Misophonia and the Trapezius Muscle:
The Effect of Misophonic Sounds on Trapezius Muscle Activity in Individuals with Misophonia
Bachelor Thesis
Nora van Wassenaer
University of Amsterdam
Student number: 10581685
Supervisor: Dr. Romke Rouw
Date: June 2nd 2017 Number of words: 5842
Abstract
Misophonia is a chronic condition in which individuals experience a
disproportionately aversive reaction to certain, usually human-made, sounds. In this EMG study, trapezius muscle activity was measured in misophonic versus control conditions, in misophonic and non-misophonic individuals. Both subject groups displayed more EMG activity during misophonic sounds than control sounds.
Misophonic subjects displayed more activity than non-misophonics when listening to misophonic sounds than when anticipating these sounds and displayed more activity in anticipation of misophonic than control sounds. There was no relation between the EMG activity and the rated aversion to sounds for misophonics, but misophonics gave a higher rating to misophonic sounds than control sounds. These results
demonstrate the importance of the physical response in misophonia and further help understand this relatively unknown disorder.
Misophonia is the name for a chronic condition in which certain auditory stimuli, so-called “triggers” cause a strong negative emotional reaction within an individual (Schröder, Vulink & Denys, 2013). The condition was first named “misophonia” in a review by
Jastreboff & Jastreboff (2001). Since then it has received a growing amount of attention, but the literature on this topic is still sparse. Previously it has been shown that individuals with misophonia report a strong physical reaction in response to trigger sounds (Edelstein, Brang, Rouw & Ramachandran, 2013). These physical effects have not been specifically investigated yet. To further study the physical response in misophonia, this study looked at the influence of misophonic sounds on the EMG activity of the trapezius muscle in misophonic individuals.
The auditory stimuli that cause a negative emotional reaction in individuals with misophonia are most often human-made noises, with eating, chewing and crunching sounds reported as the worst triggers (Edelstein et al., 2013). Between individuals it can differ which specific sounds set off a reaction (Edelstein et al., 2013). The symptoms of misophonia appear similar across studies and can be more specifically summed up as the following. 1) The individual experiences a strong feeling of anxiety, disgust, irritation or anger in anticipation of or in response to the trigger (Schrӧder et al., 2013; Edelstein et al., 2013). In addition, as reported by Edelstein et al. (2013), a physical reaction is experienced by misophonic
individuals as tense muscles and as pressure in chest, arms, head or whole body region. 2) The individual is afraid to lose control in anticipation of the trigger with potential aggressive outbursts as a result (Schrӧder et al., 2013). 3) The individual knows his or her behavior is
disproportionate and unnecessary (Schrӧder et al., 2013). 4) To deal with their symptoms, misophonic individuals tend to avoid social situations in which their trigger could possibly occur (Schrӧder et al., 2013). 5) The symptoms cause significant distress to the individual experiencing them, with social isolation as an underlying problem (Schrӧder et al., 2013). For example, the avoidance of social situations can make it difficult for misophonic individuals to meet new friends or attend social gatherings. 6) The symptoms cannot better be explained by any other already existing disorder (Schrӧder et al., 2013). There is a high co-morbidity with OCPD, but this disorder does not account for all symptoms (Schroder et al., 2013) and it is therefore apparent that misophonia is a disorder on itself.
