Negative practice

1. Introduction

Version followed: Knight Dunlap's negative practice (beta hypothesis, 1928; Habits: Their Making and Unmaking, 1932; 1942 technique paper) as operationalised for tics in A. J. Yates's (1958) massed-practice paradigm, with parameter guidance from the Weems & Costa SAGE encyclopedia entry. Dunlap's own procedural chapter (ch. X, "The Breaking of Specific Bad Habits", 1932 book) is not drawn on; his conditions are reported here via secondary summaries and are marked accordingly. [source: Weems & Costa, SAGE; Irwin 1933 review; Robertson 2019; Dunlap 1932 ch. I]

What it is. Negative practice is a single behavioural technique, not a complete therapy: the client deliberately and repeatedly performs an unwanted habitual response (a tic, a habitual error, a habit such as teeth grinding) in order to eliminate it. Dunlap defined it as the elimination of habitual responses by their conscious repetition. "Massed" negative practice is the version in which repetitions are packed into long blocks with as little rest as possible. [source: Irwin 1933; Weems & Costa; Romanczyk & Gillis, SAGE]

How change is supposed to happen (two accounts).

• Dunlap (beta hypothesis): repetition does not by itself strengthen a habit; what matters is the person's intention during practice. Practising a response with the aim of getting rid of it, while bringing it under voluntary control, weakens the involuntary habit. He stated that "the determining factors are the thoughts and desires in the practice." [source: Robertson 2019 quoting Dunlap 1932; Wikipedia: Knight Dunlap]
• Hull/Yates (learning theory): massed repetition builds reactive inhibition (fatigue-like work inhibition) that forces the person to stop; the act of not-performing the response is then reinforced as that fatigue dissipates, building conditioned inhibition, a learned "not doing the tic" habit. [source: Weems & Costa; Yates 1958 summary via Brainscape; Wikipedia: Massed negative practice]

Status today (context only). For tics, massed negative practice has been superseded by habit reversal training (HRT) and the Comprehensive Behavioral Intervention for Tics (CBIT); guideline reviews list it among techniques with little supporting evidence. Dunlap also applied it to sexual behaviours (masturbation, homosexuality) that are not legitimate treatment targets today; those uses are excluded from this guide. [source: Wikipedia: Massed negative practice citing Andrén et al. 2021; Frank & Cavanna 2013 abstract; Irwin 1933] [uncertain: guideline full text]

2. Framework

The sources describe no formal phases. Dunlap and Yates describe a procedure (rationale, voluntary reproduction, massed trials, rest, repetition across days); the phase structure below is my organisation of that procedure for clinical use. [synthesis, built from: Weems & Costa; Yates 1958 summary; Robertson 2019; Azrin, Nunn & Frantz 1980]

Session count and length. No standard exists. The SAGE entry states that the therapist must set the number or duration of repetitions and rest intervals, that general rules may not help, and that parameters should be tailored to each individual; in principle more repetitions with less rest produce the largest reduction. Yates (1958) compared different practice and rest lengths with one patient and found very prolonged massed practice followed by prolonged rest worked best. Reported single-case work ranges from a few sessions to several weeks. [source: Weems & Costa; Yates 1958 summary] [uncertain: Yates's exact durations; commonly reported as blocks ranging from a minute up to about an hour]

Indications described in sources: motor tics, including tics in Tourette syndrome (single cases; inconsistent evidence); nocturnal teeth grinding; habitual performance errors in typing, shorthand, spelling, piano or organ playing; maze-learning errors; pica; nail biting; thumb-sucking; stammering. The SAGE entry notes it works less well where the problem involves inhibition (speech blocking, stammering). [source: Weems & Costa; Irwin 1933; Robertson 2019; Azrin, Nunn & Frantz 1980]

