Skip to content

Research — Breathing patterns + sleep/relaxation audioGuided sessions

Date: 2026-06-10 Scope: Net-new breathing patterns and sleep/relaxation audioGuided sessions to close two gap-analysis findings: "relaxation thin on active stress management" and "sleep missing actionable wind-down". These are the doing exercises (paced or narrated), distinct from written insight cards (handled elsewhere). Origin tag: research:breathing-sleep-relaxation.

This doc grounds the rows in sessions/sessions-backlog.csv. IP/copyright status is load-bearing — every audio row carries an ip_status. Where a script-detail or clearance question is involved, cross-reference ../session-audio-protocols-ip-guide.md.

Why these, and why distinct from what exists

Existing breathing rows: diaphragmatic-breathing, 4-7-8-breathing, crisis-breathing, box-breathing. Existing relaxation/sleep-relevant audioGuided: progressive-muscle-relaxation, autogenic-training, body-scan, guided-imagery-visualisation, 3-minute-breathing-space.

The gap is active stress-physiology downregulation (a standalone, repeatable sympathetic-reset tool rather than a 15-minute relaxation protocol) and actionable sleep wind-down (CBT-I behavioural components and a sleep-onset attention task — the existing sleep variants are only modifications of relaxation scripts, not sleep-specific sessions in their own right). Each row below is genuinely distinct from the existing set; duplicate-slug check done against the lists above.

The breathing-vs-audioGuided boundary follows the authoring standards: a breathing session is pacer-driven (a visual breath-pacer does the work; copy frames the mechanism), while an audioGuided session is narrated (the user closes their eyes and a voice runs a structured protocol). Sessions whose core is a counted breath pattern are breathing; sessions whose core is narrated guidance are audioGuided.

Breathing patterns (pacer-driven)

1. Physiological Sigh — standalone (physiological-sigh)

  • Technique: Double inhale through the nose (a second short top-up inhale on top of the first), followed by a long, slow exhale through the mouth. Repeated for ~1–3 minutes. The second inhale re-inflates collapsed alveoli and the extended exhale offloads CO₂, driving the fastest known voluntary shift toward parasympathetic dominance.
  • Distinct from crisis-breathing: crisis-breathing is the spike bundle (sigh + box + 5-4-3-2-1 grounding for acute distress). This is the standalone, repeatable micro-tool — a 60–90s reset usable proactively (pre-meeting, mid-commute, on waking), not only at a spike. Different framing, different use-case cluster, separate pacer entry.
  • Evidence: Balban, Yilmaz Balban et al. 2023, Cell Reports Medicine (the cyclic-sighing RCT, n=111): a daily 5-min exhale-focused cyclic-sighing practice produced greater mood improvement and respiratory-rate reduction than mindfulness meditation over 28 days, with measurable effect after a single session. Same study indexed as PMC9873947 in the IP guide.
  • IP: Physiological mechanism — unprotectable / public domain. The "physiological sigh" is a descriptive physiological term, not a brand. Write own narration; do not reproduce Huberman-Lab scripted text. ip_status: public-domain technique / original script.
  • Sources:
  • Balban et al. 2023, Cell Reports Medicine — PMC9873947 (open dataset: Stanford/Dryad 10.5061/dryad.mpg4f4r0v)
  • Stanford Medicine summary — https://med.stanford.edu/news/insights/2023/02/cyclic-sighing-can-help-breathe-away-anxiety.html

2. Extended-Exhale / Coherent Breathing (extended-exhale-breathing)

  • Technique: Slow, even breathing at ~5.5–6 breaths/min (the population-average resonance frequency), with the exhale equal to or longer than the inhale (e.g. 5-in/5-out for coherence, or 4-in/8-out for an exhale-weighted calming variant). At this rate the cardiovascular and respiratory systems enter resonance, maximising heart-rate variability and baroreflex gain.
  • Distinct from existing rows: box-breathing is 4-4-4-4 with breath holds; diaphragmatic-breathing is the foundational belly-breath taught first; 4-7-8 is a fixed bedtime ratio with a 7s hold. This row is the hold-free, paced resonance/slow-breathing pattern — the evidence-based "active stress management" daily practice the gap analysis flagged as thin. It is the pattern with the strongest HRV-biofeedback literature.
  • Evidence:
  • Steffen et al. 2017, Frontiers in Public Health 5:222 (open access, CC BY) — resonance- frequency breathing (~6/min) produced higher positive mood, elevated LF/HF, and lower systolic BP reactivity vs other rates. PMC5575449
  • Lehrer & Gevirtz resonance-breathing review — PMC7578229 (also cited by the IP guide and breathing standard).
  • IP: Coherent / resonance breathing (Lehrer & Gevirtz lineage) is public-domain technique; "Coherent Breathing®" as a branded product name (Stephen Elliott) should be avoided — use the generic descriptor "coherent" or "resonance" breathing and write own copy. ip_status: public-domain technique / original script; avoid the "Coherent Breathing®" brand name.
  • Tinnitus framing: extended exhale interrupts the tinnitus→anxiety→more-tinnitus loop (same frame as diaphragmatic), positioned as a daily resilience practice rather than acute relief.

