EEG Correlates of Middle Eastern Music Improvisations on the Ney Instrument

The cognitive sciences have witnessed a growing interest in cognitive and neural basis of human creativity. Music improvisations constitute an ideal paradigm to study creativity, but the underlying cognitive processes remain poorly understood. In addition, studies on music improvisations using scales other than the major and minor chords are scarce. Middle Eastern Music is characterized by the additional use of microtones, resulting in a tonal–spatial system called Maqam. No EEG correlates have been proposed yet for the eight most commonly used maqams. The Ney, an end-blown flute that is popular and widely used in the Middle East was used by a professional musician to perform 24 improvisations at low, medium, and high tempos. Using the EMOTIV EPOC+, a 14-channel wireless EEG headset, brainwaves were recorded and quantified before and during improvisations. Pairwise comparisons were calculated using IBM-SPSS and a principal component analysis was used to evaluate the variability between the maqams. A significant increase of low frequency bands theta power and alpha power were observed at the frontal left and temporal left area as well as a significant increase in higher frequency bands beta-high bands and gamma at the right temporal and left parietal area. This study reveals the first EEG observations of the eight most commonly used maqam and is proposing EEG signatures for various maqams.

Introduction

Human creativity has been the focus of thousands of studies and is still a topic of debate with considerable heterogeneity of evidence in brain research. Often, creativity is placed in the right hemisphere (Joseph, 1988; Hoppe, 1998; Demarin et al., 2016), but contradicting theories have been proposed to describe the underlying processes: (1) the dominance of right hemisphere activity (Joseph, 1988), (2) the low cortex activity (Carlsson et al., 2000; Starchenko et al., 2014), (3) the high neural connectivity (Razumnikova and Yashanina, 2014; Mayseless and Shamay-Tsoory, 2015), and (4) the prefrontal and frontal brain activation (Arden et al., 2010). Music improvisation refers to both the process and product of spontaneous creativity of music and constitutes a good model to study neural correlates of creative processes (Liu et al., 2012). Music improvisations, although spontaneous, are constructions resulting from successive decision-making processes. Nevertheless, research studies exploring the neural activity that underlies the creative process of music improvising remain scarce. Moreover, no study has explored the neural correlates of creating Middle Eastern Music, which involves the use of more than the major and minor scales.

Indeed, Middle Eastern Music is characterized by the use of maqams (literally “place” and “position”), which are recognized as the system of melodic tunes used in traditional Middle Eastern Music from Turkey, Azerbaijan, Israel, Iran, and all Arab countries from Middle East to North Africa. These maqams are not exactly the equivalent of “scales” in Western Music and are characterized by defined tonal–spatial factors, while the rhythmic-temporal features are free (Touma, 1971). On the contrary, a waltz would have a fixed rhythmic-temporal organization, while the tonal–spatial component (the melody) is free. As a result, maqams differ in their intervals between the first few notes and are characterized by combinations of phrases that can include microtones such as quartertones, for example. Therefore, maqams resonate as various melodic tunes, that can be classified based on their characteristics into more than 50 families and subfamilies. The names given to each of these maqams are not subject to consensus, thus creating confusion in the literature. For this reason, it is always preferable to define the maqam by its set of intervals signatures as described in Table 1.

TABLE 1

Table 1. Interval signatures of the eight Maqams studied.

The music improviser builds the scales by using combinations of these families and subfamilies of maqams. In addition, associations of maqams to different emotional status are well rooted in Middle Eastern cultures (Kligman, 2001; Powers, 2005). Indeed, since the 9th century, philosophers and scientists such Al-Farab (870–950) have associated different maqams to different sets of emotions (Naroditskaya, 2009; Yöre, 2012). For example, the maqam called Rast is believed to trigger happiness, the maqam called Saba, is associated to sadness. Kligman (1993) has published a first trial of canonization of the maqams in the prayers of Syrian Jews in Brooklyn, New York, during their Shabbat morning service, where he described the association of maqams to various prayers. In his research, Kligman (2001) details how each maqam is said to convey a unique emotion and therefore preferable to use in specific prayers. In Turkey, research and clinical trials involving the use of maqams have been flourishing in the past years, and maqams are still believed to be associated to specific emotions (Özaslan et al., 2012; Bekiroğlu et al., 2013; Tumata, n.d.). Studies that support association of these maqams to specific emotions remain extremely scarce and none of them has explored the possibility of EEG-based emotion recognition. While Self-Assessment Mannequins have become a standard for emotion detection, a growing number of studies are using the development of machine learning algorithms and brain computer interfaces to propose novel methods for emotional assessments using brain signals, that are not be dependent on individual’s ability to express themselves or grasp of their mental state (Suhaimi et al., 2020; Torres et al., 2020). These studies have benefited from previous EEG explorations of neural correlates on Western Music and instruments. Lopata et al. (2017) compared improvisation-trained vs. non-improvisation-trained western musicians and showed an increase in right frontal upper alpha-band activity during more creative tasks such as improvisation, and their results suggested that creativity is probably a trainable mental state. Sasaki et al. (2019) showed in a study involving 14 male guitarists that improvisation over scale is characterized by an increase in power of theta, alpha, and beta bands in prefrontal and motor regions. Dikaya and Skirtach (2015) reported a distinguished EEG pattern in professional musicians during improvisation which was the predominant activation of the left-hemispheric cortical regions simultaneously with high interhemispheric integration in the high-frequency band.

Unfortunately, evidence on how humans perceive and process other modes of music such as the tonal–spatial system of Maqams is almost inexistent. Despite a recent study published in scientific reports by Teixeira Borges et al. (2019) that supported the role of scaling behavior of music in determining the emotions elicited, no study has yet explored and compared the neural correlates of different maqams. The present single subject study is the first exploration of EEG correlates to the eight most commonly used maqams and is believed to allow the expansion of this area of research to include Middle Eastern Music. Prior to this study, no research has focused on how Ney playing translates into EEG correlates.

