Iso principle

1. Introduction

Version followed: the iso principle as music therapists use it now: begin with music that matches the client's present mood or arousal, then shift it step by step toward a target state. The guide uses the two clinical forms described in the sources. (a) The receptive mood-modulation sequence and playlist (Starcke et al. 2021; Ratcliff et al. 2014; Zybura & Head 2024). (b) Live tempo entrainment to physiology (practitioner account, Leone 2024). Altshuler's original text is not drawn on [uncertain].

What it is. The iso principle is a technique principle: a rule for ordering and shaping music. It is not a stand-alone therapy, school or manual. Music therapists of many orientations use it inside receptive (listening), active and improvisational work. It also appears in music therapy method texts (Wigram, Pedersen & Bonde 2002, as cited by Starcke et al. 2021; not drawn on) and has been adapted to dance therapy (Trautmann-Voigt & Voigt 2020, as cited by Starcke et al.). [source: Starcke et al. 2021]

Origin. The principle is credited to the psychiatrist Ira M. Altshuler, who used music with psychiatric inpatients at Eloise Hospital, Michigan. The sources give different dates:

• Starcke et al. cite Altshuler (1944), Am J Psychiatry, "Four years' experience with music as a therapeutic agent at Eloise Hospital".
• Ratcliff et al. say he introduced it in 1945.
• Heiderscheit & Madson say it was first described in 1948.
• Murphy (2021) cites Altshuler's 1948 chapter in Schullian & Schoen (eds), Music and Medicine.

The date (1948) matches the book chapter. The concept may have appeared earlier in the 1944 paper [uncertain]. "Iso" means "equal": the opening music is "equal to" the listener's mood. [source: Ratcliff et al. 2014; Starcke et al. 2021; Murphy 2021; Heiderscheit & Madson 2015 abstract]

How change is thought to happen. Music that fits the client's current state first earns acceptance and synchrony (being "met"). Gradual changes in tempo, mode, rhythm and content then carry mood and arousal along with them.

• Two accounts exist of what has to be matched. Ruppenthal (1954) stressed tempo and the listener's "mental tempo". From Shatin (1970) onward the focus moved to emotional state.
• Physiological versions rely on rhythmic entrainment: body rhythms such as heart rate and breathing synchronise with the beat and then follow it as it slows.
• Starcke et al. state that no empirically based theory yet explains the emotional shift.

Later writers have extended the principle through entrainment (Dileo 2018) and resonance (Ridder 2019), both as cited by Murphy 2021 and not drawn on. [source: Starcke et al. 2021; Murphy 2021; Leone 2024]

Contrast. The compensation principle (also called the contrast approach) jumps straight to music of the target state. Goldschmidt (2020) compared the two and found no physiological difference between them in a small study. [source: Goldschmidt 2020 abstract]

Professional frame. In the US, "music therapy" is delivered by board-certified music therapists (MT-BC, through AMTA-approved training and the CBMT exam). In the UK it is delivered by HCPC-registered music therapists (BAMT is the professional body). The clinicians in the studies were board-certified music therapists. [source: Ratcliff et al. 2014; Leone 2024; credentialing detail is general professional knowledge, synthesis] [uncertain]

• The principle itself is not a regulated technique. A psychotherapist can use mood-matched listening sequences and client-built playlists as an adjunct within their own competence.
• Do not call this "music therapy".
• Leave these uses to trained music therapists or medical teams: live entrainment with medically unstable patients (ICU, ventilated, cardiac), improvisation-based work, and work with severe psychiatric presentations. [synthesis]

2. Framework

There is no official phased protocol. Starcke et al. note that the clinical literature gives no step-by-step procedure. Ratcliff et al. note that ideal parameters, such as how long to stay on each track, are unknown. The structure below combines the published operational versions. Each element is tagged. [source: Starcke et al. 2021; Ratcliff et al. 2014]

Phase A: Assessment of current state and goal (session 1, about 10–20 min within a session) [synthesis, informed by the pre/post ratings in Zybura & Head 2024 and Starcke et al. 2021]

• Rate the client's present mood (valence) and arousal, for example on a 0–10 scale.
• Name the desired state. Typical directions are anxious/tense → calm, sad/low → energised, and agitated → settled.
• Take a music history: preferences, music the client dislikes, and pieces with traumatic or grief associations.
• Criterion to move on: the current state and a realistic target are named, and trigger music is identified.

