A Systematic Review of Scientific Studies on the Effects of Music in People with or at Risk for Autism Spectrum Disorder

The prevalence of autism spectrum disorders (ASD) is globally increasing, and the current available interventions show variable success. Thus, there is a growing interest in additional interventions such as music therapy (MT). Therefore, we aimed to provide a comprehensive and systematic review of music and people with, or at risk of, ASD. We used the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines and used PubMed, PsycINFO, and Web of Science as databases, with “music”, “music therapy”, “autism spectrum disorder”, and “ASD” as search terms. Among the identified and screened articles, 81 out of 621 qualified as scientific studies involving a total of 43,353 participants. These studies investigated the peculiarities of music perception in people with ASD, as well as the effects of music and MT in this patient group. Most of the music-based interventions were beneficial in improving social, emotional, and behavioural problems. However, the availability of studies utilizing a rigorous randomized controlled trial (RCT) design was scarce. Most of the studies had a small sample size, and the applied therapeutic and scientific research methods were heterogeneous.


Autism Spectrum Disorder: Symptoms and Diagnosis
The article, "Autistic Disturbances of Affective Contact", written by Leo Kanner in 1943, presented 11 children who showed an affinity for loneliness and obsessive behaviours but displayed an intact intelligence. This clinical phenomenon is known as autism spectrum disorder (ASD) today [1]. ASD refers to complex neurodevelopment conditions characterised by some degree of impairments in behaviour, communication, and social functioning [2]. The overall incidence has shown that ASD frequently arises during childhood and persists in adolescence and adulthood. It typically becomes visible in the first five years of life [3]. Over the past 50 years, epidemiological studies have confirmed that the prevalence of ASD is increasing globally. The current prevalence of ASD reported by the Center for Disease Control is estimated to be about one in 160 children. ASD is more than four times more prevalent among males than females [4].
According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), ASD is characterised by a persistent deficit in social communication, restrictive and repetitive patterns of behaviours, social and occupational impairments, and other areas of functioning [2].
While the exact cause for ASD remains undetermined, literature on developmental disorders indicate that brain abnormalities could explain the reason for ASD in structure, function, and genetic susceptibility [5]. The literature further suggests that ASD is genetically determined. Additionally, environmental influences might also be considered as risk Music might also be used as a tool to enhance the interpretation and communication of emotions [16]. A study by Katagiri (2009) presented four teaching conditions to determine how people learn emotional concepts: non-purposeful teaching, teaching with verbal intrusions, teaching while background music representing the emotion was played, or teaching while, singing [22]. The findings highlighted that all children showed greater emotional understanding when teaching was accompanied by background music representing the emotion, enhancing empathy and social understanding. Additionally, biological studies have reported that, when music is created or listened to in a social context, some neurohormones, such as oxytocin and neuropeptide, are released by the posterior pituitary gland, which promotes "mind-reading" and empathy in people with ASD [23].
Moreover, it has been shown that music helps to alleviate pain, anxiety, agitation, and depression [24]. Pervasive behaviours, such as developing rigid routines, are typical behaviours of individuals with ASD, and changes in these routines might cause significant stress and aggression. It has been demonstrated that participation in MT reduces anxiety and aggression [25].
Researchers have examined the effects of MT on individuals with ASD with regards to behaviour, psychosocial, intellectual, and interpersonal parameters. However, there is a need for a comprehensive review of the published benefits and potential side effects of the use of music in people with ASD [26][27][28]. This comprehensive overview is the aim of our systematic review.

Search Strategy
This systematic review was conducted following the recommendations outlined in the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) [29].
We performed the systematic literature search using the medical databases PubMed, PsycINFO, and Web of Science. The search of the literature was conducted from inception until February 2022. We used the following key search terms: autism spectrum disorder/autism/ASD in conjunction with music/music therapy. The specific search algorithm in PubMed was: (("autism spectrum disorder"[Title/Abstract]) OR ("autism"[Title/Abstract]) OR ("ASD"[Title/ Abstract])) AND (("music"[Title/Abstract]) OR ("music therapy" [Title/Abstract])). The search terms for PsycINFO via Ovid were ((autism spectrum disorder, or autism, or ASD) and (music, or music therapy)). The search terms for Web of Science were also ((autism spectrum disorder, or autism, or ASD) and (music, or music therapy)).

