Why Loudness Changes What You Hear

Episode 2 · How Hearing Works · Published Aug 13, 2026

Episode purpose

These five articles are deliberately connected rather than treated as separate lessons. The episode's central idea is that changing playback level changes the way the listener perceives the same physical signal, which has direct consequences for mixing decisions. The bass-at-low-volume phenomenon is treated as an important extension of that idea, not a separate topic.

Script

Opening

Welcome back. Last episode I made the case that your ears aren't a measurement system, that perception and signal aren't the same thing. Today I want to push that idea somewhere I think you'll find genuinely useful: what happens when you just turn the volume knob.

Turn your mix down. Not just a little, turn it down until you're listening quietly enough that the music no longer feels impressive. Now listen. Something strange happens.

The mix you were just hearing at a comfortable level may suddenly feel different:

• The bass seems to disappear
• The kick may feel less substantial
• The vocal may become more prominent
• The top end may seem different
• A dense arrangement may suddenly feel much thinner

And you may find yourself thinking, "what happened to my mix?" Usually, nothing happened to the mix. You changed the listening level, and your hearing changed with it.

This is one of the most important things to understand about mixing, because we often talk about a mix as though it were a fixed object. We say "the bass is too loud," "the vocal is too quiet," "the top end is too bright." But those statements are incomplete. Too loud at what listening level? Too quiet relative to what? Too bright under which listening conditions?

The physical signal coming out of your speakers can stay exactly the same while the listener's perception of that signal changes. That's what we're going to explore in this episode.

1. The Same Mix Can Sound Like a Different Mix

Let's start with something simple. Imagine you've finished a mix. You play it at a comfortable level. The bass feels balanced. The vocal sits nicely. The cymbals have enough energy. The kick feels solid. Everything seems to make sense.

Then you turn the monitor level down. Suddenly the bass feels weaker. You turn the bass up. You return to the original listening level. Now the bass is too loud. So you turn it back down. Then you turn the volume down again, and the bass disappears. This can become an endless loop.

The temptation is to believe your mix has a problem that needs fixing. But sometimes the problem isn't in the mix, it's in the assumption that your hearing behaves the same way at every playback level. It doesn't. Human hearing is level-dependent, and this is where equal-loudness contours become extremely useful.

2. Equal-Loudness Contours: Hearing Does Not Treat Every Frequency Equally

An equal-loudness contour is a way of representing how much sound pressure is required at different frequencies for those sounds to be perceived as equally loud. The important part is that phrase, equally loud.

Imagine you have a low-frequency tone, a mid-frequency tone and a high-frequency tone. You can adjust their physical levels until they're perceived as having approximately the same loudness. The resulting relationship isn't flat. Our hearing doesn't have equal sensitivity at every frequency, and that sensitivity changes with overall listening level.

That gives us a useful mental model. A frequency response graph describes the signal. An equal-loudness relationship describes something about how that signal is perceived. Those are different things. This is why a perfectly flat frequency response doesn't guarantee the listener perceives every part of it as equally prominent, and why the same spectral balance can produce a different subjective experience when you change the playback level.

3. Turn the Volume Down and Your Perception Changes

Here's a useful experiment. Take a mix you know well. Listen at a moderate level. Don't try to mix, just listen. Notice:

• How much bass you perceive
• How prominent the vocal feels
• How present the cymbals are
• How much low-mid density you notice
• How large the mix feels

Now turn the monitoring level down. Don't change anything in the session. Don't touch the EQ, the faders, or the master. Just reduce the playback level. Listen again.

You may notice the tonal balance doesn't feel identical. That's the important observation. The signal hasn't changed. The listener has changed. More precisely, the relationship between the physical stimulus and the listener's perception has changed because hearing is level-dependent. That means listening level isn't just a comfort setting, it's part of the conditions under which you evaluate a mix.