Phase B: Music selection and sequencing (in the HSCT trial, 2 sessions of 50 min, one per direction) [source: Ratcliff et al. 2014]

• The client chooses about 15 tracks per direction.
• Relaxing sequence: from fast tempo (allegro, about 120–168 BPM) and anxious/tense character to slow tempo (largo, about 40–60 BPM) and relaxed character.
• Energising sequence: from slow to fast, and from sad to energised/joyful.
• The therapist puts the tracks in order. In the trial each sequence became a 30-minute compilation.
• Criterion to move on: the sequence's first piece fits the client's state and its last piece fits the target. Each step feels like a small change rather than a jump.

Phase C: Review and refine (session 3) [source: Ratcliff et al. 2014]

• The client listens and asks for changes.
• Criterion to move on: the client accepts the order, and nothing in it jars or triggers.

Phase D: Guided in-session listening (session 4) [source: Ratcliff et al. 2014]

• The client listens to the whole sequence with the therapist, then debriefs with before and after ratings.

Phase E: Self-management at home (ongoing) [source: Ratcliff et al. 2014; Zybura & Head 2024]

• The client uses the sequence when needed and logs mood before and after.
• In the HSCT trial, patients listened about 10 times in a month, about 39 minutes each time.
• Review criteria: a consistent change in the right direction across the logged uses, and the client can now build or adapt sequences without help.

Typical dose. 4 × 50-minute individual sessions in the one randomised trial cited. Inpatient psychiatry uses single-session playlists facilitated by a music therapist, with pre/post ratings. The experimental minimum is two pieces of about 90 seconds each, but Starcke et al. suggest more pieces for a gradual shift. [source: Ratcliff et al. 2014; Zybura & Head 2024; Starcke et al. 2021]

Live entrainment format (medical settings, one session of about 30 min) [source: Leone 2024]

• Match the observed heart rate or breathing.
• Hold the match until the client is in sync.
• Slow the tempo gradually. The practitioner reports heart rate falling from over 100 to the 80s over about 30 minutes.

3. Guiding principles

• Meet before you move. Start where the client is, not where you want them to be. In practice, the first piece should feel "like me right now". [source: Starcke et al. 2021; Ratcliff et al. 2014]
• Change gradually. Shift through intermediate steps in tempo, mode, rhythm and content. In practice, avoid big jumps between adjacent pieces. Two pieces is a minimum; more steps make a smoother path. [source: Starcke et al. 2021; Iowa M+M+M 2025]
• Tempo and mode are the main levers. Fast tempo and major mode are linked with happy music, slow tempo and minor mode with sad music (Juslin & Laukka 2004, as cited by Starcke et al.). In practice, order tracks by BPM and mode, not only by title. [source: Starcke et al. 2021; Ratcliff et al. 2014]
• Client preference matters. Ratcliff et al. suggest that a fixed order may have weakened preference. Their proposed future design: the client chooses the music and the therapist arranges it by BPM. In practice, the client picks the tracks and you sequence them. [source: Ratcliff et al. 2014]
• Be clear which variable you are matching. Name the target: mood, arousal, or a physiological rhythm. Scholars disagree about which states and musical elements define the principle. [source: Goldschmidt 2020 abstract; Starcke et al. 2021]
• Allow enough time to entrain. Short exposure (one four-minute track) may not shift physiological arousal. In practice, give each stage enough time, and judge by the client's response rather than the clock. [source: Ratcliff et al. 2014]
• Measure every time. Take a quick rating before and after. In practice, you get evidence for the client and for you, and the client learns which sequences work. [source: Zybura & Head 2024; Ratcliff et al. 2014]
• Build self-regulation. The goal is a tool the client owns and uses "on their own terms". [source: Zybura & Head 2024 abstract, paraphrased]
• Treat it as an adjunct in medical care. Entrainment work complements medical and nursing assessment. Heart rate alone is not a full measure of pain or anxiety. [source: Leone 2024]
• Watch for gender and population differences. In the one controlled study, the iso effect showed in women but not in men. The principle was developed for psychiatric inpatients and has not been well studied elsewhere. In practice, check that it works for this client rather than assuming it will. [source: Starcke et al. 2021; Ratcliff et al. 2014]
• Do no harm with music. Screen for trauma-linked music. Keep volume safe. Respect a client's choice to stop. [synthesis]