Inclusion and Exclusion Criteria
Inclusion Criteria:

•
The studies are published in English or have an English-language abstract available; participants of studies who were diagnosed with or had symptoms of ASD; studies must be ASD and music-related.

•
The full text of the article is available.

•
The studies used music in an experimental or observational study design • Studies with measurable results, or outcomes were reported.
Exclusion Criteria: • Studies were excluded if they were systematic reviews or meta analysis'. • Studies that were not ASD related were excluded. • Studies were excluded if there was a non music related intervention • Articles were excluded if they were protocols, hospital reports, evaluation papers, editorials, qualitative studies, syntheses, case reports, personal reviews or essays. • Studies were excluded if they were not published in English.

•
Feasibility studies were excluded. • Studies without a clearly defined control group were excluded.
• Articles were excluded if there was a mixture of diagnosed groups not including those with ASD.

Study Selection and Data Extraction
We screened articles reporting studies of any design that assessed the use of music or MT with people with ASD. All articles were included if the full text was available. Articles were included if they described the methodology and measurable results or outcomes were reported. We only included studies in which music was a part of an experimental or observational study design. Only studies published in English were included. Articles were excluded if music was not applied, and measurable outcomes or effects of music were not reported. Only original publications were included, and reviews, meta-analyses, case reports, protocols, editorials, syntheses, qualitative studies, evolution papers, personal essays, feasibility reports, phenomenological studies, and hospital reports were excluded, and duplicates were removed. Articles that mainly dealt with the mental health of musicians, music students, and music therapists were excluded. Only human studies were included, and animal studies were removed. All articles where only hospitals or therapy programmes were described were excluded. All articles were screened and categorised as "included", "excluded", and "unclear". The titles and abstracts of all identified articles were screened independently by another two reviewers. In total, three reviewers were involved.
The data from all included studies were extracted into an electronic summary table, as displayed in Table 1, by specific methodological characteristics: author, year of publication, sample and group size, total participants, study design, questionnaires and research methods, main outcomes, and the significance of main outcomes. The data extraction was based on the study results investigating the effect of music or MT on ASD. All articles were then thematically presented, and the findings were accordingly reported in a matrix. All the extracted data described were critically discussed in the results section.

Data Analysis
After the extraction of study details, the articles were thematically arranged based on the study design and types of intervention by both B.A, Z.C., and H.H. The findings were then reported accordingly.

