Trait Anger and Music Perception

Mark C. Gridley

ABSTRACT – The Spielberger Trait Anger test was administered to 287 undergraduate college students enrolled in courses in jazz appreciation. The recording of a jazz saxophone improvisation was played for the students, and they were asked to rate its emotion. The mean trait anger score for listeners who rated the music as angry was significantly higher than the mean trait anger score of those who rated it as friendly. A small but significant correlation was found between perceptions of anger in the music and listener scores on the trait anger test. Results suggest that personality may influence perception of emotion in music and that jazz improvisation may not be reliable for communicating emotions because of wide individual differences in how its emotional content is perceived.

Noticing perceptions of anger in published reviews of a few jazz saxophone performances, 1 particularly John Coltrane’s, Gridley and Hoff (2006) speculated that the activity of mirror neurons in the premotor cortex might be leading the minds of the reviewers to equate the sounds of vigorous saxophone improvising with the anger that would cause reviewers to make similar sounds themselves. Previously, jazz performing experience and technical sophistication had been considered factors that might distinguish those who perceived anger from those who did not (Gridley, 1984, 1986). The results of nine surveys (Gridley, 2009), however, revealed that only about 13% to 22% of listeners perceived anger in the playing of Coltrane, and neither student musicians, veteran jazz critics, nor professional jazz musicians were much less likely than nonmusicians to perceive anger in Coltrane’s playing. Therefore, because (a) everyone has mirror neurons in the premotor cortex of their brains, (b) yet only a few listeners perceived anger in these saxophone performances, and (c) technical expertise had not made a huge difference in degree of anger perceived, this study was conducted to determine whether personality trait anger in the listener played a role in perception of anger in jazz improvisations.

Trait anger can be distinguished from state anger by its enduring, rather than temporary, character. An angry temperament is the general tendency to be aroused to anger and to get angry frequently. Angry personalities tend to perceive a wide range of situations as anger provoking. They also experience more intense elevations of anger whenever annoying conditions are encountered. To further clarify the difference between state anger and trait anger, consider the types of questions typically included on self-report inventories measuring it. For example, the statement ‘‘I am furious’’ may appear on a measure of state anger whereas the statement ‘‘I am generally hot-tempered’’ may appear on a measure of trait anger.

Gridley and Hoff (2007) recruited 205 highly creative high school students and played a John Coltrane saxophone solo for them. Students’ perceptions of emotion in the music were then obtained via a ‘‘Perception of Emotion Survey’’ (e.g., ‘‘Circle the number that best indicates your perception of emotion in the saxophone solo you heard on the recording: friendly 1 2 3 4 5 6 7 angry’’), after which trait anger was assessed via the Multidimensional Anger Inventory (Siegel, 1985). A marginally significant relation (p<.056) between trait anger and emotion perception was found. 2 These findings provided encouragement to conduct another study but with college students, instead of high school students, and to recruit participants who had not been selected for their creativity, obtain their perceptions of emotion in the same music, and then assess trait anger from the students’ scores on a shorter, more widely used instrument. Perceptions of anger in Coltrane’s playing were expected to relate to levels of trait anger in the listeners.



The entire enrollments of three jazz appreciation classes at two Midwestern universities were employed in this study. The sample included 287 undergraduate students, 161 men and 126 women, with a mean age of 20.11 years (SD¼2.58). Questionnaire responses indicated that 41% had played a musical instrument and 35% had some familiarity with jazz. Louis Armstrong, Kenny G, and Miles Davis were the names most frequently mentioned by students who indicated having familiarity with jazz. Only 12 students (4%) mentioned John Coltrane among the musicians they recalled having heard before enrolling for this course.


During the first week of class, in an auditorium classroom, the instructors played the 2 min 42 sec recording of Coltrane’s solo on ‘‘Two Bass Hit’’ from the 1958 Newport Jazz Festival performance of the Miles Davis Sextet (Davis, 1958). This was the longest solo available from the performance that had evoked reporter Don Gold’s (1958) characterization of its player as an ‘‘angry young tenor’’ (p. 16). Neither the instructors nor their students knew the hypotheses of the researcher or what they were hearing. Each student then completed a rating of Coltrane’s emotion by indicating a position on a seven-point ‘‘Perception of Emotion Survey’’ in which ‘‘friendly’’ was position ‘‘1’’ and ‘‘angry’’ was position ‘‘7.’’ (‘‘Circle the number that best indicates your perception of emotion in the saxophone solo you heard on the recording: friendly 1 2 3 4 5 6 7 angry’’) Each student then completed the 10-item version of the trait anger scale from the State-Trait Anger Scale (Spielberger, 1999). 3

Four levels of anger are indicated for each question on the inventory, providing a total score range of 10 to 40.


