MMP7C002R~001 HAS20078373 Critical Perspectives on Music Production

Abstract

Exploring the ever evolving role of Artificial Intelligence inside of my personal music production practice and more widely how its effects play out within the music industry as a whole. Its purpose is to provide a framework for my own artistic development, allowing me to articulate and think more deeply about the role that AI plays within my work. This has been through a mixture of primary and secondary research that I have organised into sub sections within this thesis, resulting in allowing myself a deeper understanding of how Artificial Intelligence interplays alongside human expression and performance.

The results of this investigation have helped inform me about my personal bias towards what I consider ethical use of AI and has served to help me to challenge some of those beliefs to see a different perspective where certain types of generative AI music can actually provide a way in for some people to begin making music. To conclude, my findings have been interesting to me and have helped me to navigate any strong feelings I’ve had either for or against the use of AI within music, especially as I now have a deeper understanding of the different types of AI, such as machine learning, algorithmic and generative AI, for which before I didn’t have distinctions for.

Where it all began

Artificial Intelligence in music has been around in terms of computer based algorithms for longer than I first realised when I started to research this area of music production history, with one of the most notable being the first instance of an actual computer, known as the ILLIAC I (Innovation 2025), which composed an entire piece of music on its own back in 1957.

The composition created entirely by the computer was known as the “Illiac suite,” which was later retitled to “String Quartet No. 4” (Illiac Suite 2025). The piece consisted of parts programmed by ILLIAC I which determined the dynamics, patterns and rhythms of the piece of music without any human intervention at all.

Fast forward nearly seventy years, and instead of the ILLIAC I, we have a much more refined kind of AI now available to us in the shape of software and platforms such as Suno, Audimee, Ultimate Vocal Remover and Sonible’s Smart EQ to name but a few that I personally have used or are products I use regularly as a producer. These AI tools are algorithmic, machine learning, utility tasking and generative in nature, each having their strengths and weaknesses and all have relevant placement within my own music practice and workflow, which I will get into in more detail.

Over the years as an artist, I have used many different types of computer software and hardware within my music production and in recent years I have been amazed at the growth of different brands of VSTS, hardware emulators, plugins and platforms that all have unique abilities to take over the workload for the human producer and keep us more productive and arguably therefore more creative.

Digital Audio Workstations (DAWS) such as the birth of Ableton that give the ability to make your own devices in Max for Live (2009), or similarly the use of Patcher in FL Studio to create personalised effects chains. Additionally, the age of generative AI, machine learning and algorithmic based plug-ins has continued to make music production more and more accessible and aids the user to be more creative by presenting us with options that we could never conceive of ourselves, in which to add and accompany our own initial ideas to.

One thing has always been very important to me when it came to making creative or workflow decisions; whatever I was using had to be feeding my inspiration. It has to be adding something to what was already existing, from my own ideas and my own artistic expression. It would be a disaster to me personally if I handed over that part of being creative completely to a computer.

AI Vocals

Although a quite divisive topic, especially for singers themselves, I have on occasion used AI vocals for my productions. I’ve achieved some good results and although wherever possible, I will try to outsource vocals from an actual human being, I can see the benefit of using AI vocals as place holders or for ideas on top lines, depending on the project I am creating. I recently got asked to do a remix for Big Speaker Records which within the stems had an AI vocal of a man speaking. The track was called “I am a Twitch Streamer”. I wanted my remix to incorporate this vocal, but as I am a female Twitch streamer, I thought it would be more authentic to morph the male vocal to a female vocal. I used Audimee to convert the vocals (see Appendix MMP7C002R~001_HAS20078373_AI_Vocal).

This shows the male and female versions after exporting from Audimee. The great thing about doing this with the AI, was the way that the AI made an exact copy of the cadence and expression in the voice in the exact same way that the original male vocal did. This made layering it, and morphing from the male version into the female version within the project very easy to do.

I do still think that you can absolutely tell that the vocal is AI, and to a degree it did feel like it distracted from the rest of the remix, however with it being a remix and not an original production of mine, I felt like I had done what I could to make this remix more authentic to my usual stylistic choices. So far, I’ve not (to my knowledge) used an AI vocal in any of my tracks. I think that when I have a “real” voice I can connect more to the project and feel like I am more invested in it. To this day, I still employ the use of a real singer and get them to sing the top line if I ever need one for a project. I feel like this is my small way of ensuring that human singers are reminded that they are not replaceable by AI. Human singers have natural flaws to their voices, those flaws are what make them human. Sometimes it’s the breaking of the voice in an emotional rendition of a song that truly makes it beautiful. AI cannot replicate that kind of nuance again, because it doesn’t feel the expression of those words in the way that a singer can emulate that.