Ethical Considerations
As our systematic review did not require the recruitment of probands or patients and did not use data or specimens of individuals, ethical approval is not required. However, we checked the ethical aspects and approval of all included studies. Mozart's simple minuet, semi-structured parental interview, harmonic, and inharmonic version of a head-turn preference procedure, four musical pieces of Mozart sonatas, ADI-R, spatial tasks, AP Both groups preferred the original version of the piece over the inharmonic version. Children with ASD tended to show preference for aesthetic quality of the high dissonant music compared to TD children. AP task: Some children with ASD showed extraordinary musical memory. Children with ASD activated bilateral temporal brain networks during sung-word perception; ASD: spoken-word perception right-lateralised and reduced IFG activity; diffusion tensor imaging: reduced integrity of the left hemispheric frontotemporal tract in the ASD group.
∆ decreased IGF activity during sung words: p = 0.042 Children with ASD (n = 25) TD children (n = 25) 50 Experimental design Music Game 1, brief IQ measure of the Lieter-R paired single-tone pitch discrimination task, melodic memory encoding task, Music Game 2 Improved pitch discrimination ability in the single-tone and melodic context, as well as superior memory for melody and a positive correlation between pitch memory and performance on non-verbal build reasoning ability was found for children with ASD. Adolescents with ASD (n = 33) TD adolescents (n = 26) 59 Experimental design WASI, Digit Span and Letter-Number Sequencing subtests ofthe WISC-IV, SAMMI, SCQ, SRS Adolescents with ASD less likely to make social attributions, especially for those animations with the most complex social interactions. When stimuli were accompanied by music, both groups were equally impaired in appropriateness and intentionality.
∆ longer description time for animation with complex social interactions in ASD: p < 0.01; ∆ appropriateness: p = 0.14; ∆ intentionality: none reported. Children with ASD assessed most music fragments similarly to their TD peers, with likelihood of EEG oscillatory patterns closely corresponding to emotion self-reports. In S2 fragments, "sad" was a reported emotion in TD children and adult neurotypical raters, but "angry and frightening" were emotions elicited by children with ASD. In S2 fragments, EEG oscillatory response showed greater cortical activation in the right hemisphere.
Emotional response to music self-report data: p < 0.00001 Children with ASD had higher reaction times than controls; accuracy differed when incongruent or no music was played.
Significantly negative correlation between AQ-10 and emotion recognition accuracy. 13 Weiss BVAQ-B, AQ, 12 music items Adults with alexithymia and ASD did not differ based on physiological responsiveness; ASD group significantly lower on the verbal measure; significant effect of mood with music in response between two groups but the maximum difference was relieved for scary music.  The initial outcomes of brain responses and behaviours showed a positive effect of MT on selective attention skills. In both conditions, participants improved their pre to post-test verbal production. No significant difference between low and high functioning in improving speech production after both training.
∆ improvement in pre to post-test production: p < 0.001; ∆ n.s difference between low and high functioning in speech production: p = 0.053 23 Lim & Draper (2011), USA [86] Children with ASD (n = 22); 22 Cross-sectional design VPES Both music and speech training were effective for production of the four ABA verbal operants; difference between music and speech training not significant. Music incorporated ABA VB training.   Significant improvement in NGF level in participants of music condition, but not significantly different to the placebo group. The level of NGF in participants in vitamin D condition showed a significant increase compared to participants in music condition.
The level of NGF in music and vitamin D condition was significantly higher compared to only music, only vitamin D and placebo condition.
∆ improvement in NGF level in music condition: p = 0.001; ∆ no difference between music condition and placebo group: p = 0.07; ∆ participants' higher NGF level in vitamin D condition compared to music condition: p = 0.001; ∆ NGF level in both music and vitamin D condition compared to only music only vitamin D or placebo group: p = 0.001. Adults with ASD (n = 20) 20 Experimental design (2 studies were conducted; MT was used in the second study) sAA Occupational therapy leaded a significant increase in sAA levels while MT significantly deceased baseline sAA levels indicating that the ability of receding stress in both interventions and by proxy contribute to improve overall well-being.
∆ sAA levels in occupational therapy: p < 0.05; ∆ sAA levels in MT: p < 0.05 There was no significant difference between the sung and spoken conditions. There was a significant increase in receptive labelling skills after both conditions were enacted and these results were maintained at follow-up.
A difference in group performance was found.
n. s. The presence of rhythmic cueing and tempo adjustment correlated with social skills, providing a strong rationale for the use of dyadic drum playing to address social skills. Children shows a reduction in asynchrony when tapping with a partner at adjusted tempi after the rhythm-mediated intervention and showed a greater engagement in joint action following the intervention.
None reported.

Included Studies
Following three searches using PubMed, PsycINFO, and Web of Science databases, 887 candidate papers were identified, and two studies were added through hand-searching and reference-chaining. After the removal of duplicates, a total of 621 articles were assessed for inclusion. A total of 81 studies met the full eligibility criteria and were chosen for analysis. These studies contained data of n = 43,353 patients or study participants. Figure 1 shows a PRISMA diagram describing the results of the search strategy and reasons for exclusion. Table 1 summarises all the publications that met the eligibility criteria and were included in this systematic review.

