Sound · Mind · Meaning
Amaç Erdem
one question — six disciplines
What is Music?
For 15 years the question would not leave me.
No satisfying answer from music theory, art, or philosophy.
How does the brain process music?
For an answer, a year of reading and research became unavoidable.
Perception, prediction, reward, neurochemistry — twenty-one papers, load-bearing.
A theory of cortical responses
Free-energy formulation of perception as approximate Bayesian inference — formalizes precision-weighted prediction errors that drive belief updates throughout cortex. The mathematical scaffold underpinning every Bayesian-update step in MI's F1 cycle.
Anatomically distinct dopamine release during anticipation and experience of peak emotion to music
[11C]raclopride PET in 8 listeners: caudate DA release precedes nucleus-accumbens DA release during musical “chills.” The anatomical-temporal dissociation is the load-bearing falsifiable prediction MI re-derives at +0.9 s on 52/56 DEAM tracks.
Dopamine modulates the reward experiences elicited by music
Within-subject pharmacological design (n=27): levodopa enhanced hedonic ratings and skin conductance; risperidone blunted both — causal evidence for DA in music-evoked reward.
Music in the brain
Comprehensive review framing music perception as predictive coding across cortical hierarchies. Precision-weighted prediction errors organize melodic, harmonic, and rhythmic processing — the conceptual ground for MI's F2 prediction function.
Pleasurable music activates cerebral μ-opioid receptors
[11C]carfentanil PET-fMRI: pleasurable listening shifts μ-opioid binding across cingulate, OFC, NAcc, thalamus — validating MI's OPI channel (7/7 regions).
Tonal consonance and critical bandwidth
Local consonance and the relationship between timbre and scale
Local consonance curve as sum of dissonance from partial-pair interactions — predicts scale construction from arbitrary inharmonic timbres. R³'s BCH backbone.
Timbral effects on consonance disentangle psychoacoustic mechanisms
N=11,754 dyad ratings across 5 timbral conditions: pure tones show no preference for traditional consonance ratios; complex tones do. Disentangles innate vs. statistical components — the load-bearing Phase 6 anchor for MI's consonance battery.
Uncertainty and surprise jointly predict musical pleasure
Inverted-U relationship between Bayesian uncertainty × surprise and pleasure ratings. Yields the constants behind MI's F6 reward formula (surprise × resolution).
Auditory expectation: the information dynamics of music perception
Brain correlates of music-evoked emotions
Maps emotional response to hippocampus, insula, OFC, mPFC — separates core arousal/valence circuits from refined aesthetic emotion. MI's F5 anatomy reference.
When the brain plays music: auditory-motor interactions in music perception and production
The rewards of music listening: response and physiological connectivity of the mesolimbic system
fMRI demonstrating coordinated mesolimbic activation (NAcc, VTA, hypothalamus, amygdala) during pleasurable music listening. Reference for MI's F6 reward network.
Interactions between the nucleus accumbens and auditory cortices predict music reward value
On Repeat: How Music Plays the Mind
Cognitive-science monograph: repeated listening drives parsing, anticipation, and aesthetic depth. Connects repetition to memory consolidation — informs MI's F4 recurrence handling.
Music, Language, and the Brain
Cross-domain analysis: music and language share processing resources for syntax, rhythm, and prosody. Reference for MI's F3-F4-F5 boundary decisions.
Indifference to dissonance in native Amazonians reveals cultural variation in music perception
Tsimané (Bolivia) listeners show no preference for consonance over dissonance — cultural-statistical origin of consonance preferences; boundary on universalism claims.
Subcortical sources dominate the neuroelectric auditory frequency-following response
The statistical structure of human speech sounds predicts musical universals
Simultaneous consonance in music perception and composition
Bayesian psychophysics on inharmonic / stretched-octave stimuli: Western harmony judgments depart from pure ratios; spectral interference dominates. Anti-overfit anchor.
Musical sound features & auditory cortical processing
Maybe to truly understand it, I should build one.
Build a musical brain.
Starting as a composer with zero technical training in April 2025, an enormous amount of work had to be internalized in 11.5 months.
-
Computer Science & Software Architecture
- Python — 756 files, 2.81M lines of code
- Modular contracts, dataclasses, ABCs, type hints
- Three-layer pipeline: R³ → H³ → C³
- Git, frozen SHA pinning, reproducibility audits
-
Signal Processing & Psychoacoustics
- DSP — STFT, MFCC, spectral flux, F0 estimation
- Critical bandwidth models (Plomp-Levelt, Sethares)
- Roughness, dissonance, harmonic-product spectrum
- 97-dimensional perceptual feature vector design
-
Cognitive Neuroscience
- F1–F8 functional architecture (sensory → learning)
- 89 mechanisms, 131 beliefs, Bayesian update cycle
- Predictive coding hierarchies (Vuust 2022)
- 26D Regional Activation Map, 529 RegionLinks
-
Neurochemistry
- Dopamine, Norepinephrine, μ-Opioid, Serotonin dynamics
- Caudate → NAcc dopamine sequence (Salimpoor 2011)
- Pharmacological cross-validation (Ferreri 2019)
- 54 NeuroLink call sites, anatomically grounded
-
Bayesian Statistics & Validation
- Precision-weighted Bayesian belief update
- Hierarchical Benjamini–Hochberg FDR correction
- Permutation null testing (p<0.0001 anatomy)
- Bit-exact reproductions across 9-dataset battery
-
Philosophy of Science
- Zero-calibration doctrine — 16,191 untuned constants
- Pre-registration, hypotheses frozen before data
- Falsifiability — load-bearing, kill-switch predictions
- 21 papers internalized as mechanisms, not citations
Musical Intelligence — a system that hears like a mind.
Three layers: R³ perception → H³ memory → C³ cognition.