Of particular interest to this study is the physical reaction that misophonic individuals report in reaction to their triggers (Edelstein et al., 2013; Schrӧder et al., 2013). A study by Dozier (2015) found that in a counterconditioning treatment of one misophonic individual, the reaction to low-intensity triggers was purely physical. It was suggested that the physical reaction in misophonia consists of a conditioned physical reflex response to a stimulus, which in turn evokes the emotional response.Thus the physical reaction seems to play an important role in the expression of misophonia. A couple of studies have indeed found that trigger sounds increase heart rate and skin conductance response in individuals with misophonia, similar to a fight-or-flight response (Kumar et al., 2017; Schrӧder et al., 2013; Edelstein et al., 2013), yet no study to date has examined EMG activity in response to triggers in misophonic individuals. Electromyography (EMG), in this case surface EMG, uses electrodes placed on the skin to measure the electrical current generated by skeletal muscles during contraction (Raez, Hussain & Mohd-Yasin, 2006). Tense muscles is one of the core physical reactions reported by individuals with misophonia (Edelstein et al., 2013). Previous research by Rouw & Erfanian (2017) even found that tense muscles are the most often reported physical
complaint in misophonic individuals. As such, investigating this specific symptom with EMG could be extremely helpful in understanding some of the underlying problems in misophonia. EMG studies have shown that specifically the trapezius muscle, a muscle in the higher back and shoulder region, is more activated in response to physical and mental stress tasks than during rest (Krantz, Forsman & Lundberg, 2004; Wijsman, Grundlehner, Penders & Hermens, 2011). Also, Lundberg et al. (2002) found that even in the absence of any physical demands, the trapezius muscle is activated in response to mentally stressful tasks. Furthermore, it has been found that compared to other muscles, the trapezius muscle is most activated in response to mentally stressful tasks (Roman-Liu, Grabarek, Bartuzi & Choromański 2013; Willmann &
Bolmont, 2012). Based on these studies, an EMG measurement of the trapezius muscle in misophonic individuals should be extremely helpful in understanding the physical reaction in misophonia. The triggers in misophonic individuals could evoke a similar reaction as the mentally stressful tasks in healthy individuals from previous studies, such that the triggers alone should evoke a higher motor response in the trapezius muscle of misophonic individuals compared to healthy individuals.
Another topic of interest within this study contains the short term effects that triggers have on trapezius muscle activity in misophonic individuals. It is unclear at what time point the muscle response in individuals with misophonia is most prominent. Misophonic
individuals have reported a physical response both in anticipation of the trigger as during the occurrence of the trigger (Schrӧder et al., 2013). In a previous EMG study, the short term effects of a nociceptive stimulus (a painful stimulus, in this case an electric shock) on the trapezius muscle in healthy individuals have been investigated (Luijcks et al., 2014). Participants in this study responded with higher trapezius activity in the anticipation phase, when the participant knew the stimulus was coming but not exactly when it was coming, than during the stimulus phase. Relating these results to the current study and assuming that the nociceptive stimulus of the previous study is similar to a trigger stimulus in misophonic individuals, it could be that the EMG activity in misophonic individuals is higher during the anticipation of the trigger than during the occurrence of the trigger itself. This effect should then be higher for misophonic sounds than control sounds. Luijcks et al. (2014) further demonstrated that the activity of the trapezius muscle was higher in the anticipatory phase, when the self-reported anticipatory stress level was high, compared to when it was low. In line with this finding, it is interesting to investigate whether this also holds for misophonic individuals. It might be that the EMG activity in individuals with misophonia is higher in anticipation of misophonic sounds, when the anticipatory stress level is high, than in anticipation of control sounds. Such findings could be of beneficial use in the treatment of misophonia. For example, one might think of a treatment in which the physical response is tackled at the time point when it is most prominent.
A final question investigated within this study is the relationship between the
subjective response and the physical response to a trigger in individuals with misophonia. In a study by Edelstein et al. (2013), a positive correlation was found between the subjective rating of a stimulus and the skin conductance response for individuals with misophonia and controls. Linking these results to the current study, it can be expected that if the subjective response to
a trigger is high, the corresponding EMG activity should be high as well. Further in the same study by Edelstein et al. (2013), a positive correlation between the subjective ratings of misophonic individuals and controls on triggers was found, although the misophonic rating was higher on average. Thus it was suggested that the misophonic reaction to an aversive stimulus is an extreme of a similar reaction in controls. In the current study, it can be investigated whether similar results can be found in the ratings of misophonic and control subjects.
