Audiences See Before They Hear: The Research on Visual Perception in Musical Performance
- Gökçe Kutsal
- May 7, 2024
- 10 min read
Updated: Jun 16
* Where I describe clients in my blog posts, I've changed names, instruments, and other identifying details to protect their privacy. Sometimes I'm describing one person, sometimes a pattern I've seen across several. The essence of the experience and what we worked on together remain true.
In 2013, a researcher named Chia-Jung Tsay published a paper in the Proceedings of the National Academy of Sciencesthat quietly upset a lot of people in classical music.
What she did was simple...
She took recordings of finalists from prestigious international music competitions and split them into three formats: Sound only, video only with no sound, and both together.
Then she asked participants, including professional musicians, to identify which finalist had won the competition.
The professional musicians had insisted, in interviews, that they relied on sound. Sound is what we care about. Sound is what we evaluate. Sound is what matters.
But the data didn't agree.
Participants who watched the silent video clips identified the winners at a rate significantly higher than chance.
Participants who listened to sound only did no better than guessing. Yes, even the trained musicians.
And when participants had access to both sound and video, their judgments were still anchored to the visual information. The sound, available right there alongside the picture, didn't pull their evaluations back toward what they consciously believed they were evaluating.
The visual side of performance, in other words, isn't a small contributor to how performances are perceived. It's often the dominant one. Even for the people whose entire professional identity is built around listening.
This finding has been replicated and extended several times since 2013. Waddell and Williamon (2017) showed that the same effect persists across full-length performances, not just brief excerpts... and that a musician's stage entrance behavior and facial expressions significantly affect continuous quality ratings throughout the performance, not only at the start.
The body-movement literature (Wapnick, Davidson, Broughton & Stevens, and many others) has confirmed the basic pattern across instruments, audience types, and styles.
I want to talk about what this research means. But more importantly, I want to talk about what it doesn't mean.
Because the natural reading of these findings can take a musician somewhere unhelpful, and the right reading can change how you think about being on stage.
The Wrong Takeaway
The obvious lesson, if you were skim-reading the Tsay paper, is something like: Audiences are shallow. Look more confident. Perform a version of confidence you don't feel, because that's what gets rewarded.
This is the fake it till you make it version of the research, and I want to name it directly because it's the version most people land on. It's also the version that does not work, for at least three reasons.
First, it doesn't actually work as performance technique. Faking a confident face while feeling internally anxious is exhausting, and most audiences and panels can tell the difference between performed composure and the real thing. The performed version usually reads as slightly wooden, slightly disconnected, slightly off. Because the human brain is genuinely good at distinguishing a mask from a state. The musician working this hard to look calm tends to read as trying to look calm, which is the opposite of what they were going for.
Second, it intensifies the very anxiety it's trying to manage. A musician on stage who's monitoring their own facial expression and body language for confidence cues has added a second cognitive load on top of the first. They're now performing the music and performing themselves, simultaneously, with the constant secondary worry that the second performance might be failing. That doesn't reduce anxiety. If anything, it compounds it.
Third, it doesn't fit how the visual dimension actually transmits. The research isn't really showing that audiences are rewarding the appearance of confidence. It's showing that audiences are reading something more specific... a quality the literature variously calls presence, commitment, expressiveness, embodied engagement with the music. Those qualities don't come from facial control. They come from the performer's underlying orientation toward what they're doing.
Which brings me to the part of this research I think is genuinely useful.
The Right Takeaway
What the visual research is really pointing at, in my reading of it, is this:
The version of you that shows up on stage transmits visually whether you intend it to or not.
Your commitments, your orientation, what you understand yourself to be doing in the room... all of that is encoded in how you walk on, how you settle, how you breathe before the first note, how you hold yourself between phrases, how you handle a moment that didn't go quite as planned.
The audience isn't reading confidence as a costume. They're reading who you are showing up as. They're picking up your inner state through the only channel they have access to before the sound starts, which is what they can see.
This is why the fake confidence approach fails. It tries to manipulate the outer transmission without changing the inner orientation. The audience sees the mismatch.
So in my experience, the more useful question — the one that actually changes things — isn't how do I look more confident. It's who am I deciding to be in this performance.
