Schema’s and Stereotypes: The Effect of Music
on How we Perceive Film Characters
Author: Chris Peeters
Student-ID: 10530819
Master Thesis
Graduate School of Communication
Track Entertainment Communication
Supervisor: Dr. Marlies Klijn
Abstract
The current study investigated if music alters the interpretation of different character aspects from a schema theory background. Either relaxing or tense music was played in ambiguous dialogue scenes that were scored on relaxing/tense thoughts and good/bad intention.
Additionally, a rock song and a pop song were matched with opening scenes to investigate the music’s effect on how the audience perceives the featured characters based on feminine and masculine personality traits. Results showed that while the relaxing/tense music did not seem to affect interpretation, characters featuring a rock background song were perceived as more masculine and characters featuring a pop background song were predicted to be more feminine. Furthermore, the effect on feminine traits was only significant for featured female characters, indicating that male characters are not easily regarded as being more feminine. The results implicate that background music evokes gender stereotypes that are applied to the character present in the scene.
Keywords: film music, gender stereotypes, schema theory, relaxing music, tense music, pop, rock.
Music plays an integral part in communicating something in almost every story in film and television. Most of the research on film music focuses on how it enhances emotions (e.g. Bullerjahn, 2001; Cohen, 2001; Steffens, 2018), investigating how music amplifies what is already happening on the screen. However, the relationship between visuals and sound primarily becomes complex when the film music does not duplicate what’s already on the screen (Kalinak, 1992). That’s how music provides unique information. It can and always has served various meaningful functions, shaping how the audience perceives a story in one way or another. It may be because of this complex nature that this area has received so little academic attention.
Luckily, a recent systematic review (Herget, 2019) of most studies that investigated meaningful music effects in film identified these research gaps, such as in the relation between music and how the audience perceives a character. This area suffers from the one dimensional use of stimulus music and the lack of research on stereotypes and gender. Until now most research only compares positive with negative music as opposed to other types of music and nothing is known about how music and stereotypes relate to each other. The current study is for a large part a response to the review and aims to investigate these exact gaps. Continuing on the theoretical foundations of schema theory (Fiske & Taylor, 1991), the current study investigates with an experiment how music influences the perception of characters in film. Schema’s play a large role in interpreting situations where limited information is available. Cues like music activate
concepts that are then applied to the situation. Connecting schema’s to characters is important, because characters are the primary element in most films and the audience is generally
transported in the story through connection with a character (Tal-Or & Cohen, 2010). In addition, research in this field consistently fails to structure both the music and character
elements (Herget, 2019). By defining precisely what type of music and film scenes were used, it becomes easier to build on the results of earlier research. The current study aims to correct this by defining these constructs, investigating if music directs the interpretation of a character’s mood, intentions and personality traits.
Research in this field serves multiple purposes. It gives insight in how an audience perceives a film. Since films are usually a (exaggerated) reflection of life, findings apply to how people consume information in general too. If applied correctly, research can tell something about which aspects of the music are important in communicating something. This can add to the knowledge of how music affects us the way it does. If music is consistently matched with characters in the same way, this could contribute to schema’s such as stereotypes. And finally, this study field shows how music can be used in innovative ways in an industry that relies so on creativity. As stated by Boltz (2001), music can have various functions, informing without explicitly mentioning it on screen, setting expectations or increasing suspense.
Theoretical Framework
When you watch a film scene, you are processing information. Schema theory posits that in order to quickly evaluate information, certain stimuli and attributes are grouped together based on experience to form an interpretative framework (Fiske & Taylor, 1991). These so-called schema’s will usually be activated when limited information is available. It’s a subconscious process and way of categorizing the world around you. This means that especially when a film scene is ambiguous (positive nor negative), ‘its evaluation is more likely to be influenced by
active schemas, such as those primed by music’ (Shevy, 2007). Depending on several music
factors this can have an effect on how the audience interprets the situation.
As described in full by Herget (2019), music can relate to film in four ways. It can trigger associations through memory when the music is known, called a musical quote. This means that the song is associated with the context in which it was earlier heard. An example from film is the leitmotif, a technique where a distinctive melody is introduced early on and repeated throughout the film to create a connection between the sound and the protagonist, certain scenes or the film as a whole.
