emotional response to music

5 downloads 0 Views 738KB Size Report
You are going to hear a 20-minute excerpt from Act I of Puccini' s La Boheme. This study is an attempt to provide ongoing information concerning what.
Psychomusicology, 16, 59-67 © 1997 Psychomusicology

EMOTIONAL RESPONSE TO MUSIC Clifford K. Madsen Center for Music Research The Florida State University This paper concerns issues regarding ongoing measurement of emotional response to music. It provides a truncated overview of related research concerning individual investigations using the Continuous Response Digital Interface. In this series of investigations results indicate that emotional response to music can indeed be measured and that highly trained musicians and those having little formal training have extremely similar responses to a variety of Western art music selections. Introduction There are several questions concerning emotional response to music that should be of interest to researchers: Can we construct a mode of response that can serve as an analogue of the intensity of listeners' emotional response to music? Does the output obtained with this mode of responding reflect what subjects believe to be their response(s) to music? How consistent are subjects' expressions of their responses, and how similar are these responses between subjects? This paper deals with the above questions and presents an overview of research findings, most of which have been summarized elsewhere (Madsen, 1996; Madsen, in press; Madsen & Gregory, 1997). All reported studies used the Continuous Response Digital Interface (CRDI) technology developed in the Center for Music Research at The Florida State University (Madsen, 1990). This technology represents extended development of previous continuous response measuring devices (Clynes & Nettheim, 1982; Greer, 1981; Hatoh, Kato,Kuwano,&Namba, 1989;Kuwano&Namba, 1985; Nielsen, 1987). The CRDI permits both discreet and/or continuous non-verbal measurements of subjects' responses to whatever variable or variables the investigator chooses to study. It is capable of sampling from one to 10,000 times each second, and various configurations of the user interface make it possible to study many aspects of music for both categorical and continuous data. Since its initial development in 1988 (Robinson, 1988), we now have four separate devices that have been used in a variety of studies. The latest is a two-dimensional CRDI that projects x and y coordinates (in the shape of an open box) on a computer monitor. Reliability has been assessed across all of the devices and has been determined to be very high (Capperella, 1989; Gregory, 1989; Gregory, 1995).

Madsen

59

The Initial Investigation In an initial study we attempted to ascertain just what represents the "it" that people refer to when they speak of aesthetic responsiveness or the aesthetic experience in relationship to music. This area of investigation seems important because aesthetic responsiveness has often been used to justify the place of music in the lives of children and adults (Madsen & Yarbrough, 1980, p. 18). The first study (Madsen, Brittin, & Capperella-Sheldon, 1993) was designed to assess aesthetic responsiveness of extremely sophisticated musicians in a large school of music. Subjects (N= 30) were given the following instructions: You are going to hear a 20-minute excerpt from Act I of Puccini' s La Boheme. This study is an attempt to provide ongoing information concerning what you define as the aesthetic experience. As you listen to the music, move the dial corresponding to your aesthetic response. Subjects were seated before a CRDI dial with the words negative at one end and positive at the other. At the conclusion of the listening session, subjects completed an exit questionnaire. Results indicated that all subjects reported having an aesthetic experience(s). All respondents also indicated that their movement of the CRDI dial roughly corresponded to the variations of their aesthetic experience. No subject indicated that this "experience" lasted throughout the entire act. Analyses of the graphic plots of dial movement across time, combined with questionnaire results indicated that there were several "peaks and valleys" relating to greater/lesser aesthetic interest which was evident across all subjects and may correspond to theories initially advanced by Meyer (1956). Test-retest reliability, which was .90, also indicated that most people have the same peaks at the same points in time across repeated listening, even after several weeks, which is consistent with data reported by Sloboda (1991). In the second set of investigations (Madsen, Byrnes, & Capperella-Sheldon, 1993) we hypothesized that sophisticated musicians, whose training includes many years of study in the Western art tradition, would have very different specific and collective responses when compared to nonmusicians. In the next study, which used the same Puccini excerpt, a group of sophisticated musicians was compared to listeners with little formal training (i.e., people with less than three years of formal music study). The overall response of both groups was very similar. This surprising result stimulated another series of studies using different music stimuli. We used a wide variety of examples in the next studies in order to sample across traditional composers and performing media: the last 13 minutes ofR. Strauss' Death and Transfiguration', G. Hoist's 1st Suite for Military Band, 1st Movement; Haydn's Symphony #104,1st Movement, and Mozart's "Laudate Dominum" from Vesperae Solennes de Confessore in CMajor. Sophisticated musicians were compared with listeners with little formal study (Madsen, 60