AI Stem Separation

Another major advancement in recent years has been the birth of stem separation tools. FL Studio now has its own built-in stem separation software, as do many other DAWS. The AI itself is trained on contemporary music (again this limitation makes it better for 4/4 dance music over other more complex genres or non-western music), and it analyses the spectral frequencies and rhythmic qualities in order to decide how to separate the stems. (Swingle, 2023)

As an FL Studio user, I have used this feature within the DAW, however I usually seem to get better results from using Ultimate Vocal Remover. UVR is a standalone stem separation application, which has many more options than those found on FL Studio. There are different models which bring up different results for separation of percussive elements, bass or vocals. I like to try maybe two or three different ones and then compare the results. UVR is completely free as it is open source. I believe this is a reason why it is so good in comparison to other paid for stem separation services, as the code is always being tweaked and updated.

Artificial Intelligence Replacing Human Artists?

An area of great controversy at the moment is the notion that AI will take over all creativity from the artist, and replace the artist entirely. A thought that I, myself had not too long ago. However, in virtue of doing this research and delving into the world of AI a little deeper I have come to realise that AI is here to help us to be more productive with our creativity, and will not inherently take over from us.

On the subject of emotive music in his book, Artificial Intelligence and Music Eco System, Martin Clancy expresses that “it can recollect, can determine immediate connections and is as mental as it is physical. It’s a process to communicate even if that communication is different for each individual. There are some pieces of music that will regenerate memories and others that can change a listener’s life forever. Its effects on our minds and bodies can be astounding and sometimes equally devastating, but no matter what the results, music changes everyone and does so by being itself, a devastating and timeless art, that living without it, at least for many, is like living without life itself”. (AI Music, 2022)

This made me think about my experiences with using Suno as an ideas starter, although it can be used to generate an entire track, I wondered why that had never been appealing to me. The answer I found myself going back to was because I wouldn’t feel anything in making it. I wasn’t involved enough in the process of using my imagination, hearing the melody in my mind that wasn’t yet recorded down, yet somehow lived somewhere between physical reality and consciousness. I find myself waking up in the mornings and having those melodies looping around my mind in those early moments between dreams and the physical realm. Those are the ideas that want to be brought forward to be created.

“We are all translators for messages the universe is broadcasting. The best artists tend to be the ones with the most sensitive antennae to draw in the energy resonating at a particular moment” (Rubin, 2023)

There is something very other-worldly about this experience as a human to be involved in the creative act, yet nothing is more human than to use our minds to create. Everything we see around us once started as a thought in the human mind. To create is to feel and to express emotion through our music and then for others to hear that music and find their own interpretation of said music for themselves. Artificial Intelligence would only get so far in terms of creating one hundred percent generative music, however there will always be that one missing aspect of human made music that translates to other humans, making us feel something.

If I used generative AI for my music, and couldn’t feel anything from making it with prompts, then perhaps also my audience would feel anything from it either. It’s just regurgitated ideas that have already come to pass, which don’t ultimately represent me, my sound, my authenticity or me as a human being. If you want to touch someone’s heart, and most importantly your own, this can only be achieved by harnessing and expressing that energy into your music. Even if that message doesn’t translate fully to the listener, it doesn’t even matter, because we, the artist, already got what we needed in virtue of making that piece of art; we got the ideas out of ourselves and externalised them in the physical world.

Personal Expression

My experiences with using Suno have been interesting when considering the notion of personal expression. I have used Suno to create a full piece of music and also as a means to help me out of a creative rut and get an idea going in the studio, which I would always then recreate myself within my DAW.

When using Suno to create an entire song, it was always something for “fun” or for a quick turnaround for use on my social media posts as content, and always in a funny, comedic way which feels like the only way I really can use generative AI in this context (see Appendix MMP7C002R~001_HAS20078373_Name That Riff).