Methodological aspects
The design of the studies, their main results, and their statistical significance are presented in Table 1. Of the 81 identified studies, 3 were surveys, 45 were experimental studies, and 25 studies were longitudinal studies or randomised controlled studies (RCTs) investigating the effect of music and MT. Most of the studies were relatively small with ~50 participants. However, we identified four meaningful studies with more than 100 study participants each.
The largest survey was performed by Ruan et al. (2018) including 34,749 parents of children around the age of three years [94]. Researchers found that antenatal music training and maternal talk to the foetus were associated with a reduction in autistic-like behaviours in children, with a dose-dependent relationship [94].
A large and well-designed experimental study by Goris et al. (2020) with 161 participants from university students investigated music preference in relation to autistic

Methodological aspects
The design of the studies, their main results, and their statistical significance are presented in Table 1. Of the 81 identified studies, 3 were surveys, 45 were experimental studies, and 25 studies were longitudinal studies or randomised controlled studies (RCTs) investigating the effect of music and MT. Most of the studies were relatively small with 50 participants. However, we identified four meaningful studies with more than 100 study participants each.
The largest survey was performed by Ruan et al. (2018) including 34,749 parents of children around the age of three years [94]. Researchers found that antenatal music training and maternal talk to the foetus were associated with a reduction in autistic-like behaviours in children, with a dose-dependent relationship [94].
A large and well-designed experimental study by Goris et al. (2020) with 161 participants from university students investigated music preference in relation to autistic traits by presenting tone sequences that varied in predictability [37]. They found a positive correlation between autistic traits and a preference for predictability in music [37].
The RCT with the most advanced study design was the TIME-A randomized clinical trial by Bieleninik et al. (2017), which was conducted in nine countries and enrolled N = 364 children, aged four to seven years, with ASD [70]. Patients were either allocated to enhanced standard care (n = 182) or to enhanced standard care plus improvisational music therapy (IMT) (n = 182). In IMT, trained music therapists sang or played music with each child, attuned and adapted to the child's focus of attention, to help children develop affect sharing and joint attention. The primary outcome measure was symptom severity over 5 months, based on the Autism Diagnostic Observation Schedule (ADOS). However, the mean ADOS social affect score changes did not differ significantly between groups [70].

Content of Included Studies
In the following paragraphs, we provide a comprehensive summary of the content of the studies that fulfilled the inclusion criteria. Table 1 contains further details about each study. Content-wise, the studies fell into four categories: studies on the specific perception of music and music preferences in people with ASD; studies on the effect of music in people with ASD; studies investigating the effect of MT and musical training in people with ASD and their caregivers; studies reporting combined creative arts therapies, including MT, in patients with ASD.

Music Perception in ASD
People with ASD have been found to use music for various purposes such as to address their cognitive, emotional, and social needs including mood management, personal development, and social inclusion [30]. They were reported to have a strong preference for music over verbal material [31], superior pitch memory, recognition of changes of the pitch [34,42,45,48], and to enjoy dissonant music more than controls with typical development (TD) [53]. It's been shown that people with ASD were physiologically more responsive to their preferred music than those in the comparison group [42]. In a peerassisted learning study by Johnson and LaGasse (2021), music appeared to increase prosocial skills in ASD children that were paired with neurotypical children [45]. People with ASD appeared to prefer more predictable [37] and upbeat music [31]. The latter may be related to increased activity in dorsolateral prefrontal regions in response to happy music [36].
In contrast to their ability to understand musical cues, as well as the affective and emotional content of music, comparable to non-autistic people [40,43,50], and unlike their similar magnetic resonance imaging (MRI)-measured activation pattern in cortical and subcortical brain regions after hearing music [33], impairments in prosodic language processing were found [35,41,47,51], suggesting alternate mechanisms of speech and music processing in ASD. MRI studies by Sharda et al. (2015), Lai et al. (2012), and Hesling (2010) indicate that the difficulties in prosodic language processing may be due to reduced integrity of the language-processing brain networks, such as the frontotemporal tract, the inferior frontal, and the supramarginal gyrus [35,41,47,51]. However, people with ASD seem to activate their bilateral temporal brain networks during sung-word perception similar to healthy people [51].
Interestingly, compared to TD children, children with ASD do not have a strong preference for a human therapist or human voice over interacting with a robot while dancing or hearing an artificial voice [31,46]. Whipple et al. (2015) compared recognition of symbolic representations of emotions or movements in music between children with severe to profound hearing loss or ASD and TD children, but no significant difference between the ASD, the TD, or the hearing loss group were found [53].