One of the specificities of Middle Eastern Music is that it can include microtones such as quartertones, used in many maqams, but not all instrument can play those tones. This case study exploration relied on a performance using the Ney instrument, an end-blown flute, that has been played for more than 4500 years in the Middle East and is still an important component of today’s Middle Eastern musical ensemble. Playing the Ney requires a close coordination of mouth and jaw embouchure, lip contraction, diaphragm, and breathing control, in addition to fingering control. Moreover, similarly to the Clarinet, the Ney is known to be one of the most difficult Middle Eastern instruments to play, not only because it does not overblow in the octave, so almost every note has its own fingering and embouchure, but also because the Ney comes in thirteen different sizes, each one allowing the musician to play different ranges of octaves. These characteristics of the Ney instrument result in the necessity to coordinate multiple cognitive processes, including integration of sensory feedback, attention, working memory, decision making, movement, etc. (Ninaus, 2011).

Ney players and teachers often mention that playing the Ney is similar to singing and refer to the Ney as a continuum of their body into which they blow not only air, but also meaning and “words.” Although no research has previously focused on the cognitive demand of performing with this instrument, evidence supports the parallel of woodwind instruments and speech. Because woodwinds are hold at the continuity of the body and playing such instruments involves several parts of the body involved in singing (jaw, lips, tongue, mouth muscles, and breathing), it is expected that similarities are found between the cognitive demand of Ney and speech or singing (Zarate, 2013). Indeed, the musician’s lips function as a valve might for a woodwind instrument, just as the vocal folds need to be controlled to produce various resonances that are important for timbre, loudness, shape, and sharpness (Wolfe et al., 2009).

In addition to investigating Ney improvisations, this single-subject study is the first to explore EEG commonalities and differences among various music modes or maqams of Middle Eastern Music. The EEG is an excellent method to run a first exploration on this original topic, as it gives an overview of several cognitive processes in real time. Those cognitive processes include: working memory, retrieval, focus, concentration, relaxation, planning, decision making, motor planning, and arousal. Therefore, the aims of this short study are as following: (a) explore the EEG commonalities and specificities of Ney performance of various modes or maqams; (b) identify the EEG correlates of improvisations on Ney and comparing it with singing or spontaneous speech; (c) identify interesting patterns or signatures of maqams to explore further. This EEG-based study will set the ground for further explorations finding applications in musicology, music psychology, music performance, neurofeedback training, and music-based therapeutic interventions.

Materials and Methods

Materials

The EEG was recorded using the EMOTIV EPOCx headset (EMOTIV, San Francisco, CA, United States), a wireless headset that consists of 14 saline-based electrodes, recording at 14 sites according to the international 10–20 locations (AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, and AF4) and two reference electrodes CMS/DRL at P3 and P4 (Figure 1). The headset also includes 9-axis motion sensors and detects head movements. The EMOTIV measurement system recorded in a sequential sampling at a sampling rate of 128 Hz, for a band width of 0.16–43 Hz. EEG signals were collected at a resting state with eyes closed, then relaxed with eyes opened. Signals includes digital notch filters at 50 and 60 Hz and built-in digital fifth order Sinc filter. Although the device is able to record performance metrics and facial expressions through its sensors, only EEG signals are analyzed and presented in this present paper. The data were wirelessly recorded using EMOTIV-PRO software running on the experimental computer.

FIGURE 1

Figure 1. Study design. Detail of the 24 improvisations played and the 14 electrode positions, as well as the variables observed for each improvisation.

Experimental Set-Up and Recording

The subject is a healthy right-handed professional Ney player in the age range 30–40 years. Subject was comfortably sitting on a chair in a classroom with dim light and no other electronic than the recording computer. Subject was asked to play a total of 24 improvisations in a row, in the following melodic tunes: Kurd, Saba, Bayati, Hijaz, Huzam, Ajam, Nahawand, and Rast. Each melodic tune or maqam is defined by the intervals depicted in Table 1. The 24 tasks were listed in front of the performer. Each improvisation was played for 1 min at three different tempos: 60, 100, and 120 bpm, before moving on to the following maqam.

Another recording, referred to as the baseline (eyes opened, relaxed, and non-blowing status) was performed before starting the series of improvisations. The EEG was continuously recording throughout the experiment, which lasted 34 min and 32 s, including the setting and baseline recording as well as the time in between improvisations. To identify timestamps of beginning and end of each improvisation, markers were added by the experimenter using the keystroke marker function of the EMOTIV-PRO as described in their guidelines (Emotiv-Pro, 2013) and confirmed post-experimentally by visual observation of the timeline on the screen recording video of the entire experiment. A total of 25 min was included in the present analysis.

Improvisations were recorded using Audacity software and the EEG was recorded as described below. Figure 1 shows a summary of the design. The 24 improvisational audio clips can be found in Supplementary Files.

Data Processing and Analysis

Using a band pass filtering system, the EMOTIV-PRO pre-processes and extracts the power spectra of the following frequency bands: theta (4–8 Hz); Alpha (8–12 Hz); low beta (12–16 Hz); high beta (16–25 Hz); and gamma (25–45 Hz). The EMOTIV-PRO provides the power spectra for all timestamps and electrode location in a csv file. All data recorded between improvisations were eliminated, and data were organized in an excel file based on the following variables: maqam, tempo, frequency band, and electrodes. Markers on timestamps were used to define beginning and end of each improvisation during the recording, allowing the calculation of mean power spectra in excel for the entire duration (1 min) of each improvisation. Data were uploaded into IBM-SPSS (version 26.0) to perform descriptive and statistical analysis. For the descriptive analysis, the mean powers of each frequency band per maqam, regardless of the tempo, as well as the mean powers of each frequency band per tempo, regardless of maqam, were compared. In addition, the mean powers of each electrode separately were compared between each maqam regardless of tempo, and between each tempo regardless of maqam.

