Fifths

Science blog ·

Cadence first, scale second: how to set up your ear before every drill

How you establish a key before an ear-training question changes what your ear is actually trained to do. The case for using a cadence rather than a scale.

cadence tonality key-establishment practice krumhansl

Open most ear-training apps — Tonedear, EarMaster, Functional Ear Trainer, Fifths — and you’ll see the same opening move on a scale-degree question. The app plays an ascending major scale, then a single note, and asks you to identify which degree you heard.

This works. It establishes the key. Most learners never question it.

There is a slightly different approach that, on the available evidence, trains a stronger and more transferable kind of musical hearing: establish the key with a cadence rather than (or before) a scale. The difference seems small. It is not.

What a cadence does that a scale doesn’t

A scale gives the listener the pitch material of the key — the seven notes you can use. A cadence (typically I–IV–V–I, ii–V–I, or I–vi–IV–V) gives the listener the functional structure of the key. A cadence is what tells the ear, harmonically, what the tonic is — not just where it lives in the alphabet of the scale.

Carol Krumhansl’s foundational tonal-hierarchy work used both scale-only and cadence-based key-establishment paradigms. Listeners produced the same probe-tone profile — the stable, robust hierarchy where the tonic rates highest, then dominant, mediant, the rest of the diatonic notes, then chromatic notes [1]. But cadence-based contexts produced stronger and more rapidly stabilized hierarchy ratings. The ear locks onto a key faster when it has heard functional motion.

This is intuitive once stated. A scale is a list. A cadence is a sentence. You can identify “is this English?” from a list of words, but you understand English faster from hearing a complete grammatical statement.

What this means in practice

For ear training, the implication is direct: a question presented after a cadence places the listener in a stronger functional state than the same question presented after a bare scale. The trained skill is not just “identify the pitch” — it is “identify the pitch in a key that has been firmly established harmonically.” That is what music does in real listening: melodies do not arrive after a scale, they arrive after a chord progression.

Two further benefits:

1. Cadences train chord-function hearing as a side effect. Every time the app plays I–IV–V–I before the question, you are getting a free, low-stakes exposure to the cadential motion. Over hundreds of questions across weeks of practice, this exposure compounds. By the time the explicit chord-progression lessons come around, the cadential ear is already partly built.

2. Cadences disambiguate major and minor. A C major scale and an A natural minor scale share exactly the same notes. A scale-only primer in C major could plausibly be heard as A minor by a listener who is not yet sure where the tonic sits. A cadence — say, C–F–G–C — leaves no such ambiguity. The functional hierarchy makes the tonic obvious in a way that a scale, on its own, does not.

The progression: cadence → scale → drone → silence

There is a useful pedagogical progression of how much help you give the ear before each question, ordered from most supportive to least:

  1. Cadence + scale. Maximum support. Best for early learners.
  2. Cadence only. Slightly less support. The learner has to retain the pitch material from the chord progression alone.
  3. Scale only. A different kind of support — pitch material is given but functional structure is not. Useful for learners who already have strong functional ears and want to focus on pitch-discrimination per se.
  4. Drone (sustained tonic) only. Holds the tonal center but provides no functional context. The learner must internally generate the key from the drone.
  5. No primer. Most demanding. The learner must hold the key from a previously established context. This is the closest to actual real-world musical hearing.

A learner can usefully spend weeks at level 1, weeks at level 2, and so on. Each downward step asks the ear to do more of the work the previous primer used to do. This is consistent with the broader perceptual-learning principle that gradually fading scaffolding produces more independent skill than removing it abruptly [2].

What the apps currently do

Most ear-training tools default to either “scale only” (Functional Ear Trainer, classic Tonedear) or “drone only” (some advanced modes). Cadence-only or cadence+scale options are less common, even though they correspond to the strongest setup for the ear’s perceptual task.

If your tool gives you a choice, start at cadence+scale, work down through cadence-only and scale-only, and end at drone-only and silence as your skill grows. If your tool does not give you that choice, you can self-administer: hum or play a I–IV–V–I before each session, even if the app only plays a scale. The few seconds spent on the cadence reset the tonal frame for everything that follows.

A note on the science of cadences themselves

The pull from V to I is not a Western cultural convention — it is a perceptual phenomenon supported by acoustic and statistical factors. The leading tone (7) is a half-step below the tonic; that proximity creates a perceptual tension that resolves to 1. The dominant chord (V) contains the leading tone and the second scale degree, both of which want to resolve inward. Krumhansl-Schmuckler key-finding algorithms, which match heard pitch-class distributions to learned key profiles, find cadential progressions to be high-information events for key identification [1:1]. Modern statistical learning accounts treat cadence-recognition as one of the earliest tonal skills infants acquire, well before they can name a key [3].

In other words: the cadence is not just a more elaborate way to set up a key. It is the natural way the auditory system extracts a key from a stream of music. Using cadences in practice aligns the trained skill with the way real music presents itself.

A small adjustment, a meaningful return

If you take one thing from this article: when you sit down to ear-train tomorrow, play (or sing, or hum) a I–IV–V–I cadence in your practice key before you start. If your app supports a cadence primer, switch it on. If it does not, do it yourself. The cost is five seconds. The benefit is that every question that follows is being asked of an ear that has been put into the same functional state real music puts it in.

References

  1. Krumhansl, C. L. (1990). Cognitive Foundations of Musical Pitch. Oxford University Press. The probe-tone profile and key-finding algorithm: Krumhansl, C. L., & Schmuckler, M. A. (1986). Key-finding algorithm. In Cognitive Foundations of Musical Pitch. See also: Krumhansl, C. L., & Kessler, E. J. (1982). Tracing the dynamic changes in perceived tonal organization. Psychological Review, 89(4), 334–368. ↩︎ ↩︎

  2. Wulf, G., & Shea, C. H. (2002). Principles derived from the study of simple skills do not generalize to complex skill learning. Psychonomic Bulletin & Review, 9(2), 185–211. On scaffolding fade as a general perceptual-learning principle, see also Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In Metacognition: Knowing about knowing (pp. 185–205). MIT Press. ↩︎

  3. Trainor, L. J., & Trehub, S. E. (1992). A comparison of infants’ and adults’ sensitivity to Western musical structure. Journal of Experimental Psychology: Human Perception and Performance, 18(2), 394–402. https://trainorlab.mcmaster.ca/publications/pdfs/critical_periods.pdf ↩︎

← Back to all articles