A screening tool of amusia: a pilot study

Introduction

People with amusia from infancy have difficulty with musical perception despite normal hearing, language, and intelligence (Ayotte et al., 2002). More specifically, amusics understand everyday environmental sounds, but unfortunately, do not recognize familiar melodies. They do not detect the difference between melodies, which differ only in one or two notes out of scale (Peretz & Hyde, 2003), or off-key notes in unknown melodies (Ayotte et al., 2002; Hyde & Peretz, 2005; Peretz, Cummings and Dubé, 2007). For this reason, some investigators refer to the disorder as tone deafness, with difficulty in pitch discrimination and frequent comorbidity with learning difficulties, mainly dyslexia, due to the main and common deficit of pitch discrimination. This impairment in auditory accuracy and all of the above deficits are related to cognitive deficits that constitute determinants off both language learning difficulties and tone deafness (Couvignou & Kolinsky, 2021). In addition to these basic features, a secondary deficit is the inability to understand the completion of a musical phrase, which is still being researched (Koelsch et al., 2000; Koelsch & Jentschke, 2010). Τhe present disorder with the above deficits, has been investigated since the 19th century (García-Casares et al., 2013), but only recently the characteristics and diagnostic tests have been studied (Vuvan et al., 2018). In 2003 the diagnosis of the disorder was executed through “The Montreal Battery for the Evaluation of Amusia” (MBEA, Peretz et al., 2003), and it is a reference point and quantitative tool for identifying congenital amusia globally. Accordingly, six subtests were created, including the off-tone, key, rhythm, interval, measure, and memory test (Vuvan et al., 2018). A few years later, Peretz (2008) developed a corresponding short online test of 15 minutes (Peretz et al., 2008). Studying the above cues, this screening tool was created in the present study for the first time briefly and simultaneously based on laboratory evaluation, as described below.

Methods

In the present study, quantitative research methodology was applied using a new screening tool that contained 55 melodies and 12 pairs of notes. These are framed in 7 trials such as “dissonant intervals”, “out of tone”, “contour”, “rhythm”, “memory”, “integration” and “emotion”. The main trials were based on the global valid «Montreal Battery for the Evaluation of Amusia (MBEA)» due to the deficit that the amusics have a difficulty in distinguishing or remembering two melodies that differ in only one note out of tone or out of key. Melodies are based on different rhythmic alternations and emotions. In addition to these aspects of amusia the present screening tool contains two additional trials. First, the dissonant intervals that refer to the difficulty of recognizing these sounds as unfamiliar and second the completion of sheet music which is referred to as a secondary deficit by recent researchers. All music stimuli were presented to a piano timbre and constructed in major mode according to Western tonal-harmonic conventions. Moreover, melodies were recorded in wav format at adobe audition 3, to improve sound quality. Using a metronome, the tempo of the musical stimuli was 80 beats/min, and the procedure was based on laboratory evaluation lasting 15 min¹. Participants who obtained a score below 29,4 (out of 42 total responses) were considered as possible amusics because it is 2 SD under the mean of non amusics, as it is applied to the MBEA. Therefore, if the participants scored above the baseline, they were considered to have passed the test, whereas if they scored below it, a suspicion of amusia was noted and the application of the valid diagnostic tool was encouraged. After the test was administered, factors influencing amusia were investigated (Manika, 2024a). The t-test statistical control was used to compare the averages between two groups, whereas when there were more than two averages for comparison, the statistical method of analysis of variance was applied. To neutralize the effects of other factors, regression analysis was used for gender, age and musical training as explanatory factors. In all tests, a 5% level of statistical significance was used, and the results were analyzed using SPSS (Underlying data) (Manika, 2024b).