In the current study, a surface EMG measurement was taken of the left trapezius muscle while individuals with misophonia and control subjects listened to different
misophonic or control sounds. Misophonic sounds were for example the sound of chewing and sniffing. Control sounds were sounds such as shaking a bottle of water or cutting paper. Before the sound was heard, participants waited ten seconds, during which participants read the sound that they were going to hear on the screen. The sound itself lasted twenty seconds. After each sound, participants gave a rating of how they felt on a scale of minus five to plus five.
Based on the aforementioned findings and theories, a couple of expectations were made. First of all, an interaction effect was expected of subject group and sound on EMG activity. The mean EMG activity was expected to be significantly higher during misophonic sounds than control sounds for misophonic individuals compared to controls. Secondly, in individuals with misophonia, the mean EMG activity during the anticipation phase was expected to be significantly higher than during the sound phase for misophonic sounds. Thirdly, it was expected that in individuals with misophonia, during the anticipation phase, the mean EMG activity would be significantly higher in anticipation of the misophonic sounds compared to the control sounds. Furthermore, it was expected that the mean EMG activity during misophonic sounds would correlate positively with the rating of the
corresponding sounds in individuals with misophonia. Finally, it was expected that the subject group would have an effect on the rating, such that the mean ratings of individuals with misophonia would be higher than those of controls for misophonic sounds compared to control sounds.
Methods Subjects
43 subjects took part in this study (2 men and 41 women). Their age ranged from 17 to 72 (M= 38, SD=14). Misophonic individuals were recruited online from a Dutch social
network community for misophonia. Control participants were recruited by inviting
misophonic individuals to bring a non-misophonic friend or family member to the experiment and through other online social networks. Participation was rewarded with 12.50 euros an hour. There were 21 misophonic subjects and 22 control subjects. Misophonic subjects were matched on age and gender to control subjects. As triggers in misophonia differentiate between misophonic individuals, a preliminary study tested whether a description of the misophonic sound that would be presented during this experiment caused any distress. If subjects rated seven out of eight descriptions of the sounds as a trigger, they were invited to participate.
Materials
The sounds that participants listened to during the experiment were divided into
misophonic sounds and control sounds. There were eight misophonic sounds and eight control sounds. The sounds that were presented in the experiment are listed in Table 1. The sounds were shuffled in a random order. Matched participants received the same order of sounds. The same sound was not presented twice in a row nor was a misophonic sound presented three times in a row. Each misophonic and control sound was presented two times over the whole experiment. Loudness of the sounds ranged from 53.7 dB to 71.9 dB (M=61.5 dB). Subjective responses to the sounds were measured by presenting a rating scale after each sound, in which participants were asked how they felt during the sound. Participants rated the sounds on a scale from minus five to plus five. Activity of the trapezius muscle was measured using EMG. The experiment was performed using Presentation® software (Version 18.3, Neurobehavioral Systems, Inc., Berkeley, CA, www.neurobs.com).
Table 1
Misophonic and control sounds presented during the experiment.
Misophonic Sounds Control Sounds
Eating an apple Pealing an apple Eating a carrot Pulling a zipper
Procedure
Participants were seated in front of a computer screen and received headphones to put on. Electrodes were placed on the left trapezius muscle. ECG electrodes and an EDA meter were also placed on the participants, but both of these measurements are not taken into account in this study. The left arm was positioned resting on a towel on the table in order to minimize the amount of movement possible. Each trial, of either misophonic or control sound, was divided into four phases. During the wait phase, participants waited twenty seconds, during which there was no stimulus or anticipatory text displayed on the screen. During the anticipation phase, which lasted ten seconds, a text on the screen displayed which sound was going to be heard next. The sound itself lasted twenty seconds. This was the sound phase. After each sound, a rating was given by pressing on the left and right arrow keys on a
keyboard with the right hand. The left arrow key moved the rating to the negative values and the right arrow key to the positive values. After the rating phase, the wait phase started again. At the start of the experiment and during the breaks, the participant was reminded to sit as still as possible. There were three breaks in the experiment. The experiment lasted around 30 minutes.