Then there is the musical stereotype. Not only is this the standard for employing music in film, it’s also the focus of the current study. The musical stereotype doesn’t rely on having to know the specific music, but more on what the type of music represents. The tradition of film music has learned the audience for instance to connect the accordion to Paris or harbor folk and bongos to a jungle setting (from Wüsthoff, 1978, cited in Herget, 2019). It also makes it easy to connect new wave to the 80’s or hip-hop to urban culture. Each time the music plays this evokes the schema of connected stereotypes that are in turn related to whatever happens on the screen. Research has tested this in various different settings and found for instance that certain
background music changes the film genre perception of a movie trailer (Strobin, Hunt, Spencer, & Hunt, 2015). Especially this could relate to gender stereotypes as well. If there is a tradition of pairing certain musical genres with either males or females, then these musical genres alone might activate schema’s or stereotypes based on gender.
The third and fourth element in music conveying meaning relate to connotation and the
interaction between music and film. Connotation is mainly positive or negative and has a lot to do with the biological foundations of music, as demonstrated by the seminal works of Hevner
(1937). This means that certain properties of music, like tempo, pitch and mode, influence whether the music piece is perceived as having a positive or negative feeling. All the studies on the emotion enhancing effects of music (for an overview, see Cohen, 2001) investigate the exact requirements for conveying emotion through music, but the current study focuses on the basic structure of music and the combination between the musical stereotype and connotation. This means audiences have an idea of what music is ‘supposed to’ make us feel happy, nervous or sad because of previous movie experiences where this music was present during scenes that evoke those emotions. Finally, the when/where music is played during the film relates to whether the tempo matches between audio and visual, to the exact moment a tune is introduced and to the origin of the sound, coming from within the film world or as score. These elements can convey meaning too, but are very dependent on the specific context of the scene.
The importance of these musical qualities is best understood through reception theory (Hall, 2001). As an audience varies on cultural and individual background, the interpretation of a message changes as well. This makes it harder to convey a certain meaning to a large audience. The four musical qualities previously described make it easier to choose music that is interpreted in the same way by the entire audience, because they define which aspect of the music conveys a message.
Previous Studies
As arguably the most important aspect of any story, the protagonist or main character has received considerable attention in the field of film music. For instance, a lot of the studies focus
on how film music influences what emotional state is ascribed to a character in scenes that were either neutral or ambiguous (e.g. Bullerjahn & Güldenring,1994; Tan, Spackman, & Bezdek, 2007). These studies find that the music played alongside (or partially before or after) a character influence the ascribed emotions in a music-congruent way. Character mood or emotion is often chosen because the music shares some of the same attributes as the emotions. It also fits in the natural function of music in emotion regulation (Saarikallio & Erkkilä, 2007), although not that of the audience, but of the film character, in this setting.
However, emotion is not the only studied character aspect. Bullerjahn & Güldenring (1994) investigated character motivations in relation to different musical pieces. They combined a short film with different music and found that a ‘crime’ soundtrack led to significant more ‘bad’ interpretations of the character’s intentions (murder, adultery, ‘settling the score’) than a ‘melodrama’ soundtrack, where interpretations were mainly positive. A recent study (Tan, Spackman & Wakefield, 2017) found something similar. They showed a scene where a man and a woman tried to escape from enemies and paired it with different songs. With a romantic song, the couple was more likely to be perceived as romantically involved than with ‘chase music’. In addition, the male character’s ascribed emotions (fear and excitement in particular), were much higher when the ‘chase music’ played.
And even besides temporal evaluations relating to mood or intention, music also seems to affect the evaluation of a character’s personality. While the research is more scarce, Boltz (2001) found that when combined with positive or negative music, the onscreen character was congruently evaluated in traits as caring, manipulative, evil and playful. Positive music made the character seem nice, where negative music caused the character to be viewed as bad.
These studies show that the score, depending on its qualities, can convey meaning to multiple aspects of a story at once. This is in line with findings that the evaluation of a character is correlated to the evaluation of the whole scene (Shevy, 2007) and it fits with schema theory. As people watch a scene, all the different components of the scene activate concepts related to these components. These concepts are then used to make sense of whatever is happening in the scene. Music can fill in the gaps wherever information is limited through one or more of the four previously mentioned musical elements.
Even though these studies provide a great deal of information, there are some general limitations. Most studies use existing scores or don’t go beyond distinguishing positive from negative music. This makes some of the music too complex to replicate, such as with scores. They vary to an amount where it makes it difficult to say what aspect of the music caused an effect. On the other end of the spectrum, defining music as only positive or negative provides limited information. After all, music is known to evoke thoughts and feelings that have dimensions that extend beyond basic positive/negative affect. In addition, the film scenes used feature no dialogue as to make it more ambiguous, so very little is known about the effect of music in dialogue scenes. The current study aims to complement these shortcomings. First of, instead of positive versus negative music, relaxing versus tense music will be tested to influence the perception of character inner thought and intention. Secondly, the ambiguous scenes used will feature dialogue. Finally, to examine personality evaluation from a new angle, the effect of music on gender stereotypes will be investigated.