4. Intervention inventory

No source gives a complete technique list for the iso principle. This inventory covers every concrete procedure in the sources (4.3–4.11), plus clearly labelled supporting steps that are my own synthesis (4.1, 4.2, 4.12, 4.13). All therapist wording is illustrative.

4.1 Current-state and target rating

• Purpose / when: Use at the start of every iso session. It defines the starting point the music must match and the goal, and it provides the before-and-after comparison. [synthesis; pre/post ratings from Zybura & Head 2024, Starcke et al. 2021, Ratcliff et al. 2014]
• Steps:
  1. Ask for a 0–10 rating of mood (low to good), and separately of tension or arousal (calm to wound up).
  2. Ask for a few words describing the feeling.
  3. Agree a realistic target, for example "from 3 to 5", not "from 3 to 10".
  4. Write both numbers down.
• Example language (illustrative): "Before we choose any music: on a scale of 0 to 10, where is your mood right now, and how wound up do you feel? Where would you like to be by the end, not perfect, just a bit better?"
• Watch for: a gap between how the client says they feel and how they look, such as reporting calm while visibly agitated. Match what you observe as well as the report. Clients with flat affect may need the arousal rating more than the mood rating.

4.2 Music history and trigger screen

• Purpose / when: Use before any music is chosen or played. Personal associations can override a piece's tempo and mode. [synthesis]
• Steps:
  1. Ask about preferred genres and artists, music that soothes and music that energises.
  2. Ask about music the client dislikes.
  3. Ask about music linked to loss, trauma, abuse, a perpetrator, an accident, a relationship break-up, or substance use.
  4. Ask about hearing problems (tinnitus, hyperacusis, hearing aids).
  5. Record a "do-not-play" list.
• Example language (illustrative): "Some songs carry strong memories, good or bad. Is there any music that reminds you of something painful, or that you'd rather I never play? You can add to that list at any time."
• Watch for: pauses or a change of expression when a genre or era is mentioned. The music of a particular period of life can be a trigger in itself.

4.3 Two-step iso listening sequence (sad → happy)

• Purpose / when: The simplest form of the principle. Use it as a brief in-session mood lift, or to show a client how the principle works. It was tested experimentally after sadness induction. [source: Starcke et al. 2021]
• Steps:
  1. Rate mood (4.1).
  2. Play a piece that matches the current low mood: slow tempo, minor mode. In the study each piece was about 90 seconds.
  3. Follow it with a piece of the desired state: faster tempo, major mode. Keep tempo differences between the paired pieces modest; the study adjusted one piece to keep tempi similar.
  4. Rate mood again.
  5. Ask what the client noticed.
• Example language (illustrative): "The first piece will probably sound close to how you feel now, and that's on purpose. Then the second one will move somewhere else. Just notice what happens."
• Watch for: In the study, this sequence gave the highest positive affect and valence of the four sequences tested. It was clearly better than sad-sad and happy-sad, but not clearly better than happy-happy, and the effect appeared in women only. Students listened to classical pieces, not music they chose, and only short-term change was measured. If one switch feels too abrupt, add intermediate pieces (4.4/4.5). [source: Starcke et al. 2021]