This video is one of a series of short form video content I made for my social media accounts where I played the riff to a well known Dance music tune on a recorder (very badly) and then my followers would comment what song they thought it was. You can absolutely tell that the intro song was AI generated, which unironically fits the aesthetic and throw-away, disposable nature of this kind of music for short form content creation. This is an example of when generative AI music can have a use and a distinct role. Another example would be for singers or instrument players who are in need of a backing track to perform to, in order to show what they are capable of for possible job opportunities or to see how ideas work without needing to employ other musicians or producers at their expense or if they are in a remote part of the world, they won’t need to wait to find someone to collaborate with. Generative AI can have a positive effect in diminishing gatekeeping in that aspect.

Ultimately, the artist’s act of making music is quite sacred and something that helps us to express our emotions and trauma to process them in a positive way, in quite the same way that alchemy changes base metals into gold. (Chrysopoeia 2025)

It helps us to transcend this world and our life situation and allows us to enter this alternative reality, where all things can be possible. If Artificial Intelligence could do this, it would indeed be sentient. I believe we are still a long way from that being a reality. As it stands for now, generative AI can only regurgitate ideas that have already come to pass, and only with ideas based on a very narrow window of music types that it has been exposed to, which is mostly Western music which limits it even further.

This is a limitation of the AI and the ways it listens to music in order to interpret it. Indeed, “music notation can be difficult if not impossible to interpret” (Clancy, 2022) and AI can misinterpret it quite wildly. An example of this is when Clancy talks about when AI was asked to interpret certain movements in classical music, namely the conclusion of Mahler’s Fifth Symphony, the AI was unable to interpret the notation correctly, with the AI calling the piece “lazy and dissonant”. (Clancy, 2022)

This demonstrates a lack of nuance that the AI is unable to detect or value in more complex music; a nuance which is very much connected to human evolution, cultural influences and our life’s experiences which are interwoven into music on a much deeper level than the AI is currently able to both determine or express in of itself. How could it value something that it has never experienced itself? For example, Blues, Gospel and later on Hip Hop were all derived from slave songs and hymns of the early 1700’s in America as a way to express the intense hardship they faced. (Clytus, 2016)

Even without having to have experienced those things myself, I cannot help but feel moved by songs such as “Swing Low, Sweet Chariot” especially once understanding the underlying meaning for those songs. AI would never be able to replicate that kind of lived experience.

Live Performance in the age of AI

“Very little performance escapes some form of influence from electronic technologies. Auslander argues that even our interpretation of supposedly live performance is highly influenced by the extent to which cultural practice is now completely embedded in some form of electronic mediation” (Sanden, 2019)

Many of my peers and artists I work with have created and performed “live” versions of their music that has been previously created and released on physical media such as vinyl, or online on digital download sites, however fans will always buy tickets to see those same tracks performed in a live, club environment.

Sanden continues in his book to say “live recordings carry meaning as a type of live event because, despite the fact that they present highly mediated musical experiences, their apparent fidelity to an actual live performance carries meaning for many listeners that is absent from a studio recording… the concept of liveness continues to carry great meaning for many musickers, even in a cultural environment of extreme digital saturation.” (Sanden, 2019)

This is exactly my experience as a music lover and also as an artist and DJ myself. Going to a live event to see one of my favourite artists perform their own music live or indeed DJ their already recorded music in a live setting at a festival or a nightclub is something that is deeply embedded into us as human beings. That musical connection is more than just to do with the music, it’s the community aspect that physically going to an event also carries with it. For twenty five years I have DJ’ed playing my own music and that of my peers all over the world, and I have made a fiercely loyal fan base from these interactions and live performances.

Further to the live events aspect of being an artist and a DJ, there is also the “live”streaming aspect of what I personally do as an artist; four times a week I “go live” on an internet platform called Twitch and I perform live to hundreds of viewers. Usually, I will be performing live to my audience as a DJ, using a mixture of vinyl, digital and also a hybrid set up consisting of a Behringer TD-3 and a Virus Ti-snow. Both of these are synthesisers. I use these synths to add melodic and rhythmic patterns to the existing recorded music that I am playing on the CDJ’s, which adds another live element to the already finished recordings.

This transforms the performance for myself and my viewers. It shows an extra level of skill and showmanship that has taken me a quarter of a century to perfect to the best of my abilities (see Appendix MMP7C002R~001_HAS20078373_Td-3_Virus).