Psychomusicology • Spring/Fall 1997

Byrnes, Brittin, & Capperella-Sheldon, 1993). In all these additional studies graphic results were much the same as with the Puccini excerpt; trained musicians' responses were not very different from those of nonmusicians. Although the responses of individuals within each group differed from one another, musicians and nonmusicians evidenced extremely similar group responses. Analyses of exit questionnaires indicated that 90% or more of all subjects, across all studies, had had what they considered to be an aesthetic experience, and 90% or more also indicated that movement of the dial roughly corresponded to their experience. "Aesthetic experience" was purposefully not defined in any of the above studies, yet some tentative answers emerged and one aspect seemed consistent. The aesthetic experience, according to subjects' responses in these studies, does not appear to be an encompassing attribute of music that begins with the start of a piece of music and continues in the same way throughout. In no case did an individual set the dial of the CRDI and leave it there throughout the listening experience. Also, no one reported in the exit questionnaires an aesthetic experience that lasted the entire excerpt. In light of these results, it seemed that the responses of musically trained adults do not differ appreciably from responses of musically untrained adults. There are, however, important and interesting differences when comparing graphs of different music selections. These results seem consistent with those reported by Waterman (1996). There seem to be certain musical events that predictably elicit an "aesthetic response," if focus of attention is maintained. Also, each separate piece of music has a very distinctive aesthetic response profile which we are calling a musical "footprint." Additionally, it seems that subjects in all these studies were responding to emotional aspects elicited by the music, regardless of the "aesthetic" definition. Coggiola (1997) attempted to extend the music samples by using excerpts from the jazz idiom. Inexperienced and experienced jazz musicians responding to four jazz excerpts were chosen for various levels of melodic advancement from simple to complex. Only during the most melodically advanced jazz excerpt did subjects evidence a differentiated response, with the experienced jazz musicians rating the performance as having higher aesthetic interest compared to the nonexperienced group. It is important to note that this singular jazz excerpt appears to be the only music selection where experienced musicians evidenced a higher differentiated response when compared to their nonexperienced counterparts. However, in a study by Goins (in press) in which jazz selections from the music of Pat Metheny were used as stimuli, experienced musicians did not evidence listening differences when compared to inexperienced nonjazz musicians. Focus of Attention In an attempt to ascertain what element(s) in music contribute to listeners' aesthetic response we designed a series of studies concerning what musicians attend to (or focus on) while listening to various musical excerpts. In these studies we used the following elements of music: melody, dynamics, Madsen

61

timbre, rhythm and a category that we called "everything" which represented subjects listening to all elements at once without specifically focusing on any one element (Madsen & Geringer, 1990; Geringer & Madsen, 1996). These elements certainly are not exhaustive as there are many ways that various taxonomies can be structured (Brittin, 1991; Madsen & Kuhn, 1993. p. 74). This type of taxonomic organization did permit comparison with results of the previous aesthetic research. We compared results obtained from a group of subjects who responded by tracking their degree of attend veness to individual music elements during the La Boheme excerpt to the overall aesthetic response for this same selection. Comparisons were made via Pearson correlations for each one-second time interval sampled. The Melodic element correlated .67 with the composite aesthetic graph; the Dynamic correlation was .59 and the combination of Everything correlated .53. Interestingly, the element of Timbre correlated -.14 and the Rhythmic element revealed a -.58 correlation. Additionally, exit questionnaire responses that asked subjects if they had had an aesthetic response were almost identical to results from similar response forms used with the previous aesthetic studies (Madsen, 1997). An additional study compared focus of attention to these same musical elements while subjects listened to the entire 1st Movement of Haydn's Symphony #104 (Madsen, Geringer, & Fredrickson, 1997). In this study only the element of rhythm evidenced a moderately strong relationship to the previous aesthetic response. All the other elements demonstrated relatively low relationships when compared to the aesthetic response profile. Pearson correlations were: rhythmic .69, everything .34, melodic .23, timbral .14, and dynamic -.004. Although subjects were asked to track their aesthetic responses to the above music selections, we assumed that subjects were actually tracking their emotional responses on the above CRDI studies. Adams (1997) eschewed the aesthetic terminology and asked subjects to track their felt emotional response during the last 9 minutes of Mahler's Symphony #2 (Resurrection). Adams also tested the visual aspect of felt emotional response using a Leonard Bernstein television recording of the Mahler. He enlisted three groups of subjects: One group engaged in listening only, another group only watched the television monitor without sound, the third group both watched and listened. All the group's graphic data were similar when analyzing the graphic peaks and valleys. Although exhibiting much less overall felt emotional response, the group who only watched the television had the same CRDI graphic contours compared to the other listening groups. Lychner's (in press) research provides even more interesting results and raises even more troublesome questions concerning how to label music responses. Lychner attempted to investigate the terms that had been used in our previous research with four new selections: "Nessun dorma" from Puccini's Turandot; Fugue from the Toccata and Fugue in d minor by Bach; Scherzo from Beethoven's 3rdSymphony and Sousa' s Stars and Stripes Forever. Lychner compared four different instructional groups. One group was asked to track 62