RQ: Can music influence the evaluation of a character’s thoughts, intentions and personality traits (e.g. sex roles) through schema-evoking attributes of the music?
The Present Study
Relaxing versus Tense Music
Relaxing, but especially tense music plays a major part in most movies, functioning as a signal of oncoming danger, creating suspense and grabbing the audience its attention (Shevy, 2007). Still it has been by large disregarded in research. The music in this study was chosen according to the four guidelines discussed in the theory. The musical quote however will be less important as the tracks are unfamiliar to the audience. The connotation aspects will follow the basic guidelines from Hevner (1937) in what makes music carry positive or negative affect, adjusted for the musical stereotype to make them more relaxing/tense. The properties of the music pieces can be found in Matrix 1.
Matrix 1
Properties of the musical tracks
Relaxing Music Tense Music Musical Stereotype the instrumentation will feature
ambient sounds as are often used in big scenery scenes, as well as in yoga/meditating background music.
the music will mimic the ominous sounds in films when something bad is about to happen and the tension builds. Recognizable bass strings.
Connotation Positive; major mode, higher tempo/tone, bright timbre.
Negative; minor mode, low tempo/inconsistent beat, low tones.
Interaction with Film Starts playing midway through the scene.
Starts playing midway through the scene.
These tracks will be played over four scenes chosen for character inner thought and intention. These two aspects determine for a large part if the audience connects with a character (Chen, Bell & Taylor, 2016). Two scenes are chosen with fewer dialogue and ambiguous setting to direct the audience attention towards the internal life of the main character. The music should then influence how the audience perceives the inner thought of the character as follows:
H1: Relaxing and tense music leads to music-congruent interpretations of the character’s thoughts.
Two scenes feature a conflict between two characters of some kind with an ambiguous ending. One of the characters in each clip is urged to do something by the other, but it remains unclear whether this will happen. These scenes will be used to asses character intention. The music should influence the perception as follows:
H2: Relaxing and tense music leads to music-congruent interpretations of the character’s intentions.
Gender Stereotypes
To continue the work on music and personality evaluation, the current study will look at gender stereotypes (sex roles). There are two aspects of gender stereotypes in film. On the one hand,
media plays a role in forming gender stereotypes through the representation of males and females. This means that if a gender is consistently portrayed in a certain way, in terms of
personality, clothing style or how they behave, this can reinforce the audience’s ideas of how sex roles are supposed to be divided (Lauzen & Dozier, 2005). On the other hand, existing
stereotypes alter how the audience interprets a certain character, based on gender-related cues and individual differences ( Hall, 2001). This study shows the latter process, but a similar design could be used to measure stereotype formation.
Previous studies found that gender stereotypes are commonly displayed in the cinema, such as in Disney movies (England, Descartes & Collier-Meek, 2011). Most of the work on music and gender focuses on ascribing gender to instruments (for an overview, see Stronsick et al, 2018). The one study that looked at gender, music and visual information used abstract half-balls as protagonists (Brosius & Kepplinger, 1991). It is not surprising that Herget (2019) notes that the research into gender stereotypes is scarce to non-existent.
However, there are some hints of a connection between certain musical styles and gender. While the research is scarce, one study found that when asking children ‘who likes this music?’, rock music was almost exclusively ascribed to males and the same was true for music with flutes and females (Marshall & Shibazaki, 2011). This study will compare a rock song with a pop song. Looking at schema theory and musical stereotypes, these songs will either directly activate gender-related schema’s that will be projected onto the character, or the music might activate schema’s of other movies wherein such songs are used. Even though there is no research on this, rock songs seem to be more prevalent in action movies directed at a male audience, where pop songs are more common in romantic comedies directed at women. Since the characters in these
movies are usually more rough and more caring, respectively, this might affect how they perceive a character when the music plays.
The scenes accompanying these songs are opening credits of two movies, one featuring a male and one a female. Again, there is no research on this, but opening scenes from a schema point of view might be associated with first impressions and cause the audience to be more focused on assessing the character than in other scenes. It also allows for longer music play time and it is convenient to edit in new songs, as there are no sound effects or dialogue in these scenes. By asking the audience about what gender-specific traits they ascribe to these characters, the influence of the music on gender stereotypes should be as followed:
H3a: Pop background music will lead to more feminine trait expectations in a character than rock music.
H3b: Rock background music will lead to more masculine trait expectations in a character than pop music.
Additionally, it’s expected that the change in music has the same effect with a male film character as with a female film character, as there is no research contradicting this.