Procedure

The entire procedure was carried out with laboratory evaluation and the duration was 15 min. Before the inception of screening tool, was checked the basic condition for identifying amusia which is the measurement of hearing thresholds because, as it was mentioned, the deficits of this disorder such as difficulty in pitch discrimination, memory, and rhythm occur even without hearing insults. All participants had hearing within normal limits bilaterally, with pure-tone hearing thresholds equal to or less than 15 dBHL. Afterwards, each participant went into a special chamber and was properly configured without being affected by environmental noises, wearing appropriate headphones to achieve clear listening to melodies. Before each test the participants received one instance and were provided with feedback after each practice trial. For example, in the first test two pairs of notes were given each time and participants were asked to answer which of the two sounds was wrong. In the next three tests individuals should detect whether the two melodies are the same or because the critical tone is altered either in time or pitch. Afterwards, in the fifth and sixth trials the participants’ task was to recognize the famous part of the song and decide which of two melodies were complete or incomplete respectively. Finally, they determined the emotion of each melody. Answers were recorded in a completely anonymous questionnaire, and as justified in the Montreal Battery for the Evaluation of Amusia, the disorder was detected when the final score was less than 2 SD from the global mean of non-amusic individuals. For this reason, in the present and brief screening tools, success appears when the score ranges from 4 to 6 (pass), while detecting the suspicion of amusia with a score below grade 4, recording the final score as pass or fail, respectively. The research findings analyzed in SPSS, are featured in the results and the content of the present screening tool is explicitly described below:

• 1. Dissonant intervals

  It is noteworthy that people with amusia do not distinguish dissonant musical intervals as unfamiliar (Goswami, 2011; Anthony, 2016). Therefore, the innovative first trial is applied, and it includes individual dissonant intervals, to determine whether each participant recognizes which musical stimulus sounds discordant. Based on mathematics, the dissonant intervals studied by Pythagoras in ancient times are second, augmented fourth, tone and semitone (Anthony, 2016). These musical stimuli sound as «wrong»² (Tramo et al., 2001). Six examples with two pairs of notes were presented each time. In each example one pair is unfamiliar (dissonant interval) whereas the other sounds good to the human ear (consonant interval). The «wrong» intervals that were selected are Do+ and Fa# ( Figure 1), Re and Sol#, Sol b and LA, Do+ and Re#, Sol # and Do+, Fa # and Sol. Therefore, does the listener identify which of the two examples sounds unfamiliar?

• 2. Out of Tone

  The basic deficit in amusia is the difficulty in distinguishing the note from the tone. Therefore, the listener compares melodies that differ by only one note (Vuvan et al., 2018). Six pairs of melodies were created based on the rules of harmony in Western classical music, and the scales used were Do major, Mib+ and Fa major. The tone that changes is located at the strong part of the last or before the last measure of melody. Therefore, the two melodies are almost identical and present a foreign note, which is one semitone ( Figure 2) or a tone away from the previous note, because amusics do not distinguish this difference. Lastly, only one pair of melodies is identical to stimulate the listener’s interest in more careful listening.

• 3. Contour

  Six pairs of melodies were presented, almost identical, with only a change in one or two notes. This change concerns the distance of the tone or semitone ( Figure 3). The difference between “out of tone” and “contour” is that in the previous trial, only one note changed and sounded false. Although in contour, two melodies differ in one or two notes that come from another scale and the different sounds are correct. More specifically, the difficulty in recognizing almost identical melodies is caused by a lack of distinguishing a tone or a deficit in memory capability (Lu et al., 2017). The first case concerns the precise pitch discrimination of the difference of tones, while the low score in the second case is created by having difficulty in memory. Therefore, this trial is applied to detect amusia, researching the difficulty in pitch discrimination or remembrance of the changes in melodies (Liu, Patel, Fourcin & Stewart, 2010). In both cases, whether the sound that differs is correct (the trial of contour) or wrong (out of tone), the amusics cannot discern whether the melodies are the same or different.

• 4. Memory

  This trial is applied to detect people with amusia, because they do not recognize well known melodies, as they have a deficit in memory of melodies, which reflects the difficulty of perceiving tone. It is of great importance that amusics do not distinguish similar melodies because of the difficulty of acoustic tone accuracy and they cannot recognize them in the future (Graves et al., 2019). These song elements are encoded in memory and rely on neurological functions in the auditory cortex (Allen et al., 2019, 2017). Consequently, famous melodies are given to check if the participant can identify which song, it is such us “twinkle little star”, “Bella Chiao”, “The little rooster”, “Jingle bells” ( Figure 4), “Rudolph the red nose reindeer” and “The drummer.”

