I. Presentation of the study
Eschrich, Münte, & Altenmüller (2008) analysed that specific types of songs can produce strong emotions. This strong link between songs and the different emotions it provokes can stay embedded in one’s mind for a relatively long period of time. The authors of the study had two purposes: the first one was to demonstrate that emotional music is remembered better than less emotional music. The second was to prove that musical structure has an influence on memory performance. To do so, the present study proposes to examine “the influence of emotional properties of musical pieces on subsequent recognition for these pieces” (p.2).
II. Procedure
Twenty-four non-musicians, twelve women and twelve men, all undergraduate or graduate students from the University of Hanover participated in this study. The average age of the participants was about 25.5 (range 19 to 44 years old) and none of them had hearing disabilities.
The authors runned a series of three experimental sessions. Each one was separated by a day and every session proceeded as follows: every student sat in a chair and listened to stimuli (excerpts of symphonic film music) via headphones which were connected to a computer. At the end of each excerpt, questions with several answer options appeared on the computer screen and participants answered. They were not timed.
Besides, before the first two sessions, participants had to fill out a short mood questionnaire which indicated their emotional state before the task.
During the first session, the participants were divided into two groups: firstly the “emotion group” whose task was to rate the emotionality of each piece from the forty excerpts regarding the arousal (from very relaxing to very exciting), the valence (from neutral to strongly positive) and the emotional intensity (from no emotion to highly emotional) and secondly the “time-estimation group” whose task was to estimate the total length of each excerpt of the same forty excerpts. In addition, the latter group also had to judge the loudness of each excerpts. The length of the forty musical pieces was about twenty to thirty seconds.
In the second experimental session, all participants had to listen to the same forty excerpts which were presented in the first one. However, this time, they were not required to rate the pieces, but their only task was to listen to the music. The purpose of this session was to facilitate memory consolidation.
For the third session, all participants came back to the laboratory and listened to forty old music pieces (the ones from session one and two) and forty new pieces. For this session, the length of the musical excerpts was about ten, twenty and thirty seconds. The pieces were mixed and presented randomly. After each piece each participant had to rate the familiarity of the pieces; “Is it a new piece of music?” or “Is it the same as what I have heard in the previous sessions?”.
III. Results
The results show that strong positive valence ratings seem to be associated with better recognition memory which means that they are remembered better. However, the authors did not prove that high arousal music is remembered better than the less one. They did not demonstrate that the emotion group showed a better recognition performance about the musical pieces than the time-estimation group either.
Thus, Eschrich, Münte, & Altenmüller (2008) concluded that very positive valence is associated with better memory performance of music in a recognition task while arousal seems to be a less important variable in recognition memory processing emotional events. As no significant differences have been found between the emotion group and the time-estimation group, it has been concluded that emotions induced by music are likely to be processed in an automatically manner and yet influences recognition as both groups performed well in the recognition task.
IV. Significance of the study
Why are some songs able to elicit strong emotions such as shivers and chills to the bone and why others don’t? How is it possible that these songs that trigger in us such emotions do not only provoke a somatosensory response but also take place into our memory and stay there for quite a long time?
It seems like this phenomenon can be linked to a form of non-associative learning called sensitization . In fact, we have seen that some stimuli whether noxious or arousing ones could be the cause for a heightened response which in this case is represented by any kind of emotions: joy, gladness, sorrow, sadness etc. However, it does not completely match with the concept of sensitization because one song can elicit a more or less intense emotional response, but it does not mean that every song that could follow will induce an emotional response as well. Indeed, only those songs which share some features with the song that is responsible for the previous galvanic response will provoke “peak experiences” (Eschrich, Münte, & Altenmüller 2008, p.1). As for us, it is only in that sense that we can link the research led by Eschrich, Munte and Altenmuller (2008) with the concept of sensitization. Moreover, it has been observed that one song “can be remembered even years later, possibly due to the strong emotions it first elicited.” (p.1). Thus, it means that a single exposure can be enough to store music into the long-term memory just as the concept of sensitization suggests.
Furthermore, the fact that there were no differences between the “emotion group” and the “time-estimation group” of the study and that both performed well in the recognition task show that whether it is implicitly or explicitly, musical information is still processed. We can link this to the phenomenon of priming in that sense that the participants in the time-estimation group might have encoded the musical pieces in an implicit way which explains their feeling of knowing them although astonishingly it was unconscious. The first and second exposure biased their behaviour while they had to judge the pieces.
V. Critics
The research led to show “the role of emotion in episodic long-term memory for music” (p.1) evinces positive aspects as well as negative ones. We will begin with the positive ones. First of all, the researchers filled the criteria of a good study in psychology, in terms of equality: twelve women, twelve men. Besides, they removed everything that could skew the experiment. For example, they made sure to use non-musicians or at least non-experimented ones for this study. On the one hand, we found that the researchers did great efforts in trying to neutralize the possible interferences related to the methodology of the experimentation which could have distorted the results. Their first initiative was to ensure that none of the participants knew the selected musical pieces. Even for the two participants who were familiar with three of the eighty pieces, the researchers found that there was no difference in the ratings of valence. Thus, it did not disqualify the value of the experiment. Furthermore, they gave a mood questionnaire to each of the participants in order to neutralize the potential effect of emotional state which could have influenced their answers and thus compromised the experiment. Another fact to consider is the well-chosen design of the study. Indeed, all the participants were able to listen to the same excerpts of music at each session which meant that all participants could see all the modalities of the experiment. We noticed that this choice was only made possible because of the shortness of the musical pieces. However, if the excerpts lasted longer, they should have used different methodological methods known as “full counterbalancing” in order to counterbalance the effect of habituation for example or a “random partial counterbalancing” if the variable had more modalities. Finally, another interesting point was to reduce the length of all the eighty excerpts in the third session in order to eliminate the effect of fatigue.
On the other hand, some flaws can be found. The main and most common bias in psychology researches is that they originate from WEIRD (Western Educated Industrialized Rich Democratic) societies. In fact, 96% of participants come from Western and industrialized countries. More problematically, 70% of participants are usually psychology undergraduates . Unfortunately, WEIRD countries are not representative of the world’s population because they only make up for 16% of it (Heine, 2011). Thus, the results of such psychology study can only be true for, at least 16% of the human world. Besides, the hypothesis stating that stimuli which induce high arousal are remembered better was not confirmed, is another drawback and is likely to reflect a weakness in the choice or use of the material. At last, we can notice that the sample of participants is undersized as to reflect a representative estimate of the general population.
References:
• Eschrich, S., Münte, T. F., & Altenmüller, E. O. (2008). Unforgettable film music: the role of emotion in episodic long-term memory for music. BMC neuroscience, 9(1), 48.
• Heine, S. J., (2011) Cultural Psychology (2nd éd.). New York: W. W. Norton & Company