H3c: The effect of music on character won’t be dependent on the gender of the character.
Finally, gender stereotypes are typically present in both females and males (Ellemers, 2018), so there are no differences expected in the responses:
H3d: There will be no gender differences in the participants in gender stereotyping.
Method
Instrument
The study was conducted using an online experiment survey tool (Qualtrics), chosen because of its ease of access and distribution. Participants were free to fill it in at their own convenience, as long as they were able to hear the audio (as instructed at the beginning of the survey).
Design and Participants
The experiment had a mixed design with two main conditions (rock vs. pop music for the character introduction video, relaxed vs. tense music for the character thought and character intention videos). However, since there were two versions of each video, one with a female protagonist and one with a male one, each participant was randomly assigned to one of four possible conditions, in a way that the three videos watched were from the same film, to mimic real story progression. Participants also answered questions related to the film scenes.
Participants were mainly college students from a large Dutch university. They were recruited through a combination of convenience sampling and by handing out cards with links to the survey on them. There were no exclusion criteria other than understanding English, being over the age of 18 and having access to an internet connection and phone/pc, necessary to access the survey. There were three weeks of data collection to get as many participants as possible.
Stimulus Material Film Scenes
The six different scenes (see Appendix A) were obtained from two different movies that were available on Youtube. They were chosen for ambiguity and similarity between the two scenes for each character aspect. In addition, both were ‘made for tv’ movies. This made it more probable the participants were unfamiliar with them and didn’t know anything about the characters
beforehand. The first film is the Marla Hanson Story (1991) and portrays an aspiring model. The second one is Comfort (2016) and is about a courier meeting a clients daughter. The latter was rendered in lower quality than the original to match the quality of the 90’s movie.
Music
For the character introduction scenes popular music was used, a masculine rock song and a feminine pop song. The rock song is Sharp Dressed Man by ZZ Top. the song is supposed to be representative of the rock genre. The pop song is 9 to 5 by Dolly Parton, in the version of the Dollies due to copyright difficulties. The song is practically identical to the original and hopefully associated with more feminine traits, although there is no prior research to back that up. However, a stimulus test showed the songs were connected to the expected gender
categories.
The instrumental songs used in the character thought and action scenes were chosen according to previously discussed literature on musical structure qualities and relaxing/tense associations (see Appendix A for the songs). In addition, the music shouldn’t distract from the scenes as there was dialogue, so strong melody lines were avoided. The relaxing tune was obtained from a royalty free youtube video called ‘ambient film music’, whereas the tense tune was made in Ableton
Live. Since one participant could hear either of the instrumentals twice, different parts were chosen for the thought and action scene, to avoid conscious recognition.
Procedure
The opening screen of the survey viewed an introduction and informed consent, found in
Appendix B, which had to be agreed upon to continue. Participants would then be directed to the first video, after which the first set of questions followed. The next screen asked the gender of the main character to ensure the second and third scene were of the same movie. after each of those scenes they again would be asked a couple questions. After the third set of questions the survey ended.
Measures
All the scales used can be found in Appendix C. In order to measure gender stereotyping the short form of the Bem Sex Role Inventory (BSRI-12; Vafaei et al, 2014) was adapted. The scale includes six feminine (e.g. ‘affectionate’) and six masculine (e.g. ‘acts like a leader’) personality traits that are to be scored on a 7-point Likert scale. Each participant was asked to what extent the character portrayed in the scene possessed these traits. Both feminine and masculine traits showed high internal consistency, respectively Cronbach’s alpha = .89 and .74.
The operationalization of the inner thought construct was made based on the music. Since the music was either relaxing or tense, the questions consisted of these aspects along with closely related constructs (e.g. disturbed). For the intention scale, the questions were more tailored to what would happen next (what was the intention of the character). The music here is supposed to act as an indicator of everything going well (relaxed) or trouble brewing (tense). Hence the scale was made to inquire about the audience predictions regarding a good or bad ending (e.g. the
character will do the right thing). The questions were about the male characters in the scenes, because they were the ones that had to act. Both scales were validated following data collection.
Results
Participants
The final sample consisted of 113 participants ( 55% female) with a mean age of 29.3 (SD = 12.0). Qualtrics was used to gather the data and partial responses were set to be deleted after one week of inactivity. Additionally, four responses were deleted because they completed the
experiment in under five minutes, which is too short to watch all the videos. To check if gender was comparable across conditions, a chi-square test was done over all conditions and gender. The percentage of females in the pop condition (63%) was not significantly different than that in the rock condition (46%), χ2(1) = 3.19 , p =.074. For thought, the percentage of females in the relax music condition (59%) was not significantly different than that in the tense music condition (51%), χ
2
(1) = 0.74 , p = .39. For intention, the percentage of females in the relax music condition (60%) was not significantly different than that in the tense music condition (50%), χ2(1) = 1.06 , p = .303.