Psychomusicology • Spring/Fall 1997

their aesthetic response, as in previous studies; another group tracked their felt emotional response as in the Adams study above; a 3 rd group indicated their perceived tension, and the 4th group (intended as a control group) was presented with an overlay that had the word More on one end, and Less on the other. The control subjects were given no instructions except that they could move the dial as they wished. Graphic analyses indicated a strong relationship between the graphs of subjects evaluating aesthetic response and the group evaluating their felt emotional response. The perceived tension group was different from all other groups, which is a finding consistent with our previous research (Fredrickson, 1995; Fredrickson, 1997; Fredrickson & Johnson, 1996; Madsen & Fredrickson, 1993; Nielson, 1987). All of our research suggests that tension, although related to the aesthetic response, represents a unique dimension. Surprisingly, the free response condition (in which subjects were given no instructions except to move the dial) demonstrated a remarkable similarity to the aesthetic and felt emotional response conditions. Even with no instructions, except being presented with an overlay containing the words More and Less, subjects managed to indicate a response that appears to be extremely similar to their felt emotional and aesthetic response. We tentatively concluded that there are only two dimensions related to music listening: emotion and tension. The Two-dimensional Studies The two-dimensional CRDI uses a mouse with a computer monitor and provides the possibility of using the computer screen via a mouse to indicate the interrelationship of two dimensions simultaneously. Development of the two-dimensional continuous response was done in two independent settings originally based on the previous work of Russell' s circumplex model of emotions (1980), where emotions are distributed around a two dimensional space concerning "valence" (happiness or sadness) and "arousal" (passiveness or activeness). One prototype (Tyler, 1996) was developed in our Center for Music Research at The Florida State University where the first CRDI originated. Schubert (1996), using Russell's same theoretical model, developed a slightly different two-dimensional prototype at the University of South Wales, Sydney. This line of research uses two separate Osgood-like systematic differentials anchored with opposite descriptors and placed on the television screen with one word placed at the top of the screen and its opposite placed on the bottom. Similarly, another descriptor is placed on one side of the screen and its opposite on the other side of the screen. In my first two-dimensional study (Madsen, 1997) I thought it would be important to investigate applications of the two-dimensional CRDI using the same music excerpts that we had previously studied in the original CRDI aesthetic studies. I also wanted to use the two dimensions that had emerged from our previous investigations that suggested the two different aspects of emotion and arousal. Therefore, I used the same Puccini's La Boheme, Act 1 excerpt we had used in our previous investigations (Madsen, Brittin, & CapperellaMadsen

63

Sheldon, 1993). Each subject (N = 50) moved the mouse cursor tracking the dimensions of arousal, i.e., relaxing-exciting and affect, i.e., ugly-beautiful simultaneously. The vertical dimension on the computer screen listed the words Exciting at the top and Relaxing at the bottom; the horizontal axis listed Ugly on the left side and Beautiful on the opposite side. Visual and temporal analyses of this 20 minute selection indicated that there is not a consequential difference between subjects' responses to the ugly-beautiful dimension throughout the excerpt that appear substantially different from our previous La Boheme aesthetic responses. Furthermore, subjects' arousal responses (i.e., relaxing vs. exciting) did not replicate the ugly-beautiful dimension and, therefore, adds important new information to overall analysis of emotional response to music. Responses to the relaxing-exciting dimension represent degrees of arousal and suggests one needs to be somewhat aroused in order to have an emotional response. My second two-dimensional study used the same word configuration and compared musicians' responses (N = 48) while listening to the same 1st Movement of Haydn's Symphony #104, that we used in our previous research concerning aesthetic response and tension response (Madsen, in press). Not surprisingly, results indicated an inverse relationship between the two dimensions. The correlation between the two dimensions was r= -.58. Graphic analysis indicated almost a "mirror image" relationship between the two dimensions. When analyzed within the musical context, both the arousal and the aesthetic dimensions evidenced many subtle characteristics in relationship to various "peaks and valleys" that corresponded to variations in the music. In this two-dimensional study, however, the exciting/relaxing dimension did not replicate previous research that asked musicians to track their tension responses, yet the exciting/relaxing dimension was very similar to previous tension responses of nonmusicians. Obviously more research is warranted. Discussion Several issues relating to these studies seem to be important for music researchers. The use of current technology that obviates dependence on paper and pencil verbal reports is an important development in studying ongoing responsiveness to music listening. Specifically, using one of the several CRDI devices seems very appropriate for longer music examples. While various combinations of short excerpts gives us information concerning some of the many ways people respond to music, longer examples make it possible to study detailed as well as the broader responses over extended periods of time. However, in a recent study concerning measurement of short music stimuli (30 s) it appears CRDI data are extremely similar when compared to traditional paper and pencil instruments (Madsen & Geringer, in press). It should be noted that the La Boheme selection provided the greatest graphic variation between high and low "aesthetic" or "emotional" interest. This may indicate that certain music is, indeed, more "thrilling" at times (Goldstein, 1980). In the original aesthetic study (Madsen, et al., 1993) it was suggested 64