4.4 Client-curated down-regulating sequence (anxious/tense → relaxed)

• Purpose / when: Use for anxiety, tension or hyperarousal. The client builds their own relaxation sequence. [source: Ratcliff et al. 2014]
• Steps:
  1. From a library or their own collection, the client picks about 15 tracks across the range from fast (allegro, about 120–168 BPM) and tense/anxious to slow (largo, about 40–60 BPM) and relaxed.
  2. Note each track's BPM. A tap-tempo app or a published BPM database works.
  3. Order the tracks from the fastest and most tense-sounding to the slowest and most relaxed.
  4. Check adjacent tracks for jumps in tempo, loudness, key or mode, and smooth them. [step 4 synthesis]
  5. Compile about 30 minutes.
• Example language (illustrative): "Pick a few tracks that feel the way you feel when you're keyed up, a few that feel like being halfway down, and a few that feel like real rest. I'll help put them in order so each one is only a small step slower than the last."
• Watch for: The opening tracks should feel accurate to the client, not pleasant. If the client only picks calm music, the sequence cannot "meet" them. As tempo slows, tension ratings should drop, along with visible signs such as slower breathing and softer posture.

4.5 Client-curated up-regulating sequence (sad/low → energised)

• Purpose / when: Use for low mood, lethargy or flat affect, as a mood-management tool. [source: Ratcliff et al. 2014]
• Steps: The same procedure as 4.4 in reverse.
  1. The client picks about 15 tracks from slow and sad/depressed in character to faster and energised/joyful.
  2. Order them by rising tempo and from minor towards major.
  3. Compile about 30 minutes.
• Example language (illustrative): "We start with something that honours how heavy things feel, and only then begin to pick up the pace."
• Watch for: a client who stays in the sad opening and does not shift. Shorten the opening and add more middle steps, or end earlier at a neutral point (see 5). Also watch for a client who wants only sad music. That can be validating, but if it deepens rumination, see Section 5.

4.6 Review-and-refine session

• Purpose / when: Use after the sequences are built and before home use. [source: Ratcliff et al. 2014]
• Steps:
  1. The client listens to, or skims, each compilation.
  2. Ask them to flag any track that feels out of place, too sudden, or carries unwanted associations.
  3. Reorder, replace or remove tracks.
  4. Confirm the final version.
• Example language (illustrative): "Was there any point where it felt like the music jumped ahead of you, or dragged you back? Let's fix those spots."
• Watch for: clients who are too polite to criticise. Ask about specific transitions rather than asking "Is it OK?".

4.7 Guided in-session listening with debrief

• Purpose / when: Before home practice, the client experiences the whole sequence with you present. [source: Ratcliff et al. 2014; debrief structure synthesis]
• Steps:
  1. Rate mood (4.1).
  2. The client chooses one sequence and listens to all of it in a comfortable position at a moderate volume.
  3. Stay quietly present and observe breathing, posture and tearfulness.
  4. Rate again.
  5. Debrief: where in the sequence did things shift, what images or memories came up, and what would they change.
• Example language (illustrative): "Where did you notice the change start, if at all? What came up for you along the way?"
• Watch for: strong emotion or memories surfacing during the matched opening (see Section 5, overwhelm). Note which track marks the turning point. It is useful for future editing.

