The location of sense of humor within comprehensive personality spaces: An exploratory study

The present paper tries to discover the position of sense of humor in personality space. Several definitions of humor are discussed and a review of studies relating personality and humor is given. The authors’ conceptualization of humor is introduced and two studies are presented that tried to locate these humor dimensions within personality space. Our humor tests were given to two samples: these samples also had to answer several multidimensional personality inventories to cover the whole domain of temperament (e.g. EPQ, 16PF, FPI, STPI, SSS-IV and STAI). The Humor scores were correlated with each of the personality variables. The factor compositions of the best predictors for a humor category were used to arrive at a tentative hypothesis of the location of dimensions of humor in personality space. The results show that strong relationships between humor and temperament exist. Some of the hypotheses discussed in the literature were supported, but also some new hypotheses were derived from this study. All in all, the results show that the separation of appreciation of humor into the independent components of funniness and rejection is appropriate since they are located in different parts of the personality space. DOI: https://doi.org/10.1016/0191-8869(85)90081-9 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-77468 Originally published at: Hehl, Franz-Josef; Ruch, Willibald (1985). The location of sense of humor within comprehensive personality spaces: An exploratory study. Personality and Individual Differences, 6(6):703-715. DOI: https://doi.org/10.1016/0191-8869(85)90081-9 This manuscript was published as: Hehl, F.-J. & Ruch, W. (1985). The location of sense of humor within comprehensive personality spaces: An exploratory study. Personality and Individual Differences, 6, 703-715. THE LOCATION OF SENSE OF HUMOR WITHIN COMPREHENSIVE PERSONALITY SPACES: AN EXPLORATORY STUDY* FRANZ-JOSEF HEHL AND WILLIBALD RUCH Department of Experimental Biological Psychology, University of Düsseldorf, Universitätsstrasse 1, 4 Düsseldorf, F.R.G. (Received 27 February 1985) Summary-The present paper tries to discover the position of sense of humor in personality space. Several definitions of humor are discussed and a review of studies relating personality and humor is given. The authors' conceptualization of humor is introduced and two studies are presented that tried to locate these humor dimensions within personality space. Our humor tests were given to two samples: these samples also had to answer several mutidimensional personality inventories to cover the whole domain of temperament (e.g. EPQ, 16PF, FPI, STPI, SSS-IV and STAI). The Humor scores were correlated with each of the personality variables. The factor compositions of the best predictors for a humor category were used to arrive at a tentative hypothesis of the location of dimensions of humor in personality space. The results show that strong relationships between humor and temperament exist. Some of the hypotheses discussed in the literature were supported, but also some new hypotheses were derived from this study. All in all, the results show that the separation of appreciation of humor into the independent components of funniness and rejection is appropriate since they are located in different parts of the personality space. I N T R O D U C T I O N Every theory of personality that claims to be comprehensive should also have assigned a place to such basic human phenomena as humor and laughter. In fact, some personality theorists (e.g. Allport, Cattell, Eysenck, Erikson, Freud, Guilford, Maslow, Murray) have discussed or integrated humor into their theories to a certain extent. But generally, looking into psychological literature (introductory textbooks, monographs, readers etc.) one very seldom finds the corresponding keywords [humo(u)r, laughter, smiling, wit, cartoons, jokes] listed in the index. Allport (1960) complained that the majority of psychological investigators are preoccupied with decidedly unpleasant emotions at the expense of pleasant emotions. Now, a quarter of a century later, how much has changed? Have we overcome this 'tenderness tabu' (Allport, 1960) or is it still present? As far as the psychology of humor is concerned there have been some changes. Some multidisciplinary International Conferences on Humo(u)r and Laughter (Cardiff, Wales, 1976; Los Angeles, Calif., 1979; Washington, D.C., 1982; Tel Aviv, 1984; Cork, Eire, 1985) and the publication of several books (e.g. Goldstein and McGhee, 1972; Chapman and Foot, 1976; McGhee, 1979; Wilson, 1979; Bariaud, 1983; Ziv, 1984) demonstrate the increased activity in this realm. The future will show whether the results obtained will also find their way out of the humor ghetto and gain the attention of a broader psychological public. The present paper explores the relationship between sense of humor and personality. More specifically