This multivariate study used a two-way analysis of variance (ANOVA) to test the hypotheses. The outcome variable was the power, which is quantitative, while all the independent variables were qualitative such as maqam (8 levels), frequency bands (5 levels), tempo (4 levels), and electrode (14 levels). The statistical analysis was done using SPSS to see the main effects as well as the interaction effects of the independent variables on the outcome variable. The interaction term reveals whether the effect of one independent variable on the outcome variable is the same for all values of the other independent variable. Post hoc tests were administered for the pairwise comparisons. Separate plots were prepared for main effects and interaction effects. The assumption of homogeneity was tested using Levene’s test of homogeneity. The confidence level was set at 95%. Research questions were answered here using the two-way ANOVA technique. Heatmaps produced on excel were computed for each frequency band separately and included minimum and maximum values from the mean power spectra for all maqams. To better visualize eventually asymmetry, cells were then reorganized into right/left and frontal/occipital for each maqam.

Using IBM-SPSS, Kaiser–Meyer–Olkin, and Bartlett’s tests were used to measure sampling adequacy for principal component analysis (PCA) analysis. Eigen values and scree plots were computed to find the number of components that can be used. Four components were used for PCA analysis of the eight maqams. Detailed output of the analysis is available in Supplementary File 3.

Results

Changes in Slow Oscillations at the Left Frontal and Left Temporal Area

For each maqam, a different improvisation was performed at 60, 100, and 120 bpm, which allowed the calculation of a mean of spectral powers for each electrode and for each maqam, compared to the baseline (non-blowing and non-improvising resting state). Figure 2 suggests a highly significant increase of theta bands at the frontal left area F7 for almost all maqams (p-value < 0.001), except for the maqam Saba. The details of the means and p-values for all electrodes are available in Supplementary File 1.

FIGURE 2

Figure 2. Left fronto-temporal increase of low frequency bands and parietal left and temporal right increase of high frequency bands during improvisations for each maqam as compared to the baseline. (A) Theta bands at F7. (B) Alpha bands at F7. (C) Beta high bands at T8. (D) Gamma bands at T8. (E) Gamma bands at P7. (F) Beta high bands at P7. (G) Theta bands at T7.

Because these means were performed between different tempos, they reflect the effect of the maqam regardless of the tempo. There seem to be some topographical differences as well as power differences as the tempo increases. However, the effect of the tempo was not tested statistically, since only one improvisation per maqam and per tempo was performed. A Supplementary Figure that details the mean power of bands obtained at each tempo is available in Supplementary File 3.

Observations suggest some specificities of improvising on certain maqams. A highly significant increase of theta bands compared to the baseline is observed for the maqam Ajam in the frontal left area F3 (p-value = 0.000) (Supplementary Table 1) and the maqam Hijaz at the left temporal area T7 (p-value = 0.003) (Supplementary Table 1). In addition, we also observe at F7 a significant increase of alpha bands for the maqam Bayati at the left frontal area F7 (p-value = 0.039) and the maqam Rast (p-value = 0.05) (Table 1).

Changes in Fast Oscillations at the Left Parietal and Right Temporal Area

At the left parietal area P7, the two maqams Saba and Huzam showed significant increase of beta-high bands (p-value = 0.034, p-value = 0.050, respectively) and gamma bands (p-value = 0.002, p-value = 0.009, respectively) (Supplementary Table 1). Two other maqams, Nahawand and Hijaz, showed a significant increase at P7, but only for gamma bands (p-value = 0.043, p-value = 0.018) (Supplementary Table 1).

At the right temporal area T8, almost all maqams were characterized by a highly significant increase in gamma bands (p-value = 0.000). The only maqam that did not show a significant increase in gamma at T8 was the maqam Kurd. For beta-high bands, the increase was significant at T8 for the maqam Saba (p-value = 0.001), Nawahand (p-value = 0.023), Hijaz (p-value = 0.014), Huzam (p-value = 0.004), and Bayati (p-value = 0.028), but not significant for Kurd, Ajam, and Rast (Supplementary Table 1).

Data suggest that, at the contrary to beta-high bands, the activity of beta-low bands is increased not only in T8 and P7, but also in F7. However, the pairwise comparisons for beta-low were not significant (Supplementary File 4).

Improvisations Done on Different Maqams Induce Different EEG Signatures

Three different improvisations were played for each maqam, at three different tempos and the mean power spectra for each maqam were used to perform pairwise comparisons with the baseline and between all maqams. Table 2 summarizes the significant changes observed in all maqams as compared to the baseline and suggests the existence of maqam EEG signatures. Improvisations on the maqams saba and huzam both showed significant increase of beta high and gamma bands at P7 and T8. However, the maqam huzam differed by a significant increase of theta at F7. In addition, improvisations made on maqam saba produced significantly higher power spectra of gamma bands than all other maqams (p-value < 0.001) (Table 2) not only at locations P7 and T8, but also at locations T7, P8, and O1 (Supplementary Figure 4).

TABLE 2

Table 2. Proposed EEG signatures for improvisations at eight commonly used maqams.

The maqam Hijaz is the only maqam studied that includes an interval of 1.5 tone and also happens to be the only maqam on which improvisations induced a significant increase of theta at the left temporal region T7 (Table 2). Improvisations done on Hijaz showed the highest theta activity at almost all brain locations (Supplementary Figure 4). The maqam hijaz depicts significant changes at similar topographic area to improvisations done on maqam nahawand, with significant increase of theta bands at F7, beta-high at T8, and gamma at P7 and T8.

The improvisations on Kurd showed the lowest power of gamma across all locations and whatever the tempo, did not increase in T8 as in all other maqams (Supplementary Figure 4). The pairwise comparisons confirmed that the gamma activity of improvisations on Kurd in T8 are not significantly different from the baseline and significantly lower than on all other maqams (p-value < 0.001). The only location where significant changes were observed for improvisations on maqam kurd was F7 with an increase in theta bands (Table 2).

