← ProjectsLearning & pleasurePDF ↗
Auditory Learning | Music research trackv3.0 · LISTENING HISTORY & PLEASURE

RESEARCH PROPOSAL · BEHAVIOURAL EXPERIMENT AND MODEL TEST

How does learning change the pleasure
of the same musical event?

Amaç Erdem

Independent research proposal · Listening history, controlled learning and model validation

Abstract

The same musical event may feel different after different listening histories. We will combine participant-provided listening records with Musical Intelligence (MI) features from verified recordings to estimate individual exposure profiles. Recognition and continuation tasks will test these estimates. A controlled experiment will then manipulate learned expectations while keeping test audio unchanged and measure pleasure. Separate comparisons will test whether history-informed MI adds predictive value beyond self-report, listening frequency and acoustic baselines. Exposure is treated as evidence about listening experience, not a direct measurement of memory.

Full manuscript mapShared selected inputsF2 PredictionF5 EmotionF6 Reward
Original manuscript architecture with all original nodes and paths. Unselected branches are gray; selected F2 Prediction, F5 Emotion and F6 Reward paths use manuscript colours. Each selected mechanism is labelled at its original coordinate.
SPH Spatiotemporal Prediction Hierarchy
PWUP Precision-Weighted Uncertainty Processing
WMED Working Memory-Entrainment Dissociation
UDP Uncertainty-Driven Pleasure
SRP Striatal Reward Prediction
RPEM Reward Prediction Error in Music
Figure 4. The implemented routes used in the model comparison. Coloured routes use the manuscript's F2/F5/F6 palette; unused routes remain gray. Dashed lines read R³ directly; solid lines pass through T³ or upstream mechanisms. The selected paths use 17 R³ features and 49 temporal keys. PWUP and WMED supply internal values. The 28 September source audit is historical, static evidence. Counts are declarations; archived artwork values are not study results. Regional and chemical outputs are outside the experiment.

1. The unresolved question

Research connects musical pleasure to surprise and uncertainty, but reports different directions of their interaction.1,2 Learning can also create preferences for unfamiliar musical structures.3 We ask how measured listening experience and experimentally learned context relate to the response to an unchanged musical event.

We will manipulate the probability of two target continuations and the uncertainty of the cue that precedes them. Cue meanings will be counterbalanced across listeners. Thus, an identical cue–target recording can have different learned meanings without changing its waveform.

2. Two distinct tests

H1. Contextual uncertainty changes the difference in pleasure between less- and more-expected targets. The primary contrast is a difference of differences, tested in both directions.

H2 · Secondary. MI-derived exposure profiles improve held-out prediction of familiarity, expectation and pleasure beyond self-report and listening-history baselines. H1 tests experimentally assigned learning; H2 tests prediction from observed histories. Neither establishes a unique neural mechanism.

3. What is held constant?

Within each musical family, target A is the same audio file wherever it occurs; the same applies to B. Each target is heard equally often overall. Four cues assign A and B probabilities of either 1/2 or 1/8, under low or high uncertainty. Three other continuations preserve these distributions.

This separates conditional predictability from simple target familiarity. Cue acoustics and target identities are counterbalanced, with sequence and repetition effects included in analysis. Specific cue–target pair familiarity remains linked to probability and limits interpretation.

4. Scope and participants

The planning sample is 192 adult non-musicians, plus a separate pilot of 24. Twelve new musical families will be tested; each listener learns one. English comprehension, headphone screening6 and task practice are required.

Musical experience and reward sensitivity will be measured.4,5 History donation is optional; its absence does not exclude a listener from H1. H2 uses the consenting subset with usable records, with coverage and selection reported. Neither dopamine nor brain activity is measured.

Different learned probabilities, identical test audio, continuous pleasure response
Figure 1. The critical comparison holds the cue–target waveform constant across assigned histories.
7 October 2026 · Proposed design · Not preregistered · No new results1
Auditory Learning | Music research trackv3.0 · LISTENING HISTORY & PLEASURE

5. Learning, listening and reporting

Before testing, consenting listeners provide Spotify, Apple Music or equivalent track records. Names and artists locate candidate recordings; versions are verified before MI analysis. Frequency and recency are used only when present. A title list supports a repertoire profile, not a timed exposure history.