4. Why This Matters So Much in Mixing

Think about what happens when you're mixing. You make an EQ decision. You make a compression decision. You balance the vocal. You decide how much bass the song needs. You decide whether the cymbals are too aggressive. Every one of those decisions is being made through your auditory system.

If the auditory system responds differently at different levels, listening level becomes part of the decision. That's why constantly changing your monitoring level while making tiny tonal decisions can be dangerous, you aren't necessarily evaluating the same perceptual situation each time. You may think you're comparing version A against version B. But you're actually comparing version A at one listening condition against version B at another. That's not a clean comparison.

5. Why Bass Seems to Disappear Quietly

Let's focus on one of the most familiar examples. You turn the music down. The bass seems to disappear. This can be particularly obvious with a mix that felt full and powerful at a louder level. At low level, the kick may lose some of its physical weight, the bass guitar may feel less obvious, and the sub information can become much less apparent.

The natural response is "I need more bass." Sometimes you do. But this is exactly where you need to be careful. If you add bass because the mix sounds bass-light at a very quiet monitoring level, then return to your normal listening level, you may have created a mix with excessive low-frequency energy. You were compensating for the listening condition rather than correcting the mix.

This is one reason low-volume checking is valuable. But low-volume checking shouldn't automatically dictate your EQ. It's an evaluation condition, not necessarily a prescription.

6. The Quiet Mix Is Not Lying to You

This distinction is important. When you turn a mix down and the bass seems to disappear, your hearing isn't "wrong." The perception is real. The mistake would be assuming "because I perceive less bass, the mix must contain too little bass." Those aren't equivalent statements.

The quiet playback condition is revealing something about how the mix behaves when the overall stimulus becomes weaker. That's useful information. The question is what you do with it:

• Maybe the mix has enough low end and behaves normally at lower levels
• Maybe the bass really is too dependent on very low-frequency energy and needs useful harmonics or a different balance
• Maybe the kick and bass relationship isn't robust
• Maybe the arrangement depends too heavily on sub energy
• Maybe the low end is fine and you simply need to avoid judging final tonal balance at an unusually quiet level

The listening test gives you evidence. It doesn't give you the answer automatically.

7. This Is Why "Mix at 80 dB SPL" Needs Context

You may have heard the advice, "mix at 80 dB SPL." There's a reason standards and studio practice have historically cared about monitoring level, and it isn't arbitrary: it connects directly to the equal-loudness relationship we just talked about, which is formally described in the ISO 226 equal-loudness contour standard. At lower playback volumes, bass and extreme treble become less prominent to your ear; at higher volumes, your hearing's frequency response becomes more balanced.

A reasonably controlled monitoring level gives you a repeatable reference condition, and it avoids some of the extremes of very quiet and very loud listening. But the important lesson isn't "80 dB SPL is the magic mixing volume." It isn't. The more useful idea is: have a known, repeatable monitoring condition, and understand what changes when you move away from it.

If you constantly mix at whatever volume happens to feel comfortable at that moment, your judgments can become inconsistent. You make one decision while listening relatively loudly, another while listening quietly, another after turning the level back up, and each decision gets made through a somewhat different perceptual system. A reference level gives you a stable baseline.

Worth flagging directly: prolonged listening at high sound pressure levels can also temporarily reduce your hearing sensitivity, particularly in the upper midrange where your ears are naturally most sensitive, a relationship covered in the WHO's safe listening guidance. That's a real, established finding, not a FREQ opinion, and it's one more reason a moderate, repeatable reference level beats mixing loud out of habit.

8. Why You Should Not Mix Everything at One Volume

This might sound contradictory. First I'm saying consistency matters, now I'm saying you shouldn't listen at only one level. Both are true. You need a primary listening condition for making detailed decisions, but you also need to check how the mix behaves away from that condition.

Think of your main listening level as the reference, then use other levels as tests. At your normal working level: is the overall balance convincing? At a quieter level:

• Does the hierarchy still work?
• Can you still follow the vocal?
• Does the bass remain perceptually present enough?
• Does the groove survive?