• 5. Rhythm

  Some amusic people have difficulty perceiving the rhythm of melodies (Vuvan et al., 2018). This depends more deeply on temporal processing, when the listener distinguishes the intensity of the sound stimuli (Blauert, 1997). The present study focused on the general population, checking music perception in the frame of rhythm, inspired by the Montreal Battery for the Evaluation of Amusia. More specifically, the first three acoustic stimuli differed rhythmically in the value of two notes in only one meter ( Figure 5). In the fifth example the pairs of melodies have the same pitch but different rhythm (such as a beat of 2/4 in one melody and 3/4 in the other) to check if they detect rhythmic alternations (Grahn & Brett, 2009) and in the fourth and last example the melodies are presented rhythmically identical to confuse and at the same time attract listeners’ interest.

• 6. Integration

  Some people with amusia have deficits in completeness³ (Koelsch et al., 2000; Koelsch & Jentschke, 2010). In this study, we investigated the shortfalls in the music frame. Every piece of music contains phrases with endings and “breaths”, based on musical rules of harmony. However, every listener, such as a non-musician with normal acoustic processing, was located at the end of a musical phrase. Fiveash et al. (2018) examined the completion ability, interrupting the musical piece at specific points and asking listeners if they understood that the phrase ended each time. Therefore, in the present research, six melodic phrases were heard to check whether the listener understood whether they were complete. First, a pair of melodies is presented and strictly based on the rules of harmony of Western music. The first example is complete in the beginning, while the second end in the seventh step gives a feeling of non-completion ( Figure 6). There are five individual melodies, which are parts of well-known melodies, of which three are complete, and the other two do not give a feeling of integration. More specifically, the first phrase presented by “Inside this boat “in Mi minor, the 4th” Etude e-moll op.46 Nr. 7, Stephen Heller “in Mi minor” and then the“ Concerto in D minor, Bach J.S. ” in Re Elasson, they finish in the first step, giving the feeling of completion. In contrast, the melodic phrases in the second and third examples from the “Etude e-moll op.46 Nr. 7, Stephen Heller “(Minor) and” Numb (Piano), Linkin Park “respectively, have been discontinued in the third tier of harmony and both melodies are not complete. Thus, the above test on the ability of music to integrate is applied, as amusics cannot distinguish whether a melody is complete or not (Koelsch et al., 2000; Koelsch & Jentschke, 2010).

• 7. Emotion

  Apart from integration, another secondary deficit in amusia is the difficulty in perceiving emotion, which emits a musical melody (Reed et al., 2011). Thus, six melodies were chosen, and the listener was required to distinguish which one was happy or sad. For example, for happy feelings were “Sonate Allegro No 279, Mozart” ( Figure 7) and the waltz “Por una Cabeza, Carlos Gardel”. On the contrary, for sad expressiveness were chosen the sonata “Moonlight Adagio sostenuto, Beethoven,” “Notos, L. Machairitsas,” the famous song “The piano-amazing short, Yann Tierse” and “The train leaves at eight” by Manos Eleftherios.

Figure 1. Recognition of the dissonant interval.

Figure 2. Auditory discrimination of whether the melodies are the same or different.

Figure 3. Identification of whether the melodies are the same or different.

Figure 4. Recognition of well-known melody.

Figure 5. Recognition of rhythmic changes.

Figure 6. Identification of which melody is complete.

Figure 7. Recognition of melody feeling.