3. Alternate-Nostril Breathing — Nadi Shodhana (alternate-nostril-breathing)

  • Technique: Using the thumb and ring finger to alternately occlude one nostril, inhale through one side, switch, exhale through the other, and continue alternating. Naturally slows the breath and adds a tactile attentional anchor.
  • Defensibility: Included with a moderate evidence base (see below) and a clear attentional-anchor rationale that suits tinnitus (gives attention a competing, neutral somatic task). Marked needs_confirm = tier/category and priority v1 rather than MVP, because the evidence is weaker and more heterogeneous than the exhale-based patterns, and because it requires hand involvement (not eyes-open/commute compatible). Defensible to ship, but lower-confidence than rows 1–2.
  • Evidence: A randomized clinical trial in hypertensive patients found Nadishodhana (and Bhramari) improved HRV, auditory reaction time, and blood pressure (Kumari et al. 2023, J Ayurveda Integr Med) — PMC10388195, PubMed 37499590. A systematic review of 44 RCTs found reasonably consistent reductions in resting heart rate and systolic BP, with mixed HRV parameter findings — Int J Res Med Sci (msjonline.org/index.php/ijrms/article/view/3581). Evidence is suggestive, not definitive; the row's copy must not over-claim.
  • IP: Ancient pranayama practice — public domain. Write own script; attribute to pranayama roots, not to any modern teacher's branded course. ip_status: public-domain technique (pranayama) / original script.

Sleep / relaxation audioGuided (narrated)

4. CBT-I Bedtime Wind-Down (cbt-i-wind-down)

  • Technique: A narrated wind-down routine assembling the actionable behavioural components of CBT-I — the gold-standard, first-line treatment for chronic insomnia and the most evidence-backed psychological treatment for tinnitus-related sleep disruption: (a) a buffer-zone wind-down (lower arousal, dim light, off-screen before bed); (b) stimulus control framing (bed = sleep; if awake and wired, get up — narrated as guidance, app cannot enforce); (c) a brief constructive-worry / cognitive offload step (move tomorrow's worries to a fixed earlier "worry slot", not the pillow). This is the actionable wind-down the gap analysis found missing — distinct from the relaxation scripts' generic "sleep variant" modifications.
  • Distinct from existing: the existing sleep coverage is only sleep-onset modifications applied to PMR / body-scan / autogenic scripts (per the IP guide's "Sleep-onset modifications" template). No session currently delivers the CBT-I behavioural components. This fills that gap.
  • Evidence:
  • CBT-I as first-line for chronic insomnia — Sleep Foundation overview; Cleveland Clinic; summary of best evidence PMC12897499.
  • CBT-I ~80% response rate for tinnitus insomnia and sleep mediating the tinnitus→anxiety pathway — see docs/session-content-evidence-base.md §4d and §5a (vault #90, #342, #405).
  • Constructive worry as a CBT-I component with insomnia-worry as mediator — Jansson-Fröjmark et al., Behav Res Ther (sciencedirect.com/science/article/abs/pii/S1389945713015815).
  • IP: CBT-I is an academic clinical framework — components (stimulus control, constructive worry, sleep hygiene) are unprotectable and freely teachable. Write own narration; do not reproduce any specific clinician's scripted workbook text. ip_status: academic/clinical framework (CBT-I), unprotectable / original script.

5. Sleep-Onset Relaxation — Cognitive Shuffle (sleep-onset-relaxation)

  • Technique: A sleep-onset attention task: the voice guides the user through serial diverse imagining — picturing a sequence of unrelated, neutral, concrete images (one object, then an unrelated one, then another), deliberately not connecting them. This mimics the disordered, image-rich "micro-dream" mentation that naturally precedes sleep and crowds out the rumination/arousal loop (and, for tinnitus, gives attention a competing stream so the sound is not the only signal in the quiet). Applies the IP guide's sleep-onset modifications (5–8s pauses, lower pitch, no re-arousal close — ends with "let it dissolve", never "open your eyes").
  • Distinct from existing: not a relaxation protocol (no tense-release, no body sweep) and not guided imagery toward a single safe place — it is the opposite, deliberately diverse and non-narrative. No existing row delivers a cognitive-offload sleep-onset task.
  • Evidence: Beaudoin 2014, A design-based approach to sleep-onset and insomnia: super-somnolent mentation, the cognitive shuffle and serial diverse imagining (SFU Summit, open: summit.sfu.ca/item/17237); somnolent-mentation theory and the SDI task. Note the evidence base is theory-driven and early-stage (not yet large RCTs) — copy must frame it as a practical attention technique, not a proven cure.
  • IP: The technique/theory (serial diverse imagining, somnolent mentation) is academic and freely teachable — write own script. CAUTION / FLAG: "Cognitive Shuffle" and "mySleepButton" are associated with a commercial app by the same author (Beaudoin). Do not use "mySleepButton" and avoid leaning on "Cognitive Shuffle™" as a product name in-app; use a neutral descriptor ("sleep-onset imagining" / "image drift") and cite the academic paper. ip_status: academic technique (serial diverse imagining), unprotectable / original script; avoid the "mySleepButton" / "Cognitive Shuffle" product branding — needs editorial naming review.