Figure 3 depicts heatmaps for all the maqams tested. The heatmaps are organized in such a way to simulate scalp maps and show the asymmetry of brain activation for each band type at the frontal, temporal, parietal, and occipital regions of the brain. Although these do not account for the significance of the changes, these maps propose a hypothetical patterns of brainwave activation per maqam, to be explored further.

FIGURE 3

Figure 3. Heatmaps of power spectra of frequency bands at 14 electrode sites for each maqam. Heatmap were computed for each frequency band type across all maqams. Power spectra are organized in a way to simulate scalp map (frontal on top, occipital on bottom, left, and right hemispheres).

In addition, a principal component analysis was performed on the eight maqams to explore any eventual proximity between the maqams. The Bartlett’s test was significant (p-value < 0.05) and concluded that our data have enough variance to be partitioned using factor analysis. Figure 4 depicts the component plot in rotated space and suggests the presence of three clusters, with the maqam kurd being the closest to zero. Figure 4 suggests the presence of three clusters of improvisations around the maqam kurd: (1) bayati and ajam, (2) saba, rast, and hijaz, and (3) nahawand and huzam. These clusters will be interpreted and discussed in the following section.

FIGURE 4

Figure 4. Principal component analysis plot of the eight maqam commonly used in Middle Eastern Music. Detailed output available in Supplementary File 5.

Discussion

EEG Correlates of Improvisations on the Ney Instrument

Musical improvisation is a complex musical behavior that captured attention of a growing number of scientists. This single subject case aimed at exploring human creativity by exploring improvisations via the prism of Middle Eastern Music and using the Ney flute, which were not explored before. In this single case study, we observed significant changes in the powers of the low frequency bands (theta and alpha) in the left frontal and left temporal areas −F7, F3, and T7. In addition, significant increases in the powers of the higher frequency bands (beta-high and gamma) were observed in the left parietal and right temporal areas, P7 and T8. These results align with previous studies done on western music improvisation and using other instruments. Indeed, Dikaya and Skirtach (2015) showed in a cohort of 136 musicians, amateur, and professionals, that professional musicians were distinguished by a predominant activation of the left hemisphere, with a simultaneous integration between both hemispheres in the higher frequency bands, which is similar to what we have observed in this experiment done on a professional musician. Other studies showed this increase of the EEG spectral power at the prefrontal cortical area, when playing guitar (Sasaki et al., 2019), rock music (Tachibana et al., 2019), and piano (Pinho et al., 2014). The specific increase of frontal theta is facilitated by emotions, concentration, and mental tasks (Aftanas and Golocheikine, 2001; Marcuse et al., 2016; Katahira et al., 2018). While EEG allows an extremely good temporal resolution, it does not provide a good spatial resolution and as a result, studies using EEG do not inform on accurate location of the observed data. Some have used other techniques to explore improvisation cognitive demand such as Limb and Braun (2008), who used fMRI on jazz pianists improvising novel melodies using pre-existing chord patterns (equivalent to our experiment with the maqams) and highlighted an activation in lateral and prefrontal area. In a study of Faber (2014), an exploration involving 32 channels during improvisations in various settings, a strong activity of the frontal and central–temporal regions were also observed suggesting improvised music is a communicative medium. Itthipuripat et al. (2013) showed that frontal theta constitutes a signature of a successful working memory manipulation, as theta bands are also known to facilitate the encoding of temporary episodic memories into long-term memory (Baddeley, 2003; Itthipuripat et al., 2013). Interestingly, at the contrary to all other maqams, improvisations done on the maqam Saba did not induce significant increase of theta bands at the F7 frontal area, suggesting the possible use of different cognitive processes for improvisations done on this maqam.

The ability to improvise is one of the highest levels of musical achievement, as it requires from the improviser to master the music language necessary to spontaneously compose original music. The cognitive processes underlying improvisations have been compared to spontaneous speech (Peretz et al., 2015). Therefore, it is thought to be a powerful mean to express oneself and communicate with others. Nevertheless, these processes are still poorly understood. The data obtained through this single subject study corroborate the existing studies showing an increase in the EEG spectral power at the pre-frontal area (Sasaki et al., 2019), suggesting a strong role for the region F7 or left frontal area of the brain, involved in controlling language-related movement and executive functions such as planning, organizing, and self-monitoring. We have found that most improvisations in this present study resulted in significant increase of beta-high and gamma activity at parietal left P7 and temporal right T8 area. Wan et al. (2014) explored the EEG signals of a pianist on improvisations using Western Music scales have also suggested that frontal, parietal, and temporal regions play a key role in differentiating improvisations from playing composed music. Interestingly, they noted the strong involvement of T8 but not P7 (Wan et al., 2014). While T8 represents the right temporal lobe, close to the amygdala and hippocampus and involved in auditory processing and music appreciation (Stewart et al., 2006), P7 is involved in logical or verbal understanding, word recognition during auditory processing. We emitted the hypothesis that EEG signals on woodwind instrument such as the Ney would be similar to those during spontaneous speech. The strong involvement of F7, T7, and P7 that we noted for Ney improvisations were also spotted by Chengaiyan et al. (2020) in a study of speech imagery (or imagining of speaking), where they observed that left frontal and left temporal electrodes (where T5 correspond to our P7) were activated for speech and speech imagery processes (Chengaiyan et al., 2020).