Psychophysiological Recordings
EMG was measured with disposable pre-gelled Ag/AgCl electrodes (3M Red Dot), connected to a custom made bipolar EMG amplifier with an input resistance of 1GΩ and a bandwidth of 5-1000Hz (6dB/oct). Two electrodes were placed on a line between the acromion and C7. The reference electrode was placed slightly below the two electrodes to form a triangle (Seniam, 2017). Both the raw and 50Hz notch-filtered EMG were sampled. A National Instruments USB-6210 A/D converter was used to digitize the analogue data at a rate of 1000Hz/s. Vsrrp98® software (Vsrrp98 v10.4, University of Amsterdam, 1998-2017) was used to record and analyze the data. Before data analysis, the EMG signal was high-pass filtered at 28Hz (Butterworth, 4th order) to reduce both movement and ECG artefacts in the signal. A 50Hz notch filter was applied to reduce remains of noise on the signal. Root Mean
Slurping Squeezing a cloth Snoring Cutting cardboard Eating a sandwich Shaking a bottle Sniffing Pulling duct tape Chewing gum Cutting newspaper Eating chips Pouring water
Square values were calculated for each phase, except for the rating phase due to the high amount of possible movement artefacts during this phase, and during every sound within the experiment. This was done by squaring the filtered EMG data, taking the means of each phase (wait, anticipation and sound) per sound (misophonic or control) and then displaying the square roots of the resulting data (Motion Lab Systems, 2009). RMS averaging has been shown to be a reliable way of analyzing the EMG signal (Motion Lab Systems, 2009).
Results
Of the 43 participating subjects, one subject from the control group was removed from analysis. The reason for this removal was that the subject’s mean EMG values were twice as high in each phase as that of the other control subjects. The subject reported being very bored and its extreme EMG values are therefore most likely the result of not being able to sit still. The misophonia subject group did not have any obvious outliers. 21 Control subjects and 21 misophonic subjects remained for the data analysis. An independent t-test was conducted on the data to test whether the mean age was similar in both subject groups. The outcome of this test was not significant, t(40)=0.55, p=.59, indicating that there was no difference in age between the two subject groups. To control for outliers, individual EMG values that were 2.5 standard deviations above the individual means for each phase were removed. In the waiting phase, 13 values were removed for misophonic sounds and 16 for control sounds. In the anticipation phase, 6 values were removed for misophonic sounds and 14 for control sounds. In the sound phase, 13 values were removed for misophonic sounds and 16 for control sounds. Overall, 78 values of a total of 252 EMG values were removed. Table 2 shows the resulting means and standard deviations of the data per subject group, sound and phase.
Table 2
Mean EMG values and standard deviations (between brackets) during the wait, anticipation and sound phase.
Misophonic Sounds Control Sounds
Wait Anticipation Sound Wait Anticipation Sound Misophonic Subjects 3.9 (2.0) 3.9 (2.0) 4.0 (2.1) 4.0 (2.0) 3.8 (1.9) 3.8 (2.0) Control Subjects 3.6 (1.3) 3.4 (1.3) 3.4 (1.4) 3.7 (1.4) 3.4 (1.2) 3.3 (1.1)
As can been seen from Table 2, the mean EMG values during the wait phase were higher than the EMG values of most of the other phases. This result was unexpected. To further investigate the wait phase and to check whether the wait phase was the same in all conditions, a Factorial Mixed ANOVA was conducted on the mean EMG values of subject group and sound during the wait phase. There was no significant difference between the subject groups during the wait phase, F(1, 40)=.36, p=.55 or between misophonic or control sounds during the wait phase, F(1, 40)=2.33, p=.13. Further, the wait phases were checked for outliers, but no significant outliers were found to make a difference on the high EMG values in the wait phase. Therefore, no statistical explanation was found for the high wait phase.