6. Sound Enrichment for Sleep (tinnitus-specific) (sound-enrichment-for-sleep)

  • Technique: A guided bedtime session that coaches the user to set up low-level background sound (fan, soft pink noise, nature sound, or a Naluma sound album) at the bedside — then narrates settling into it. The key clinical distinction, narrated explicitly: enrichment, not masking — keep the level just below the tinnitus so it reduces the silence-vs-tinnitus contrast without drowning the sound out (over-masking does not support long-term habituation and can make tinnitus seem louder when removed). Tinnitus is louder at night because the quiet removes competing signals; bedside sound reduces that contrast and has been associated with improved sleep latency and reduced nocturnal tinnitus awareness.
  • Distinct from existing: no existing session is a sleep-specific sound-enrichment setup coach. This deep-links to the app's sound library (cross-references the sounds/ content) and is the bedtime bridge between sound therapy and sleep — uniquely tinnitus-relevant.
  • Evidence:
  • Sound therapy/masking review (Cochrane) — PMC7390392 (limited certainty; enrichment-not-masking principle).
  • Tinnitus/anxiety/sleep multimodal review — PMC12474699 (also cited in the evidence base for the stress→tinnitus link).
  • Enrichment-below-tinnitus and habituation principle — TRT counselling protocol PMC8632517; see session-content-evidence-base.md §4c–4d.
  • IP: Sound-enrichment principle is from freely available TRT/audiology literature and VA PTM materials (US-gov public domain) — describe the model, do not reproduce a proprietary protocol. ip_status: clinical principle from open literature (TRT/VA PTM, public domain) / original script.

7. Stress-Physiology Reset (stress-physiology-reset)

  • Technique: A short narrated session that names the stress-physiology mechanism (stress → sympathetic arousal → heightened tinnitus salience → more stress) and then runs an active downregulation sequence the user can do in daytime: a brief physiological-sigh burst → a slow extended-exhale stretch → a 30-second somatic check (shoulders/jaw/breath). It is the "active stress management" the gap analysis flagged as thin — a do-it-now reset, delivered as narration so it can be followed eyes-closed when the user is overwhelmed but not in an acute spike.
  • Distinct from existing: post-spike-recovery-sequence is the acute-spike assembly (sigh → box → defusion → body check) for a tinnitus spike; this is the everyday stress reset (not spike-triggered), pairs a psychoeducation hook with action, and lives in the relaxation cluster, not crisis. Different trigger, different register, different sequence.
  • Evidence:
  • Stress→tinnitus-loudness pathway — PMC12474699 (and session-content-evidence-base.md §5a, vault #405: sleep/stress mediation of the tinnitus→anxiety pathway).
  • Brief breathwork reduces physiological arousal — Balban et al. 2023 PMC9873947.
  • Resonance/extended-exhale evidence — Steffen et al. 2017 PMC5575449.
  • IP: Built from public-domain breathing techniques + freely citable stress-physiology science. Write own narration. ip_status: public-domain techniques + open-literature psychoeducation / original script.

Tier rule applied (per task LOCKED RULE)

  • free — foundational breathing + sleep wind-down: physiological-sigh, extended-exhale-breathing, cbt-i-wind-down, sleep-onset-relaxation, sound-enrichment-for-sleep, stress-physiology-reset.
  • premium — advanced/extended audio: alternate-nostril-breathing is the only non-foundational pattern here; it is set premium and flagged needs_confirm=tier (it is a single short pattern, so the premium call is a judgement — an editor may prefer free).

Outstanding clearance / editorial flags

  1. sleep-onset-relaxation naming — avoid "mySleepButton" / "Cognitive Shuffle" product branding; needs an editorial naming decision (neutral descriptor recommended).
  2. extended-exhale-breathing — avoid the "Coherent Breathing®" brand; use generic "coherent / resonance breathing".
  3. alternate-nostril-breathing tier — premium vs free is a judgement call (flagged).
  4. All seven require own-script authoring; none reproduce protected text. No fabricated citations — every source above is a real, resolvable URL/PMC ID.