I've been practicing on an Ear Trainer program for my iPhone, and I've noticed a distinct pattern. The way the program works you tell it what intervals/scales/chords you want to practice, and it'll play two variants of it and have you identify one of them. For example, if I'm practicing seconds then it'll choose a random root note, then play a minor and a major second from that root and have me identify the major second.

What I've noticed through this practice is that while I'm usually pretty consistent at identifying the intervals when ascending, even when I'm confident about my responses I seem to do little better than just random guessing when the intervals are played descending.

On the one hand, this helps explain why I've always had trouble tuning by ear, on the other hand I'm wondering why it's so hard to tell intervals I otherwise know cold when they're descending instead of ascending. Can anyone offer some insight into this?


3 Answers 3


Picking notes 'by ear' is only 50/50 if you are tone deaf. For those of us not tone deaf, this is something that will improve with practice, which is why the Ear Trainer was built in the first place.

But, you do have a slightly increased chance of getting it wrong when descending, and there is a reason. The short answer is, it's music theory, and you need to change 'how' you are hearing it when descending.

Specifically, the first (random) note you hear when ascending is always the 'root' note of your interval. Thus, picking the half step or whole step above is fairly straight-forward because the next note is the 'relative' note to the first one. In this case, you are picking either the II or II flat of the I.

However, when descending, the first (random) note you hear is actually 'relative' to the second note you need to pick. So, when descending, the second note is the I, but you were only given a relative (either II or flat II). Thus, the task of picking the 'root' is more difficult, partly because you are thinking of the first note as the 'root'.

If you do insist on thinking of the first note of your descending interval as the root, then you have to pick the VII (major or dominant) below your root, making a third inversion out of your interval, thus complicating the task.

If you hear that when descending, the first note is your relative note and not your (root) I, you'll have an easier time picking the second note of the interval.

  • what I meant was if there are only 2 choices for an answer, the random chance of being correct is 50/50.
    – Tim
    Aug 1, 2014 at 7:17

With two choices for an answer, it's 50/50. Intervals are generally considered using the lower note as a basis. Thus C-D =maj 2, C-Db= min2. Using the first note heard (which is higher) as a datum point is unusual, as ears tend to gravitate towards the lower note as a start.Thus, I would hear , say C-Bb as a minor seventh, even though it is an octave out, rather than a major 2, which could be argued- Bb-C. We're on the 'rule of 9', with maj turning into min, and aug. into dim.

There's also an extra problem likely here, where if you hear, e.g. C-Eb/D#, is it a min 3 or an aug 2 ? Sound-wise, who knows...


As said, the lower tone is the basis for working out the interval's name. If you are trying to work out the interval of a descending step e.g C followed by G below, simply reverse the order of the two tones in your mind, and you'll know your answer. Nothing more complicated than that!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.