it tries to define sense(s) of humor as personality concepts and place them within already existing, comprehensive personality systems. personality space, 2 Two problems arise. How can sense of humor be defined, and with what domains of personality might there he relationships? These problems are discussed in the next two sections. S O M E P O S S I B L E D E F I N I T I O N S O F H U M O R Humor can be defined in several ways, depending on which aspects of humor are emphasized. First we have to distinguish between trait and state aspects of humor. The state aspect (cheerful, mirthful, joyful) is usually represented in the 'Elation' or 'Surgency' scales of mood-adjective lists [for a recent survey of mood instruments see Howarth and SchokmanGates (1981)]. A further distinction can be drawn with regard to different aspects like comprehension, appreciation, expression or creation of humor. Humor comprehension refers to the often observed fact that people sometimes fail to understand the meaning of a joke for intellectual as well as emotional reasons. Similar phenomena can be found in Feingold's (1982, 1983) 'Humor Perceptiveness Test' which quantifies two components, memory for jokes and humorreasoning ability. Cattell and Warburton (1967) regarded the ability to solve riddles (MI 26) as a marker for the factor U.I.19 (Promethean Will). Expression of humor refers to differences in quality and quantity of reaction to humorous stimuli. Independent of the amount of cheerfulness experienced, people differ in the way (e.g. smile, laughter, body movements, facial responses) they express this emotion. The reaction to humor can differ depending on the degree to which different systems are involved (motor, accoustic, respiratory, cardiovascular, sympathetic, cortical etc. system). For example, one can laugh loudly or quietly, with or without vocalization, in different rhythmicity, with a certain respiratory pattern, with involvement of only a few or more muscular systems etc. Humor expressiveness could be related to temperamental variables of the introversionextraversion complex. Creation of humor denotes the ability to bring neutral stimuli or situations into a new, funny context. Creation of humor is usually assessed by confronting Ss with incomplete jokes or cartoons without captions and giving them the task of producing funny endings. Sociometric methods as well as questionnaires are also used to assess the individual's ability to create humor (Babad, 1974; Ziv, 1984). Babad (1974) showed that production and reproduction of humor should be distinguished carefully. Zero correlations between the two variables supported his suggestion. The ability to create humor is supposed to be related to creativity (Brodzinsky and Rubien, 1976; Rouff, 1973; Threadwell, 1970). Writing original and funny captions for pictures is used in creativity tests (König, 1983). Krismanic, Kolesaric # , Rohac # ek and Vlasta (1982) have designed humor tests which also distinguish between production of funny captions and selection of the right caption out of five given alternatives. Ss have to choose captions that make the joke funniest and least funny. The intercorrelation (0.65) between these two tests is higher than their separate correlation with a humor appreciation test. Cartoon tests (ordering separate pictures of a cartoon in a meaningful manner, prediction of cartoons, missing cartoons) are included in test batteries measuring social intelligence (e.g. Probst, 1982; O'Sullivan and Guildford, 1966). Appreciation of humor, the more passive aspect of humor, has been investigated most frequently. It refers to a S's susceptibility to experience amusement, mirth and cheerfulness in the presence of humorous stimuli and situations. Differences exist in the ease of experiencing cheerfulness as well as in the variety of humor categories that are enjoyed. Usually jokes, cartoons etc. are presented to the S in order to rank them on rating scales according to their 'funniness' or similar criteria. Recent investigations (Hehl and Ruch, 1984; Rath, 1983; Ruch 1981) showed that it is also useful to use criteria that cover the negative reactions to humor since they tap other sources of variance, i.e. their correlations with the Funniness scale are personality space, 3 close to zero. With regard to stimulus dimensionality it turned out that there is no general factor in humor appreciation. Differences in stimuli are due to structural properties as well as to the content of the material. Many humor tests have been constructed (e.g. Almack, 1928; Cattell and Tollefson, 1966; Krismanic #et al., 1982; Mindess, Turek and Corbin, 1984; O'Connell, 1960; Redlich, Levine and Sohler; 1951; Roback, 1943; Ruch and Hehl, 1984a) in order to measure appreciation of humor, some of them on a more intuitive basis, some employing sophisticated statistical procedures. There is empirical evidence that appreciation of humor and expressivity should be distinguished since they are often dissociated (Cupehik