“Corporeal liveness is the shading of liveness that so often concerns musicians working with new digital performance interfaces, when they want to ensure an understandable connection for their audience between their physical gestures and the electronic sounds that result from them” (Sanden, 2019)

As Sanden mentions here, this aspect of being able to document and in some cases prove the extent of our involvement within our performance is important in the age of Artificial Intelligence. The great thing about livestreaming a performance and creating short form social media content as I’ve shown in the appendix, this allows us to show our “working out” and creates that trust between artist and fan. So much content on the internet is now AI generated and fake, so as an artist, if you can create a few clips that absolutely demonstrate what you are doing when you are creating your music or performing, then it all adds to the very realness that makes that artist authentic and trustworthy.

Experimentation and Randomisation with AI

There is a ritual to music production. In my experience, some of my best ideas have come from sitting in my studio and having what I call a “jamming session” where I try out ideas, new plug-ins, playing with ideas on my midi keyboard, and recording all of the time. One of my latest remixes incorporated a lot of use of the Melda comb filter and the TAL dub delay which I used on the original vocal from the track. These audio recordings became the basis for the entire track, providing the textures and build ups within the arrangement which in turn gave me the direction on which to base the rest of the track (see Appendix MMP7C002R~001_HAS20078373_So_Long_DubDelay_CombFilter).

This shows how I used these algorithmic plugins in a way that helped inspire and give the foundation for the rest of the remix. These are sounds that I would have never have been able to create myself if it wasn’t for the use of algorithmic AI plugins. Some of my favourite plugins are by the company Melda. Melda also features some great randomisation features in their plugins (see Appendix MMP7C002R~001_HAS20078373_Randomisation_Melda), so if I am ever stuck for ideas, I can use the randomiser “dice” button and see what preset parameters it generates for me.

This feels collaborative with algorithmic AI and sits well with me to use the tools which are available to me in the 21st century. I feel that I can use these tools without it hindering my own authenticity. In fact, I feel that my authenticity actually comes from the methodology and workflow that I, specifically take to bring my ideas to fruition. AI can help to fuel my authenticity through inspiring me to take different directions that I wouldn’t have otherwise thought about.

A great example of a more “analog” version of this is Brian Eno’s Oblique Strategies, where we are given a prompt in which to get ourselves out of a creative rut or “writers block” by taking a randomised change in direction to lead us to a completely different path as an artist. (Eno, 2025)

This is what excites me so much about the creative act of making music. Just one change, distraction, comment or suggestion can take the project on a completely different trajectory! That notion of unlimited possibilities is both a comfort to me and also what overwhelms me. These prompts or randomisation features can sometimes feel like a welcome limitation in a lot of ways to stop me from spiralling into a million different ideas.

The Human Touch

Creating music is indeed a very deeply embedded part of what it is to be human. In every culture and every tradition the human species has always attributed absolute value and importance to the act of music making in a very tribal sense. It is what makes us human and it is what bonds us together. Music surpasses the need for syntax and expresses in a much more simplistic and connective way, no matter what your native language may be.

“The likely importance of social bonding via synchrony in music-based activities draws on the observation that beyond a tendency to synchronize with one another, humans have a culturally ubiquitous aptitude for entrainment to rhythmic beats.” (Tarr et al., 2014)

Even as far back as 40,000 years ago, and most likely further, Neandathals have been known to have been using musical instruments. In a time when life would have been much more difficult in terms of survival, there was this tribalistic, human instinct and ritual for music making. (Killin, 2018)

Music could be described as any sound which evokes or expresses emotion in a way that differentiates it from language that is spoken. For that reason in of itself, humans would never fully stop making music themselves by turning that job over to AI, as to do so would go against everything that fundamentally makes us human. A world without creative music makers and humans experiencing that music would collapse human interaction and interconnection with each other. It would not be possible for humans to even attempt this as it would go against thousands upon thousands of years of human evolutionary process.

From personal experience, every single one of my closest friends and I enjoy a deep love and understanding for music. Music has saved me many times from some extremely difficult times in my life. After losing my mother while I was still a dependent teenager, I remember finding friends at that time who had access to turntables and vinyl and that quickly became a way for me to channel my energy into making music and DJ performance. Due to the intense pain I had endured from the loss of my mother, I subconsciously harnessed that pain into my music and subsequently it resulted in me becoming a professional DJ which I am still doing as my main job twenty five years later. That is the power of music. Music serves as a way for us all to process our trauma and emotions and carries with it a parallel journey into self awareness and knowing who you are on a deeper level, without the labels, accolades and void of social status.