At a louder level:

• Does anything become aggressive?
• Does the vocal become overbearing?
• Does the top end become uncomfortable?
• Does the low end become excessive?

These are different questions. You aren't asking one volume to tell you everything, you're using different listening levels to reveal different behaviours.

9. Low Volume Is Particularly Useful for Hierarchy

There's another reason quiet listening is valuable. When you turn the level down, you often become more aware of the large-scale hierarchy of the mix:

• Can you still identify the lead vocal?
• Can you follow the main melody?
• Does the snare still define the groove?
• Does the bass still provide enough musical foundation?
• Do supporting instruments remain supporting instruments, or does something secondary suddenly become distracting?

This connects directly to what we discussed in Episode 1. A mix isn't simply a collection of sounds, it's an arrangement of attention, and listening quietly can expose whether that hierarchy is robust. If the mix only works when it's loud, that's worth investigating. But again, the answer isn't necessarily "make everything louder." The question is "why does the hierarchy change when the listening level changes?"

10. Quiet Listening Can Reveal Arrangement Problems

Suppose you have a very dense chorus. At a moderate level it sounds huge, everything seems to work. Now turn it down. Suddenly the chorus becomes difficult to follow. The vocal gets buried. The bass disappears. The guitars merge into a thick middle. The drums lose definition.

You could start EQing. But perhaps the more important observation is that there's simply too much information competing at the same time. The quiet listening test has revealed a weakness in the arrangement or mix hierarchy:

• Maybe the chorus needs more contrast
• Maybe an instrument should drop out
• Maybe a part should occupy a different register
• Maybe the vocal needs better dynamic control
• Maybe an element that was useful at a louder level isn't contributing enough at a quieter one

Again, the test doesn't tell you which solution to use. It tells you where to investigate.

11. Loud Listening Has Its Own Information

Now let's go the other direction. Turn the mix up, not to an unsafe level, just louder than your normal working level. What happens? Some problems become more obvious:

• Harshness becomes harder to ignore
• Aggressive upper-mid energy can become apparent
• Low-frequency imbalance can become physically obvious
• Compression artifacts can become more noticeable
• A vocal that felt comfortably present at moderate level may suddenly feel too forward
• A cymbal that sounded exciting may become tiring

This is why a mix shouldn't be judged at only one level. Different listening levels act almost like different diagnostic instruments. Quiet listening can expose hierarchy and robustness. Moderate listening can provide your main balancing reference. Louder listening can expose aggressive or excessive elements. The important thing is not to confuse the results.

12. The Loudness Trap

There's another problem that shows up constantly during mixing. You make a change. The new version sounds better. Then you discover it's slightly louder. Suddenly you have a new question: do I actually prefer the mix, or do I simply prefer the louder version?

This is why level matching is so important when comparing changes. Our perception of loudness influences our perception of quality. A louder signal can seem more exciting, more detailed or more impressive. That doesn't automatically mean it's better.

So when you're comparing processing decisions, try to make the comparison fair. If one version is noticeably louder, you may be evaluating loudness rather than the actual tonal, dynamic or spatial change, and that matters most when you're making subtle decisions. Remember what we discussed in Episode 1: a physical change and a perceptual consequence are not the same thing. Here, the same principle applies to level itself.

13. Loudness and Frequency Are Connected in Perception

This is where equal-loudness contours become more useful than simply memorizing the term. Imagine a frequency spectrum. A spectrum analyzer might show energy distributed across the entire audible range. But the listener doesn't experience that graph directly, the auditory system transforms the physical stimulus into perception, and the overall level of the stimulus affects that transformation.

So when you change listening level, you aren't merely changing "how loud the same thing is." You're changing the conditions under which different parts of the spectrum are perceived. That's why the tonal balance of a mix can appear to shift even when the mix itself hasn't changed. The consequence for mixing: don't assume a tonal judgment made at one level will remain identical at another.

