Methodology.

Everything aurakin tells you about yourself is derived from your listening data. No personality quiz, no self-report (unless you run calibration). It reads your top artists, tracks, genres, and listening behavior and pattern-matches them against decades of published research correlating music preference with personality.

The frameworks we use

three layers

MUSIC model (Rentfrow, Goldberg & Levitin, 2011)
Five sonic dimensions on a 0-100 scale: Mellow (smooth, acoustic), Unpretentious (sincere, accessible), Sophisticated (complex, cerebral), Intense (loud, forceful), Contemporary (rhythmic, current).
Big Five / OCEAN (Rentfrow & Gosling, 2003 + replications)
The gold standard in academic personality psychology. Five percentiles: Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism. More rigorous than MBTI, anchored in decades of research.
Empathy / Systemizing (Greenberg et al., Cambridge, 2015-2022)
Two orthogonal cognitive styles. Empathizers read emotional and social states; systemizers find structure and patterns. Most people lean one way; about a fifth are balanced.

How we derive your scores

the path from music to persona

Your top 50 artists, 50 tracks, 30 genres, and listening-behavior stats go to an LLM prompted with the three frameworks above plus the specific research correlations. It pattern-matches the signal against known trait-to-taste correlations. For example, Rentfrow & Gosling 2003 found that high openness correlates with complex, reflective music — sophisticated genre preferences. High extraversion correlates with upbeat, rhythmic, contemporary music. Lyric-focused listening correlates with empathy; production-focused listening with systemizing.

The LLM returns your 11 scored dimensions plus the closest of 16 archetypes synthesized from the combined score space. If your listening spans two archetypes, it returns a secondary with a strength rating.

What we're honest about

precision caveats

Ballpark, not diagnostic.
Your 72nd-percentile openness is probably in the 60-80 range if measured by a standard Big Five questionnaire. Not 50, not 95. We infer, we don't measure.
Directional, not precise.
"You lean empathizer" is a robust claim. "Your empathy is 67th and systemizing is 51st so you're 16 points more empathizer" is false precision. Trust the direction, not the exact margin.
Research effect sizes are modest.
The Big Five × music correlations sit at r = 0.2-0.4. That's meaningful but not deterministic. Lots of variation within any group.
Calibration closes the gap.
Running the calibration quiz adds your self-report to the signal. Post-calibration scores are materially more accurate than music alone.
The deep cut is interpretive.
The 10-section behavioral profile is LLM-authored prose grounded in the Big Five × behavior literature. Specific scenarios are educated extrapolation. Cite, don't swear.

Research we lean on

start here if you're curious

Rentfrow & Gosling (2003). The Do Re Mi’s of Everyday Life: The Structure and Personality Correlates of Music Preferences. Journal of Personality and Social Psychology.
Rentfrow, Goldberg & Levitin (2011). The structure of musical preferences: A five-factor model.
Greenberg, Baron-Cohen, Stillwell, Kosinski & Rentfrow (2015). Musical preferences are linked to cognitive styles. PLOS ONE.
Greenberg et al. (2016). The song is you: Preferences for musical attribute dimensions reflect personality.
Baron-Cohen (2003). The Essential Difference: The Truth About the Male and Female Brain.

What we don't do

bright lines

We don’t sell your listening data or your persona.
We don’t tell you who you “really” are. We tell you what the research says about people who listen like you do.
We don’t use pseudoscience. Everything above is peer-reviewed psychology.
We don’t make clinical claims. This isn’t therapy or diagnosis.

Derived, not measured. Calibrate to sharpen.

Try it →