4.8 Home listening with a mood log

• Purpose / when: Generalising the skill to daily life between sessions. [source: Ratcliff et al. 2014; Zybura & Head 2024]
• Steps:
  1. Agree on situations in which to use each sequence, such as before sleep, after a stressful call, or on waking when low.
  2. The client rates mood before and after each use in a simple log (date, sequence used, before and after ratings, notes).
  3. Review the log at the next session.
  4. Adjust the sequences.
• Example language (illustrative): "Each time you use it, jot down a number before and after. We're not testing you, we're testing the playlist."
• Watch for: In the HSCT trial, patients used their compilation about 10 times in a month. Patients who chose and ordered tracks by preference alone, without BPM ordering, improved just as much. So if a client's own preference-based order works, do not insist on strict tempo ordering. [source: Ratcliff et al. 2014]

4.9 Inpatient mood-modulation playlist (single session, music therapist-facilitated)

• Purpose / when: Use on acute psychiatric units, for brief mood regulation in one session. [source: Zybura & Head 2024 abstract]
• Steps (only as far as the abstract describes):
  1. A music therapist sets up an individualised playlist informed by the iso principle.
  2. The patient rates mood on a Likert scale before listening.
  3. The patient listens.
  4. The patient rates mood again after.
• Details such as track number, scale range and session length are in the full text, which is not drawn on [uncertain].
• Example language (illustrative): "Let's build a short playlist that starts where you are today and moves toward where you'd like to be."
• Watch for: In 125 routine pre/post pairs, mood ratings rose. There was no control group, so the design cannot show that the playlist caused the change. [source: Zybura & Head 2024]

4.10 Live heart-rate entrainment (medical adjunct)

• Purpose / when: Reducing anxiety and pain-related arousal in critically ill adults, such as ventilated patients who cannot speak. This is music therapist work within medical care. A psychotherapist should not do it at the bedside of unstable patients. [source: Leone 2024, practitioner account]
• Steps:
  1. Take the patient's heart rate from the monitor as the starting tempo.
  2. Improvise live (the author uses guitar) at that tempo, with some light purposeful tension in the music, until music and body are in sync.
  3. Slow the tempo gradually while moving the music toward softer, more consonant, more harmonic material.
  4. Use familiar structured songs only when family can confirm the patient's preferences. Otherwise improvise rather than guess.
  5. Over about 30 minutes, track heart rate alongside nursing pain and anxiety scales.
• Example language (illustrative, to the patient): "I'm going to play along with your rhythm for a bit, then slow things down. You don't need to do anything."
• Watch for: heart rate following the tempo downwards. The author reports a fall from over 100 to the 80s. Heart rate alone is not a full measure of pain or anxiety, so coordinate with the nurses. If heart rate does not follow, or vital signs change, stop and inform the medical team [synthesis]. The evidence here is anecdotal. [source: Leone 2024]

4.11 Breath-paced singing and accompaniment

• Purpose / when: For shallow, anxious breathing, including breathlessness in medical patients. [source: Leone 2024]
• Steps:
  1. Set the musical pulse slightly slower than the client's current breathing.
  2. Slow it further in small steps.
  3. Where possible, sing a familiar song, such as a hymn the patient knows, so that breathing follows the song's phrasing.
• Example language (illustrative): "Let's sing this one together, nice and easy. Just breathe where the line ends."
• Watch for: breathing lengthening with the phrases, and visible relaxation. If singing worsens breathlessness, stop and switch to quiet listening [synthesis]. In respiratory illness, follow the medical team's guidance [synthesis].

4.12 Iso-informed active music-making (matching then leading)

• Purpose / when: For trained music therapists in improvisation-based work. The therapist first matches the client's playing or vocal energy (tempo, loudness, density), then gradually leads toward the target state. Sources describe the iso principle as gradually changing a music's tempo, rhythm and content, and as later extended through entrainment and resonance. No step protocol for improvisational use is described. [synthesis; Iowa M+M+M 2025; Dileo 2018 and Ridder 2019 as cited by Murphy 2021, not drawn on; Heiderscheit & Madson 2015 full method not drawn on]
• Steps (synthesis):
  1. Join the client's playing at their tempo and loudness.
  2. Hold the match for long enough that the client feels met.
  3. Introduce one small change at a time, such as slightly slower, softer, or fewer notes.
  4. Check whether the client follows. If they do not, return to matching.
  5. Close at the target state and debrief.
• Example language (illustrative): "I'm going to play along with you. If I go somewhere you don't want to go, just keep doing your thing and I'll come back."
• Watch for: the client following your changes. If they resist, you are leading too fast. This needs musical skill and clinical training (see Section 1, Professional frame).