14. Why This Doesn't Mean You Should EQ for Equal-Loudness Contours

Here's where a useful psychoacoustic concept can become a terrible mixing recipe. Someone learns that human hearing is less sensitive to certain frequencies at lower levels. They think, "then I should boost those frequencies when I mix quietly." That can be useful in a very specific context if you're deliberately compensating a monitoring system for low-level listening. But it's not a general mixing rule.

You're not supposed to EQ your mix to compensate for the ear's equal-loudness characteristics. Your listeners will experience the same perceptual system. The point of mixing isn't to create a signal that looks perceptually flat according to a hearing curve, it's to create a musical result that works for the listener.

Equal-loudness contours explain why your perception changes. They don't tell you what EQ curve your mix should have. This distinction is crucial: psychoacoustics explains behaviour, it doesn't automatically prescribe processing.

15. The Problem With Mixing at Extremely Quiet Levels

Very quiet monitoring can be useful as a test. But if you spend the entire session working at an extremely low level, you may start making decisions to compensate for the way your hearing behaves under that condition. You add bass. You add top end. You push elements that feel less prominent. Then you turn the level back up and discover the mix has become exaggerated.

This is why a controlled reference level is useful. You need a condition where your judgments are consistent, then you deliberately step away from it to test translation. Think: reference, then test, then return to reference. Not: random volume changes leading to random corrections.

16. The Problem With Mixing Extremely Loud

The opposite problem exists too. If you mix loudly for long periods, the presentation can become exciting: the bass feels powerful, the vocal feels immediate, the cymbals sound impressive, everything seems larger. But your judgment can become biased by the physical excitement of the playback level, and prolonged loud listening can contribute to auditory fatigue, which we'll discuss in the next episode.

So loud listening is useful as a check. It isn't necessarily the best environment for making every tiny decision. The purpose of a loud check is usually "what becomes objectionable when I turn this up," not "what should I make everything sound like at this level."

17. Listen for What Survives

Here's a useful way to think about monitoring-level checks. Don't ask "does the mix sound identical at every level?" It won't. Instead ask "does the musical intention survive?" That's a much better question.

At lower level, the bass may feel different. Fine, but does the song still feel grounded? The vocal may become somewhat more prominent. Fine, but does the hierarchy still make sense? The chorus may lose some physical impact. Fine, but does it still feel bigger than the verse? The cymbals may become less noticeable. Fine, but do they still contribute to the sense of movement? You're checking robustness, not demanding identical perception.

18. The Mix Should Change Without Falling Apart

A good mix doesn't necessarily sound identical at every volume. In fact, expecting identical perception is unrealistic. What you want is a mix that changes gracefully. Turn it down, something changes. Turn it up, something changes again. But the core musical relationships remain intact:

• The listener can still identify what matters
• The groove remains understandable
• The emotional centre remains convincing
• The arrangement remains intelligible
• The tonal balance remains believable

That's a much more useful definition of translation.

19. A Listening Experiment

Let's make this practical. Take a finished mix. You're going to listen three times.

Pass one: your reference level. Set your normal working level and listen to the entire song. Write down:

• What is the focal point?
• Where does the low end feel strongest?
• Which section feels largest?
• What sounds closest? What sounds distant?
• What is supporting the main idea?

Don't change anything, just establish your reference.

Pass two: quiet. Turn the level down substantially. Don't compensate, don't touch the mix. Listen again, then ask:

• Does the vocal remain understandable?
• Does the song still groove?
• Does the bass still contribute musically?
• Does the chorus still feel like a chorus?
• Has something secondary become disproportionately distracting? Has the hierarchy changed?

Write down what changed.

Pass three: louder. Return to your reference, then go somewhat louder. Again, don't touch anything. Ask:

• Is the vocal now too aggressive?
• Is the top end becoming tiring?
• Is the low end excessive?
• Are compression artifacts becoming obvious?
• Does anything suddenly demand too much attention?

Now return to your reference level. Only after returning should you make decisions.