Psychomusicology • Spring/Fall 1997

that in order to have the "peak experience" or "thrilling" experience of which some researchers speak (Goldstein, 1980; Panksepp, 1995; Sloboda, 1991) one must spend several minutes in highly concentrated focus of attention, especially the 30-45 seconds immediately proceeding the peak experience. If concentration is broken (especially by a competing overt experience such as talking), the listening, while pleasant, does not seem to evoke an intense aesthetic response. Thus, if one is distracted by anything, this may represent a competition for focus of attention (Madsen, 1987). Perhaps the most important unifying aspect of these studies concerns focus of attention. There appear to be events in ongoing music that predictably elicit emotional responses, but only if focus of attention is maintained, especially during the 30-45 seconds that precede the emotional response. The similar methodology of these studies, which instructs subjects to listen attentively and indicate via a CRDI dial aspects relating to their music listening, has the common variable of keeping people on task. Indeed, it seems that many people do not really listen to music for more than a few seconds, unless they are listening for something specific (Madsen & Madsen, 1997, p. 45). In summary, the research reported above suggests that music researchers should continue to establish a firmly researched taxonomic basis for the descriptors used to assess music listening as well as ways to indicate reactions; moods, thoughts, preferences, and/or other responses to music. While extremely short acoustic stimuli can be more readily controlled in tight experimental investigations (Madsen & Geringer, in press; Madsen, Edmonson & Madsen, 1969), we also need to investigate longer musical constructs, complete melodies, phrases, and even complete music works (Aiello, 1994) in order to acquire more knowledge concerning emotional response to music. The CRDI seems very well suited for these longer musical excerpts.

References Aiello, R. (1994). Can listening to music be experimentally studied? In R. Aiello & J. A. Sloboda (Eds.), Musical perceptions (pp. 273-282). New York: Oxford University Press. Brittin, R. V. (1991). The effect of overtly categorizing music on preference for popular music styles. Journal of Research in Music Education, 39, 143-15. Byrnes, S. R. (1994). Elementary school, middle school, high school, and trainable mentally handicapped students' response to four western art selections. Unpublished doctoral dissertation, Florida State University, Tallahassee. Capperella, D. A. (1989). Reliability of the Continuous Response Digital Interface for data collection in a study of auditory perception. Southeastern Journal of Music Education, 1, 19-32. Clynes, M., & Nettheim, N. (1982). The living quality of music. In M. Clynes (Ed.), Music, mind, and brain: The neuropsychology of music. New York: Plenum Press. Coggiola, J. C. (1997). The effect of conceptual advancement in jazz music selections andjazz experience on musicians' affective response. Doctoral dissertation, The Florida State University, Tallahassee. Fredrickson, W. E. (1995). A comparison of perceived musical tension and aesthetic response. Psychology of Music, 23, 81-87. Madsen