To check if age was comparable across conditions, an independent samples t-test was conducted with conditions as independent variable and age as dependent variable. In order of the previous analyses, age was not significantly different across pop/rock, t(105.3) = -1.68, p = .096. It was
not significant for the thought conditions, t(111) = -0.76, p = .939 or the intention conditions,
t(107.7) = -0.83, p = .406. This indicates the randomization of gender and age worked.
Stimulus Test
To test if the pop song was more perceived as feminine and the rock song more as masculine, 10 separate participants were recruited through convenience sampling and asked after hearing the songs: Do you think this song is more masculine of feminine? Participants then checked a box with either masculine, neither or feminine. All ten participant rated the pop song as more feminine and the rock song as more masculine. This served as a manipulation check for popular song choice.
Construction of Variables
Character Thought
This scale measured the prediction of relaxing to tense thoughts in the film character. PCA was used to check latent variables and check the quality of the scale. PCA was chosen because the communalities were all > .5.
KMO was good with .74. Correlation between variables was high enough with all correlations >.3. Bartletts test of sphericity was significant, χ2 (6) = 93.9, p < .001.
Based on the eigenvalues and the scree plot one component was selected, eigenvalue = 2.39, explained variance = 60%. This means the four questions all seem to measure inner thought on the relaxed-tensed dimension. In order to check internal consistency the items relaxed and happy were recoded to fit the other two variables. Internal consistency was acceptable, Cronbach’s
alpha = .77. Deleting one of the items would not increase the consistency so all items were kept. The mean of the items was taken to measure inner thought. Thought (M = 4.30, SD =1.11) has a minimum of 1.75 (indicating relaxing thoughts) and a maximum of 7 (indicating being tense).
Character Intention
This scale measured the prediction of character intention. PCA was again used as communalities were all > .5. KMO was adequate with .65. Correlations were > .3. Sphericity test was
significant, χ 2
(3) = 58.78, p < .001.
Based on the eigenvalues and scree plot one component was selected with eigenvalue = 1.94, variance explained = 64.76%. All the three questions seemed to measure the latent variable of character intention on a good-bad dimension. Reliability analysis was conducted to check internal consistency and was found acceptable, Cronbach’s alpha = .72. By deleting the second item the alpha would go up by .015, but the item was kept as character personality contributes to perceptions of character intention. The mean of the items was taken to measure intention.
Intention (M = 3.60, SD =1.30) has a minimum of 1 (indicating bad intentions) and a maximum of 6 (indicating good intentions).
Main Analyses
The hypotheses are as follows:
H1: Relaxing and tense music leads to music-congruent interpretations of the character’s thoughts.
H2: Relaxing and tense music leads to music-congruent interpretations of the character’s intentions.
H3a: Pop background music will lead to more feminine trait expectations in a character than rock music.
H3b: Rock background music will lead to more masculine trait expectations in a character than pop music
H3c: The effect of music on character won’t be dependent on the gender of the character. H3d: There will be no gender differences in the participants in gender stereotyping.
Assumptions are only reported where violated. Effect size interpretations are taken from Field (2013, p. 332, p.389).
To test H1 an independent t-test was done with music (relaxing versus tense) as independent variable and inner thought perceptions as dependent variable. All the means and standard deviations for the hypotheses H1 and H2 can be found in Table 1. There was no significant difference between the conditions, t (111) = -1.40, p = .164, d = 0.26 (small effect). This means H1 was not supported by the data.
To test H2 an independent t-test was done with music (relaxing versus tense) as independent variable and intentions as dependent variable. There was no significant difference between the conditions, t (111) = 0.78, p = .437, d = -0.14 (small effect). This means H2 was not supported by the data.
Means and (Standard Deviations) of Thought and Intention across Conditions
Relaxing Tense
Thought 4.15 (1.07) 4.43 (1.07) Intention 3.69 (1.36) 3.50 (1.22)
Note. A score of 4 represents neutral interpretations, with a higher score respectively indicating
more tense thoughts or a more positive intention.
To test H3a, an independent t-test was done with music (pop versus rock) as independent variable and feminine traits as dependent variable. The means and standard deviations can be found in Table 2. There was a significant effect of music on feminine traits, t (111) = 2.42, p = .017, d = -0.46 (medium effect), meaning that the pop song caused higher feminine trait
expectations than the rock song. This means that H3a is supported by the data.