Next, the main expectation was analyzed. The mean EMG activity was expected to be significantly higher during misophonia sounds than control sounds for misophonic individuals compared to controls. To test this, a Factorial Mixed ANOVA was conducted on difference scores during the two sound phases of misophonic and control sounds for both subject groups. Difference scores were created to control for individual differences in EMG activity. To create the scores, EMG values during the anticipation and sound phase were subtracted from the values during the wait phase. The difference scores were computed for each trial in every condition, after which the mean of each difference score was calculated for every condition. The resulting scores are summarized in Table 3.
Table 3
Difference scores and standard deviations (between brackets) of the anticipation and sound phases subtracted from the wait phase, for both subject groups.
Misophonic Sounds Control Sounds
Anticipation Sound Anticipation Sound Misophonic Subjects -.02 (.17) .08 (.27) -.19 (.19) -.18 (.21) Control Subjects -.22 (.34) -.17 (.37) -.25 (.40) -.37 (.59)
Assumptions were checked before testing the main expectation. The assumption of homogeneity of variance was met for the misophonic sounds, F(1, 40)=.28, p=.60, but not for the control sounds, F(1, 40)=6.53, p=.02. Yet because ANOVA is a robust measurement and
the sample sizes were equal in both groups, the data is fairly unaffected by this violation (Field, 2013). When analyzing the main expectation, the difference scores were on average higher for misophonic sounds (M=-.04, SD=.34) than for control sounds (M=-.27, SD=.45) for both subject groups. This difference was significant, F(1, 40)=9.53, p<.05. The effect of subject group on EMG activity was also significant, F(1, 40)=5.62, p<.05; there was a significant difference in EMG activity between the misophonic subjects and control subjects, where the misophonic subject group had a higher EMG average than the control subject group for both misophonic and control sounds. Unfortunately, there was no significant interaction effect of sound and subject group on EMG activity in the sound phase though, F(1, 40)=.16, p=.69. Figure 1 depicts these results.
Figure 1. Decrease in EMG activity of the Trapezius muscle during the control sounds compared to the misophonic sounds for both subject groups. Error bars indicate 95% confidence intervals.
To look at the different phases separately, an exploratory Factorial Mixed ANOVA was conducted on the mean EMG values for each sound (misophonic, control) and phase (wait, anticipation, sound) per subject group (misophonic, control). These values are indicated in Table 2. The assumption of homogeneity of variance was met for all phases, but the
assumption of sphericity was not met, W(2)=0.58, p<.05 and therefore a correction test was conducted on the results. The Greenhouse-Geisser correction was below .75, so this correction was used. The effect of sound (misophonic, control) was not significant, F(1, 40)=.66, p=.42 and the effect of subject group (misophonic, control) was not significant, F(1, 40)=.72, p=.40. There was no significant difference in mean EMG activity for both subject group or sound, in contrast to the previous analysis on the difference scores in which there was a significant difference in subject group and sound. Though the effect of phase (wait, anticipation, sound) was significant, F(2, 80)=12.54, p=.001; the mean EMG activity was significantly different for each phase for both misophonic subjects and control subjects, where the EMG activity was highest during the wait phase compared to the other phases, except for the misophonic sounds in the misophonic subject group. Yet caution has to be taken when interpreting these results, as the difference scores are not used and thus individual differences in EMG activity are not taken into account. Figure 2 shows the difference in phases during misophonic sounds and Figure 3 shows the difference in phases during control sounds.
Figure 2. Mean EMG activity for both subject groups during the different phases of misophonic sounds. Error bars indicate 95% confidence intervals.
Figure 3. Control Sounds. Mean EMG activity for both subject groups during the different phases of control sounds. Error bars indicate 95% confidence intervals.
It was further expected that, in individuals with misophonia, the mean EMG activity during the anticipation phase would be significantly higher than during the sound phase for misophonic sounds. To test this expectation, a paired samples t-test was conducted on the difference scores during the anticipation and sound phase for misophonic sounds in
individuals with misophonia. The difference score was higher during the sound phase (M=.08, SD=.27), than during the anticipation phase (M=-.02, SD=.17). This difference was
significant, t(20)=-2.57, p<.05, although not in line with the expectation.