Effect of Music on People with ASD
Studies on the effect of music in people with ASD found that music increases their emotional recognition and comprehension [22,57,64], including increased sign and spoken word imitation [58], help with process socially significant auditory signals [65], and focusing attention [63]. Music was reported to increase motivation for physical exercise [66]. Additionally, adults with ASD reported poor auditory imagery in comparison to health controls [54]. Heaton (2003) found that children with ASD had enhanced pitch memory and tone labelling [59]. Stephenson et al. (2016) tested the response of children and adolescents with ASD to music-evoked emotions using skin conductance and found that participants with ASD showed a reduction in the skin conductance response [62]. However, there was a significant interaction effect with the age group, and the authors interpreted their findings with caution. Lundqvist et al. (2009) found that vibroacoustic music reduced challenging behaviours, in children with ASD, as well as reduced the frequency of self-injurious behaviours [60]. A study conducted by Portnova et al. (2018) found that children with ASD perceive most pieces of music similarly to neurotypical children except in one musical fragment, where children with ASD found the musical fragment to be "angry and frightening", whereas neurotypical children perceived the fragment as "sad" [61]. Boorom et al. (2020) examined preschoolers with ASD and their parent's responsiveness comparing musical and non-musical engagement in 12 parent-child dyads [56]. However, they did not find a significant difference between parental responsiveness in the playing with musical toys condition compared to the non-musical condition [56]. Bhatara et al. (2009) investigated the impact of musical animation videos on social attribution in adolescents with ASD or TD [66]. In their study, adolescents with ASD perceived and integrated music soundtracks with visual displays similarly to TD individuals [55].

Effects of MT and Musical Training on People with ASD
A large survey study showed that regular antenatal exposure to music and talking to the baby might prevent traits of ASD [94]. In small studies, MT was shown to be beneficial regarding the bond between ASD children and their parents [84], movement coordination [73], social communication, interaction and attention [21,26,28,70,71,75,82,90,92,93,96,98], and their overall ASD symptoms [71,75,92,93]. Listening to musical excerpts increased vocal recognition and ability to not hide emotions [91].
However, these positive effects could not be replicated in a large RCT, with 362 participants testing the specific effects of IMT on ASD [70]. The findings of Bieleninik et al.'s study were reported by Crawford et al. (2017), and they were later used by Mössler et al. (2020) to examine the relationship between attunement and changes in the above-mentioned outcomes. However, such a relationship could not be demonstrated.
A study by Simpson et al. (2013) found that children with ASD were more engaged in the sung intervention compared to the spoken condition [20]. Additionally, the use of infant-directed singing was more engaging for children with ASD over infant-directed speech. However, in a follow-up study, Simpson et al. (2015) found there was no significant difference between the sung and spoken conditions [96], although there was a significant increase in receptive labelling skills after both the sung and spoken conditions were enacted [96].
The level of functioning in individuals with ASD [85] and the length of the music intervention might play a significant role in its effectiveness [87].
Active music-making in music therapy sessions seemed to produce significant improvements in social skills in children with ASD [69]. MT also seems to have an impact on biological parameters of neuroplasticity and stress, such as levels of nerve growth factor (NGF) and salivary α-amylase (sAA) [88,91]. Musical training has been found to have a beneficial impact on sound sensitivity [68], sensory gating, and attention [83].
IMT was shown to produce improvements in communicative behaviours [76], as well as their self-regulation, engagement, behavioural organization, and two-way purposeful communication in children with ASD [72]. IMT has also been shown to increase emotional attunement and behavioural synchrony [99]. However, in a study using Relational Music Therapy (RMT), the effects on communication skills were inconclusive [77].
A study by Kern and Aldridge (2006) found that, when music was utilized on the playground, there was an improvement in peer interaction in children with ASD; however, there was no increase in social interactions [81].
A recent study by Jin et al. (2020) examined the efficacy of acupoint therapy on joint attention and social communication [79]. MT was part of both the control and the observation group. Researchers found that both interventions led to an improvement of autistic symptoms in several domains [79]. However, the specific effect of the MT element could not be determined, as MT was included in both treatment conditions [79]. Mössler et al. (2019) investigated the effect of therapeutic relationships, during MT, on changes in social skills at 5 and 12 months [89]. They detected that the therapeutic relationship might be an essential predictor of the development of social skills, language ability, and social communication [89]. However, Mössler et al. (2019) did focus on the effect of MT on social communication. Lim and Draper (2011) compared music and speech training in children with ASD and found that both were effective but not statistically significantly different [86]. Kalas (2012) ascertained that joint attention responses of children with ASD depended on the complexity of the music and their level of functioning [80].