Testing H3b another independent t-test was done, now with masculine traits as dependent
variable. There was a significant effect of music on masculine trait expectation, t (111) = -3.49, p = .001, d = 0.65 (medium effect), so that people who heard the rock song attributed higher masculine traits to the characters than those who heard the pop song.
Table 2
Means and (Standard Deviations) of Feminine and Masculine Traits across Conditions
Pop Rock
Masculine 3.95 (0.94) 4.61 (1.09)
Note. A score of 4 represents neutral trait expectations, with a higher or lower score indicating
respectively higher or lower trait expectation.
Hypothesis 3c tests whether there were any differences between the female and male character in the effect the music had. In other words, it tests any interaction effect between the gender of the character and the music on trait expectations. Univariate analysis of variance (ANOVA) was used to test this with feminine traits as dependent variable. There was an interaction effect of character gender and music, F (1, 109) = 7.23, p = .184, η2 = .06 (medium effect). Figure 1
Figure 1: Representation of feminine trait scores across music condition and character gender
with 95% confidence intervals
The figure shows that the music only had an effect on feminine trait expectations when the character was female. When the character was male, there was no change in trait expectations caused by the music. To check if a similar effect was found with the masculine traits, another ANOVA was done with music and character gender as independent variables and masculine traits as dependent variable. However, there was no interaction effect here, F (1, 109) = 0.46, p = .498, meaning the masculine trait scores didn’t change as a result of the character gender. This means H3c was violated. While music changes predictions of both feminine and masculine traits in female characters, it only changes masculine traits in male characters.
In order to test H3d, two covariate analyses (ANCOVA) were conducted with music as independent variable, traits as dependent (separately) and gender as covariate. For masculine traits, the effect of gender was not significant, F (1, 109) = 1.79, p = .184. For feminine traits it wasn’t significant either, F (1, 109) = 0.42, p = .520. This means H3d was supported, as there were no differences between male and females on the responses on traits.
Discussion
The current study aimed to contribute knowledge about the relation between music and film character, focusing on how music plays a role in how the audience perceives a character. Examining the three area’s most important in connecting with a character, namely through their
thoughts, action intentions and personality (Chen, Bell & Taylor, 2016), the study found mixed results. It seems that the general mechanism of delivering music-congruent meaning in
ambiguous situations not always has the desired effect. While background music indeed plays a part in how an audience forms their impression of an on-screen character’s personality, this study couldn’t confirm music shapes the audience perception of character thoughts or intentions. However, overall this study contributed in two major ways to the existing literature.
Firstly, no prior research examined the role of background music in opening scenes, nor did they investigate the effect of music on gender stereotyping film characters. Presumably, having an opening scene song that is clearly more connected to either females or males, as with the pop and rock song respectively, causes the activation of a gender-related schema, which is then applied to the character. Even more interesting, while the female character was perceived as more feminine or masculine depending on the song, the male character was only perceived as more or less masculine, while no effect on feminine traits was found. This could indicate that males in film are perceived more static than females. To make an analogy, even though it’s accepted for women to wear masculine clothing, it is less accepted for men to wear feminine clothing (Lynn, 1959). Implications of these findings do fall in line with the use of the musical ‘leitmotif’ in music, as simply pairing certain musical pieces with a character influences inferences about their personality, based on structural qualities of the musical piece. Even though this study focused on gender-related traits, based on the success of leitmotifs, it’s likely that music may impact
expectations of other personality traits as well.
The second contribution lies in how the investigated constructs are chosen and defined. As noted by Herget (2019), this field lacks comparable study results, because the materials often vary too much. By clearly defining the types of music used as well as the film scenes, all from a
comprehensive theoretical framework, it will become easier to continue on previous work in this area of research. At the same time it’s important to acknowledge the variations in scenes.
One limitation of this study is that no manipulation check was done to ensure the ambiguousness of the scenes. It could be that some of the scenes, especially the thought and intention scenes, gave more away than intended. Even though the predictions tended towards the music congruent interpretation, the results were not significant. This may be why H1 and H2 were not supported. Instead the scenes were chosen to relate to thought and action intention respectively, with two scenes featuring unsure inter-character relationships (triggering thought predictions) and two scenes featuring inter-character conflict (triggering action intention predictions).
A second explanation of H1 and H2 not being supported is the order of the survey. Since the opening scene came first, followed by the gender trait question and then the other scenes, participants might not have been watching attentively anymore.
A third explanation could be that the music played doesn’t activate schema’s relating to character thought or intention, but only towards story progression in general. To give an example: the audience might hear tense music and implicitly know things are about to take a turn for the worse, but this doesn’t mean the character on screen realizes it too. This could relate to film traditions where the audience knows more than the character.
