design but with another kind of stimuli: women and men crossed with negative and positive. For now it is important to have some proof of concept regarding the methodology.
2.4 Data analysis
In a preliminary analysis the results of facial coding system will be explored to find out which blends of positive and negative emotions are observed most frequently. If almost no such emotions are identified in the pilot, another set of artistic stimuli needs to be found. If the frequency (how many times) is high enough, a score will be determined per pair of a subject and a painting based on the three most frequently found blends. This score will be the basis for mixed modeling, of the linear kind (ignoring the scores are frequencies) and with a loglink (recognizing the scores are frequencies). A Poisson model is often used to for frequencies implying that the expected value of the log is modeled. In a first analysis the blended emotion score is the dependent variable. The model will include a random intercept (for the individual differences) and fixed effects of gender and of the paintings. Both a linear and a loglink (Poisson) model will be used in order to check the sensitivity of the effects for the assumptions of the model. For the loglink the equation is:
log(Freqpa) = cp + ba + bgender + epa
where Freqpa is the frequency of a blended emotion experienced by person p as a spectator of
art piece a,
cp is a random person intercept,
ba is the fixed effect per piece of art,
bgender is the effect of gender,
epa is the residual.
The second analysis focuses on the artistic appreciation ratings and will again have a random intercept and a fixed gender effect. For the paintings the blended emotion score (or the log) per painting will be introduced as a painting covariate with a fixed effect to explain
appreciation differences between the paintings and with a random slope in order to investigate the individual differences in the relationships between blended experiences and appreciation.
Ideally, the blended score covariate can be used per pair of a subject and a painting, but that would lead into complicated types of analysis. For the appreciation ratings, the mixed model is linear:
Ypa = cp + bXa + wpXa + epa
where Ypa is the appreciation of piece of art a by person p;
cp is a random person intercept;
Xa is the blended emotion score per piece of art, and b is its fixed effect;
wp is the random person slope of the blended emotion score of the pieces of art (expressing
how much blended emotions affect appreciation); epa is the residual.
2.5 Interpretation
1. The results of the facial coding system will tell us whether blends of positive and negative emotions actually occur in spectators of paintings that express something negative. 2. The results of the mixed modeling with and without loglink can tell us about individual differences and painting differences with respect to blended emotions of art spectators. For example models with and without these differences can be compared. 3. The results can tell us about gender differences in experiencing blended emotions. Due to the limitation of the experimental set-up, we cannot know whether a possible gender difference is an overall gender difference or a difference that is induced by the paintings depicting female subjects. The latter would be in line with the theoretical expectations because female subjects most likely experience less of the distance that is needed for the blending. However, the former explanation cannot be excluded. 4. The results of a linear mixed model will tell us about the relationship between blending and artistic appreciation and individual differences in that relationship.
References
Ben-Zeev, A,( 2010). Are Negative emotions more important than positive emotions: The average person considers herself happier than the average person, found online April 2015
https://www.psychologytoday.com/blog/in-the-name-love/201007/are-negative-emotions-more-important-positive-emotions.
Van Bree, L. (2002). De Rol van Gevoelens in Esthetische Appreciatie: een onderzoek over aria’s uit De Toverfluit van W.A. Mozart. Niet –gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
Brughmans, L. (2004). De Rol van Gevoelens in Esthetische Appreciatie: een onderzoek over lithografieën uit ‘Dali illustre Casanova’ van Salvador Dali. Niet -gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
Burke, E. (1757). A Philosophical Enquiry into the Origin of our Ideas of the Sublime and Beautiful, London: Oxford University Press
Cauwenbergh, M. (2007). De Rol van Negatieve Gevoelens in Esthetische Appreciatie: een onderzoek over lithografieën uit ‘Dali illustre Casanova’ van Salvador Dali. Niet –
gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
De Boeck, P., & Rosenberg, S. (1997). Emotions reflected in Picasso’s Portraiture of Four Women in His Intimate Life. In Perevia F. (Ed.), Proceedings of the Fourteenth International Conference on Literature and Psychoanalysis (p.293-298). Lisbon: Instituto Superior de Psicologia Aplicada. online: http://ufdc.ufl.edu/UF00019686/00007 April 2015
De Bonis, M., De Boeck, P., Pérez-Diaz, F., & Nahas, M. (1999). A Two-Process Theory of Facial Perception of Emotions. C.R. Acad. Sci. Paris, Sciences de la vie / Life sciences (pp. 669-675).
Demandt, T. (2003). De Rol van Gevoelens in Esthetische Appreciatie: een onderzoek over de Goldbergvariaties van J.S. Bach. Niet- gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
Du, S., Tao, Y., & Martinez, A.M. (2014). Blended Facial Expressions of Emotion. Department of Electrical and Computer Engineering, and Center for Cognitive and Brain Sciences. The Ohio State University, Colombus, OH 43210.
Goris, S. (2007). De Rol van negatieve Gevoelens in Esthetische Appreciatie: een onderzoek over oorlogsfoto’s van James Nachtwey. Niet –gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
Kotsch, L. (2007). De Rol van Gevoelens in Esthetische Appreciatie: een onderzoek over rap muziek. Niet –gepubliceerde licentiaatverhandeling Katholieke Universiteit Leuven, Faculteit Psychologie en Pedagogische Wetenschap, Leuven.
Lewinski, P., Uyl, den, T.M., & Butler, C.(2014) Automated Facial Coding: Validation of Basic Emotions and FACS AUs in FaceReader. Journal of Neuroscience, Psychology and Economics (ISSN;1937321X), Vol.7(4), (pp.227-236).
Meulders, M., De Boeck, P., & Van Mechelen, I. (2005). Probabilistic Feature Analysis of Facial Perception of Emotions. Appl.Statist. 54, Part 4, (pp. 781-793).
Rosenberg, S., & De Boeck, P. (2001). The Role of Emotion in Aesthetic Preference. Submitted to Cognition an Emotion.
Rosenberg, S., & De Boeck, P. (1997). Emotions depicted and experienced: Picasso’s portraiture. In L. Dorfman, C. Martindale, D. Leontiev, G. Cupchick, V. Petrov, & P. Machotka (Eds.). Emotion, Creativity, & Art. Vols. 1 & 2 (Vol. 2, pp. 371-386). Perm, Russia: Perm State Institute of Arts and Culture.