20. What You Should Actually Fix

Suppose your bass disappears at low volume. What should you do? Don't automatically boost the bass. First ask: is the bass supposed to remain strongly audible at that level? If yes, investigate why it isn't:

• Maybe the bass contains too much energy in a region that becomes less perceptually useful at lower level
• Maybe it needs harmonic information that lets its musical identity remain apparent without depending entirely on very low-frequency energy
• Maybe the kick and bass relationship needs work
• Maybe the arrangement is too dense
• Maybe the bass is actually fine and the quiet monitoring condition is simply exposing the limits of your hearing

The answer depends on the record. There's no universal "quiet listening EQ."

21. This Is Why Bass Can Be Felt at One Level and Heard at Another

Low-frequency content is particularly interesting because our experience of bass isn't simply one thing. A sub-heavy signal can provide physical weight at higher playback levels; at lower levels, that same signal may become less perceptually prominent. But a bass instrument can also contain higher harmonics that communicate its pitch and identity.

So two bass sounds with similar fundamental energy can behave differently when the playback level changes. One may seem to disappear more dramatically. Another may remain recognizable because its higher-frequency information keeps contributing to its perception.

That's an important production insight. When you want a bass part to remain meaningful across listening conditions, the question isn't necessarily "how much sub do I have." It can also be "what information lets the listener recognize this bass?" That distinction becomes particularly important later when we talk about harmonic structure, masking and translation.

22. Why the Quietest Listening Test Can Be So Revealing

There's a reason engineers have used quiet monitoring as a practical test for decades. At low level, you can strip away some of the excitement created by sheer playback energy. You're forced to hear the hierarchy: what survives, what disappears, what still feels important, what becomes irritating, what was only impressive because it was loud.

Those are valuable questions. But remember, quiet listening is a test, not necessarily your target listening condition. If the mix only works at a loud level, investigate it. If it works at quiet level but feels wrong at normal level, investigate that too. The point is to understand the behaviour.

23. Why Consistency Matters More Than a Magic Number

If you remember only one thing from this discussion about 80 dB SPL, let it be this: a repeatable listening level is more useful than worshipping a particular number. A reference level lets you make comparisons. You can return to the same condition after making a change, compare today's judgment with yesterday's, and establish a stable baseline. Then you can deliberately listen at other levels to test the mix.

That's much more useful than constantly changing your monitor level because one moment feels more exciting than another.

24. The Deeper Lesson

Let's connect this back to Episode 1. We started this podcast with the idea that your ears are not a measurement system. Now we can extend that idea: your ears are also not a fixed measurement system. Their perception depends on the conditions under which they're listening. The physical signal may stay unchanged; the listener's experience can change.

That's why mixing isn't simply about finding the "correct" frequency balance or "correct" level. You're trying to create relationships that remain musically convincing under changing listening conditions. That's a much more sophisticated goal.

25. A Better Way to Think About Monitoring

Instead of thinking "what volume should I mix at," think "what is my reference listening level, and what does my mix reveal when I move away from it?" That's the useful question.

Your reference level is where you make detailed decisions. Your quiet level is where you inspect hierarchy and low-level robustness. Your louder level is where you inspect aggressiveness, excess energy and fatigue. Returning to your reference level lets you decide what, if anything, actually needs changing. That's a workflow, not a rule.

26. Don't Correct the Listener

There's a subtle but important philosophical point here. You cannot mix around every possible listening condition. Someone might hear your record on earbuds. Someone else on a large speaker system, in a car, quietly at night, loudly at a party, on a phone speaker. The listener's auditory system and playback system will always be part of the final experience.

Your job isn't to make the record behave identically everywhere. Your job is to create a signal whose important musical relationships survive those changes. That means thinking about robustness rather than perfection.

27. Listening Experiment: The Bass Test

Here's one final experiment specifically for the bass. Take a mix with a clear bass part. Play it at your normal reference level. Listen to the bass, don't judge whether it's "loud enough," listen to what role it plays. Does it establish pitch? Does it provide weight? Does it reinforce the kick? Does it connect the rhythm section?