From a biological perspective, listening to music lights up significant parts of the brain, pointing to music being more of a human necessity than simply for recreational pleasure alone. Sensory, emotional, memory and motor skill functions are all activated when humans listen to and partake in music based activities, either making music, collaborating in music live and dancing together to it. (T Zaatar et al., 2023)

People need people. People crave human connection. People will support the artist whom they know their story, who they are invested in and who they have seen grow as a person and as an artist. I certainly feel that this is true for me personally. I am very fortunate to have a small but very loyal fan base who have followed me through many versions of myself over the past twenty five years. For this reason, I don’t feel threatened by AI potentially taking over my role as an artist. Take the “band” The Velvet Sundown, for example; this band amassed over one million streams on Spotify since June 2025 allegedly using the platform Suno to create their music with.

Although I do not personally feel threatened by the age of generative AI in music, I can see why other musicians might feel that this is a threat. A few months previously to The Velvet Sundown controversy, there was another conversation around the potential threat of AI to artists and musicians, which compelled some very big names in the industry such as Annie Lennox, Damon Albarn and Kate Bush to release a “silent album” entitled “Is This What We Want?” to share awareness to the government’s plan to introduce an “opt out” for artists who do not want AI to be trained on their art to develop their models. “Under the new proposals, AI developers will be able to use creators’ content on the internet to help develop their models, unless the rights holders elect to “opt out”. (Glynn, 2025)

The argument here was that it would not be feasible for any artist to be about to opt out of the many thousands of AI platforms that exist, with many more models and platforms being produced every year.

Connection with Fans

We are living in an age where we have high speed internet, access to computers and the ability to livestream to thousands of people and show our audience who we are as a human being. We can even sit with them on an evening, while we are in the studio making music, and they can feel like they are a part of the process, and indeed they are.

I believe that in the future, artists will adapt to the rise of generative AI music saturating streaming sites such as Spotify and differentiate themselves from that type of music and their demographic. They will be funded directly from their fanbase through platforms that incorporate livestreaming and a paywalled community such as Bandcamp and Patreon. This frees the artists from the old model of the record label essentially “choosing” who will be the next big thing, and owning the artist in contracts such as 360 deals which are now becoming less commonplace now that we are in the age of alternative means for artists to support themselves financially. (Kjus, 2025)

As I mentioned earlier, I have been a successful livestreamer for many years now and I have built up a community of 25,000 followers and have partner status. I usually stream three to four times a week and have built up a loyal following which has given me the ability to fund myself entirely on my own as a completely independent artist and content creator. I also can advertise myself as a DJ and Music Producer teacher to my audience which all supplements my income. This online presence also puts me in front of more promoters so it also serves as a way for me to get more DJ gigs around the world.

I see that in the future, this is the model that artists will be getting on board with. Fans who want to see their favourite artist’s process and have deep and meaningful interactions with the artist. The great thing about this model is you are not bound by an infinitely changing algorithm on Spotify or Instagram or locked into a ten year contract with a manager or record label that essentially owns you and your art.

Collaboration

Occasionally, I will go live when making music in my studio as a way to involve my audience, and show them my process. This is interesting to both non producers and other music producers, and frequently changes how I would be creating my music if it was just me on my own. The audience interferes with the process in a way that changes the outcome. Kind of like a human random generator in a sense, as mentioned earlier. The audience throws out ideas, sparks new inspiration or perhaps another music producer in the chat will mention a new plugin they’ve had good results with, or suggests a change I could make within the project at that moment.

In my BA Honours in Music Production in 2022, for an assignment, I requested the assistance of another livestreamer on Twitch named “Lady Paintsalot”. While we were both livestreaming on our respective channels, we brought our viewers together to be a part of this collaboration.

We used Midjourney to generate a picture based upon a theme of “PLUR”, the universal dance music ethos of Peace, Love, Unity and Respect” (see Appendix MMP7C002R~001_HAS20078373_Collaborative_AI_Human_project).