3. Guiding principles

• Intention governs the effect (Dunlap). Practise with the explicit aim of eliminating the response; practice aimed at forming or perfecting it would strengthen it. In practice: restate the purpose at the start of every block. [source: Robertson 2019 summarising Dunlap 1932]
• Reproduce the real response. The voluntary version should copy the habitual response as accurately as possible. In practice: compare the practised tic or error with the one recorded at baseline and correct the copy. [source: Robertson 2019] [uncertain: secondary]
• Bring the involuntary under voluntary control. The technique works by turning an automatic act into a deliberate one. In practice: the client starts, performs and stops each repetition on purpose. [source: Weeks & L'Abate via Wikipedia: Knight Dunlap]
• Mass the trials; minimise rest within a block. Reduction is expected to be greatest with many repetitions and little rest. In practice: keep inter-repetition gaps short within the scheduled block. [source: Weems & Costa]
• Follow massed practice with real rest. Yates found prolonged rest after prolonged practice gave the best result and no recovery of the tic. In practice: schedule a genuine break after each block and do not test for the habit immediately. [source: Yates 1958 summary]
• Individualise parameters. No general rule for practice and rest duration; the therapist sets and adjusts them. In practice: start with short blocks, record frequency, lengthen or shorten based on data. [source: Weems & Costa]
• Measure, do not assume. Results are inconsistent across behaviours. In practice: keep baseline and ongoing counts; judge effect on out-of-session frequency, not in-session fatigue. [source: Weems & Costa; Azrin, Nunn & Frantz 1980] [synthesis: emphasis on out-of-session measure]
• One target at a time. Yates treated four tics but with independent practice per tic; the most-practised tic changed most. In practice: allocate practice to the target you most want to change. [source: Yates 1958 summary]
• Attend to motives and replacements. Address what the habit does for the client and have a correct or replacement response ready (e.g. the right spelling or pronunciation). In practice: combine with function-based work where the habit serves a purpose. [source: Robertson 2019] [uncertain: secondary; not confirmed as Dunlap's own rule]
• Prefer the better-supported method. For tics and body-focused habits, habit reversal outperformed negative practice in direct comparisons. In practice: offer HRT/CBIT first and use negative practice only by informed choice or as a time-limited trial. [source: Azrin, Nunn & Frantz 1980; Frank & Cavanna 2013 abstract] [synthesis: first-line ordering]
• Do no harm. Do not apply the technique to behaviours that cause tissue damage when repeated, and never to sexual orientation or other non-pathological behaviour. [synthesis]

4. Intervention inventory

Negative practice is one technique with a small set of components and named applications. Every component the sources describe is listed below; steps, wording and timings not given in the sources are marked [synthesis]. No published manual or scripted protocol exists. [source: Weems & Costa; Romanczyk & Gillis; Robertson 2019; Yates 1958 summary]

4.1 Target definition, baseline and self-recording

• Purpose / when: Before any practice. Defines what will be practised and gives the measure against which change is judged. Azrin, Nunn & Frantz's (1980) negative practice condition included self-recording, and they noted that any procedure with self-recording produced some reduction. [source: Azrin, Nunn & Frantz 1980 via Quaderns review; Yates 1958 summary]
• Steps: (1) Ask the client to describe and demonstrate the response; write an operational definition (body part, movement or sound, duration, what counts as one occurrence). (2) Choose a recording method: tally counter, phone tally or paper sheet, sampled in fixed daily windows. (3) Collect several days of baseline before starting practice. (4) If there are several habits, pick one target. (5) Screen for contraindications (see 4.10). [synthesis: steps 2–4 standard behavioural assessment]
• Example language (illustrative): "Show me the tic exactly as it happens. I'll write down what counts as one, so we're counting the same thing."
• Watch for: reactivity (frequency drops from recording alone, which is fine but must be separated from the practice effect); vague definitions that make counts meaningless; tics that change form (common in tic disorders). [synthesis]