That decision, made deliberately, shapes the inner state. The inner state then transmits visually, the way the research describes. And the visual transmission shapes how the audience receives what you're playing.
You're not performing confidence. You're performing as the version of yourself that already has access to it.
You Already Have This
Here's what I mean:
You already have access to a confident performer-self. You demonstrate this routinely, in contexts you've probably never thought of as performing.
When you sit down to play for a five-year-old, you're not worried about making a mistake. You're not monitoring your facial expression. You're not running an internal commentary about whether you sound good enough. You're just playing, with your full attention on the music and the person you're playing for. The five-year-old gets your real playing, your real presence, your real musicianship. Often it's some of your best.
The same thing happens when you play for a dog. For a loved one in your living room. For a teacher you respect and feel close to. For an esteemed colleague who's a friend.
That musician — the one who shows up in those rooms — is not a different person from the one walking into an audition. It's the same person under different conditions.
The musical capacity is identical. The technique is identical. The artistry is identical.
What's different is the threat response, which doesn't fire in those low-stakes contexts and does fire in the high-stakes ones.
This is important because it means you don't have to learn a new skill when you work on stage confidence. You are learning to access a familiar state under unfamiliar conditions. The capacity is already in you.
The question is how to bring it with you into the rooms that currently trigger you out of it.
A Story I Want to Tell You
I've been working with a brass player who, before our work together, had what they called a one-down feeling in every audition. The internal commentary was something like: I don't matter. No one wants me here. I might as well pack up and go home.
They were technically excellent, they'd been hired into good orchestras. By every external measure they were a serious professional musician.
But in the audition room, they showed up as someone trying not to take up space.
When we talked about it, I asked what they were like in orchestral rehearsals and concerts, you know, settings where their professional life worked.
The answer was completely different.
In those rooms, they were curious, engaged, committed. The internal orientation was let's see what I can bring to this piece today. They were interested in the music, focused on what they could offer, present in their body, comfortable in their role.
So we made the orchestral version of them the version that would walk into the audition room. Notice how this is not faking it.
It's just naming what was already true: That musician exists. They're you. They show up in concerts. They can show up here too.
We worked on the orientation specifically. Deciding before the audition: Who am I showing up as?
The musician who walks into a concert thinking let's see what I can bring. What am I bringing to this audition? My years of work with this repertoire. My ear. My musicianship. My genuine curiosity about how it'll sound in this room today.
The next audition, they walked in feeling just as nervous at the start. The nerves didn't disappear because we'd made a decision about orientation — that's not how nerves work, and the post would be lying if I told you it was :)
But something else happened:
The chosen orientation gave them an anchor to keep returning to. The nerves were there. They just didn't get to decide how they showed up in the room.
By their own account, the playing was some of their best... and within the first few minutes, they stopped noticing the nerves at all.
The orientation had taken over the cognitive real estate that the anxiety had been occupying. The nerves weren't suppressed, they were just displaced by something more substantial.
What My Own Research Showed
This isn't only what I've seen in coaching. It's also what the data showed when I measured it.
In 2022, as part of my MA in Voice Pedagogy, I ran a mixed-methods study with four professional singers, applying ACT (Acceptance and Commitment Training) coaching across a six-week intervention.
(The full methodology and findings are on my supporting research page if you'd like to see them properly.)
But the finding most relevant to this post is this:
After the coaching, the singers' performances were rated meaningfully higher by a blinded adjudication panel. Not because the singers had become technically better in six weeks or stopped being nervous, because they hadn't. They reported still feeling anxiety during the assessment performances.
What had changed was the version of themselves they were showing up as. The panel, who didn't know which recordings were before-coaching and which were after, could hear the difference. Or rather, given what the visual-perception research shows, they could probably see and hear the difference.
The singers were more committed in the music, more in character, more present with their material. The nerves were still present, but they just didn't get to determine how they showed up.
This, to me, is the practical answer to the Tsay research. So the audience is reading your visual presence, yes. But you can't try and influence what they're reading.
The only thing you can influence is what version of you is being read. Choose deliberately, and you can choose what gets transmitted.
A Short Honest Aside
I want to touch on something briefly, because the visual-perception research is also clear about it and it would feel dishonest to skip.