Now turn the level down. Listen again. Ask: what part of the bass disappeared first? Was it the physical weight, the sense of pitch, the attack, the relationship with the kick? Or did the bass remain perceptually understandable even though it became less physically prominent?

Now return to the original level. This experiment teaches you something much more useful than simply asking whether your bass is loud enough. It tells you which information in your bass survives changes in listening level.

28. The Connection to Everything Coming Later

This episode might seem like a lesson about volume. It isn't. It's really a lesson about perception as a moving target. In the next few episodes, we're going to build on this. We'll talk about auditory fatigue and adaptation, then about how the brain separates simultaneous sounds into auditory objects, then frequency masking, and eventually we'll connect all of that to practical mixing decisions. The common thread is that the signal itself isn't the whole story. What matters is what the listener can actually perceive.

29. What I Want You to Remember

So if you're mixing today, remember these principles:

• Your mix does not have one perceptual state. It can feel different at different listening levels.
• Don't automatically EQ for what you hear at extremely quiet levels. Use quiet listening as information.
• Establish a repeatable reference listening level. Consistency makes comparisons meaningful.
• Use different listening levels deliberately. Quiet, normal and louder listening can reveal different problems.
• Don't expect the mix to sound identical everywhere. Expect the important relationships to survive.

And finally: when something changes at a different listening level, don't immediately ask what you should fix. Ask what changed in your perception. That question will lead you to a much better answer.

Closing

A mix isn't heard in a vacuum. It's heard by a human auditory system, and that system responds differently depending on the level of the stimulus. So when you change the monitoring level, you aren't simply turning the same mix up and down, you're changing the conditions under which you perceive it.

That's why bass can seem to disappear quietly. That's why tonal balance can appear to change. That's why low-volume checks can reveal hierarchy. That's why a stable reference level is useful. And that's why there's no single monitoring volume that can tell you everything you need to know about a mix.

The goal isn't to make your mix sound identical at every level. The goal is to make the important musical information survive the change. And once you start listening that way, volume stops being just a control on your monitor knob. It becomes another way of asking the mix a question.

Next time, we're going to look at what happens when you keep asking your ears those questions for too long. Because eventually, the listener changes again, not because the volume changed, not because the mix changed, but because your auditory system adapted to what it had been hearing. And that brings us to auditory fatigue.

Practical takeaways

1. Establish a repeatable reference monitoring level for detailed mixing decisions.
2. Use quiet listening as a diagnostic, not an automatic EQ instruction.
3. Check whether the mix's hierarchy survives at low volume.
4. Check louder playback for harshness, excessive low end and aggressive elements.
5. Level-match comparisons when evaluating subtle mix changes.
6. Don't assume that a bass disappearing at low level automatically means the mix needs more bass.
7. Ask which information in a sound survives changes in listening level.
8. Judge whether the mix's important musical relationships remain convincing rather than demanding identical tonal perception at every level.
9. Return to your reference level before making final corrective decisions.
10. Treat psychoacoustic principles as explanations of perception, not rigid processing recipes.

Episode summary

Why Loudness Changes What You Hear explores the relationship between listening level and auditory perception. The same physical mix can be perceived differently when playback level changes because human hearing does not respond equally to all frequencies at all sound levels. Equal-loudness contours, formalized in the ISO 226 standard, provide a useful framework for understanding this level-dependent sensitivity. In practical mixing, this explains why bass may seem to disappear during quiet listening, why tonal balance can appear to shift, and why different monitoring levels reveal different aspects of a mix.

The episode also explains why a reference monitoring level is useful without treating any particular SPL value as a universal rule. Quiet and louder listening should be used as deliberate diagnostic conditions rather than as automatic instructions for EQ or balance.

The central principle: a mix does not need to sound identical at every listening level. Its important musical relationships need to survive the change.