Results

The present quantitative screening tool was used to identify amusia in the general population and examine which factors affect the appearance of this disorder. According to the results and as confirmed by other surveys, only 4% of the participants suffered from this disorder ( Figure 8). The averages of their success in each trial are shown Table 1. The independent samples t-test was used to compare the means between gender and the score of the present screening tool. Based on this analysis it was found that t-test=1.587, p=0,114 and this means that we do not reject the initial hypothesis Ho which concerns that the means of females and males are the same. Thus, sex did not affect the occurrence of amusia. Similarly, for the second variable of age we applied the t-test sample and specifically t (198) = 6.73, p=0, so in this case the null hypothesis is rejected and the mean values differ between students and adults. However, for greater accuracy, regression analysis was applied where t (199) = -0.443, p = 0,085, a margin greater value from the level of significance, so we accept the null hypothesis that age does not affect the appearance of amusia. To ensure validity the present screening tool should be applied to a larger sample. Afterwards, the variable of music education was examined through one-way ANOVA analysis F (23, 173) =1.108, p=.336>0.05 and the original hypothesis Ho was rejected. In addition, according to the regression analysis R2 = 0.24, F (4,195) = 2.979, p=0.20 and t (199) = 3.422, p = 0.011 <0.05. Thus, according to the finding the average performance of people who had no music education and those who had up to two years were the same. More specifically, people with more than two years of music knowledge showed a higher performance in the present detection test than those with less than two years of music training. Therefore, in contrast to other investigations, this variable seemed to influence the success of the individual in the present screening tool ( Table 2). This is also indicated in Figure 9, where individuals with music education for more than two years presented aggregate higher test scores above average levels than the others. Finally, it is noteworthy to refer to a general context of the positive relationship observed between the intervals, tone, contour and memory tests, indicating their potential for detecting amusia, in contrast to the performance of individuals in emotion and integration, which indicate their presence as secondary deficits. Similarly, it is ascertained in a rhythmical frame because amusia presents basic difficulties in pitch and memory of melodies and not always in perception of rhythm (Figure 10).

Figure 8. Suspicion of amusia-Test performance.

Figure 9. Boxplot of the influence of music training on test performance.

Figure 10. The positive relationship between intervals, tone, contour and memory tests.

The variables of gender, age and music education were examined in relation to the suspicion of amusia. Individuals with more than two years of music lessons had higher performance in the present test. However, sex and age did not affect the existence of this disorder. More specifically, people who had no music education and those who had up to two years were the same. More specifically, people with music knowledge of more than two years showed higher performance in the present detection test than those with music training for more than two years.

Discussion

The present screening tool could detect amusia quickly and in a pleasant manner. This neurological disorder exists in children and more widely in adults with deficiencies enclosing the cognitive network. Owing to the basic deficits of the present disorder, trials such as out of tone, contour and memory have been shown to be decisive for detection, having a common basis of difficulty in distinguishing pitch between melodies. As mentioned previously, the tests of integration and emotion are secondary deficits that give a beautiful nuance to the present test. More specifically, the basic aspects of amusia are examined in the MBEA diagnostic test (Peretz et al., 2003) and online test (Peretz et al., 2008), but the present research also includes a trial of integration. The sheets were taken from well-known songs, and this is appealing to the participants’ listeners, as found when administering the present test.

Additionally, regarding the factors influencing the onset of the disorder, as confirmed by other surveys, a very small percentage suffered from amusia, and the variables of age and sex did not affect the occurrence of this musical disorder. Unlike previous research, the present study suggests that music education lasting more than two years has an influence the presence of amusia. However, further validation with a larger sample size is required. Listening to the melodies took place in a specially configured space to reduce external factors that would affect the attention and concentration of the participants. A limitation of the present screening tool is that it is insufficient to establish a diagnosis. For this reason, after the application of the detection tool, it is suggested that the subsequent administration of the globally valid Montreal Battery of Evaluation of Amusia (MBEA) is the most valid and demanding worldwide diagnostic test to ensure validity of amusia.

Finally, an important cue that will be examined subsequently is the comorbidity of amusia and learning difficulties (such as dyslexia). This could be useful in the future, not only as a tool for detecting auditory difficulties but also as a basis for the creation of a therapeutic tool for amusics or individuals with learning disabilities with the result of improving the individual level in acoustic accuracy and rhythmic perception of musical phrases and language sentences. Simultaneously, this test will help improve your knowledge of musical processing and its relationship with other cognitive processes and it might inform new studies that will investigate the mechanisms of cognition in relation to language and music, opening new windows in the neurology, music, and pedagogical fields.

Ethics and consent

There is consent for publication of the participants’ details and consent of the children’s parents. Also, the Directorate of Primary Education of Achaia, which belongs to the Ministry of Education of Greece, approved the research for the application of the test to adults and students aged six to twelve. The approval number of the document is 4888 dated 04-05-2022. Simultaneously, each parent signed a form for their child’s consent to participate in the listening test. Moreover, parents or adults without children’s participation consented to participate in the survey by filling out a written form just before the amusia test (consent form).

Contributorship

All authors were involved in the literature review and the design of the study experiment. Moreover, the second author offered guidance on the clinical application of the acoustic screening test.