The mean trait anger score for the total sample was 18.99 (SD¼4.85). In response to the music, the mean perception of emotion score for the total sample was 3.49 (SD¼1.79), which is very close to the scale’s midpoint of 4. Omitting scores for listeners who endorsed the midpoint of the Perception of Emotion Survey, the mean trait anger score (20.57; SD¼6.16) for listeners who rated the music as angry (above the midpoint) was significantly higher (t¼3.31, p<.001, df¼221, d¼.44) than the mean trait anger score (18.23; SD¼4.06) for listeners who rated the music as friendly (below the midpoint). A small but significant correlation (r¼.185, p<.05; df¼285, d¼.41) for the total sample of listeners was also found between listener perceptions of anger in the saxophone solo and listener scores on the trait anger inventory.


Although only small differences were found between trait anger scores for listeners who perceived anger in the music and those who did not, statistical analysis showed that they were not likely to be purely the result of chance. Moreover, the direction of the differences is consistent with the expectation that the listeners most likely to perceive anger in John Coltrane’s playing are themselves angry. Although not entirely conclusive, the results of this study and those that led up to it have certainly made further investigations promising. More than 700 listeners had been polled, and an assortment of recordings from a 6-year period in Coltrane’s output had provided the stimuli. Similar percentages of listeners perceived anger, and the trend among the 493 who also completed measures of trait anger was that angry individuals tended to perceive anger in the music.

The findings suggest a contributing factor for a few journalists’ perceptions of anger in the playing of jazz saxophonists whose music was not perceived as angry by most other listeners. 4 This is important because the anger perception of those few journalists came to be regarded by some readers as a fact about Coltrane’s inspiration, not as just a projection of anger in the journalists. If Coltrane were not angry, these remarks misrepresented him. Such remarks about anger have continued to appear in accounts of Coltrane’s playing (Blumenthal, 2001), and they could bias the perceptions of listeners who have not already formed their own opinions of the saxophonist or his music. Another implication of these findings is that the wide individual differences in how the music’s emotional content is perceived demonstrate jazz improvisation’s lack of universality for communicating emotions.

The data in this study are consistent with the direction that trait anger affected perception in several studies that did not use music as stimulus. For instance, Guyll (1998) found that participants who were high in trait hostility judged social interactions with more negative emotion. Alfred and Smith (1991) engaged high trait anger and low trait anger participants in a discussion task with a confederate who acted in either a hostile or ambiguous manner and later engaged them in reading about a person who behaved in a manner ambiguous with respect to hostility. By comparison with low hostile participants, high hostile participants rated both the confederate and the hypothetical stimulus person more negatively. This suggested that hostile participants encoded the same information in a biased manner. Similarly, Parrott (2005) found that individuals who reported high levels of trait anger displayed facilitative biases in the processing of semantic anger-related stimuli. In social interaction scenarios, Hazebroek, Howells, and Day (2001) found that, by comparison with low trait anger participants, high trait anger participants blamed the antagonist more, responded more angrily to the same events, and more readily identified another person as an antagonist.

A stronger case could be made for the influence of personality on perception of emotion in music if more examples of music were played and more personality traits were assessed for the listeners. A predisposition to anger had been chosen as the personality trait to be investigated for the three Gridley studies because anger was the emotion perceived by a few journalists, affecting public opinion about one of America’s greatest musical figures. The reasons that only one example of music had been used in the two most recent studies were that a recording had been available for the same performance perceived as angry by one journalist and that, because the recording was not widely known, it provided a stimulus that minimized previous exposure and past associations for listeners, thereby overcoming a serious methodological challenge. Three recordings other than the ‘‘Two Bass Hit’’ had been used in previous studies by Gridley on jazz-naive listeners, but no measures of trait anger had been obtained on those listeners.

To further investigate the effects of personality on perception of emotion in music, similarly unfamiliar performances could be played for listeners who complete measures of agreeableness. The least angry ratings of the performances should be found for listeners scoring highest on agreeableness, for instance. Similarly, melancholic personalities would be expected to rate the piece’s emotion as sadder than sanguine personalities would rate the music.


