Playlist Specifications

On this page

Ten playlists, one clear job each. Every design rule traces to a verdict in 01-evidence-review.md (cited as §-references).

Two kinds of spec:

Global rules (all playlists): - Volume: background music means background - just above audibility for work playlists. Salience scales distraction (§1.3). - Familiarity: work playlists should be familiar (novelty captures attention, §1.2); emotional and creative playlists can afford novelty. - Don't mix jobs. The moment a focus playlist becomes emotionally interesting, it has failed at its job.


1. Deep Work - Verbal [ENGINE] - writing, reading, hard debugging

The strictest spec. Verbal working memory must be protected: unpredictable sound and lyrics disrupt it even when ignored (Salamé & Baddeley 1982; Perham & Currie 2014; §1.2, §2.1).

# recipe parameters
tempo_bpm: null            # unpulsed; motion from slow LFO drift only
mode: ionian_or_lydian     # consonant, low-tension
attack_floor_ms: 150
dynamic_range_max_db: 6
melodic_salience: minimal  # no foreground melody
texture: 2-3 sustained layers (pad, sub drone, sparse shimmer)
lufs_target: -23

[PERSONAL] equivalents / search terms: "ambient drone," "textural ambient," Stars of the Lid, Eluvium, Brian Eno's ambient series, Max Richter's quieter work, "deep focus instrumental." Avoid "epic study music" - cinematic dynamics violate the spec.

2. Deep Work - Analytical [ENGINE] - routine coding, math, spreadsheets

Lower verbal load tolerates a steady pulse; the pulse aids time-on-task feel. Evidence status: extrapolated from the task-complexity findings (Gonzalez & Aiello 2019, JEP: Applied); there is no strong coding-specific literature (§2.2).

tempo_bpm: 95-115
mode: dorian_or_aeolian
attack_floor_ms: 30       # soft ticks allowed, no sharp snares
dynamic_range_max_db: 8
texture: pulse + bass + pad, one slowly-evolving voice
section_change_every_bars: 32   # slow, predictable evolution
lufs_target: -20

[PERSONAL] equivalents: minimal techno (Kompakt label), dub techno, Tycho, Bonobo instrumentals, "minimal techno focus," "dub techno mix."

3. Admin / Shallow Work [PERSONAL] - email, chores, expense reports

The one work context where stimulating music with lyrics is evidence-aligned: salient music helps simple, boring tasks (Gonzalez & Aiello 2019; Kämpfe et al. 2011; §2.3).

4. Creative / Diffuse Thinking [PERSONAL, ENGINE-assisted]

Mood scaffolding around ideation, not during verbal drafting. Evidence status: mixed; background music has even impaired verbal insight problem-solving (Threadgold et al. 2019, Applied Cognitive Psychology; §2.4).

5. Downshift - Stress Reduction [ENGINE]

The strongest clinical literature: slow, predictable, low-dynamic music reduces physiological and psychological stress (de Witte et al. 2020 meta-analyses, Health Psychology Review; §2.5).

tempo_bpm: 60-80          # felt as slow swells, not a beat
mode: ionian
attack_floor_ms: 300
dynamic_range_max_db: 5
arc: descending           # spectral centroid + density fall over the render
duration_default_min: 25
lufs_target: -24

[PERSONAL] equivalents: "slow ambient," Hammock, Ólafur Arnalds' quiet pieces, Hiroshi Yoshimura, "60 bpm relaxation instrumental."

6. Emotional Processing [PERSONAL]

Deliberate, time-boxed listening - never background. Sad music reliably provides consolation and emotion regulation for many listeners, moderated by empathy and rumination tendency (Taruffi & Koelsch 2014, PLOS ONE; §2.6).

7. Walk / Recover [PERSONAL]

The best-supported positive effect in the whole literature: better affect, performance, and lower perceived exertion across 139 studies (Terry, Karageorghis et al. 2020 meta-analysis, Psychological Bulletin; §2.7).

8. Sleep Wind-Down [ENGINE]

Routine + relaxation, honestly framed: music improves subjective sleep quality in insomnia with moderate-certainty evidence (Jespersen et al. 2022 Cochrane review; §2.8). Part of a consistent pre-bed sequence, not an all-night stream.

tempo_bpm: 50-65          # barely-felt motion
mode: ionian_low_register
attack_floor_ms: 500
dynamic_range_max_db: 4
arc: descending_to_silence
duration_default_min: 40
lufs_target: -28          # markedly quiet; player volume low too

[PERSONAL] equivalents: "sleep ambient no melody," Green-House, Grouper's quieter side, "drone sleep music" (skip anything labeled with Hz claims - see §3.2).

9. Mood Lift [PERSONAL]

Deliberate short-term mood raising. The evidence needs both halves: upbeat music and the intention to feel better; passive play didn't move mood (Ferguson & Sheldon 2013, two experiments; moderate evidence for lasting effects; §2.9).

10. Intimacy [PERSONAL]

Honest evidence status: thin and this doc says so. One lab-demonstrated mechanism, excitation transfer (Marin et al. 2017, PLOS ONE), plus conditioning and self-consciousness masking; no validated genre/tempo prescription and absolutely no "aphrodisiac frequency" (§2.11).


Quick-reference

State Playlist Vocals Pulse The one rule
Writing/reading 1 Verbal never none protect verbal memory
Routine coding 2 Analytical never steady predictability
Email/chores 3 Admin yes any raise arousal
Ideation 4 Creative breaks only loose mood, not soundtrack
Stressed 5 Downshift no slow swells descend
Heavy day 6 Processing yes any time-boxed, on purpose
Walking 7 Walk yes fast motivation is legal here
Pre-bed 8 Sleep never minimal same time, every night
Slumped 9 Mood Lift yes upbeat intention + music, then move on
Intimate 10 Intimacy yes slow groove shared history beats acoustics