4.13 Compensation (contrast) sequence as a comparison or alternative

• Purpose / when: Going straight to music of the target state, without matching first. Starcke et al. tested a happy-happy sequence and Goldschmidt a "compensation principle" song. In Starcke et al., happy-happy came out close to the iso sequence, and Goldschmidt found no physiological difference. Use a direct target-state sequence when time is very short, or when matching music risks deepening a dangerous state. Agree with the client which works better for them. [source: Starcke et al. 2021; Goldschmidt 2020 abstract; clinical use synthesis]
• Steps:
  1. On different occasions, try the iso sequence (4.4/4.5) and a direct target-state playlist.
  2. Rate before and after each.
  3. Keep whichever works better for this client.
• Example language (illustrative): "Some people do better easing in, others prefer to jump straight to the calm music. Let's test both and see what your numbers say."
• Watch for: a consistent difference across several logged uses, not just one.

Not described in the sources: the procedures in Altshuler's original paper or chapter, or in the Heiderscheit & Madson mood-management method. Further procedures from those texts may exist [uncertain].

5. Handling problems

The sources say almost nothing about handling problems. Unless a source is tagged, everything below is my own synthesis.

• No shift (stuck in the matched mood). [synthesis; first point informed by Ratcliff et al. 2014 on exposure length]
  • Shorten the matched opening, add more intermediate steps, and lengthen the time at each step.
  • Lower the target to neutral rather than happy.
  • Check whether the client chose music they actually like. Ratcliff et al. suggest imposed order may weaken preference.
  • Try the contrast approach (4.13).
  • Remember that the controlled effect was absent in men in one study [source: Starcke et al. 2021]. Do not assume the principle works for everyone.
• Rumination or deepening low mood (the client wants to stay with sad music). [synthesis]
  • Validate first.
  • Limit the matched portion to one or two tracks.
  • Agree a rule in advance: the sequence always ends at least one step lighter than it started.
  • Note when sad music is being used to stay stuck rather than to be met, and discuss it as a pattern.
• Overwhelm or traumatic memories during listening. [synthesis]
  • Stop or turn the music down immediately.
  • Orient the client to the room.
  • Use grounding, such as naming five things they can see, feeling their feet on the floor, and slow breathing.
  • Add the piece to the do-not-play list (4.2).
  • Debrief without pushing for the memory's content.
  • Remember that matching music to a distressed state can intensify it before it changes it. With PTSD, dissociation or recent trauma, keep the matched opening very short, or start from a neutral "safe" piece.
  • Musical trauma triggers include music linked to abuse or a perpetrator, music played during traumatic events (including music used coercively), funeral or grief music, music of a lost relationship, music from periods of substance use, and some religious music.
• Agitation or mania. [synthesis]
  • Matching high-energy music may escalate agitation. Match energy only slightly, with lower volume and simpler texture, and slow down sooner.
  • Stop if arousal rises.
• Suicidal clients. [synthesis]
  • A mood-matching sequence is not a crisis intervention.
  • If low-mood music brings out hopelessness or suicidal thoughts, stop and complete a standard risk assessment and safety plan.
  • Do not leave the client with a home playlist that ends low.
• Non-adherence to home use. [synthesis]
  • Simplify: one short sequence of 10–15 minutes, one trigger situation, and a single-number log.
  • Load the sequence onto the client's own device.
  • Ask what got in the way, for example forgetting, or music feeling too painful when low.
• Ruptures (the client feels the therapist's choice "wasn't me" or felt manipulative). [synthesis]
  • Acknowledge the mismatch openly and give control back: the client chooses and you sequence.
  • Explain the rationale openly. The method is transparent, not a trick.
• The client's music unfamiliar or uncomfortable to the therapist (genre, explicit lyrics, culture). [synthesis]
  • Respect the client's music.
  • Screen lyrics for content that reinforces self-harm or violence, and discuss rather than ban.
  • Do not impose classical or "relaxation" music. The experimental evidence used classical pieces only, and its authors call for self-selected music. [source: Starcke et al. 2021]
• Medical settings. [source: Leone 2024; Ratcliff et al. 2014]
  • Entrainment complements medical care and does not replace it.
  • The HSCT trial excluded patients with known psychotic disorders, and the principle is not well studied in medical populations.
• Hearing safety. [synthesis]
  • Keep the volume moderate. Speech at normal volume should still be audible over the music.
  • With headphones, use a lower volume and limit long sessions.
  • Ask about tinnitus and hyperacusis, and reduce volume or avoid sharp-attack music if present.
  • For live percussion or amplified instruments in small rooms, keep levels low and offer ear protection.
  • Stop if the client reports ringing, pain or muffled hearing.
• Children and young people. [synthesis]
  • Get parental consent and the child's assent.
  • Use shorter sequences and age-appropriate music the young person has chosen.
  • Check lyrics.
  • Follow safeguarding and mandatory-reporting duties if a disclosure arises.
  • Do not interpret music choice or listening response as evidence of abuse or disorder.
• Scope. [synthesis]
  • Present this as music-based mood regulation within psychotherapy, not as music therapy.
  • Refer to a credentialed music therapist (MT-BC, or HCPC-registered) for improvisation, medical entrainment, neurological rehabilitation or complex presentations.

Sources

Sources:

• Emotion Modulation through Music after Sadness Induction: The Iso Principle in a Controlled Experimental Study (Starcke, Mayr & von Georgi 2021, IJERPH 18(23):12486)
• Music Therapy for Patients Who Have Undergone Hematopoietic Stem Cell Transplant (Ratcliff et al. 2014, eCAM 2014:742941)
• Use of the Iso Principle as a Central Method in Mood Management: A Music Psychotherapy Clinical Case Study (Heiderscheit & Madson 2015, Music Therapy Perspectives); repository record, abstract only
• The effectiveness of iso principle mood modulation playlists in inpatient psychiatry: A post-hoc analysis (Zybura & Head 2024, Music and Medicine 16(2)); abstract only
• Investigating the iso principle: the effect of musical tempo manipulation on arousal shift (Goldschmidt 2020, MM thesis, Colorado State University); abstract only
• Book review: Early theoretical writings on music therapy 1951–1960 (Murphy 2021, Approaches 13(2))
• Music Therapy Part 1: What is Music Therapy? M+M+M, University of Iowa (2025)
• How Our Bodies Work in Rhythm (Leone 2024, Psychology Today)
• Music and Medicine, Schullian & Schoen (eds) 1948: Wellcome Collection catalogue record

Further reading (not drawn on) [uncertain]:

• Altshuler, I. M. (1944). Four years' experience with music as a therapeutic agent at Eloise Hospital. American Journal of Psychiatry (as cited by Starcke et al. 2021).
• Altshuler, I. M. (1948). A psychiatrist's experience with music as a therapeutic agent. In D. M. Schullian & M. Schoen (Eds.), Music and Medicine. New York: Henry Schuman (as cited by Murphy 2021).
• Shatin, L. (1970). Alteration of mood via music: A study of the vectoring effect. Journal of Psychology.
• Ruppenthal, W. (1954). Experimental determinants of perception: some considerations for music therapy. Music Therapy.
• Wigram, T., Pedersen, I. N., & Bonde, L. O. (2002). A Comprehensive Guide to Music Therapy.
• Blocked: the full text of Heiderscheit & Madson (2015) on the Augsburg repository, and the PDF of Goldschmidt (2020).