EEG Signatures of maqams in Middle Eastern Music

Musicology research studies on maqams are extremely limited and almost exclusively run in Turkey and Israel. However, no study has explored yet the neural correlates of these maqams in the same way studies have explored the EEG correlates of major and minor chords (Petsche et al., 1996; Virtala et al., 2013). When having a closer look at the EEG signals elicited by improvisations at each maqam, we observed that each maqam was characterized by a topographically unique combination of significant electroencephalographic changes, suggesting the existence of what we would call maqam EEG signatures, as presented in Table 2. Jenni et al. (2017) explored processing of western tonal music major and minor EEG signals and observed an increased activity of higher frequency bands for the minor scale. Interestingly, in the present study, we observed the same pattern between improvisations on Nahawand (minor) and the those on Ajam (major). But because Middle Eastern Music is also using other tones than the whole tone and semi-tone, a greater number of scales is possible yielding the maqams families studied. It is therefore of great importance that we understand the interactions between the maqams, in order to build future studies on emotional correlates of these maqams. Therefore, we have opted for a principal component analysis to extract eventual clusters that could help us understand which feature in these improvisations seem to play a key role. The principal component analysis suggested the presence of three clusters of improvisations around the maqam Kurd: (1) Bayati and Ajam, (2) Saba, Rast, and Hijaz, and (3) Nahawand and Huzam. The main challenge in studying the maqam system resides in the terminology used, as these names actually refer to the first sets of intervals described in Table 1.

The term jins (or plural ajnas) refers to the building blocks of a maqam scale, which always has a lower and an upper jins. By convention, maqams are classified based on their lower jins, and the first note of the second jins is called the dominant and is the second most important note after the tonic.

By comparing Table 1 and the PCA plot, we understand that the clusters we see could correspond to the sets of intervals they have in common. Indeed, Ajam – characterized by the intervals 1-1-1/2 – is also present in within the unfolding of the Bayati scale 3/4 - 3/4 - 1- 1- 1/2 - 1- 1. The maqam Saba includes some Kurd (1/2 - 1- 1) and some Hijaz (1/2 - 1 1/2 - 1/2), while the maqam Hijaz includes some Rast (1- 3/4 - 3/4), possibly explaining the proximity of these three maqams on the PCA. As for the Kurd set of intervals 1/2 - 1- 1 is retrieved in Kurd, Saba, Ajam, Bayati, Nahawand and could explain why this particular maqam is not fitting in any of the clusters.

These interactions between maqams are well known by professional musicians and structural proximity of these scales are used to create improvisations that increase in complexity by mixing closely related maqams. Dikaya and Skirtach (2015) suggested that high-frequency coherent connections increased with the level of difficulty of the musical improvisation.

This result is important because it highlights the importance of considering intervals, tones, and microtones, in studies on processing of music and emotional correlates.

Implications for Studies on EEG-Based Detection of Emotions

Gu (2014) in her book, “Cultural history of Arabic Language,” mentioned the emotions that are commonly associated to each of the presently studied maqam. Kurd evoking freedom, romance, and gentleness; Saba evoking sadness or pain; Ajam evoking strength; Nahawand evoking drama and emotional extremes; Hijaz evoking desert, solitude, and enchantment; Huzam evoking old days; Bayati evoking femineity, joy, and vitality; and Rast evoking pride and power. Although there is no consensus among musicologists on what the mood each maqam is associated with, it is surprising to see to which extent these various mood-maqam associations have been passed on throughout history without any scientific methodology or validated emotional assessment to support these claims (Kligman, 2001; Powers, 2005; Gu, 2014; Marks, 2019).

EEG signals play an important role in research on human emotions, which in turn are involved in cognitive processes such as memory, learning, and decision-making (Zhang et al., 2020). There are converging evidence from the literature that gamma bands, in addition to being associated with focus and concentration, are associated with negative emotions such as sadness and worry (Oathes et al., 2008; Yang et al., 2020). A recent study using functional network analysis by Yang et al. (2020) exposed native Chinese individuals to 180 pictures selected from the Chinese Affective Picture System (CAPS) (Lu et al., 2005) and recorded their EEG responses as well as their self-assessment Manikin rating scales (Morris, 1995). While no significant difference in brain network were found at low frequency bands, significant differences were observed between positive and negative emotions In the high gamma bands. They concluded on the existence of neural signatures for emotional states in the high gamma bands, particularly against negative stimuli (Yang et al., 2020).

Interestingly, the Saba maqam is consistently called by musicians of the Middle East the “sad maqam” and is also depicting the highest gamma spectral power in this case study. This result represents the first piece of EEG evidence that could eventually support the historical claim that Saba maqam is associated to negative emotions. However, since we did not perform any emotional assessment during the experiment, further studies are needed to conclude on this. In addition, as this results from a single subject study, more explorations on larger and culturally diverse samples need to be performed.

The Hijaz maqam showed the highest theta activity across all brain locations. The theta bands have been shown to increase during sleep, deep meditation, and spiritual awareness. Some have described the Hijaz maqam as being “snake charming music.” Powers (2005, p. 9) writes that “this Maqam is associated with the lonely treks of the camel caravans and with fascination and enchantment.” Lee et al. (2018) have reviewed the neural signals underlying several meditation practices including focused attention, open-monitoring, transcendental attention and loving kindness meditation. It is only during focused attention that a significant increase of theta was observed across both anterior and posterior parts of the brain. While further characterization of the oscillatory activities during improvisation on Hijaz are necessary, the present data suggest similarities between improvisations on Hijaz and focused attention practices.

Music and perception have been substantially researched in the field of music psychology. Through existing neuronal measuring methodologies (EEG, fMRI, and MEG), studies shed light on brain functionality and mechanisms involved when passively listening or playing an instrument (Limb and Braun, 2008; Virtala et al., 2013; Almudena et al., 2021). It is generally agreed that music stimulates a combination of different processes including short-term memory, the nature of different emotions produced by music, concentrations, pleasure and non-pleasure, and self-reflection (Almudena et al., 2021). Studies of Almudena et al. (2021) and Virtala et al. (2013) support that different mechanisms are involved in the perception of major vs. minor, and consonant vs. dissonance chords in infants, adults and school-age children, which correlate with findings in the present study. Furthermore, Limb’s study (Limb and Braun, 2008) tries to break the code of spontaneous music performance, with the assumption that this creative music process is predicative on novel combination of ordinary mental processes. It was hypothesized that short term memory would be associated with hierarchical top-down subtle changes in other systems, such as sensorimotor area and limbic structures used to regulate memory and emotional tone. The study suggests that the prefrontal cortex is of critical importance for processes which include self-reflection, and sensory processes as integral component.