1 In his Down Beat magazine review of the 1958 Newport Jazz Festival performance by jazz tenor saxophonist John Coltrane, Don Gold (1958) termed the musician an ‘‘angry young tenor’’ (p. 16). In a Seattle newspaper account of another performance, Ed Baker (1965) stated that Coltrane’s music ‘‘could be described as angry’’ (p. 48). In summarizing Coltrane’s style for his book Jazz Masters of the Fifties, Joe Goldberg (1965) referred to ‘‘the rage in his playing’’ (p. 209). In a retrospective of Albert Ayler’s impact, Derek Van Pelt (1978) noted perceptions of anger attributed by listeners (p. 43). In an article about Peter Brotzmann’s music, David Sowd (1994) wrote that ‘‘rage is the primary emotion’’ (p. 14).

Note that Goldberg (1965) based his ‘‘the rage in his playing’’ remark on Coltrane’s pre-1963 music, having last listened to the 1962 Coltrane album (Impulse AS-21, commonly known as ‘‘the blue album’’ because of its cover color) before composing the Coltrane passage in his Jazz Masters of the Fifties book (J. Goldberg, personal communication, July 14, 2006). This music occurred long before anyone heard Coltrane’s turbulent, high-density collective improvisations that are documented by such albums as Ascension, Meditations, and Live in Seattle, and are easier to perceive as angry-sounding. Goldberg said that he still detects anger in recordings of the saxophonist’s pre-1963 playing and that he does not agree with other journalists who believe that the anger that they detect in Coltrane’s playing reflects Coltrane’s protesting treatment of African Americans during the civil rights movement of the 1960s. Goldberg did say, however, that he had met Coltrane several times, gotten to know him, and not found him to be an angry person. He said that he perceived the anger only in Coltrane’s music.

2 At least five aspects should be taken into consideration when puzzling motives and associations underlying Gold’s (1958) ‘‘angry young tenor’’ (p. 16) characterization of Coltrane’s motivation. Each of them helps us rule out alternative explanations for his ‘‘angry young tenor’’ remark and leaves us with accepting that Gold genuinely perceived anger in Coltrane’s playing. (a) Gold was writing before the time that civil rights struggles in the United States and Black Nationalism were receiving their widest media attention and motivating a few musicians to link politics with their jazz. (Even when such politics reached their peak, Coltrane and Ayler pointedly disavowed civil rights anger as motivating their music.) (b) In his review of the Miles Davis Sextet performance, Gold also expressed displeasure with Cannonball Adderley’s playing style and linked it to what he disliked about Coltrane’s, yet he did not term Adderley as ‘‘angry’’ or ascribe any other emotion in that semantic space to Adderley. In other words, although he identified similar stylistic aspects in both saxophonists and disliked those aspects, he did not attribute such sounds from Adderley to anger in Adderley. This leads us to believe that Gold distinguished style from emotion. (c) Gold did not question the technical competence of either saxophonist, as other well-publicized journalists later did. So this review does not represent a dismissal of Coltrane’s work on technical grounds, combining an emotion perception with a faulting of Coltrane’s competence, as other journalists later did. (d) That Gold misperceived vigor for anger is unlikely because in Gold’s record reviews of the time he occasionally mentioned vigor but he never attributed any other musician’s work to anger. This leads us to suspect that he could distinguish vigor from anger and was not prone to perceive anger in the musicians he reviewed, even though he was usually quite critical, not withholding negative remarks. (e) Gold (personal communication, October 26, 2006) said that he stood by his ‘‘angry young tenor’’ remark from 1958. Additionally, he reported that he had perceived Coltrane as ‘‘a difficult child’’ in the jazz world. This also suggests that his published words were reliable indications of what Gold perceived in the music and not likely to be a reflection of state anger.

The State-Trait Anger Expression Inventory-Professional Manual (Spielberger, 1999) reports that inter-item consistency, as measured by item remainder correlations, ranged from .45 to .74 for women and from .59 to .78 for men. Alpha coefficients were .88 for women 16–19 years of age, .89 for men 16–19 years of age, .85 for women 20 to 29 years of age and .82 for men 20 to 29 years of age. The manual reports construct validity, as determined by correlations with other measures of trait anger, of .71 with the Buss-Durkee Hostility Inventory and .59 with the MMPI Hostility Scale. Deffenbacher et al. (1996) reported favorably on eight studies that validated the utility of Spielberger’s trait anger scale. Jacobs, Latham, and Brown (1988) reported favorably on the test–retest reliability of Spielberger’s inventory. Cullari (1994) found relatively high reliability for trait anger scores on a one-year retest of psychiatric patients.