and Leventhal, 1974; Young and Fry, 1966; Osborne and Chapman, 1977). Many investigations have been carried out to explore the relationship between this aspect of humor and personality and intelligence variables, attitudes, sex, age and other variables characterizing the individual. There are still other aspects of humor that remain to be mentioned. One can use humor actively (e.g. frequency of joke-telling, amusing others, inducing humor in others, making others laugh). These aspects are exemplified in personality scales like 'Play' and 'Exhibition' (PRF) or 'Surgency '(16PF) which are subfactors of extraversion. In the EPQ (Eysenck and Eysenck, 1975) agreeing to the item 'Do you like telling jokes and funny stories to your friends?' is scored for extraversion. There are also differences in the frequency with which individuals actively seek out sources that make them laugh (TV shows, cinemas, cabarets, circuses, carnivals etc.). Humor can also be used as a tool to neutralize or inhibit other emotions (e.g. fear, sadness). Thus, humor in the form of cynicism, irony or gallows humor can serve as a coping mechanism (Ziv, 1984; Dixon, 1980; O'Connell, 1966). Differences also exist in memory for jokes or funny events. Other definitions of humor describe it as a certain ability to see the world, to see the funny side of even tragic events, as a socialized way to release suppressed energy or as a way of gratifying socially unacceptable impulses. There is also the statement that humor is a myopic illusion (La Fave, Haddad and Maesen 1976). A T A X O N O M Y O F J O K E S A N D C A R T O O N S It would be useful to have a taxonomy of stimuli that are considered humorous. Independent of the aspect (e.g. appreciation, creation) of humor we are interested in, one must take into consideration the category to which the joke that is created, appreciated, or not comprehended, belongs. There are several ways to establish categories and there have been numerous attempts to establish a taxonomy empirically or theoretically. We used factor analysis and found a three-dimensional system (Ruch and Hehl, 1984a; Ruch, 1981, 1984). In short, we found that jokes and cartoons differ with respect to two properties, structure and content. Our two basic structural dimensions are formed by combinations of the two most fundamental ingredients in jokes, 'incongruity' and 'resolution'. There are jokes that contain punchlines that are incongruent and surprise the recipient but can be resolved completely afterwards (incongruity-resolution jokes). The other type of joke is characterized by impossible incongruities that cannot be resolved or not resolved completely (nonsense jokes). These two factors are heterogeneous with regard to content, whereas the third category (sex jokes) is dominated by a common theme. [For a more elaborate discussion of the origin of the three-dimensional system and its relation to other systems see Hehl and Ruch (1983a), Ruch (1981,1984) and Ruch and Hehl (1983a, 1984a).] In the future we will make extensions of the present model by systematic variations of the two modes content and structure. personality space, 4 P E R S O N A L I T Y C O R R E L A T E S O F H U M O R A P P R E C I A T I O N In this paper we discuss some results which reveal a relationship between humor appreciation and individual differences in the field of personality. A hypothesis set forth in the first half of this century with a lot of supporting results (but also failures) stated that extraversion goes along with liking of orectic (sexual and aggressive) humor. One of the earliest indications for the relevance of extraversion was offered by Kambouropoulou (1930); extraverts preferred superiority humor. Later on Cattell, as well as Eysenck, provided support for the view that extraversion is positively related to appreciation of sex jokes [for a recent survey of Eysenck's contributions to this field see Nias (1981)]. Luborsky and Cattell (1947) showed that some of their humor categories correlated with factors of the Guildford and Zimmermann STDCR-GAMIN factors. Their humor test and other variables dealing with humor were included in the 1967 collection of objective tests of personality (Cattell and Warburton, 1967). A low amount of overt laughter at jokes (MI 157) was an index for a high U.I.21 (Exuberance) score, Criticalness in humor (MI 29) loaded in U.I.17 (InhibitionTimidity). Later, 13 dimensions of humor were listed in the 'IPAT Humor Test of Personality' (Cattell and Tollefson, 1966) in which the first factor ('anxious considerateness vs debonair sexual and general inhibitedness') was thought to be synonymous with the second-order factor of extraversion. This could be inferred because of correlations with the markers of extraversion F (Surgency), A (Cyclothymia) and H (Venturesome). One scale was considered to provide a rough estimate of general intelligence. Also some of the other humor scales offer information not only about a S's sense of humor but also allow