An added recognition and continuation session validates the frozen profile, using separate calibration and held-out items. The controlled experiment retains its planned 50–60 minutes; the pilot will determine the added history-task burden. Pleasure and later secondary responses remain in separate blocks.

CheckConsent · sound · practice→Learn128 listening trials→Test128 pleasure trials→Follow upResolution · replay
Four planned cue distributions over five musical continuations. Targets A and B swap probabilities one half and one eighth. Entropy is 1.41 or 2 bits.
Figure 2. Probability and uncertainty are separated by design. A and B are the focal targets; C–E are fillers. Across four cues, A and B each occur 40 times in both exposure and test. Each test yields 80 target trials: 32 more-expected and 8 less-expected trials at each uncertainty level. Test frequencies preserve training probabilities; cells are not artificially balanced by overplaying rare targets. Cue-to-distribution assignments vary across listeners.
Nine-second trial: three-second cue, one-and-a-half-second target, four-and-a-half-second silent response interval. Baseline is one second before target; response window is one-and-a-half to four-and-a-half seconds after onset.
Figure 3. An event-aligned outcome. Pleasure is sampled against the audio clock at 20 Hz. The fixed response window allows for reporting delay; it does not identify the exact onset of an internal feeling. Interrupted trials and uninitialised responses are flagged.

Continuous pleasure · primary

“How much pleasure do you feel right now?”

0 · none100 · very strong

Only this response is requested during the primary test. A stationary, initialised slider is a valid response.

Predicted continuation · learning check

“Out of 100 occasions, how often would each ending follow this cue?”

ABCDEΣ = 100

Four cue-specific allocations before exposure, after exposure and after the primary test. No outcome feedback.

Perceived resolution · secondary

“Press when the musical idea feels settled.”

◇R · mark a moment

Sixteen later trials. Closure is kept distinct from pleasure; non-response is not automatically failed attention.

Replay preference · secondary

“Which sequence would you like to hear again?”

▷ Sequence 1▷ Sequence 2

Eight pairs; one choice is implemented at the end. Curiosity and pleasure may make different predictions.

7 October 2026 · Proposed design · Not preregistered · No new results2
Auditory Learning | Music research trackv3.0 · LISTENING HISTORY & PLEASURE

6. What Musical Intelligence can explain

The original audio architecture is retained. R³/T³ descriptors and six existing C³ outputs form candidate representations. A separate, proposed listener-profile layer combines these measurements with observed listening history; it is not an implemented memory mechanism.

Table 1. Prespecified MI candidates. Incremental prediction is tested jointly; named-mechanism interpretations remain exploratory.
SPH.P1.prediction_errorAudio-derived prediction-error candidateUDP.P0.context_assessmentAudio-derived context candidateUDP.P2.reward_computationPleasure-related candidate
SRP.P0.wantingApproach-related candidateSRP.P1.likingPleasure-related candidateRPEM.current_rpeReward-error candidate, not measured RPE
Audio representation + listener history. Frozen MI returns the same scores for the same recording. The proposed profile layer adds track identity, repeat exposure, recency and feature similarity where available. Its aggregation rules are fitted in development data and frozen before held-out testing. Profile scores are hypotheses about familiarity, not memories or autobiographical diagnoses. The current MI comparison remains offline.
7 October 2026 · Proposed design · Not preregistered · No new results3
Auditory Learning | Music research trackv3.0 · LISTENING HISTORY & PLEASURE

7. Analysis, decisions and next steps

7.1 One primary behavioural test

For each target trial, we calculate the change in pleasure specified in Figure 3. A mixed-effects model estimates probability, uncertainty and their interaction, accounting for listeners, musical families, recordings and trial order. H1 tests the interaction at two-sided α = .05; effect sizes and 95% intervals are primary reporting units.