To conclude, this present case explored the power of low and high frequency bands across 14 cortical locations during Middle Eastern Music improvisations played using the Ney instrument using the tonal–spatial system of Maqams. This case provides further support to the already published studies on the important role during musical improvisations of the left hemisphere with the significant increase of the low frequency bands at the frontal and temporal left area, as well as the more integrated activity in both hemispheres at higher frequency bands.

In addition, this case introduces, for the first time in neuromusicology, the question of EEG Maqam signatures, where signatures found seems to follow the maqam’s intervals signatures, supporting the necessity of referring to Maqams by their intervals rather than their names.

Single case studies are being used by many across multiple sessions to obtain consistent results using brain activity measurements (Farrugia et al., 2021).

Finally, this case’s results can be used as ground study to design further studies, including: (1) establishing the cognitive demand for each mode or maqam on professional’s vs. amateur musicians or improvisation vs. composed music, (2) exploring the listener’s and the performer’s perception of intended emotions by using the present recordings of various maqams and combination of self-assessment mannequin and EEG-based emotion detection, and (3) increasing the sample size in order to confirm the proposed correlates and explore the possibilities of neurofeedback training to improve performance on every maqam.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

GB worked on the conceptualization, methodology, analysis, investigation, interpretation and writing, and supervised all the work. MY worked on the methodology, ran the investigation, managed software processing of data, did the analysis, and wrote parts of the manuscript. PS worked on the methodology, managed software processing of data, did the analysis, and wrote parts of the manuscript. SR participated in running the investigation, writing, reviewing, and editing. IK ran the investigation, participated in interpretation, writing, reviewing, and editing. ZP and AC participated in writing, reviewing, and editing. All authors have read and agreed to the published version of the manuscript.

Funding

The publication of this article was funded by the Weill Cornell Medicine – Qatar Distributed eLibrary.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2021.701761/full#supplementary-material

Supplementary File 1 | Mean power spectra of theta, alpha, beta-low, beta-high, and gamma at each of the 14 electrodes: pairwise comparisons between baseline and each maqam.

Supplementary File 2 | Mean power spectra of theta, alpha, beta-low, beta-high, and gamma bands at the 14 electrode sites and for all maqams.

Supplementary File 3 | Detail of the power spectra of theta, alpha, beta-low, beta-high, and gamma bands at the 3 different tempos tested, at the 14 electrode sites and for all maqams.

Supplementary File 4 | Mean power spectra of theta, alpha, beta-low, beta-high, and gamma bands at the 14 electrodes organized by maqam.

Supplementary File 5 | Detailed output of principal component analysis.

Supplementary Audios | Improvisations played are available in Supplementary Files.

References

Aftanas, L. I., and Golocheikine, S. A. (2001). Human anterior and frontal midline theta and lower alpha reflect emotionally positive state and internalized attention: high-resolution EEG investigation of meditation. Neurosci. Lett. 310, 57–60. doi: 10.1016/S0304-3940(01)02094-8

CrossRef Full Text | Google Scholar

Almudena, G., Manuel, S., and Julián Jesús, G. (2021). “EEG analysis during music perception,” in Electroencephalography-From Basic Research to Clinical Applications, ed. H. Nakano (London: IntechOpen), doi: 10.5772/intechopen.94574

CrossRef Full Text | Google Scholar

Arden, R., Chavez, R. S., Grazioplene, R., and Jung, R. E. (2010). Neuroimaging creativity: a psychometric view. Behav. Brain Res. 214, 143–156. doi: 10.1016/j.bbr.2010.05.015

PubMed Abstract | CrossRef Full Text | Google Scholar

Baddeley, A. (2003). Working memory: looking back and looking forward. Nat. Rev. Neurosci. 4, 829–839. doi: 10.1038/nrn1201

PubMed Abstract | CrossRef Full Text | Google Scholar

Bekiroğlu, T., Ovayolu, N., Ergün, Y., and Ekerbiçer, H. Ç (2013). Effect of Turkish classical music on blood pressure: a randomized controlled trial in hypertensive elderly patients. Complement. Ther. Med. 21, 147–154. doi: 10.1016/j.ctim.2013.03.005

PubMed Abstract | CrossRef Full Text | Google Scholar

Carlsson, I., Wendt, P. E., and Risberg, J. (2000). On the neurobiology of creativity. Differences in frontal activity between high and low creative subjects. Neuropsychologia 38, 873–885. doi: 10.1016/S0028-3932(99)00128-1

CrossRef Full Text | Google Scholar

Chengaiyan, S., Retnapandian, A. S., and Anandan, K. (2020). Identification of vowels in consonant–vowel–consonant words from speech imagery based EEG signals. Cogn. Neurodyn. 14, 1–19. doi: 10.1007/S11571-019-09558-5

PubMed Abstract | CrossRef Full Text | Google Scholar

Demarin, V., Bedekoviæ, M. R., Puretiæ, M. B., and Pašiæ, M. B. (2016). Arts, brain and cognition. Psychiatr. Danub. 28, 343–348.

Google Scholar

Dikaya, L. A., and Skirtach, I. A. (2015). Neuropsychology neurophysiological correlates of musical creativity: the example of improvisation. Psychol. Russia 8, 84–97. doi: 10.11621/pir.2015.0307

CrossRef Full Text | Google Scholar

Emotiv-Pro (2013). Emotiv Gitbook. San Francisco, CA: Emotiv-Pro. Available online at: https://emotiv.gitbook.io/emotivpro/event_markers/keystroke_markers (accessed August 3, 2021).