inferences about more general personality traits (especially about G, I, M, Q1, Q2). Unfortunately, the actual size of the intercorrelations is not reported in the handbook. However, the reliabilities appear to be small (<0.58). This test was used in several studies (e.g. Breme, 1975; Mones, 1974; Saper, 1984). Terry and Ertel (1974) used the 16PF as a predictor for liking of a priori defined cartoon categories. Sex jokes were preferred by toughminded (16PF I), group dependent (Q2) males. However, the correlations to the markers F, A and H could not be replicated. Hostile jokes had no correlation with any of the 16PF scales. Eysenck's (1942) early study suggested a relationship between sex jokes and self-rating factors of extraversion and masculinity. Wilson and Patterson (1969) found a positive relationship between the Extraversion scale of the EPI and sexual jokes. Some studies supported the hypothesis; others found no relationships (Babad, 1974; Ehrenstein and Ertel,1978; Godkewitsch, 1972; Koppel and Sechrest, 1970; Ruch, 1981; Verinis, 1970). Neuroticism does not seem to be relevant to the area of humor research (Verinis, 1970; Wilson and Patterson, 1969). Recently, Ziv (1984) offered some very complex hypotheses about the relationship between various types of humor and extraversion and neuroticism; these remain to be tested. Given the strong relationship between neuroticism and anxiety one would also expect little relationship between anxiety and humor. Goodkind (1976) reported a positive relationship between anxiety and hostile humor, but there was no correlation with sex jokes. On the other hand Doris and Fierman (1956) showed that Ss with low anxiety levels preferred aggressive jokes. Studies with repression-sensitization (which is also related to neuroticism) produced inconsistent results (Babad, 1974; Burns and Tyler, 1976; Byrne, 1958; Goodkind, 1976; Rath, 1983; Ruch, 1981; Ullmann and Lim, 1962; Unterweger, 1983). Repressors were supposed to show a lower appreciation of sex jokes than sensitizers. This turned out to be true only for females (Burns and Tyler, 1976), but some of the replications failed. Depressivity does not seem to be related to low sense of humor (Scogin and Merbaum, 1983). Other personality variables investigated are agression (e.g. Byrne, 1956; Grziwog and Scodel, 1956; Murray. 1934; Ruch, 1981), locus of control (Lefcourt, Antrobus and Hogg, personality space, 5 1974; Prerost, 1983), dogmatism (Mones, 1974; Rouff, 1973; Ruch and Hehl, 1983b; Smith and Levenson, 1976) and intolerance of ambiguity (Ruch and Hehl, 1983b). Eysenck's third dimension, psychoticism (P), might be related to humor. Since low scorers are characterized by superego strength and impulse control they could represent the kind of people who-in Freud's terminology-'repress' unacceptable impulses more than the 'acting-out' type of high scorers. Freud (1905) suggested that socially unacceptable impulses are repressed but can find relief in humor. The more one has repressed the more one will find funny. Therefore, applying the Freudian model, one would expect negative correlations between P and humor categories (especially aggressive and sexual). Applying the trait theory, Eysenck and Wilson (1975) expected a positive correlation between hostile humor and P. Kline (1977) offered some hypotheses about individual differences in humor that are based on the Freudian theory. The problem with most of the studies in humor is that there is a lack of comparability with regard to the humor categories used. How do we know that the sex jokes used in a certain study are comparable to the jokes used in another. study? Our results show that incongruityresolution sex jokes and nonsense sex jokes yield quite different results. Also, Nias and Wilson (1977) show that sex jokes are not unidimensional. Some studies employ jokes that are only used in that single study and then never again. For better comparability dimensional investigations should be carried out. Sense of humor inventories that are controlled for crossnational stability should be constructed. Only by using a standardized assessment method can we really compare the different results obtained in different cultures. We tried to explore the cross-national stability of our sense of humor inventory (3 WD-K) and the results look promising (Ruch and Hehl, 1984a); but hitherto only the results from two nations are available. Since the scores obtained by the 3 WD-K are sufficiently reliable we can speak of appreciation of humor as a trait. In this paper we tried to find out how our dimensions of sense of humor relate to other personality traits. A couple of questionnaires were employed to cover the whole domain of personality (temperament). Humor scores were then related to these dimensions of temperament to see what traits influence sense of humor. Based on the correlation between personality traits and types of humor a positioning of each type of humor in personality space has been attempted. The second problem concerns the selection of an appropriate personality space. The discussion about the exact number of dimensions of personality is still going on and therefore we cannot rely on a single model, e.g. Eysenck's model or Cattell's model. But much work has been done to relate the different models [for recent surveys see Kline and Barrett (1983) and Royce and Powell (1983)], so that lately it has become possible to transform the results found with one model into the other. The question arises, at what level in the hierarchy of personality traits humor should be expected. It seems necessary to use variables of different levels, i.e. 