Aside from manipulation checks, this study has a few further limitations. To mimic real films, each scene for the separate character aspects was taken from one film. This brings about two problems. There were two films, one with a female character and one with a male, and each type of scene might not be perfectly comparable, because there were limited scenes to choose from. As a result, for example the intention scenes, one scene was a relatively passive aggressive conflict, while the other was an all-out open fight. On one side this might mean that the impact of
the music is different for the two scenes, but it is also a more realistic representation of the different situations in which action intentions are predicted by the audience.
The second problem relates to scene order. The scene order was the same for each participant, so there might have been some carry-over effects from previous scenes. This is not a big deal, because real movies also consist of this. However, it’s still possible that having a tense tune in scene two and then again in scene three has a unique effect. This was not checked.
This study has several practical implications. It gives insights in how stereotypes play into media even through the music. It shows that simple background music affects how the audience
perceives personality aspects of a character. These results can be used to investigate stereotypes in an unobtrusive way, but they can also be used by filmmakers to shape what the audience perceives even further.
Future research can continue on the constructs defined in this study. By comparing character elements with other story elements the influence of relaxing and tense music can become clearer. Since no prior research investigated dialogue scenes, this study can serve as an example of how to approach these type of scenes. Finally, knowing the influence music has on personality
predictions, this can now be tested with other types of music or on other personality traits. Music has for such a long time been prevalent in film and learning more about how it affects the
audience will not only inspire filmmakers, but also teaches something about how the audience thinks about and perceives a story.
Boltz, M. G. (2001). Musical soundtracks as a schematic influence on the cognitive processing of filmed events. Music Perception: An Interdisciplinary Journal, 18(4), 427-454.
Brosius, H. B. & Kepplinger, H. M. (1991). Der Einfluß von Musik auf die Wahrnehmung und Interpretation einer symbolisierten Filmhandlung [The influence of music on the
perception and interpretation of a symbolized film plot]. Rundfunk und
Fernsehen, 39, 487-505.
Bullerjahn, C. (2001). Grundlagen der Wirkung von Filmmusik [Fundamentals of the effect of film music]. Augsburg: Wissner.
Bullerjahn, C., & Güldenring, M. (1994). An empirical investigation of effects of film music using qualitative content analysis. Psychomusicology: A Journal of Research in Music Cognition, 13(1-2), 99.
Chen, M., Bell, R. A., & Taylor, L. D. (2016). Narrator point of view and persuasion in health narratives: The role of protagonist–reader similarity, identification, and self-referencing. Journal of Health Communication, 21(8), 908-918.
Cohen, A. J. (2001). Music as a source of emotion in film. Music and Emotion: Theory and Research, 249-272.
Ellemers, N. (2018). Gender stereotypes. Annual Review of Psychology, 69, 275-298. England, D. E., Descartes, L., & Collier-Meek, M. A. (2011). Gender role portrayal and the
Disney princesses. Sex Roles, 64(7-8), 555-567.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. sage, p. 332, p. 389. Fiske, S. T., & Taylor, S. E. (1991). Social cognition. Mcgraw-Hill Book Company. Hall, S. (2001). Encoding/decoding. Media and Cultural Studies: Keyworks, 2.
Herget, A. K. (2019). On music’s potential to convey meaning in film: A systematic review of empirical evidence. Psychology of Music, 0305735619835019.
Hevner, K. (1937). The affective value of pitch and tempo in music. The American Journal of Psychology, 49(4), 621-630.
Kalinak, K. (1992). Settling the score: Music and the classical Hollywood film. Univ. of Wisconsin Press.
Lauzen, M. M., & Dozier, D. M. (2005). Maintaining the double standard: Portrayals of age and gender in popular films. Sex Roles, 52(7-8), 437-446.
Lynn, D. B. (1959). A note on sex differences in the development of masculine and feminine identification. Psychological Review, 66(2), 126.
Marshall, N., & Shibazaki, K. (2011). Two studies of musical style sensitivity with children in early years. Music Education Research, 13(2), 227-240.
Saarikallio, S., & Erkkilä, J. (2007). The role of music in adolescents' mood regulation. Psychology of Music, 35(1), 88-109.
Shevy, M. (2007). The mood of rock music affects evaluation of video elements differing in valence and dominance. Psychomusicology: A Journal of Research in Music
Cognition, 19(2), 57–78.
Steffens, J. (2018). The influence of film music on moral judgments of movie scenes and felt emotions. Psychology of Music, 0305735618779443.