Effects of combined MT and Dance Movement Therapy
A few studies utilized the combination of MT and dance movement therapy (DMT) with ASD populations. A study by Bergmann et al. (2021) utilized the Autism-Competence-Group (AutCom), which includes a combination of psychoeducation with music and dance movement interventions in adults with ASD [101]. The combined intervention produced significant improvement in social competence compared to the control group and emotional competence in the pre-post self-assessment on the AutCom questionnaire [101].
One study found a reduction in compulsive and stereotyped behaviours in a combined music and dance intervention [102]. Another found that a combination of MT and DMT reduced ASD scores in both the experimental and control group [103].

Overview of the Main Results of This Systematic Review and Their Significance
The main topics covered by the studies that fulfilled the inclusion criteria of this systematic review were the specific perception of music and music preferences in people with ASD, the effect of music in people with ASD, and the specific effect of MT and musical training in people with ASD and their caregivers. Studies revealed that people with ASD use music for various purposes [30].
In the first category of identified studies, which investigated music perception in ASD, we found that people with ASD have a superior ability of pitch memory and recognition compared to normal controls. Individuals with ASD were also reported to have a strong preference for music over verbal material. In contrast to their deficits in interpreting the prosody of spoken word, people with ASD have no problems in recognizing and understanding the emotional, communicative, and social aspects of music [36,40,43,50]. Additionally, they seem to have a superior pitch memory and recognition of pitch changes [34,38].
In the second category of articles which investigated the effects of music, studies reported that listening to music can increase emotional recognition and understanding [22,57,64], focus attention [63], and motivate them to do physical exercise [66].
In the third category of articles, which focused on the effects of MT and musical training on people with ASD, MT was shown to be beneficial for the child-parent relationship [84], movement coordination [73], social communication, interaction, attention [15,21,28,74,78,82,87,97,100], and overall ASD symptoms [71,92,93]. However, these effects of MT could not be replicated in an RCT with 362 participants [70]. Studies focusing on biological outcomes indicated that MT has a beneficial effect on biological parameters of neuroplasticity and stress [88,91]. In contrast to language-processing brain circuits, brain areas, connectivity, and function related to the perception and understanding of music seem to be intact in people with ASD [41,47,51].
Only 10 out of the included 81 studies focused on adults and the effects of music, MT, and creative arts therapies on ASD [40,41,43,47,48,62,66,70,80,88]. However, similar results were observed in comparison to adolescent ASD studies. For example, adults with ASD had a reduction in challenging behaviours associated with ASD [62] and significant improvements in social competence compared to control groups [80]. Thus, there is clearly more research available on music and ASD in children and adolescents than in adults. Future research should, therefore, include more samples of adults or studies involving both children and adults with ASD. This way, comparisons could be made about the effects of music and MT in different age groups.
Studies focusing on a mixture of both MT and DMT constitute the fourth category of identified articles. These studies showed that a combination of MT and DMT produced an increase in social competence, as well as a reduction in compulsive and stereotyped behaviours and an overall reduction in ASD scores. This underscores the potential of not just music therapy but other creative arts therapies being possibly effective therapies for individuals with ASD. However, further exploration should be conducted into the use of other creative arts therapies in those with ASD and their potential combined effects.