The same research base that shows visual presence matters also shows that audiences are influenced by visual factors that have nothing to do with musicianship: Dress, attractiveness, race, gender and body type.
Wapnick's work in the late 1990s, Goldin and Rouse's famous 2000 study on blind orchestral auditions, and a substantial follow-up literature have documented these biases directly. The blind audition reform, which dramatically increased the hiring of women into major orchestras once panels could no longer see candidates, exists because the industry was forced to acknowledge that its visual judgments were not actually about the music.
(Fun fact: Even after the blinding process, the panel could still differentiate between genders from their heel sounds. So some orchestras had to put carpet down in audition rooms to dampen these. When they did, the hiring process actually became equal. )
I think it's worth naming, because the audiences see before they hearresearch isn't a clean story. Some of what audiences see is bias. The question of how the profession should handle that is a separate conversation, and a long one. So I'm not going to make that the main point of this post... because it's largely outside any individual musician's control.
What you can influence is the part the research calls expressiveness, presence, commitment, embodied engagement. That part is yours.
The rest is structural, and focusing on things you can't control wouldn't help you show up as your confident self.
What to Actually Do
If you want to put this research to use without sliding into performing confidence you don't feel, the work is this:
Before any high-stakes performance, ask yourself: Who am I showing up as? We're not building a character to play. Think of it as a set of commitments: I'm here to communicate this music. I'm here to take this audition seriously and do my best work. I'm here to bring something to this piece, the way I would in a rehearsal. I'm here as the musician who already plays this music well when I'm not being watched. Make the choice deliberately. Write it down if it helps. Say it to yourself in the green room.
Borrow the orientation from a context where it already exists. The five-year-old. The dog. The trusted colleague. The orchestral rehearsal you'd walk into without a second thought. Identify a specific room where you already perform with full presence, and bring that orientation into the high-stakes one. In this audition, I'm being the musician who shows up in the orchestra. The one whose internal commentary is let's see what I can bring to this piece today.
When the nerves arrive on stage, don't try to make them go away. Return to the chosen orientation. The nerves will keep arriving. That's not the problem. The problem is letting them decide who you're being in the room. The chosen orientation is the anchor. Yes, I'm nervous. And I'm still here as the musician who's committed to playing this music well.Both can be true. Both will be true.
Trust that the visual side takes care of itself when the orientation is right. You don't have to manage your facial expression. You don't have to perform composure. If you're genuinely in the music, genuinely committed to what you're doing, genuinely present with what you're playing — that transmits. The research is clear about that. The version of you in the room becomes visible to the audience without you having to costume it.
If you've read the post on performing freely and confidently, this is the same underlying argument extended into the visual dimension. The performer-self isn't a mask. It's a facet of you that's always there... the question is whether you're letting it run the room, or letting your nerves run it instead.
And if you'd like company in finding the version of yourself that already knows how to do this, and learning to bring it into the rooms where the stakes feel highest, private coaching is where we'd do that together.
About
I'm Gökçe Kutsal, a music performance coach for professional orchestra musicians and opera singers, with an MA in Voice Pedagogy, plus coaching and teaching experience since 2017. I work with orchestral musicians, soloists, and audition candidates across Europe, the UK, North America, and Australia — and I write about performance anxiety, audition preparation, and the craft of practice for musicians who already have the technique and are trying to work out why it doesn't always hold up under pressure.
Research cited
Tsay, C.-J. (2013). Sight over sound in the judgment of music performance. Proceedings of the National Academy of Sciences, 110(36), 14580–14585. https://doi.org/10.1073/pnas.1221454110
Waddell, G., & Williamon, A. (2017). Eye of the beholder: Stage entrance behavior and facial expression affect continuous quality ratings in music performance. Frontiers in Psychology, 8, 513.
Wapnick, J., Mazza, J. K., & Darrow, A.-A. (1998). Effects of performer attractiveness, stage behavior, and dress on violin performance evaluation. Journal of Research in Music Education, 46(4), 510–521.
Goldin, C., & Rouse, C. (2000). Orchestrating impartiality: The impact of "blind" auditions on female musicians. American Economic Review, 90(4), 715–741.
Kutsal, G. (2022). Supporting research: A 2022 study with professional singers.