4.2 Rationale and intention setting

• Purpose / when: Before the first practice block and repeated at the start of each block. In Dunlap's account the effect depends on the person's thoughts and desires while practising: practice must be done to get rid of the habit. [source: Robertson 2019 summarising Dunlap 1932]
• Steps: (1) Explain the paradox plainly: deliberately doing the habit, with the goal of stopping it, is expected to weaken the automatic version. (2) Give the learning-theory account if it suits the client (repetition builds fatigue; resting from it builds a "not doing it" habit). (3) Have the client formulate a short statement of purpose to say or think before each block. Dunlap reportedly gave stammerers a statement linking stammering well now to breaking the habit sooner. (4) Check understanding and consent. [source: Robertson 2019; Weems & Costa] [synthesis: steps 1, 2, 4 wording]
• Example language (illustrative): "We're going to do the habit on purpose, again and again, not to get better at it but to wear it out. Before each round, remind yourself: I'm doing this to be rid of it."
• Watch for: the client treating practice as rehearsal or performance; scepticism or feeling mocked (see section 5); shame about displaying the habit. [synthesis]

4.3 Accurate voluntary reproduction

• Purpose / when: First in-session practice. The practised response should match the habitual one so that the inhibition attaches to the right response. [source: Robertson 2019 summarising Dunlap] [uncertain: secondary]
• Steps: (1) Ask the client to produce the response on purpose once. (2) Compare with the baseline description or with the involuntary version observed; give specific feedback ("tilt further", "same sound, shorter"). (3) Repeat single trials until the copy is accurate. (4) Have the client state when each voluntary repetition starts and ends, to emphasise voluntary control. [synthesis: steps 2–4]
• Example language (illustrative): "Do it once now, deliberately, as close to the real thing as you can. Was that the same as when it happens by itself?"
• Watch for: exaggeration that risks strain (neck, jaw, eyes, voice); a copied tic that differs from the involuntary one; urges or anxiety rising sharply. [synthesis]

4.4 Massed practice blocks

• Purpose / when: The core procedure, once reproduction is accurate. Repetitions are performed continuously, with minimal rest, to build reactive inhibition. [source: Weems & Costa; Romanczyk & Gillis; Yates 1958 summary]
• Steps: (1) Set the block length and the inter-block rest; the SAGE entry says this is the therapist's critical decision and must be individualised. (2) Start the timer; the client repeats the response as quickly and accurately as they can sustain, pausing only as needed. (3) Count or log repetitions per block. (4) At the end of the block, stop completely and begin the scheduled rest. (5) Across sessions, lengthen blocks if out-of-session frequency is not falling and the client tolerates the work; shorten them if distress or strain appears. Start with short blocks (around one to five minutes) and build up. [source: Weems & Costa for steps 1–2; Yates 1958 summary for lengthening] [synthesis: starting lengths and adjustment rule] [uncertain: Yates's actual block lengths]
• Example language (illustrative): "For the next five minutes, do it over and over, as fast as you comfortably can. Don't stop to check whether it's working. When the timer goes, stop completely."
• Watch for: genuine fatigue (expected and intended) versus pain or strain (stop); rising anxiety, which in one single-case study was associated with tic frequency increasing after massed practice; the involuntary tic emerging during practice. Signs of working: falling out-of-session counts across days, the response feeling effortful or tedious to produce on demand. [source: Crawley & Powell 1986; Robertson 2019 on awkwardness/tedium] [synthesis: pain vs fatigue]

4.5 Prolonged rest after massed practice

• Purpose / when: After each block or block series. In Yates's study the best condition was very prolonged massed practice followed by prolonged rest, after which the tic did not recover. In Hullian terms, the not-performing response is reinforced as reactive inhibition dissipates. [source: Yates 1958 summary]
• Steps: (1) After the block, the client stops all deliberate practice and rests (sit quietly, unrelated activity). (2) Do not ask the client to monitor or test the habit immediately after the block. (3) Schedule longer breaks (days) between series once frequency has fallen, and re-measure after the break to check for recovery. [synthesis: steps 1–3 operationalise Yates's finding]
• Example language (illustrative): "That's the end of the round. Rest now; no checking, no practising. The rest is part of the treatment."
• Watch for: spontaneous recovery after rest (if counts rebound, the parameters or the method need changing); clients cutting rest short to "get more done". [synthesis]

4.6 Daily home practice periods

• Purpose / when: Once in-session practice is accurate and tolerated. Dunlap reportedly used two 15-minute periods a day for three to four weeks for some habits. [source: Robertson 2019 summarising Dunlap 1932] [uncertain: secondary]
• Steps: (1) Agree fixed times and a private place. (2) Client restates the purpose (4.2), practises for the agreed block length, then rests. (3) Client logs practice done and out-of-session frequency. (4) Review logs at each session and adjust. [synthesis: steps 1, 3, 4]
• Example language (illustrative): "Twice a day, morning and evening, set a timer and do your practice round, then write down how it went and your usual count."
• Watch for: skipped practice, practice done half-heartedly or in public settings where it causes embarrassment; children needing a parent to prompt but not to police. [synthesis]

4.7 Deliberate repetition on spontaneous occurrence

• Purpose / when: During the home-practice phase. When the habit occurs involuntarily, the client immediately repeats it on purpose, linking each spontaneous occurrence to voluntary practice. [source: Robertson 2019 summarising Dunlap 1932] [uncertain: secondary; not confirmed in Dunlap's own text]
• Steps: (1) Agree in advance a short burst (a few deliberate repetitions) to follow each noticed occurrence, in private settings only. (2) Pair it with the purpose statement. (3) Log occurrences and whether the burst was done. [synthesis: steps 1–3]
• Example language (illustrative): "When you catch yourself doing it, and you're somewhere private, do it three more times on purpose, reminding yourself why."
• Watch for: bursts becoming frequent enough to add up to many more repetitions than intended; social embarrassment; it turning into a compulsion-like ritual. [synthesis]

4.8 Pairing with the correct or replacement response

• Purpose / when: For habitual errors and habits that have a correct alternative (a misspelling, a mispronounced word, a misplayed passage). Have the replacement response available once the error has been negatively practised. [source: Robertson 2019] [uncertain: secondary; whether Dunlap required this is not confirmed]
• Steps: (1) After a negative-practice block of the error, the client produces the correct form a few times, slowly and attentively. (2) In daily use, the client aims for the correct form. [synthesis]
• Example language (illustrative): "You've typed it wrong on purpose twenty times. Now type the correct word three times, slowly."
• Watch for: positive practice crowding out the negative practice; for tics, this component overlaps with the competing response in HRT, which is a different and better-supported technique. [synthesis]

4.9 Applications described in the sources

The core steps (4.1–4.8) are the same across applications; only the target and practicalities change. [source: Weems & Costa; Irwin 1933]

• Motor tics, including Tourette syndrome. Yates (1958) treated four tics in one adult woman (breathing, nasal, cough, eye-blink), practising each independently; the most-practised tic fell most and the patient reported improvement outside sessions. Later single cases and a 1985 two-case report (one also receiving psychotherapy, with a suggestion that medication plus massed practice may suit some cases) reported success; other single cases did not. Vocal tics: practise only in private; for socially offensive vocalisations, consider whether practising them aloud is acceptable to the client at all. [source: Yates 1958 summary; 1985 Psychotherapy abstract; Crawley & Powell 1986] [synthesis: vocal tic advice]
• Habitual errors in skills (typing, shorthand, spelling, keyboard instruments). The client deliberately produces the error many times (e.g. typing a habitual transposition), with the intention of eliminating it, then the correct form (4.8). Dunlap noted that even a small amount of practice of the wrong form could remove the error; Holsopple & Vanouse (1929) reported negatively practised shorthand errors fell to zero. [source: Robertson 2019; Weems & Costa]
• Stammering. Dunlap had stammerers stammer deliberately, framed by a purpose statement. The SAGE entry notes negative practice is less effective for problems involving inhibition such as speech blocking and stammering. Refer to or coordinate with a speech and language therapist; modern stuttering therapy uses related "voluntary stuttering" within broader programmes. [source: Robertson 2019; Weems & Costa] [synthesis: referral and voluntary-stuttering link]
• Teeth grinding (bruxism). Daytime voluntary clenching and grinding practice has been used for nocturnal grinding. Coordinate with a dentist first because repeated forceful grinding can damage teeth and jaw joints. [source: Weems & Costa] [synthesis: dental coordination] [uncertain: specific study parameters]
• Nail biting and thumb-sucking. Dunlap applied negative practice to thumb-sucking; Azrin, Nunn & Frantz (1980) found habit reversal more effective than negative practice for nail biting. If used, the client goes through the biting or sucking motion without causing tissue damage. [source: Irwin 1933; Azrin, Nunn & Frantz 1980] [synthesis: no tissue damage]
• Pica. Listed among reported applications; no procedural details found. Do not have the client ingest non-food items; use current function-based behavioural approaches instead. [source: Weems & Costa] [synthesis]

4.10 Safety screen and informed consent

• Purpose / when: Before Phase C, and whenever a new target is chosen. Negative practice can provoke anxiety and has been followed by tic increases in single cases. [source: Crawley & Powell 1986] [synthesis: the screen itself]
• Steps: (1) Exclude targets whose repetition can injure (forceful neck or head jerks, self-hitting, skin picking, hair pulling, biting that breaks skin, forceful grinding without dental clearance). (2) Check for high general anxiety, OCD or a compulsive quality to the habit, where massed repetition may reinforce rather than inhibit. (3) Explain the alternatives (HRT/CBIT for tics and body-focused repetitive behaviours), the limited evidence for negative practice, and the possibility of temporary worsening. (4) Agree stop rules: pain, marked distress, or a sustained rise in out-of-session frequency over the baseline. (5) For children, obtain caregiver consent and the child's assent, and keep blocks short. [synthesis]
• Example language (illustrative): "This method has less research behind it than habit reversal, and sometimes a tic goes up before it goes down. If it hurts, or your counts rise for more than a week, we stop and change plan."
• Watch for: consent obtained under pressure (e.g. a parent's wish); targets that are not disorders. Never use it for sexual orientation, gender expression or other non-pathological behaviour. [synthesis]

5. Handling problems

The sources say almost nothing about clinical difficulties beyond inconsistent results and anxiety-linked worsening. Everything below except where tagged otherwise is my synthesis.

• The habit gets worse. Crawley & Powell (1986) found a facial tic rose after massed practice and attributed this to the patient's general anxiety and to anxiety produced by the procedure itself. Response: stop massed blocks, return to baseline recording for several days; if the rise persists, switch to HRT/CBIT. If anxiety drove the rise, treat anxiety first (relaxation lowered tic frequency during its own block in the same study). [source: Crawley & Powell 1986] [synthesis: response]
• No change after a fair trial (stuck). Results are inconsistent across behaviours. Response: check that practice was massed (short gaps), accurate (4.3), done with the eliminating intention (4.2), and followed by real rest (4.5). Adjust parameters once (longer blocks, longer rest). Set a review point in advance, e.g. two to three weeks of consistent practice with no downward trend, then switch method rather than escalate. [source: Weems & Costa; Yates 1958 summary] [synthesis: review rule]
• Gains in session, none outside. Yates reported generalisation only by subjective report, mainly for the most-practised tic. Response: rely on out-of-session counts; move practice into the settings where the habit occurs (privately); add home practice (4.6) and occurrence-linked repetition (4.7). [source: Yates 1958 summary] [synthesis: response]
• Resistance: "This is absurd" or "You're making fun of me." Response: validate that it sounds backwards; explain the two rationales (4.2); frame it as a short experiment with data; offer HRT as an alternative. Do not push a paradoxical method on an unwilling client. [synthesis]
• Shame or embarrassment about performing the habit in front of the therapist. Response: model the response yourself once, normalise; let the client face away or practise alone in the room for the first block; agree that home practice is private. [synthesis]
• Overwhelm, panic or strong urges during blocks. Response: stop the block, use grounding or slow breathing, shorten future blocks, and reconsider suitability. Clients with marked anxiety may do better with relaxation or a comprehensive tic programme. [source: Crawley & Powell 1986] [synthesis: response]
• Pain or physical strain. Response: stop immediately; reduce force and range; for neck, jaw, eye or voice targets, get medical, dental or speech-therapy clearance before resuming; drop the target if strain recurs. [synthesis]
• Non-adherence to home practice. Response: problem-solve timing and place; shorten to one period a day; use reminders and a simple log; for children, caregiver prompts plus praise for practising (not for low counts). If practice remains rare, the method is not being delivered; switch rather than blame. [synthesis]
• Practice turns ritualistic or compulsive. Some clients start repeating the habit beyond the schedule, or feel unable to stop. Response: return to strict time-limited blocks; if this persists, stop and assess for OCD-related problems, which need ERP rather than negative practice. [synthesis]
• New tics or habits appear. Tic disorders wax and wane and tics change form. Response: keep recording; do not chase each new tic with negative practice; re-evaluate the overall plan, typically CBIT with psychoeducation and function-based components. [synthesis] [uncertain: symptom substitution with this method]
• Rupture (client feels blamed, criticised, or that the therapist does not take the problem seriously). Response: name the strain, ask directly how the method feels, acknowledge the method's limits and the client's effort, and collaboratively decide whether to continue or switch. [synthesis]
• Medication. One case report suggested combining massed practice with medication for some Tourette patients. Response: coordinate with the prescriber; this guide gives no dosing. [source: 1985 Psychotherapy abstract] [synthesis: coordination]
• When to stop using negative practice altogether. For tics, nail biting and other body-focused habits, habit reversal has outperformed negative practice, and HRT/CBIT is now the standard behavioural treatment. Use negative practice only as an informed, time-limited choice, or for habitual skill errors where the risk is low. [source: Azrin, Nunn & Frantz 1980; Frank & Cavanna 2013 abstract; Wikipedia: Massed negative practice] [synthesis: decision rule]

