GOR 2001 Homepage Contributions Abstracts Articles article-info Indices first-author-index complete-author-index Management - Staff only Infos

GOR 2001 - content This is the http://kiwi.uni-psych.gwdg.de/congress/gor-2001/contrib/horstmann-gernot/horstmann-gernot Document. Main Author: Horstmann, Gernot Co-Authors: ; Institution: Fakultät für Psychologie und Sportwissenschaft, Universität Bielefeld Contribution Title: Einschätzungen von Typikalität und Häufigkeit eines mimischen Ausdrucks. Authors Email: gernot.horstmann@uni-bielefeld.de URLs: http://www.uni-bielefeld.de Abstract German (version: 25/06/2002 - 07:47, size: 1270) German: Die Untersuchung des mimischen Ausdrucks von Emotionen gehört zu den klassischen Gebieten der Emotionspsychologie. Hierbei wurden in der Regel als Stimuli Bilder verwendet, die einen sehr starken mimischen Ausdruck einer bestimmten Emotion zeigten. Im Unterschied dazu wird in der vorliegenden Online-Studie die Intensität des mimischen Ausdrucks in fünf Stufen zwischen einem neutralen Ausdruck und einem maximalen Ausdruck variiert. Damit sollte untersucht werden, wie Einschätzungen der Typikalität (eines Ausdrucks für eine Emotion) und der Häufigkeit (mit der ein Ausdruck bei Vorliegen der Emotion gezeigt wird) in Abhängigkeit von der Intensität des dargestellten Ausdrucks variiert. Es wird angenommen, daß die Typikalität eine monoton steigende Funktion der Intensität des Ausdrucks ist, wobei maximale Intensitäten maximale Typikalitätsurteile hervorrufen. Im Unterschied dazu sollte die Häufigkeit einen umgekehrt U-förmigen Zusammenhang zur Intensität aufweisen, wobei mittlere Intensitäten als am häufigsten auftretend eingeschätzt werden. Die Ergebnisse werden im Zusammenhang mit Theorien zur Natur und Funktion von mentalen Repräsentationen diskutiert. Article (version: 25/06/2002 - 07:47, size: 7086) Judgements of Typicality and Frequency of Facial Expressions of Emotion Gernot Horstmann University of Bielefeld, Germany OBJECTIVE The study of facial expressions has a long tradition within psychology. In this tradition (e.g. Darwin, 1965; Ekman, 1972), certain facial expressions have been widely regarded as prototypical displays of a few basic emotions, in particular of happiness, sadness, fear, anger, disgust, and surprise. However, critics (i.e. Fridlund, 1994; Russell, 1994) have contended that the proposed prototypes, which are typically quite intense expressions, are (a) exaggerated and may (b) occur only infrequently in everyday life. Presumably, those critics implicitly assume that emotional expressions are common taxonomic categories, where the prototype represents a central tendency and is encountered frequently. However, expressions may be variants of goal-derived categories (Barsalou, 1985), where the prototype represents an ideal, which often is an extreme value. The aim of the study was twofold. The first aim was to empirically assess the relationship between judged typicality and judged frequency of occurrence. The second aim was to find out what determines judged typicality, following a procedure used by Barsalou (1985). METHOD Participants 85 participants (age: M= 27.9, SD= 10.6 yrs., gender: 23 men, 56 women, 6 reported no gender) completed the study. The experiment was freely accessible from the internet. Most of them, however, probably responded to flyers that were distributed on the university campus. Design. 36 participants rated the typicality, and 33 participants rated the frequency of occurrence of facial expressions. 16 additional participants rated the ideal of a clear communication of emotion. Stimuli 5 sets of stimuli were used, one set for each emotion (anger, disgust, happiness, sadness and surprise). Each set consisted of 5 stimuli: one neutral face, one full expression according to the criteria of Ekman and Friesen (1975), and three faces that were linear extrapolations between the neutral and the full face (see, for example, Figure 1). The stimuli were computer-generated line-drawings published by Musterle and Rossler (1986).   Figure 1. Screenshot from the experiment. Procedure and Materials The experiment was controlled by a CGI-script written in PERL, and the materials were displayed via the internet browser as a HTML document. After reading the general instruction and reporting age and gender, each participant completed five pages, each of which was headed by the emotion name and contained the corresponding stimulus set. For each face, a judgement had to be made with HTML-style radio-buttons forming a five-point rating scale (see Figure 1). The order of the emotions within the five screen pages was varied randomly between participants. When the participants submitted the data, the PERL script checked for completeness of the data. If the data were incomplete, the same page was sent back to the participant. If the data were complete, the next page was presented until all five pages had been completed.   RESULTS As can be seen in Figure 2, the typicality and the ideal ratings are nearly perfectly linearly related to expression intensity. In contrast, the frequency of occurrence showed an inverted-U shaped relationship to intensity. Figure 2. Mean typicality, frequency and ideal ratings for each of the five intensities of each of the five emotions. Table 1 shows the first order correlations between the three judgements, the intensity of the expressions, and the CT-score representing proximity to central tendency. The CT score was 1 for the neutral and the full face, 3 for the face with medium intensity, and 2 for the remaining two faces. Further analysis paralleled that of Barsalou (1985) and was intended to determine which variable – central tendency, ideal, or frequency of instantiation – predicts the judgement of typicality best. It focussed therefore on the partial correlations between typicality and the other variables. Table 2 shows the partial correlations between typicality and each of the other variables, statistically controlling for the influence of the remaining variables.   CONCLUSION The data further support the assumption that the prototype of diverse categories of facial expressions of emotion is an intense, "full" facial expression. The novel finding that rated typicality is linearly related to the intensity of the expression, is in full accord with this assumption. Furthermore, the results support the notion that these full faces are not necessarily those that are most often displayed in everyday life when people experience an emotion. Third, when asked how good a given expression is suited to express the corresponding emotion, the ratings were again linearly related to the intensity of the expression. This corroborates the notion that the function of emotional expressions is the communication of emotion. As to the second aim of the study, the data show that judged typicality is predominantly related to the ideal of emotion communication, and only to a moderate degree to frequency of occurrence (88% vs. 17% of unique explained variance). The unique variance explained by central tendency was nearly zero. These results suggest that facial expressions are not common taxonomic categories, but a variant of an ideal-based goal-derived category.   References Barsalou, L. W. (1985). Ideals, central tendency, and frequency of instantiation as determinants of graded structure in categories. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 629-654. Darwin, C. (1965). The expression of the emotions in man and animals. Chicago: Chicago University Press (Original work published 1872). Ekman, P. (1972). Universals and cultural differences in facial expressions of emotion. In J. Cole (Ed.), Nebraska Symposium on Motivation (Vol. 19, pp. 207-283). Lincoln: University of Nebraska Press. Ekman, P., & Friesen, W. V. (1975). Unmasking the face: A guide to recognizing emotions from facial clues. Englewood Cliffs, NJ: Prentice-Hall. Fridlund, A. J. (1994). Human facial expression: An evolutionary view. San Diego: Academic Press. Musterle, W. & Rossler, O. E. (1986). Computer faces: The human Lorenz matrix. BioSystems, 19, 61-80. Russell, J. A. (1994) Is there universal recognition of emotion from facial expression? A review of the cross-cultural studies? Psychological Bulletin, 115, 102-141.   Note If you like to contact me, please send a mail to gernot.horstmann@uni-bielefeld.de