MusIQ is split into five short sections, each aimed at a different facet of musical perception. You start with tempo and the rest unlock as you go. Together they take most people about eight to twelve minutes. Here's what each section is actually measuring and how your score comes together at the end.
Two of the sections — pitch sensitivity and counting voices — depend on you hearing fine spectral detail. Laptop speakers smear the highs, Bluetooth earbuds introduce compression artifacts and latency, and a noisy room swallows the subtler differences between tones. Wired over-ear or in-ear headphones in a quiet room give the test a fair chance to do its job. If you're on speakers or Bluetooth, expect your score to under-represent what you can actually hear.
The two rhythm sections are latency-sensitive in the other direction: Bluetooth audio is often tens of milliseconds late, which will make your taps register as consistently "early." Wired is better for rhythm sections too.
You hear a short musical excerpt at a specific tempo (120, 85, 150, 96, or 172 BPM). The music fades out, and you keep tapping the beat in quarter notes until you hear a ding. We measure how steady and how accurate your taps are relative to the true tempo over the silent window.
What's being tested: internal pulse stability. A steady internal clock is foundational for ensemble playing, conducting, and just being able to groove. It's surprisingly uncorrelated with instrument-specific skill — plenty of accomplished players struggle when the metronome goes silent.
You hear two tones in sequence. You decide whether the second is higher or lower than the first. The intervals start obvious and get progressively finer — eventually well under a semitone. Desktop users press p for higher and l for lower; mobile users tap the top or bottom half of the screen.
What's being tested: raw frequency discrimination, independent of musical context. This is closer to a psychoacoustic threshold than a music-theory question. Absolute pitch is not required; the test is purely relative.
You hear a short chord or cluster. You decide how many distinct simultaneous voices are sounding (one to six). You can tap Listen again once per clip if you need it.
What's being tested: harmonic parsing — the ability to hold multiple simultaneous pitches as separate objects in your head rather than fusing them into a single timbre. Trained ears do this automatically; untrained ears tend to underestimate, especially as density grows. This is where headphones matter most.
You hear a short melody exactly once. You then reconstruct it by clicking cells on a grid — pitch on one axis, time on the other. You can play your own submission back as many times as you want, but the original only plays once, so you have to lock the shape of it in working memory on the first pass.
What's being tested: auditory short-term memory for pitch contour. This is the ability that lets you hum a song back after hearing it once, and it's a strong predictor of ear-training outcomes.
You hear four count-in clicks followed by a rhythmic pattern. The clicks continue, and when the window turns green, you tap the pattern back. A visualization shows your taps in blue over the original in black so you can see your own timing.
What's being tested: rhythmic transcription — locking a heard pattern into your motor system and reproducing it in time. It combines memory, pulse, and motor control in a way that isolates none of them individually. Drummers and good accompanists tend to shine here.
Each section produces its own subscore. The final MusIQ score is a weighted combination of the five, plus a small amount of cross-section consistency (getting a similar rank in every section is slightly rewarded over being a one-trick specialist). Your percentile is computed against the full leaderboard — so "top 10%" means you outscored 90% of people who have finished the full test.
Your archetype is assigned from the shape of your subscores, not the total. Two people with identical final scores can land on different archetypes if one is a pitch specialist and the other is a rhythm specialist.
Once you finish all five sections, you'll see a prompt asking whether you'd like to view a short rewarded ad in exchange for revealing your final score. You can always say no and still see your score — rewarded ads are opt-in, and that's the whole point of the format. The ad is served by Google and is subject to their privacy practices and ours.
If something didn't score the way you expected, or you're curious about the methodology, drop me a line on the contact page.