65

Fredrickson, W. E. (1997). Elementary, middle, and high school student perceptions of tension in music. Journal of Research in Music Education, 45, 626-635. Fredrickson, W. E., & Johnson, C. M. (1996). The effect of performer use of vibrato on listener perception of tension in Mozart. Psychomusicology, 75(12), 78-86. Geringer, J. M., & Madsen, C. K. (1996). Focus of attention to elements: Listening patterns of musicians and nonmusicians. Bulletin of the Council for Research in Music Education, 127, 80-87. Goins, W. E. (in press). The effect of moodstates: Continuous versus summative responses, Journal of Music Therapy. Goldstein, A. (1980). Thrills in response to music and other stimuli. Physiological Psychology, 8, 126-129. Greer, R. D. (1981). Response. In Documentary report of the Ann Arbor Symposium: Applications of psychology to the teaching and learning of music (pp. 329-332). Reston, VA: Music Educators National Conference. Gregory, D. (1989). Using computers to measure continuous music responses. Psychomusicology, 8, 127-134. Gregory, D. (1995). The Continuous Response Digital Interface: An analysis of reliability measures. Psychomusicology, 14(1 & 2), (Spring/Fall, 1995): 197208. Hatoh, T., Kato, M., Kuwano, S., & Namba, S. (1989). Assessment of musical performance using the method of continuous judgment of selected description. Proceedings of the First International Conference on Music Perception and Cognition (pp. 33-36), Kyoto, Japan. Kuwano, S., & Namba, S.(1985). Continuous judgment of level-fluctuating sounds and the relationship between overall loudness and instantaneous loudness. Psychological Research, 47, 27-37. Lychner, J. A. (in press). An empirical study concerning terminology relating to aesthetic response to music. Journal of Research in Music Education. Madsen, C. K. (1987). Background music: Competition for focus of attention. In C. K. Madsen & C. A. Prickett (Eds.), Applications for Research in Music Behavior, (pp. 315-325). Tuscaloosa, AL: The University of Alabama Press. Madsen, C. K. (1990). Measuring musical response. Music Educators Journal, 77(3), 26-28. Madsen, C. K. (1996). Empirical investigation of the aesthetic response to music: Musicians and nonmusicians. Proceedings of the 4th International Conference on Music Perception and Cognition, Montreal, Canada, 103-110. Madsen, C. K. (in press). Research in Music Behavior. In J. W. Richmond & P. Webster (Eds.), "A Festschrift for Bennett Reimer. " Northwestern University Press. Madsen, C. K. (1997). Focus of attention and aesthetic response. Journal of Research in Music Education, 45, 80-89. Madsen, C. K., Brittin, R. V., & Capperella-Sheldon, D. A. (1993). An empirical investigation of the aesthetic response to music. Journal of Research in Music Education, 41(1), 57-69. Madsen, C. K., Byrnes, S. R., Capperella-Sheldon, D. A., & Brittin, R. V. (1993). Aesthetic responses to music: Musicians vs. nonmusicians. Journal of Music Therapy, 30(3), 174-191. Madsen, C. K., & Fredrickson, W. E. (1993). The experience of musical tension: A replication of Nielsen's research using the continuous response digital interface. Journal of Music Therapy, 30, 46-57. Madsen, C. K., & Geringer, J. M. (1990). Differential patterns of music listening: Focus of attention of musicians versus nonmusicians. Bulletin of the Council for Research in Music Education, 105, 45-47.

66

Psychomusicology • Spring/Fall 1997

Madsen, C. K., & Geringer, J. M. (in press). Comparison of good versus bad tone quality/intonation of vocal and string performances: Issues concerning measurement and reliability of the Continuous Response Digital Interface. Bulletin of the Council for Research in Music Education. Madsen, C. K., Geringer, J. M., & Fredrickson, W. E. (1997). Focus of attention to musical elements in Haydn's symphony #104. Bulletin of the Council for Research in Music Education, 133, 57-63. Madsen, C. K., & Gregory, D. (1997). "Use of Technology in Understanding 'Aesthetic Response' in Music." Proceedings of the 3rd International Symposium, Research Alliance for Institutes of Music Education, Tallahassee, FL. Madsen, C. K., & Kuhn, T. L. (1993). Contemporary Music Education (2nd ed.). Raleigh, NC: Contemporary Publishing, Inc. Madsen, C. K., & Madsen, C. H., Jr. (1970). Experimental Research in Music. Englewood Cliffs, NJ: Prentice-Hall Publishing, Inc. Madsen, C. K., & Yarbrough, C. (1985). Competency Based Music Education. Raleigh, NC: Contemporary Publishing, Inc. Meyer, L. B. (1956). Emotion and meaning in music. Chicago, IL: The University of Chicago Press. Nielsen, F. V. (1987). Musical tension and related concepts. In T. A. Sebeok and J. Umiker-Sebeok (Eds.), The semiotic web '86. An international yearbook (pp. 491-531). Berlin: Mouton de Gruyter. Panksepp, J. (1995). The emotional sources of "chills" induced by music. Music Perception, 13, 171-207. Robinson, C. R. (1988). Differentiated modes of choral performance evaluation using traditional procedures and a Continuous Response Digital Interface device. Doctoral dissertation, The Florida State University, Tallahassee, FL. Sloboda, J. A. (1991). Music structure and emotional response: Some empirical findings. Psychology of Music, 79(2), 110-120. Waterman, M. (1996). Emotional responses to music: Implicit and explicit effects in listeners and performers. Psychology of Music, 24(1), 53-67.

Madsen

67