Stronsick, L. M., Tuft, S. E., Incera, S., & McLennan, C. T. (2018). Masculine harps and feminine horns: Timbre and pitch level influence gender ratings of musical instruments. Psychology of Music, 46(6), 896-912.
Tal-Or, N., & Cohen, J. (2010). Understanding audience involvement: Conceptualizing and manipulating identification and transportation. Poetics, 38(4), 402-418. Tan, S. L., Spackman, M. P., & Bezdek, M. A. (2007). Viewers' interpretations of film
characters' emotions: Effects of presenting film music before or after a character is shown. Music Perception: An Interdisciplinary Journal, 25(2), 135-152.
Tan, S. L., Spackman, M. P., & Wakefield, E. M. (2017). The effects of diegetic and nondiegetic music on viewers’ interpretations of a film scene. Music Perception: An Interdisciplinary Journal, 34(5), 605-623.
Vafaei, A., Alvarado, B., Tomás, C., Muro, C., Martinez, B., & Zunzunegui, M. V. (2014). The validity of the 12-item Bem Sex Role Inventory in older Spanish population: An
examination of the androgyny model. Archives of Gerontology and Geriatrics, 59(2), 257-263.
Appendix A
Links to the Stimulus Materials
These are the scenes used for the experiment and feature the edited music. Scenes from the Marla Hanson Story (1991).
Opening Scene:
(pop) https://www.youtube.com/watch?v=BOOEe3rWTt4 (rock) https://www.youtube.com/watch?v=jUawDItg2uE&t=8s Thought Scene:
(tense) https://www.youtube.com/watch?v=O0MU7GwLFiY&t=1s&frags=pl%2Cwn Intention Scene:
(relaxed) https://www.youtube.com/watch?v=52zCJzSTpG8&t=6s (tense) https://www.youtube.com/watch?v=HZqGbWy_nHA
Scenes from Comfort (2016). Opening Scene: (pop) https://www.youtube.com/watch?v=gagYRV6sGJo&t=1s&frags=pl%2Cwn (rock) https://www.youtube.com/watch?v=iyDQuQ2jcIU&t=1s Thought Scene: (relaxed) https://www.youtube.com/watch?v=uZCRjrSS80M&t=7s (tense) https://www.youtube.com/watch?v=LWanXrH2Sqk&t=1s Intention Scene: (relaxed) https://www.youtube.com/watch?v=LouNBx5MZU8&t=24s (tense) https://www.youtube.com/watch?v=KrAeLBGHLbk&t=5s Appendix B
Introduction and Informed Consent Dear Participant,
For this study you will watch multiple movie scenes and be asked about your experience. You can watch them the same way you would watch any other movie. However, since there are no subtitles, please make sure you can hear the audio. The study shouldn’t take more than 10-15 minutes.
As this research is being carried out under the responsibility of the ASCoR, University of
Amsterdam, we can guarantee that your anonymity is guaranteed. You can quit the survey at any time or email me within 7 days to withdraw permission. There are no risks involved in doing this survey.
For more information about the research and the invitation to participate, you are welcome to contact the project leader Chris Peeters (chrispeeters40@hotmail.com) at any time.
Should you have any serious complaints you can contact the designated member of the Ethics Committee representing ASCoR, at the following address: ASCoR Secretariat, Ethics
Committee, University of Amsterdam, Postbus 15793, 1001 NG Amsterdam; 020-525 3680;
ascor-secr-fmg@uva.nl.
Any complaints or comments will be treated in the strictest confidence.
Once again, thank you so much for taking this time, together we'll hopefully make film a bit more interesting.
Kind regards, Chris Peeters
Informed Consent
I hereby declare that I have been informed in a clear manner about the nature and method of the research, as described in the email invitation for this study.
I agree, fully and voluntarily, to participate in this research study. With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the experiment at any time.
If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely safeguarded. My personal data will not be passed on to third parties without my express permission.
√ I understand the text presented above, and I agree to participate in the research study. "
Appendix C
Measures Used
Short form of the Bem Sex Role Inventory (BSRI-12; Vafaei et al, 2014)
Please rate the main character in the scene on a scale from 1 (very little) to 7 (very much):
-
Gentle-
Sympathetic-
Uncaring-
Affectionate-
Sensitive to needs of others-
Has leadership ability-
Dominant-
Strong personality-
Defends own beliefs-
Acts as a leader-
Makes decisions easilyCharacter Thought Scale
Please rate on a scale from 1 to 7 (from very little to very much) to what extent the main character from the previous scene seems:
-
Relaxed-
Tense-
Disturbed-
HappyCharacter Intention Scale
Please rate on a scale from 1 (very little) to 7 (very much) how likely:
