Movement Psychodiagnostic Inventory

1. Introduction

Version followed: Martha Davis's 2019 Guide to the Movement Psychodiagnostic Inventory (MPI), as described in the University of Maryland archive of her papers (eleven primary categories, 60 patterns), together with the full form and the short form for dance/movement therapy as presented in her 2017 lectures at the German Sport University Cologne. Earlier editions: Part 2 of Davis's unpublished Guide to Movement Analysis Methods (1991, cited as available from the author) and a 1997 edition. Sources include the archive description, the Cologne lecture page and publisher and library catalogue pages: the eleven-category, 60-pattern structure of the 2019 guide (the 2017 lecture page gives ten categories, so the count depends on the edition); what the 2019 guide contains (considerations on video length, camera placement and editing; coding forms in the appendices; coding conventions; one patient profile; NEUROGES-ELAN computer coding; YouTube-based exercises); the chapter by Davis, Lausberg, Cruz, Berger and Dulicai (2007) in Koch and Bender's Movement Analysis – Bewegungsanalyse (Logos); and the spelling Cruz and Koch (Sabine C. Koch). The manuals and the two book chapters are not openly available; category names beyond four examples, recording parameters, scoring rules and any reliability or validity coefficients are therefore not given here [uncertain]. [source: UMD Martha Davis papers; DSHS Köln 2017; Logos Verlag table of contents; Koch & Bräuninger 2006]

What it is. The MPI is an observational assessment instrument, not a therapy. It codes movement patterns linked to psychiatric disorder from video of a person in conversation. It was developed by Martha Davis, a clinical psychologist and Certified Movement Analyst (Laban/Bartenieff Institute). She built it in the 1960s from observing hospitalised psychiatric patients, drawing on research on nonverbal behaviour in face-to-face conversation and on Laban Movement Analysis. It was first used to assess patients in verbal sessions and interviews and later adapted for video analysis. Davis taught it in dance/movement therapy (DMT) training at Hunter College from 1973 and later in Montreal. [source: DSHS Köln 2017; UMD Martha Davis papers]

Core idea. Global clinical terms such as "flat affect" or "psychomotor agitation" are replaced by fine-grained, operationally defined movement patterns that trained raters can code reliably. A person's profile (which patterns occur, how often, and when) adds information to clinical assessment, and recoding over time can track symptom change. The MPI does not itself bring about change. In DMT practice, its role is to inform assessment, treatment planning and outcome monitoring. [source: DSHS Köln 2017; Linn 2015] Davis's wider research suggests that shifts in a person's movement patterns can mark changes in mental state, and that patients' pattern changes in response to therapist interventions may mark therapeutically relevant moments. [source: Lausberg 2013, summarising Davis & Hadiks 1990, 1994]

Professional frame. Full MPI coding requires formal observation training and extensive rater training, so the instrument must not be scored or interpreted without it. [source: DSHS Köln 2017; Linn 2015] No regulator credentials the MPI itself. In practice it is used by credentialed dance/movement therapists (US: ADTA R-DMT/BC-DMT; UK: ADMP UK registration, since UK dance movement psychotherapy is not an HCPC-protected title [source: UK Regulated Professions Register, whose protected arts-therapist titles are Art Psychotherapist, Art Therapist, Dramatherapist and Music Therapist]), by movement analysts (Laban/Bartenieff CMA), and by psychologists working within their own licence. The licensed diagnosing clinician, not the movement rater, is responsible for any diagnostic conclusion. [synthesis; check current requirements with your national body] Training access. Davis taught the MPI in dance/movement therapy programmes from 1973, starting at Hunter College, and until recently at the National Centre for Dance Therapy in Montreal [source: UMD]. Her 2017 Cologne series of three classes was explicitly introductory: participants practised coding part of the short form, and the page states that facility in coding needs formal observation training [source: DSHS Köln 2017]. No current public MPI course or certification is advertised [uncertain]. Ask the movement therapy institute at the German Sport University Cologne or your DMT training programme about access to the 2019 guide and to supervised rater training [synthesis].

2. Framework

The MPI is an assessment procedure rather than a phased treatment. The workflow below links the steps the sources describe; ordering and decision points are [synthesis] unless tagged.

Structure of the instrument. The MPI has primary categories, each made up of several items (patterns). Each pattern is rated for frequency, from constant through rare to absent. [source: DSHS Köln 2017] The count depends on the edition. The 2017 lecture page gives ten primary categories. The 2019 guide covers "the eleven primary categories of movement patterns" and 60 patterns within them [source: DSHS Köln 2017; UMD]. This guide follows the 2019 edition (eleven categories). If you trained on an older edition, name it in every report. Named example categories: disorganization, reduced mobility, perseveration, flaccidity [source: UMD]. The remaining category names and all definitions are in the manuals, which are not openly available, so they are not listed [uncertain]. The categories were built from research on nonverbal interaction, from Laban Movement Analysis concepts, and from observation of hospitalised psychiatric patients [source: DSHS Köln 2017]. A Spanish review describes the MPI as aimed at involuntary movement disturbances in schizophrenia and other mental disorders, mapped along a continuum from no movement to excessive movement [source: Vella & Torres Solera 2012].

Validity and reliability limits.

• Davis describes the coding as reliable, and the 2017 lectures included a discussion of validity research. The page gives no figures. [source: DSHS Köln 2017] The main published description is a book chapter, not a journal article: Davis, Lausberg, Cruz, Berger and Dulicai (2007), "The Movement Psychodiagnostic Inventory (MPI)", starting on p. 119 of Koch and Bender (Eds.), Movement Analysis – Bewegungsanalyse (Logos, Berlin). The methods chapter on movement-observation scales is Cruz and Koch, "Issues of validity and reliability in the use of movement observations and scales", in Cruz and Berrol (Eds.), Dance/Movement Therapists in Action (2004; 2nd ed. 2012). [source: Logos Verlag table of contents; BISp record; NEIT library catalogue] Neither chapter is publicly available, and this guide reports no reliability or validity coefficients [uncertain].
• A Spanish review reports that the MPI is not validated for adults without psychopathology or for children [source: Vella & Torres Solera 2012, citing Cruz & Koch 2004]. The review spells the second author "Kogh". The chapter is by Robyn Flaum Cruz and Sabine C. Koch [source: NEIT library catalogue].
• The short form had no published studies before a 2015 N=1 pilot. That pilot found similar patterns across a DMT session and a seated interview. It did not test validity, and its author calls for multi-rater, multi-participant research and published usage guidelines. [source: Linn 2015]
• Practical meaning: treat MPI results as structured clinical observation that supports a formulation. They are not a stand-alone diagnosis, never evidence of abuse or trauma history, and not a forensic tool. [synthesis, consistent with the sources above]

3. Guiding principles

• Operationalise; don't label. Record observable patterns, not global impressions like "flat" or "agitated". In practice: write what the body did, then the category, then any interpretation, in that order. [source: DSHS Köln 2017]
• Watch the movement without sound. Coding is done with the audio off. In practice: code a first pass silently so that speech content doesn't bias what you see. [source: DSHS Köln 2017]
• Frequency matters. Patterns range from constant to rare or absent. In practice: one occurrence is not a profile; note how often each pattern appears. [source: DSHS Köln 2017]
• Timing is not random. Shifts in position and movement can mark brief clinical states, and signs can vary with interview topic. In practice: timestamp patterns and line them up with what was being discussed. [source: DSHS Köln 2017]
• Context-matched forms. Use the full form on a seated interview and the short form in a DMT assessment session. In practice: don't mix contexts when comparing scores over time. [source: Linn 2015]
• Trained raters only. Facility in MPI coding requires formal observation training. In practice: don't score the instrument from this guide or from reading alone. [source: DSHS Köln 2017; Linn 2015]
• Consider medication and neurology. Davis and Lausberg address neuroleptic effects and the neuropsychology of the categories. In practice: record current medication and known movement disorders before you interpret. [source: DSHS Köln 2017]
• Population limits. The MPI is not validated for adults without psychopathology or for children. In practice: don't use it to screen general or child populations. [source: Vella & Torres Solera 2012]
• Never diagnostic of abuse. In practice: no movement pattern is evidence that someone was abused, and MPI findings must not be used in safeguarding or legal decisions as proof. Safeguarding concerns follow your usual disclosure-based procedures. [synthesis]
• Assessment serves the relationship. In practice: explain why you are filming, share findings in plain language, and let the client decline without losing therapy. [synthesis]
• Cultural and individual variation. In practice: check whether a pattern fits the client's cultural norms of gesture, posture and eye contact, or a physical disability or injury, before reading it as pathology. [synthesis]

4. Intervention inventory

The MPI has no therapeutic interventions. Its "procedures" are the steps of administering, coding and using the inventory. Below are the procedures the sources describe, plus the clinical steps needed to run them safely, which are marked [synthesis]. The licensed coding definitions, forms and items are not reproduced or paraphrased here. Get them from the manual and training.

4.1 Framing the referral question and obtaining consent

• Purpose / when: Before any recording. Clarifies what the MPI can add (for example, a detailed description of psychomotor signs, or a baseline for tracking change) and gets informed consent to video. [synthesis; the use for clinical assessment and tracking change is from DSHS Köln 2017]
• Steps: (1) Write the question in one line (e.g. "Which movement patterns are present at baseline, and how do they change over 12 weeks of DMT?"). (2) Check the question fits the instrument's populations: adults with psychiatric conditions, not healthy screening and not children [source: Vella & Torres Solera 2012]. (3) Explain what will be filmed, who will view it, that it is watched without sound, how long it is kept, and where it is stored. (4) Get written consent for recording and for each use (clinical coding, supervision, teaching or research separately). (5) Say plainly that declining does not affect therapy. (6) For clients with impaired capacity, follow your jurisdiction's capacity law and the ward's policy. [synthesis]
• Example language (illustrative): "I'd like to video one conversation with you so I can look closely at how you move while you talk. It helps me understand how you're doing and see changes over time. Only the care team and I will see it. You can say no, or stop the recording at any point."
• Watch for: paranoid or persecutory ideas about cameras, consent given only under ward pressure, and confusion about what is being assessed. It's working when the client can say back in their own words why the video is made and who sees it. [synthesis]

4.2 Recording the full-form sample: videotaped seated interview

• Purpose / when: The standard full MPI uses a person in conversation, seated, on video. [source: DSHS Köln 2017; Linn 2015]
• Steps: (1) Follow the 2019 guide's specifications for video length, camera placement and editing [source: UMD]; they are not reproduced here [uncertain], so the following are general principles only: (2) frame the whole body, including feet and hands, with the chair fully in view, and use a fixed camera; (3) use a standard interview format so repeat recordings are comparable; (4) note the time, medication and recent events on the recording log. [synthesis]
• Example language (illustrative): "Just sit however is comfortable. We'll talk for a while about how things have been. There's nothing you need to do with your body."
• Watch for: self-consciousness about the camera in the first minutes. Whether to code from a later segment is set by the manual [uncertain]. A good recording shows the whole body, has stable framing, and follows a repeatable interview structure. [synthesis]

4.3 Recording the short-form sample: DMT assessment session

• Purpose / when: The short form was created in 2007 for clinical DMT assessment and is applied to a DMT assessment session. [source: Linn 2015; DSHS Köln 2017]
• Steps: (1) Run your usual DMT assessment session structure (warm-up, movement exploration, closure). No MPI-prescribed session content is publicly described [uncertain]. (2) Film with a fixed wide shot that covers the movement space. (3) Make the space and touch safety checks in 4.12 before you start. (4) Keep the session format the same when you re-assess. [synthesis]
• Example language (illustrative): "Today is a getting-to-know-you session. I'll offer some simple movement ideas and you can take them or leave them. Go at your own pace, and you can sit down whenever you like."
• Watch for: the only published comparison is a single-case pilot, so compare short-form and full-form results with caution [source: Linn 2015]. It's working when the short-form profile is consistent with what you see clinically and with any interview-based coding. [synthesis]

4.4 Silent viewing and coding by primary category

• Purpose / when: The core of the MPI. The trained rater codes patterns in each primary category from video with the sound off. [source: DSHS Köln 2017]
• Steps: (1) Turn off the audio. (2) Use the manual's coding forms and procedures, which are in the appendices of the 2019 guide [source: UMD]. (3) For each primary category, rate its items for how often they occur, from constant through rare to absent [source: DSHS Köln 2017]. (4) Categories include, for example, disorganization, reduced mobility, perseveration and flaccidity [source: UMD]; the other category names are in the manual [uncertain]. (5) Allow several hours per full-form case [source: Linn 2015]. (6) Wherever possible, have a second trained rater code independently and resolve disagreements before reporting. [synthesis]
• Example language: none. This is rater work, with no client present.
• Watch for: drifting from definitions into impressions, bias from knowing the diagnosis (code blind to the diagnosis where you can [synthesis]), and fatigue in long coding sessions. It's working when two raters reach acceptable agreement under the manual's criteria. [uncertain: the manual's agreement criteria are not publicly available]

4.5 Timing and state-shift analysis

• Purpose / when: Patterns are not randomly timed. Shifts in position and movement can mark brief clinical states, and signs can vary with interview topic. Use this when you want to know what a pattern is linked to, not only whether it is present. [source: DSHS Köln 2017; Lausberg 2013 on Davis & Hadiks]
• Steps: (1) Timestamp each coded occurrence. (2) Then switch the sound on and mark the interview topics and interventions on the same timeline. (3) Look for clusters, such as patterns that appear when certain themes come up or that change after a therapist intervention. (4) Treat any link as a hypothesis to explore clinically, not a finding. [synthesis built on the cited observations]
• Example language (illustrative, when sharing with a client): "When we talked about your brother, your movements changed: you went very still. Does that fit with anything you noticed inside?"
• Watch for: over-reading single co-occurrences, and confronting the client with "what your body revealed". Offer the observation tentatively. It's working when the observation opens collaborative exploration. [synthesis]

4.6 Medication and neurological screen before interpretation

• Purpose / when: Before writing a profile. Davis and Lausberg address the neuropsychology of the MPI categories and how they relate to neuroleptics. [source: DSHS Köln 2017]
• Steps: (1) Record current and recent antipsychotics and other psychotropics, with doses and changes, from the chart (don't interpret doses yourself). (2) Note known extrapyramidal side effects, tardive dyskinesia, neurological illness, injuries and sensory impairment. (3) Mark any coded pattern that could plausibly be medication-related or neurological, and discuss it with the prescriber. [synthesis]
• Example language (illustrative, to the team): "Reduced mobility is prominent. Before we read it as negative symptoms, has the antipsychotic dose changed recently?"
• Watch for: attributing medication effects to psychopathology, or the reverse. It's working when the profile explicitly lists the confounds you considered. [synthesis]

4.7 Writing the patient profile

• Purpose / when: To summarise the coding for the treatment team and the DMT plan. The 2019 guide includes a patient profile. [source: UMD]
• Steps: (1) For each primary category, give the frequency level and one or two observable examples with timestamps. (2) Add the timing findings from 4.5. (3) Add the confounds from 4.6. (4) State the limits: how many raters, which form and context, and that the MPI is not validated for children or non-clinical adults. (5) Phrase conclusions as "consistent with" or "adds detail to", never as a diagnosis on its own. (6) Never include statements about abuse or trauma history inferred from movement. [synthesis; limits from Linn 2015 and Vella & Torres Solera 2012]
• Example language (illustrative, report wording): "Coded patterns of perseveration were frequent throughout and increased during discussion of discharge. This is offered as descriptive detail alongside the psychiatric assessment."
• Watch for: team members quoting the profile as a diagnosis. It's working when the profile changes concrete plans, for example the DMT focus or what to monitor. [synthesis]

4.8 Using findings to plan DMT

• Purpose / when: The short form exists for DMT assessment practice, and its stated value is helping dance/movement therapists working with people with mental illness. [source: Linn 2015; DSHS Köln 2017]
• Steps: (1) Choose one or two observable targets from the profile (for example, low mobility or disorganized phrasing). (2) Link each target to DMT goals using your DMT training. The MPI does not prescribe interventions [uncertain: none publicly described]. (3) Agree the goals with the client in everyday words. (4) Decide when to re-assess (4.9). [synthesis]
• Example language (illustrative): "From the video, it looked like your movements got very small and stuck in one place. Would it be useful to work on finding a bit more room to move?"
• Watch for: setting goals the client doesn't share. It's working when the goals are named by the client and are observable in sessions. [synthesis]

4.9 Repeated assessment to track change

• Purpose / when: The MPI can track symptom change over time. [source: DSHS Köln 2017]
• Steps: (1) Re-record in the same context, with the same form and the same set-up as the baseline. (2) Use the same rater, or blind a second rater to which recording is which. (3) Compare frequency levels category by category, considering medication changes. [synthesis]
• Example language (illustrative): "Let's do another short video like the first one, so we can see together what has shifted."
• Watch for: practice and camera-familiarity effects, and medication changes between recordings. It's working when change in the profile matches change in clinical measures and in the client's own account. [synthesis]

4.10 Computer-assisted coding (NEUROGES-ELAN)

• Purpose / when: Hedda Lausberg, a co-author of the 2007 MPI chapter, has explored the MPI as an addition to her NEUROGES analysis system. The 2019 guide looks ahead to computer coding "being accomplished by Hedda Lausberg: the NEUROGES-ELAN Method". [source: DSHS Köln 2017; UMD; Logos Verlag table of contents] NEUROGES is a separate published coding system for movement and gesture, used with the ELAN annotation tool. It has its own rater-training and rating procedures and an algorithm for interrater agreement on segmentation. [source: Peter Lang page for Lausberg 2013] The NEUROGES book's table of contents lists a section on movement psychopathology but does not name the MPI [source: Lausberg 2013 table of contents; uncertain whether its content overlaps]. Use this route in research, or where annotation software and NEUROGES-trained raters are available.
• Steps: follow the 2019 guide, and follow the NEUROGES rater-training and ELAN procedures set out in the open-access NEUROGES book [source: Lausberg 2013]. How the MPI categories map onto NEUROGES modules is not publicly described [uncertain]. Store annotation files under the same data-protection rules as the video. [synthesis]
• Watch for: treating software output as more objective than the rater's definitions allow. [synthesis]

4.11 Rater training and calibration

• Purpose / when: Before clinical use, and periodically afterwards. Facility in coding requires formal observation training. The 2017 Cologne lectures were introductory and included a first attempt at coding part of the short form. The 2019 guide covers coding conventions in depth, with examples, and its instructional exercises use online YouTube examples. [source: DSHS Köln 2017; UMD]
• Steps: (1) Complete formal training with a recognised MPI trainer. Davis taught the MPI from 1973, starting at Hunter College, and until recently at the National Centre for Dance Therapy in Montreal [source: UMD]. No current public course or certification is advertised [uncertain]. (2) Practise on training videos until your codes agree with the trainer's. (3) Re-calibrate with a colleague at intervals, both using the same manual edition. [synthesis]
• Watch for: solo coding without any reliability check. [synthesis]

4.12 Physical safety, space and touch in short-form (movement) assessment

• Purpose / when: The MPI itself is observational and requires no touch. A short-form DMT assessment session does involve movement, so these steps apply. [synthesis]
• Space and falls: (1) Clear the floor of furniture, cables and the camera tripod legs. Use a non-slip surface and have a chair available. (2) Ask about dizziness, falls, orthostatic symptoms (common with some psychotropics), pain, pregnancy and mobility aids. (3) Offer seated alternatives for every movement idea. (4) Avoid fast spins, eyes-closed travelling and floor-to-standing transitions for anyone at risk of falling. (5) Keep yourself between the client and hazards, and know the ward's emergency procedure. [synthesis]
• Touch: (1) The default is no touch. (2) If touch would help (for example, steadying), ask each time, name exactly where and why, and accept "no" without question. (3) Give a simple opt-out signal, such as saying "stop" or stepping back. (4) Touch only neutral areas (hands, forearms, upper back), and never sexual or genital touch. (5) Be sensitive to trauma and culture: same-gender expectations, religious norms, and histories of restraint or abuse. (6) Document any touch and the consent given. [synthesis]
• Example language (illustrative): "You can sit down at any time. If I ever offered a hand for balance I'd ask first, and 'no' is always fine."
• Watch for: unsteadiness, sedation, dissociation and boundary confusion. It's working when the client moves within their own limits and uses the opt-out freely. [synthesis]

5. Handling problems

The sources do not address problem handling. Everything in this section is [synthesis] unless tagged.

• Client refuses or withdraws video consent. Accept it without consequence for treatment. Offer live observation notes instead, recognising that they aren't MPI coding. Revisit consent later only if the client brings it up. If consent is withdrawn after recording, delete or quarantine the footage according to your data policy.
• Paranoia about cameras, or the camera raising distress. Postpone recording until the client is more settled. Show them the equipment, let them watch the first minute back, and use a visible off switch. Never film covertly.
• The client is "performing" for the camera, or is stiff because of it. Allow a settling period and use repeated recordings so the client gets used to the camera. Note self-consciousness as a confound in the profile.
• Overwhelm or agitation during the interview or movement session. Stop the task, slow down, sit, and ground (feet on the floor, orient to the room). Offer to end the session. Follow the ward's risk procedure if there are signs of acute risk, and assess suicide risk using your service's standard approach if it comes up. An incomplete recording is acceptable. Don't push to get a codable sample.
• Dissociation during movement. Lower the intensity, open the eyes, orient, and give the client a choice to sit. Don't interpret dissociative signs from movement alone. Refer them to the clinical assessment.
• Stuck or minimal movement (e.g. marked reduced mobility). For coding, record it as it is, and don't prompt more movement during a full-form interview. In short-form DMT sessions, offer small, seated options and mirroring, and accept stillness as data.
• The client is upset by feedback ("you're saying I move like a mental patient"). Repair the rupture by acknowledging the hurt, taking responsibility for how it was framed, and restating findings as descriptive and changeable. Ask what would be useful to look at. Use the client's own words in later reports.
• Medication or neurological confounds. Use 4.6. If you can't separate the effects, say so in the profile rather than guessing.
• Cultural, disability or age differences in movement norms. Check the client's own norms with them, or with a cultural consultant. Don't code a culturally normal behaviour as pathology without a note. The MPI is not validated for children [source: Vella & Torres Solera 2012], so don't use it with them.
• Requests to use the MPI to show abuse, trauma history, credibility or deception. Decline. Movement patterns are not evidence of abuse. Davis's separate deception research [source: Lausberg 2013 on Davis & Hadiks 1995; Davis et al. 2005] is a research line, not a validated clinical or forensic use of the MPI. Safeguarding and mandatory reporting follow disclosure and standard procedures, not movement profiles.
• The team over-reads the profile as a diagnosis. Re-state the limits in writing (rater count, form, validation status [source: Linn 2015; Vella & Torres Solera 2012]) and ask that diagnosis rest with the diagnosing clinician.
• Low inter-rater agreement. Pause clinical use. Re-calibrate on training video and check that both raters use the same manual edition. Watch especially for the ten-category version (2017 lecture description) versus the eleven-category, 60-pattern version (2019 guide) [source: DSHS Köln 2017; UMD].
• Non-adherence to re-assessment. Reduce the burden by using a shorter recording in the same format, or link re-assessment to existing review meetings. Drop it if the client no longer consents.
• Data security incident (lost or shared video). Follow your breach procedure, inform the client, and review your storage practice. Video of identifiable psychiatric patients is highly sensitive data.
• A fall or injury in a movement session. Stop, give first aid or call the ward team, document it, review the space and the client's medication for orthostatic effects, and adapt future sessions to seated work.

Sources

Web pages:

• Lectures on the Movement Psychodiagnostic Inventory (MPI) with Dr. Martha Davis (Deutsche Sporthochschule Köln, 2017)
• Martha Davis papers, Collection 0373-SCPA, University of Maryland Libraries
• Linn, A. J. (2015). Diagnostic implications from clinical assessment of movement pathology across contexts using the Movement Psychodiagnostic Inventory: a comparison study (MA thesis, Drexel University)
• Vella, G., & Torres Solera, E. (2012). Desarrollo de un instrumento de observación en danza movimiento terapia (DMT). Papeles del Psicólogo, 33(2), 148–156
• Lausberg, H. (Ed.) (2013). Understanding Body Movement: A Guide to Empirical Research on Nonverbal Behaviour. Peter Lang (open access)
• Davis, M., Dulicai, D., & Climenko, J. (1977). Movement research and movement therapist: A collaboration. American Journal of Dance Therapy, 1(1), 28–31 (bibliographic record)
• Whipple, H. (1998). A movement psychodiagnostic inventory of alcoholic clients: a pilot study (MA thesis) (record only; full text restricted)
• O'Toole, B. T. (1993). MA thesis on the MPI and the Luria-Nebraska Neuropsychological Battery (record only; full text restricted)
• Logos Verlag: table of contents, Koch & Bender (Eds.) (2007), Movement Analysis – Bewegungsanalyse
• BISp SURF record: Movement analysis – the legacy of Laban, Bartenieff, Lamb and Kestenberg (Logos, 2007)
• NEIT library catalogue: Cruz & Berrol (Eds.), Dance/Movement Therapists in Action, 2nd ed. (2012), table of contents
• Koch, S. C., & Bräuninger, I. (2006). International dance/movement therapy research: Recent findings and perspectives. American Journal of Dance Therapy, 28(2), 127–136
• Peter Lang: Lausberg (Ed.), Understanding Body Movement (NEUROGES), publisher page
• Lausberg (Ed.), Understanding Body Movement: table of contents (DNB/Lehmanns)
• Vella & Torres Solera (2012), English version, Papeles del Psicólogo
• Arts Therapist: protected titles and regulator (UK Regulated Professions Register)

Books and papers cited (details as given by the pages above; content cited only through those pages):

• Davis, M. (1991). Guide to Movement Analysis Methods (MPI is Part 2); a 1997 edition is also cited. [via UMD]
• Davis, M. (2019). Guide to the Movement Psychodiagnostic Inventory (MPI). [via UMD]
• Davis, M. (1981). Movement characteristics of hospitalized psychiatric patients. American Journal of Dance Therapy, 4(1), 52–71. [via Vella & Torres Solera 2012]
• Cruz, R. F., & Koch, S. C. (2004). Issues of validity and reliability in the use of movement observations and scales. In R. F. Cruz & C. F. Berrol (Eds.), Dance/Movement Therapists in Action: A Working Guide to Research Options (pp. 45–68). Charles C Thomas. 2nd ed. 2012. [authors verified via NEIT catalogue; Vella & Torres Solera misspell the second author as "Kogh"]
• Davis, M., Lausberg, H., Cruz, R. F., Berger, M. R., & Dulicai, D. (2007). The Movement Psychodiagnostic Inventory (MPI). In S. C. Koch & S. Bender (Eds.), Movement Analysis – Bewegungsanalyse: The Legacy of Laban, Bartenieff, Lamb and Kestenberg (pp. 119 ff.). Logos. [via Logos table of contents and BISp record]
• Davis & Hadiks (1990, 1994, 1995) and Davis et al. (2005). [cited via Lausberg 2013; full references uncertain]