Limitations
Although the current review yielded promising results, these should be interpreted with caution and considered with several limitations.
Data from RCTs were limited; 17 RCTs were included, and the sample size of most RCTs was small. Therefore, further accumulation of RCT data is necessary to draw firm conclusions. The majority of all included studies had a small sample size, resulting in greater clinical heterogeneity in patients compared to studies with large sample sizes affecting the outcome of the experimental treatment [104].
The majority of studies included child-dominant samples; few studies investigated adults or adolescents with ASD, and no studies examined the elderly population. Therefore, further studies might target different populations of varied different age groups.
Furthermore, there is a lack of longitudinal studies; only 5 out of 81 articles were longitudinal studies, indicating that only short-term effects of music-based interventions have been investigated. Therefore, long-term effects should be investigated.
There was great heterogeneity of study design, outcomes, and MT approaches. Due to different approaches used, there are no confirmatory studies available for any of the reported positive findings. Therefore, the reproducibility of the results has not been shown yet. Additionally, because of the heterogeneous study design, we didn't conduct a quality assessment which could have led to the inclusion of low-quality studies assessed in the review.
MT was used as an adjunct therapy in all the studies, and this makes it difficult to ascertain the pure effect of the MT. Furthermore, it is difficult to recommend a specific approach of MT because different approaches, as individual vs. group-based MT or active music making vs. passive MT, were conducted. Therefore, we cannot conclude that a specific approach is most effective. Moreover, studies investigated many different and unrelated symptoms, including social skills, emotional understanding, auditory processing, family well-being, and so forth, making the interpretations of the findings difficult.
Another important aspect of utilizing a music intervention with ASD populations is the music preferences of the study subjects. In one study, ASD participants cited music to be "angry and frightening", whereas neurotypical children cited the same piece of music as "sad" [61]. This highlights the fact that children with ASD perceive music differently, so in studies using a music intervention, it is important to collect the opinions of music preference and individual music sensitivities within ASD participants before beginning the intervention.
Considering the above findings in the context of the wider literature, further investigation should be conducted in different samples, including adults with ASD and elderly individuals with ASD, utilizing larger sample sizes. Additionally, more RCTs should be conducted in order to see the pure effect of MT on ASD-related symptoms. Moreover, ASD symptoms might be investigated as a comprehensive assessment in order to see the impact of music or MT on the more specific symptoms of ASD.
However, as people with ASD use music in many domains and seem to have specific talents regarding analysing and understanding music, this clinical field seems promising. Some studies explored how patients with ASD related positively to an artificial voice or a robot during a musical activity [31,46]. Their unique technical approachability might be kept in mind when designing individually tailored MT for people with ASD in the future.

Conclusions
The current systematic review demonstrates that the perception of music in people with ASD differs from the music perception in neurotypical controls. Additionally, it shows that music and MT can be used as a therapeutic tool in the management of ASD.
People with ASD seem to have a superior ability of pitch memory and recognition compared to normal controls.
When MT was utilized as an adjunct for managing ASD or ASD symptoms, recognition and understanding of the emotional, communicative, and social aspects of music was elevated, attention was heightened, and motivation for physical exercising was increased. Moreover, music and MT might prevent ASD traits during pre-birth, support the child-parent relationship, and improve movement coordination, social communication, interaction and attention, and overall ASD symptoms in people with, or at risk for, ASD. Music and MT had a beneficial effect on biological parameters. Similar positive results were reported by studies that tested MT in combination with DMT. However, these findings do need to be interpreted regarding of the methodological issues and the potential for publication bias (Thornton and Lee 2000).
These results should be interpreted with caution since findings could not be replicated in an RCT [70]. Different samples should be investigated in order to see whether music and MT might be a mediator for the relationship between age and ASD symptoms. Moreover, more RCTs should be conducted using MT as a comprehensive assessment in order to understand the impact of music and MT on the more specific symptoms of ASD.