4 Future studies might investigate these questions that were stimulated by our findings: Did the non-angry listeners filter out signs of anger in Coltrane’s music because that is how their personalities handle unpleasantness? Might listeners have imputed their own friendliness onto music that was intrinsically angry? Why was anger in Coltrane’s playing not perceived by the listeners who scored above the mean for their own trait anger who did not perceive anger in Coltrane? How are they different from the listeners who perceived anger in Coltrane’s playing and scored above the mean for their own trait anger?


Alfred, K. D., & Smith, T. W. (1991). Social cognition in cynical hostility. Cognitive Therapy and Research, 15, 399–412.

Baker, E. (1965, September 29). Coltrane sounds like nobody else in world of jazz. Seattle Times, 48.

Blumenthal, R. (2001). Liner notes to the compact disc Miles Davis at Newport 1958 [CK 85202]. New York: Columbia/Legacy.

Cullari, S. (1994). Levels of anger in psychiatric in-patients and normal subjects. Psychological Reports, 75(3), 1163–1165.

Davis, M. (1958). Two bass hit. On Miles & Monk at Newport (PC 8978, LP). New York, Columbia.

Deffenbacher, J. L., Getting, E. R., Thwaites, G. A., Lynch, R. S., Baker, D. A., Stark, R. S., et al. (1996). State-trait anger theory and the utility of the Trait Anger Scale. Journal of Counseling Psychology, 43, 131–148.

Gold, D. (1958, August 7). Newport Jazz 1958. Down Beat, 16.

Goldberg, J. (1965). Jazz masters of the fifties. New York: Macmillan.

Gridley, M. C. (1984). Unreliability of communicating emotion by jazz improvisation. Unpublished manuscript, Mercyhurst College, Erie, PA.

Gridley, M. C. (1986, April). Unreliability in conveying emotion by nonverbal auditory stimuli. Paper presented at the meeting of the Southern Sociological Society, New Orleans, LA.

Gridley, M. C. (2009). Emotion perception in jazz improvisation. In F. Columbus (Ed.), Advances in psychological research.

Gridley, M. C., & Hoff, R. (2006). Do mirror neurons explain misattribution of emotions in music? Perceptual and Motor Skills, 102, 600–602.

Gridley, M. C., & Hoff, R. (2007). Who’s angry, John Coltrane or his critics? Psychology Journal, 4(4), 153–160.

Guyll, M. (1998). Effects of trait hostility on social information processing. Dissertation Abstracts International, 59(2B), 0903. (UMI No. 9823857).

Hazebroek, J. F., Howells, K., & Day, A. (2001). Cognitive appraisals associated with high trait anger. Personality and Individual Differences, 30, 31–45.

Jacobs, G. A., Latham, L. E., & Brown, M. S. (1988). Test–retest reliability of the state-trait personality inventory and the anger expression scale. Anxiety Research, 1, 263–265.

Parrott, D. J. (2005). Effect of trait anger on cognitive processing of emotional stimuli. Journal of General Psychology, 132(1), 67–80.

Siegel, J. M. (1985). The measurement of anger as a multidimensional construct. In M. A. Chesney & R. H. Rosenman (Eds.), Anger and hostility in cardiovascular and behavioral disorders (pp. 59–82). Washington, DC: Hemisphere.

Spielberger, C. D., Jacobs, G., Russell, S., & Crane, R. S. (1983). Assessment of anger: The state-trait anger scale. In J. N. Butcher & C. D. Spielberger (Eds.), Advances in personality assessment (Vol. 2, pp. 161–189). Hillsdale, NJ: Lawrence Erlbaum Associates.

Spielberger, C. D. (1999). Inventory-2 Professional State-Trait Anger Expression Manual.

Lutz, FL: Psychological Assessment Resources.

Sowd, D. (1994, April 14). Peter Bro¨tzmann Trio: Making some noise in jazz. The Scene Magazine, 14.

Van Pelt, D. (1978, December). Albert Ayler’s Ghost. Cleveland Magazine, 43–44.

Copyright by Taylor & Francis Group, LLC
ISSN: 1040-0419 print=1532-6934 online
DOI: 10.1080/10400410802633822