'primaries' and 'higher order' factors. Furthermore, it is also possible that humor 'variables will not exactly lie close to one axis in the personality space but fall between two or more Dimensions. For example, one type of humor may not only relate to introversion but also to neuroticism, i.e. lie in the diagonal where also other concepts like anxiety, depressivity or low ego strength can be found. Many traits (e.g. impulsivity. sensation-seeking) are known to be composed by more than one higher order factor. Therefore it seems fruitful to include variables in the study that are located off the diagonal in order to have a balanced distribution of traits in the personality space. Thus, traits of different levels in the hierarchy and of mixed factor compositions will be used to facilitate the location of sense of humor. Although we already offered and supported hypotheses on links between humor and personality elsewhere (Hehl and Ruch 1983a, b; Ruch, 1984; Ruch and Hehl, 1983a, b, personality space, 6 1984b) this paper is designed as explorative. But many of the research hypotheses discussed in the literature can be found in the present design, i.e. they are tested implicitly. M E T H O D , S U B J E C T S A N D M A T E R I A L Different versions of the humor test and also some multidimensional personality inventories were given to two samples. The first sample consisted of 95 male students of all faculties except psychology. These voluntarily participating, paid Ss took part in a psychophysiological experiment and answered, among other questionnaires, also the Freiburg Personality Inventory (FPI-A; Fahrenberg, Selg and Hampel, 1978) and the German version of the STAI (Laux, Glanzmann, Schaffner and Spielberger, 1981). Sense of humor was assessed by an earlier version of our humor test, which included all, except one, of the jokes used in the latest version (3 WD-K). For a better comparabilty in the present study only these jokes were used and scored which appeared in the newer test. Ss were tested on 3 different days at 1-week intervals. The second sample consisted of 105 students (49 male and 56 female). About half of them were psychology students of introductory courses, the other half came from other faculties. The Ss answered the German version of the 16PF (Schneewind, Schröder and Cattell, 1983), the EPQ (Eysenck and Eysenck, 1975), a translation (Unterweger, 1980) of the SSS-IV (Zuckerman, 1979), the Impulsiveness, Venturesomeness and Empathy scales of the I# (Eysenck and Eysenck, 1978), the Trait form of the State-Trait Personality Inventory (STPI; Schwarzer and Spielberger, 1982), an assertiveness questionnaire (GAT; Skatsche, Brandau and Ruch, 1982) and Forms A and B of our humor test (3 WD-A, 3 WD-B). Ss participated voluntarily and were tested on 2 different days with a 1-week interval. For the 16PF and the 3 WD-B data for only 92 Ss are available. For ease of presentation and to enhance reliability, scores of both humor tests were combined. In each sample Ss were tested individually. The procedure of data analysis was the same in all samples. The reliabilities of all questionnaires and the distribution of the scores were checked and product-moment correlations between the six Humor scores (three for 'funniness' and three for 'rejection') were calculated. Of course we do not exclude the possibility that nonlinear relations might exist, or humor might be the result of higher order interactions between personality traits, but as a first step we concentrated on the evaluation of single linear relationships. A N A L Y S I S O F T H E ' F U N N I N E S S ' S C O R E S The product-moment correlations between the Funniness scores and the personality inventories are presented in Table 1. Because of the large number of tests of significance, the probability of a Type I error arises. Therefore the levels of significance are merely regarded as cutoff points for descriptive interpretation, In particular, the incongruity-resolution joke factor and the sex joke factor have a large number of predictors; but also the number of significant correlations with fur niness of nonsense jokes is above chance level. Therefore an interpretation of the data is justified Where not specified, the presented correlations reach the 5% level. Table 1. Correlations between personality measures and the Funniness scores of the humor test