Furthermore, it was expected that in individuals with misophonia, during the
anticipation phase, the mean EMG activity would be significantly higher in anticipation of the misophonic sounds than in anticipation of the control sounds. To test this expectation, a paired samples t-test was conducted on the difference scores during the anticipation phase for both misophonic and control sounds in individuals with misophonia. The difference scores during anticipation of the misophonic sounds were higher (M=-.02, SD=.17) than during anticipation of the control sounds (M=-.19, SD=.19). This difference was significant, t(20)=2.78, p<.05 and shows that the EMG activity of individuals with misophonia was
significantly higher during anticipation of the misophonic sounds than during anticipation of the control sounds.
It was also expected that the mean EMG activity during misophonic sounds would correlate positively with the rating of the corresponding sounds in individuals with
misophonia. Table 4 summarizes the mean ratings of both subject groups for misophonic sounds and control sounds. Because the EMG data in this experiment was not normally
distributed, the nonparametric Spearman test was conducted on the mean EMG activity during the misophonic sounds and the mean rating of the corresponding sounds. There was no
significant relationship between EMG activity during misophonic sounds and the rating of the misophonic sounds, r=.04, p=.88 in individuals with misophonia. This relationship is shown in Figure 4. As can be seen from the figure, the correlation between the two variables is minimal.
Table 4
Mean ratings and standard deviations (between brackets) for misophonic and control sounds for both subject groups.
Misophonic Sounds Control Sounds
Misophonic Subjects -4.24 (.14) .01 (.26)
Figure 4. Mean EMG activity for misophonic sounds in misophonic individuals in relation to the rating of the misophonic sounds in misophonic individuals.
Lastly, it was expected that the mean rating of individuals with misophonia would be significantly higher for misophonic sounds than control sounds, compared to control subjects. A Factorial Mixed ANOVA was conducted on the mean ratings per subject group
(misophonic, control) and sound (misophonic, control). Assumptions were checked
beforehand. The assumption of homogeneity of variance was met, F(1, 40)=2.16, p=.15 for the misophonic sounds and F(1,40)=.35, p=.56 for the control sounds. The effect of rating was significant, F(1,40)=228.9, p<.05. The mean rating for misophonic sounds was more negative than for control sounds. The effect of subject group was also significant,
F(1,40)=47.3, p<.05. Misophonic subjects gave a more negative rating to misophonic and control sounds than control subjects. The interaction effect of subject group and sound on rating was also significant, F(1,40)=30.64, p<.05. These results are depicted in Figure 5. As can be seen from the figure, misophonic individuals gave a more negative rating to
misophonic sounds than control sounds and a more negative rating than control subjects. Figure 5 also shows that control subjects have a similar result in rating as misophonic subjects, but to a lesser extreme.
Figur e 5. Mean rating for both subje ct group s after misop honic and contr ol sound s. Error bars indicate 95 % confidence intervals.
Discussion
The goal of this study was to investigate the influence of misophonic sounds on the EMG activity of the trapezius muscle in misophonic individuals. To summarize the results, firstly, both misophonic and control subjects activated their trapezius muscle more during misophonic sounds than during control sounds. Also, the misophonic subjects had more muscular activity in response to both sounds than control subjects. There was no interaction of misophonic sound and subject group on EMG activity though. When there was no sound and no anticipation of a stimulus, during the wait phase, both misophonic and control subjects activated their trapezius muscle the most, except for misophonic sounds in the misophonic subject group. In the misophonic subject group, muscular activity was higher during the sound itself than during the anticipation of the sound. Additionally, in misophonic subjects, the trapezius activity was higher during anticipation of misophonic sounds than in anticipation of control sounds. Finally, there was no relationship between the subjective ratings and trapezius activity for both misophonic subjects and controls, but there was an interaction effect of misophonic sound and subject group on the rating.