Google Scholar

Faber, S. (2014). The Communicative Processes Of Musicians Engaged in Synchronous Play. Finland: University of Jyväskylä.

Google Scholar

Farrugia, N., Lamouroux, A., Rocher, C., Bouvet, J., and Lioi, G. (2021). Beta and theta oscillations correlate with subjective time during musical improvisation in ecological and controlled settings: a single subject study. Front. Neurosci. 15:626723. doi: 10.3389/FNINS.2021.626723

PubMed Abstract | CrossRef Full Text | Google Scholar

Gu, S. (2014). A Cultural History of the Arabic Language. Jefferson: McFarland & Company.

Google Scholar

Hoppe, K. D. (1998). Hemispheric specialization and creativity. Psychiatr. Clin. North Am. 11, 303–315.

Google Scholar

Itthipuripat, S., Wessel, J. R., and Aron, A. R. (2013). Frontal theta is a signature of successful working memory manipulation. Exp. Brain Res. 224, 255–262. doi: 10.1007/s00221-012-3305-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Jenni, R., Oechslin, M. S., and James, C. E. (2017). Impact of major and minor mode on EEG frequency range activities of music processing as a function of expertise. Neurosci. Lett. 647, 159–164. doi: 10.1016/J.NEULET.2017.03.022

PubMed Abstract | CrossRef Full Text | Google Scholar

Joseph, R. (1988). The right cerebral hemisphere: emotion, music, visual-spatial skills, body-image, dreams, and awareness. J. Clin. Psychol. 44, 630–673. doi: 10.1002/1097-4679(198809)44:5

CrossRef Full Text | Google Scholar

Katahira, K., Yamazaki, Y., Yamaoka, C., Ozaki, H., Nakagawa, S., and Nagata, N. (2018). EEG correlates of the flow state: a combination of increased frontal theta and moderate frontocentral alpha rhythm in the mental arithmetic task. Front. Psychol. 9:300. doi: 10.3389/fpsyg.2018.00300

PubMed Abstract | CrossRef Full Text | Google Scholar

Kligman, M. (1993). Modes of prayer: the canonization of the maqamat in the prayers of the syrian jews in brooklyn, New York on jstor. JSTRO Proc. World Congr. Jew. Stud. 2, 259–266.,Google Scholar

Kligman, M. (2001). The bible, prayer, and maqam: extra-musical associations of syrian jews. Ethnomusicology 45:443. doi: 10.2307/852866

CrossRef Full Text | Google Scholar

Lee, D. J., Kulubya, E., Goldin, P., Goodarzi, A., and Girgis, F. (2018). Review of the neural oscillations underlying meditation. Front. Neurosci. 12:178. doi: 10.3389/fnins.2018.00178

PubMed Abstract | CrossRef Full Text | Google Scholar

Limb, C. J., and Braun, A. R. (2008). Neural substrates of spontaneous musical performance: an fMRI study of jazz improvisation. PLoS One 3:e1679. doi: 10.1371/JOURNAL.PONE.0001679

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, S., Chow, H. M., Xu, Y., Erkkinen, M. G., Swett, K. E., Eagle, M. W., et al. (2012). Neural correlates of lyrical improvisation: an fMRI study of freestyle rap. Sci. Rep. 2:834. doi: 10.1038/srep00834

PubMed Abstract | CrossRef Full Text | Google Scholar

Lopata, J. A., Nowicki, E. A., and Joanisse, M. F. (2017). Creativity as a distinct trainable mental state: an EEG study of musical improvisation. Neuropsychologia 99, 246–258. doi: 10.1016/j.neuropsychologia.2017.03.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Lu, B., Hui, M., and Yu-Xia, H. (2005). The development of native chinese affective picture system–a pretest in 46 college Students. Chin. Ment. Health J. 19, 719–722.

Google Scholar

Marcuse, L. V., Fields, M. C., and Yoo, J. (2016). “The normal adult EEG,” in Rowan’s Primer of EEG, eds L. V. Yoo, M. C. Marcuse, and J. Fields (London: Elsevier), 39–66. doi: 10.1016/b978-0-323-35387-8.00002-0

CrossRef Full Text | Google Scholar

Marks, E. (2019). Arab musical culture in a jewish liturgy. Musica Judaica 22, 66–86.

Google Scholar

Mayseless, N., and Shamay-Tsoory, S. G. (2015). Enhancing verbal creativity: modulating creativity by altering the balance between right and left inferior frontal gyrus with tDCS. Neuroscience 291, 167–176. doi: 10.1016/j.neuroscience.2015.01.061

PubMed Abstract | CrossRef Full Text | Google Scholar

Morris, J. D. (1995). Observations: SAM: the self-assessment manikin an efficient cross-cultural measurement of emotional response 1. J. adv. Res. 35:63.

Google Scholar

Naroditskaya, I. (2009). The philosophy of music by abu nasr muhammad al-farabi (review). Asian Music 40. 133–137. doi: 10.1353/amu.0.0030

PubMed Abstract | CrossRef Full Text | Google Scholar

Ninaus, P. (2011). The Fingering Logic and Performing of Woodwind Instruments. Sacramento, CA: GRIN.

Google Scholar

Oathes, D. J., Ray, W. J., Yamasaki, A. S., Borkovec, T. D., Castonguay, L. G., Newman, M. G., et al. (2008). Worry, generalized anxiety disorder, and emotion: evidence from the EEG gamma band. Biol. Psychol. 79, 165–170. doi: 10.1016/j.biopsycho.2008.04.005

PubMed Abstract | CrossRef Full Text | Google Scholar

Özaslan, T. H., Serra, X., and Arcos, J. L. (2012). “Characterization of embellishments in ney performances of makam music in Turkey”. in Proceedings of the 13th International Society for Music Information Retrieval Conference, ISMIR 2012, November 2014, Porto, 13–18.