While livestreaming, I wrote a song around that theme, focusing on the pictures that midjourney generated, whilst Lady Paintsalot created her own art while listening to me creating the track around this theme in real time. It was interesting to both of us how the prompt of the image from Midjourney influenced our musical and artistic decisions respectfully. Whilst we were creating our art, our audiences who were watching our livestreams also interjected ideas from the chat. Without their input and the initial artwork generated by Midjourney, the art would have no doubt come out differently.

Conclusion

Through researching valuable information and cross referencing well documented ideas and books with my own lived experience of the music industry and use of AI within my music, I find that AI absolutely has a place within our society. As time goes on, generative AI music will become non distinguishable from human made music. This in turn will actually accentuate what it is to be a human composer, and make that something that is valuable and sought after by other human beings. I see a world in which both AI generated music and human artists can live together, both providing music for completely different reasons and for different demographics.

Bibliography

Chrysopoeia (2025) Wikipedia. Available at: https://en.wikipedia.org/wiki/Chrysopoeia (Accessed: 07 December 2025).

Clancy, M. (2022) ‘AI Music’, in Artificial Intelligence and Music Eco System. Oxford: Routledge , pp. 21–21.

Clancy, M. (2022) ‘AI Music’, in Artificial Intelligence and Music Eco System. Routledge, pp. 23–23.

Clytus, R. (2016) The music and the musical inheritance of slavery (Chapter 10) – the cambridge companion to slavery in American literature, Cambridge Core. Available at: https://www.cambridge.org/core/books/abs/cambridge-companion-to-slavery-in-american-literature/music-and-the-musical-inheritance-of-slavery/40A7E44D8C24FA85ED05B024A9FCC0EA (Accessed: 07 December 2025).

Eno, B. (2025) Oblique Strategies, Oblique strategies. Available at: https://stoney.sb.org/eno/oblique.html (Accessed: 07 December 2025).

Illiac Suite (2025) Wikipedia. Available at: https://en.wikipedia.org/wiki/Illiac_Suite (Accessed: 01 December 2025).

Innovation (2025) Illinois Distributed Museum. Available at: https://distributedmuseum.illinois.edu/exhibit/illiac_and_ordvac/ (Accessed: 06 December 2025).

Killin, A. (2018) The origins of music: Evidence, theory, and prospects, Sage Journals Home. Available at: https://journals.sagepub.com/doi/10.1177/2059204317751971 (Accessed: 07 December 2025).

Kjus, Y. (2025) Twists and turns in the 360 deal: Spinning the risks and rewards of artist–label relations in the streaming era, BBC News. Available at: https://www.bbc.co.uk/news/articles/cwyd3r62kp5o (Accessed: 07 December 2025).

Max for live (2009) Ableton. Available at: https://www.ableton.com/en/live/max-for-live/ (Accessed: 01 December 2025).

Rubin, R. (2023) ‘Tuning In’, in The Creative Act. New York, New York: Penguin, pp. 13–13.

Sanden, P. (2019) in Rethinking Liveness in the Digital Age. Cambridge: Cambridge University Press, pp. 179–179. Available at: https://www.cambridge.org/core/books/abs/cambridge-companion-to-music-in-digital-culture/rethinking-liveness-in-the-digital-age/E85C9E4A2FC00FA987C52A052C7C912A (Accessed: 2025).

Sanden, P. (2019a) ‘Rethinking Liveness in the Digital Age’, in The Cambridge Companion to Music in Digital Culture. Cambridge: Cambridge University Press, pp. 182–182. Available at: https://www.cambridge.org/core/books/cambridge-companion-to-music-in-digital-culture/rethinking-liveness-in-the-digital-age/E85C9E4A2FC00FA987C52A052C7C912A (Accessed: 07 December 2025).

Swingle, E. (2023) Fl Studio 21.2 is set to include STEM Separation Tools built into the daw, MusicTech. Available at: https://musictech.com/news/gear/fl-studio-21-2-stem-separation-daw/ (Accessed: 07 December 2025).

T Zaatar, M. et al. (2023) The transformative POWER OF MUSIC: Insights into neuroplasticity, health, and disease, Brain, behavior, & immunity – health. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC10765015/ (Accessed: 07 December 2025).

Tarr, B., Launay, J. and Dunbar, R.I.M. (2014) Music and social bonding: ‘self-other’ merging and Neurohormonal Mechanisms, Frontiers in psychology. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC4179700/ (Accessed: 07 December 2025).