Based on previous research it was first of all expected that misophonic subjects would have higher trapezius muscle activity in response to misophonic sounds than control subjects, as misophonic subjects report having a strong physical reaction in response to triggers
(Edelstein et al., 2013; Schrӧder et al., 2013). This interaction effect was not found, as there was no effect of subject group and sound on EMG activity. Yet surprisingly, the control subjects activated their trapezius muscle in response to misophonic sounds as well and the difference between misophonic sounds and control sounds was significant for both subject groups. Therefore, misophonic subjects seem to be an extreme of a similar response in control subjects. This idea corresponds to a finding by Edelstein et al. (2013), in which a positive correlation between the subjective ratings of misophonic individuals and control subjects on misophonic triggers was found. Secondly, the expectation that individuals with misophonia would have more EMG activity in anticipation of the sound than during the sound itself did not hold. This expectation was supported by previous research from Luijcks et al. (2014), in which participants responded with higher trapezius activity in the anticipation phase, when the participant knew the stimulus was coming but not exactly when it was coming, than during the stimulus phase. Yet, the EMG activity during the wait phase in the current study was higher than during the anticipation phase and the sound phase. As such, the wait phase in the current study might have been more similar to the anticipation phase in Luijcks et al. (2014), in which case the second expectation would hold. The third expectation of this study does hold, as misophonic subjects in the current study had higher EMG activity during the
anticipation phase for misophonic sounds compared to control sounds. Thus it seems that the anticipation of a trigger plays an important role in the physical symptoms involved in
misophonia, though not as important as the physical symptoms during the sound itself. The expectation on the correlation of EMG activity with the ratings did not hold, but the
interaction effect of subject group and sound on rating found in this study does support the idea that misophonic subjects are an extreme of a similar response in non-misophonic individuals.
A couple of alternative reasons for these results are still possible. One might be the possibility that the experiment made participants nervous, especially during the wait phase when the participants did not know what going to come next. This might have been the cause of the high EMG activity for both subject groups during the wait phase, as there is no
statistical explanation for this result. Another cause of the high EMG activity during the wait phase might have been movement artefacts, especially since the rating, in which participant
pressed a key, preceded the wait phase. Even though the EMG electrodes were placed on the left shoulder and responses were given with the right hand, a hand movement in the rating phase might have had an influence on these results, as participants might still have been moving their hand back to its original position during the wait phase. A final alternative explanation for the results in this study is that individual differences in triggers were not taken into account when analyzing the correlation between the rating and the EMG activity in misophonic individuals. Even though a preliminary study tested whether the misophonic sounds presented in this experiment were triggers for the participating subjects, individuals with misophonia still differ individually in which triggers they find most annoying (Edelstein et al., 2013). Because of this, the correlation should have been conducted per subject and misophonic sound to be able to find significant results.
There are also a few limitations to this study. One of these is that surface EMG
(sEMG), which was used in this experiment, is not a completely reliable measuring device, as sEMG is sensitive to movement and noise. Considering it is hard to sit still for 30 minutes, movement might have had an influence on the results in this study, especially as it is impossible to filter out all movement artefacts. Also, because the wait phase was so high in this study, there is no real baseline state of EMG activity for the participants. In this way, there is no point to which the anticipation or sound phases could really be compared.
In a future study, an experiment in which responses are not given manually might be helpful in assuring the results are not influenced by hand movements. A future study could also set up better baseline measurements, in which the EMG activity is not as high as it is in this study. Furthermore, a future study could take individual differences into account when analyzing the correlation between EMG activity and triggers, as triggers are experienced differently for every misophonic individual. Another topic of interest for a future study is the sensitivity of misophonic individuals in general. From the results of this study, misophonic individuals appear to be more sensitive to sounds than non-misophonics. The EMG responses are similar, but just more extreme. It might be interesting for a future study to investigate the sensitivity of misophonic subjects in other areas as well.