Probability allocations test whether learning occurred. Main-study listeners are not excluded for weak learning or low pleasure. Failed learning limits interpretation of a null H1. Resolution and replay are secondary; neither establishes causal mediation.

7.2 Controlled-learning comparison

Four models predict the same pleasure outcome: A, R³/T³ acoustics and timing; B, A plus six MI scores; C, A plus a sequential cue–continuation learner; D, C plus the MI scores. B–A tests MI's audio representation; D–C tests its contribution beyond an explicit learning account (Figure 5, figure atlas).

The learning model receives the actual ordered exposure history. All models share matched acoustic access and fitting budgets. Entire musical families and their listeners are withheld together; tuning uses development folds only. The primary predictive metric is held-out mean absolute error. Predictive gains do not validate a named brain mechanism.

7.3 Pilot and registration gate

The pilot uses separate musical families to check learning, discrimination, timing and burden. N = 192 is a planning figure, not a power result. Simulation must support detection of a four-point interaction on the 0–100 pleasure scale before the final sample is registered. Stimuli, randomisation, exclusions, windows, code and model versions are frozen before recruitment.

7.4 Listening-history validation

Compare self-report alone, self-report plus available listening counts/recency and artist/genre baselines, then the same model plus an MI-derived exposure profile. Profile ablations use R³, R³ + T³, and R³ + T³ + the six C³ candidates, with matched fitting budgets. Freeze histories before testing; evaluation items never tune the profile.

Primary history validation is recognition discrimination on held-out familiar candidates and newly created, matched foils; continuation prediction and pleasure are secondary outcomes. "Absent from the export" does not mean unheard. Hold out listeners and recording families; report matching coverage and sensitivity to omitted tracks. Predictive improvement is not a causal effect of past listening.

7.5 Recording identity and retention

Match title, artist, recording version, duration and identifiers where available. Catalogues and public pages, including YouTube, may locate candidates; uncertain, live, cover or remastered substitutes are flagged. Extraction requires an authorised audio source and applicable-use review; academic purpose and deletion alone do not establish permission.

No audio is redistributed. Temporary analysis audio is deleted after extraction and quality checks, within a planned seven-day maximum. Retain only necessary pseudonymous features, recording identifiers, checksums, match decisions and extraction/deletion logs under the consented retention policy. Experimental playback has separate source and retention requirements.

7.6 Readiness

The history layer is proposed. Record import, matching, permitted audio access, extraction, behavioral calibration and its separate precision analysis remain to be completed. The app illustrates the design and collects no history or responses. Ethics approval and registration precede recruitment; neither memory diagnosis nor autobiographical inference is claimed.

Selected references

1. Cheung et al. (2019). Uncertainty and surprise jointly predict musical pleasure. Current Biology. 10.1016/j.cub.2019.09.067.

2. Mas-Herrero & Marco-Pallarés (2025). Predictive processes shape individual musical preferences. PNAS. 10.1073/pnas.2500494122.

3. Kathios et al. (2024; online 2023). Generating new musical preferences from multilevel mapping of predictions to reward. Psychological Science. 10.1177/09567976231214185.

4. Müllensiefen et al. (2014). The musicality of non-musicians. PLOS ONE. 10.1371/journal.pone.0089642. 5. Mas-Herrero et al. (2013). Individual differences in music reward experiences. 10.1525/mp.2013.31.2.118.

6. Woods et al. (2017). Headphone screening to facilitate web-based auditory experiments. 10.3758/s13414-017-1361-2. 7. Albury et al. (2026). Stimulus predictability and liking enhance auditory–motor encoding and memory for melodies. 10.1111/nyas.70310.

Model provenance. MI Systems Neuroscience manuscript, 22 September 2026; static source audit, 28 September 2026. The original map and channels are unchanged. The v3 listening-history specification distinguishes proposed additions from implemented audio routes. No engine or listener-profile validation is claimed.

INTERACTIVE FIGURE · DESIGN ILLUSTRATION