Google Scholar

Peretz, I., Vuvan, D., Lagrois, M. É, and Armony, J. L. (2015). Neural overlap in processing music and speech. Philos. Trans. R. Soc. B 370:20140090. doi: 10.1098/rstb.2014.0090

PubMed Abstract | CrossRef Full Text | Google Scholar

Petsche, H., Von Stein, A., and Filz, O. (1996). EEG aspects of mentally playing an instrument. Cogn. Brain Res. 3, 115–123. doi: 10.1016/0926-6410(95)00036-4

CrossRef Full Text | Google Scholar

Pinho, A. L., de Manzano, Ö, Fransson, P., Eriksson, H., and Ullén, F. (2014). Connecting to create: expertise in musical improvisation is associated with increased functional connectivity between premotor and prefrontal areas. J. Neurosci. 34, 6156–6163. doi: 10.1523/JNEUROSCI.4769-13.2014

PubMed Abstract | CrossRef Full Text | Google Scholar

Powers, C. (2005). Arabic Musical Scales: Basic Maqam Notation. Toronto, ON: G. L. Design.

Google Scholar

Razumnikova, O. M., and Yashanina, A. A. (2014). Personality specific differences in EEG reactivity on convergent and divergent thinking. Int. J. Psychophysiol. 94:160. doi: 10.1016/j.ijpsycho.2014.08.702

CrossRef Full Text | Google Scholar

Sasaki, M., Iversen, J., and Callan, D. E. (2019). Music improvisation is characterized by increase EEG spectral power in prefrontal and perceptual motor cortical sources and can be reliably classified from non-improvisatory performance. Front. Hum. Neurosci. 13:435. doi: 10.3389/fnhum.2019.00435

PubMed Abstract | CrossRef Full Text | Google Scholar

Starchenko, M. G., Kireev, M. V., and Medvedev, S. V. (2014). Brain organization in creative thinking. Int. J. Psychophysiol. 94:160. doi: 10.1016/j.ijpsycho.2014.08.703

CrossRef Full Text | Google Scholar

Stewart, L., Von Kriegstein, K., Warren, J. D., Griffiths, T. D., and Griffiths, T. (2006). Music and the brain: disorders of musical listening. Brain 129, 2533–2553. doi: 10.1093/brain/awl171

PubMed Abstract | CrossRef Full Text | Google Scholar

Suhaimi, N. S., Mountstephens, J., and Teo, J. (2020). EEG-based emotion recognition: a state-of-the-art review of current trends and opportunities. Comput. Intell. Neurosci. 2020:8875426. doi: 10.1155/2020/8875426

PubMed Abstract | CrossRef Full Text | Google Scholar

Tachibana, A., Noah, J. A., Ono, Y., Taguchi, D., and Ueda, S. (2019). Prefrontal activation related to spontaneous creativity with rock music improvisation: a functional near-infrared spectroscopy study. Sci. Rep. 9, 1–13. doi: 10.1038/s41598-019-52348-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Teixeira Borges, A. F., Irrmischer, M., Brockmeier, T., Smit, D. J. A., Mansvelder, H. D., and Linkenkaer-Hansen, K. (2019). Scaling behaviour in music and cortical dynamics interplay to mediate music listening pleasure. Sci. Rep. 9:17700 doi: 10.1038/s41598-019-54060-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Torres, E. P., Torres, E. A., Hernández-Álvarez, M., and Yoo, S. G. (2020). EEG-based BCI emotion recognition: a survey. Sensors 20, 1–36. doi: 10.3390/S20185083

PubMed Abstract | CrossRef Full Text | Google Scholar

Touma, H. H. (1971). The maqam phenomenon: an improvisation technique in the music of the middle east. Ethnomusicology 15:38. doi: 10.2307/850386

CrossRef Full Text | Google Scholar

Tumata (n.d.). Turkish Music Tonalities (Makam) and their Effects on Human Beings | Tumata. Available online at: https://tumata.com/en/turkish-music-therapy/turkish-tonalities-and-their-effects (accessed April 18, 2021).

Google Scholar

Virtala, P., Huotilainen, M., Partanen, E., Fellman, V., and Tervaniemi, M. (2013). Newborn infants’ auditory system is sensitive to Western music chord categories. Front. Psychol. 4:492. doi: 10.3389/fpsyg.2013.00492

PubMed Abstract | CrossRef Full Text | Google Scholar

Wan, X., Crüts, B., and Jensen, H. J. (2014). The causal inference of cortical neural networks during music improvisations. PLoS One 9:e112776. doi: 10.1371/JOURNAL.PONE.0112776

PubMed Abstract | CrossRef Full Text | Google Scholar

Wolfe, J., Garnier, M., and Smith, J. (2009). Vocal tract resonances in speech, singing, and playing musical instruments. HFSP J. 3, 6–23. doi: 10.2976/1.2998482

CrossRef Full Text | Google Scholar

Yang, K., Tong, L., Shu, J., Zhuang, N., Yan, B., and Zeng, Y. (2020). High gamma band EEG closely related to emotion: evidence from functional network. Front. Hum. Neurosci. 14:89. doi: 10.3389/fnhum.2020.00089

PubMed Abstract | CrossRef Full Text | Google Scholar

Yöre, S. (2012). Maqam in music as a concept, scale and phenomenon. Zeitschrift Für Die Welt Der Türken 4, 267–286.

Google Scholar

Zarate, J. M. (2013). The neural control of singing. Front. Hum. Neurosci. 7:237. doi: 10.3389/fnhum.2013.00237

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, L., Gan, J. Q., Zhu, Y., Wang, J., and Wang, H. (2020). EEG source-space synchrostate transitions and Markov modeling in the math-gifted brain during a long-chain reasoning task. Hum. Brain Mapp. 41, 3620–3636. doi: 10.1002/hbm.25035

PubMed Abstract | CrossRef Full Text | Google Scholar