Overall these results give a deeper understanding of the different aspects of misophonia. As this was the first EMG study on misophonia, all the results help in better understanding the physical symptoms of this disorder. Clinicians especially might be able to benefit from the results of this study, as they show the importance of the physical response in misophonia and the importance of the anticipation of a trigger. Furthermore, as it seems that
misophonic individuals are not that different from non-misophonic individuals, an assurance for individuals suffering from misophonia might be that their symptoms are not an
abnormality, but more the result of an extreme of a similar sensitivity in others.
This study demonstrated that the influence of misophonic sounds on the EMG activity of the trapezius muscle in misophonic individuals is very prominent. The trapezius muscle seems to play an important role in the experience that misophonic individuals have when hearing a trigger. Thus the report that misophonic individuals give of a physical response to their trigger is experimentally demonstrated (Edelstein et al., 2013; Schrӧder et al., 2013). The results of this study can be used to further help understand this still relatively unknown
disorder and its origin.
References
Dozier, T. H. (2015). Counterconditioning treatment for misophonia. Clinical Case Studies, 14(5) 374-387. doi: 10.1177/1534650114566924
Edelstein, M., Brang, D., Rouw, R. & Ramachandran, V. S. (2013). Misophonia:
Physiological investigations and case descriptions. Frontiers in Human Neuroscience, 7(296), 1-11.
Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics, 4th edn., Comparing several means: ANOVA (pp. 445). London: Sage.
Jastreboff, P. J., & Jastreboff, M. (2001). Hyperacusis. Audiology Online, 1–5.
Krantz, G., Forsman, M. & Lundberg, U. (2004). Consistency in physiological stress responses and electromyographic activity during induced stress exposure in women and men. Integrative Physiological and Behavioral Science, 39(2), 105–118.
Kumar, S., Tansley-Hancock, O., Sedley, W., Winston, J., Callaghan, M., Allen, M., Cope, T., Gander, P., Bamiou, D., and Griffiths T. (2017). The brain basis for misophonia. Current Biology 27, 1–7.
Luijcks R., Hermens H., Bodar L., Vossen C., Os J. & Lousberg R. (2014). Experimentally induced stress validated by EMG activity. PLoS One, 9(4).
Lundberg , U., Forsman , M., Zachau , G., Eklöf , M., Palmerud ,G., Melin B. & Kadefors R. (2002). Effects of experimentally induced mental and physical stress on motor unit recruitment in the trapezius muscle, Work & Stress, 16(2), 166-178. doi:
10.1080/02678370210136699
Motion Lab Systems (2009). A Software User Guide for EMG Graphing and EMG Analysis. Signal Analysis Methods (pp. 43). Los Angeles: Motion Lab Systems.
Raez, M., Hussain, M. & Mohd-Yasin, F. (2006). Techniques of EMG signal analysis: detection, processing, classification and applications. Biol. Proced. Online 8(1), 11– 35. doi: 10.1251/bpo115
Roman-Liu , D., Grabarek I., Bartuzi P., & Choromański, W. (2013). The influence of mental load on muscle tension, Ergonomics, 56(7), 1125-1133. doi:
10.1080/00140139.2013.798429
Rouw, R. & Erfanian, M. (2017). A large-scale study of misophonia. Journal of Clinical Psychology 0(0), 1-27.
Schröder, A., Vulink, N. & Denys, D. (2013). Misophonia: Diagnostic criteria for a new psychiatric disorder. PLoS One, 8(1). doi:10.1371/journal.pone.0054706
Wijsman, J., Grundlehner, B., Penders, J. & Hermens, H. (2011). Trapezius muscle EMG as predictor of mental stress. ACM Transactions on Embedded Computing Systems, 12(4). doi: http://dx.doi.org/10.1145/2485984.2485987
Willmann M. & Bolmont B. (2012). The trapezius muscle uniquely lacks adaptive process in response to a repeated moderate cognitive stressor. Neuroscience Letters, 506, 166– 169.