Sources

Sources:

• Massed negative practice (Wikipedia)
• Negative practice, Weems & Costa, SAGE Encyclopedia of Behavior Modification and CBT
• Massed practice, Romanczyk & Gillis, SAGE Encyclopedia of Behavior Modification and CBT (partial)
• Dunlap, Habits: Their Making and Unmaking, full text (Internet Archive; introduction and chs I–II only were readable)
• Irwin 1933, review of Dunlap's Habits, Psychological Clinic
• Robertson, The Paradox of Negative Practice in Behaviour Change
• Paradoxical therapies (Psychology: An Introduction, psywww)
• Knight Dunlap (Wikipedia)
• Yates: The application of learning theory to the treatment of tics (study summary flashcards)
• Crawley & Powell 1986, A comparison of the effects of massed practice and relaxation upon the frequency of a facial tic (abstract)
• Massed Negative Practice as a Behavioral Treatment for Gilles de la Tourette's Syndrome, American Journal of Psychotherapy 1985 (abstract)
• Frank & Cavanna 2013, Behavioural Neurology (abstract)
• Review citing Azrin, Nunn & Frantz 1980 (Quaderns de Psicologia)
• Shall you do it voluntarily? A behavioral approach to Tourette's syndrome (abstract)

Further reading (not drawn on unless listed above):

• Dunlap, K. (1928). A revision of the fundamental law of habit formation. Science, 67. [uncertain]
• Dunlap, K. (1932). Habits: Their Making and Unmaking. New York: Liveright. (chapter X not readable)
• Dunlap, K. (1942). The technique of negative practice. American Journal of Psychology, 55. [uncertain]
• Yates, A. J. (1958). The application of learning theory to the treatment of tics. Journal of Abnormal and Social Psychology, 56. [uncertain: summary only]
• Azrin, N. H., Nunn, R. G., & Frantz, S. E. (1980). Habit reversal vs. negative practice treatment of nailbiting. Behaviour Research and Therapy, 18, 281–285.
• Crawley, B., & Powell, G. (1986). Behavioural Psychotherapy, 14(3), 249–257.
• Andrén, P., et al. (2021). European clinical guidelines for Tourette syndrome and other tic disorders, version 2.0, Part II: psychological interventions. European Child & Adolescent Psychiatry. [uncertain]
• Woods, D. W., Himle, M. B., & Conelea, C. A. (2006). Behavior therapy: other interventions for tic disorders. Advances in Neurology, 99. [uncertain]
• Weeks, G. R., & L'Abate, L. (1982). Paradoxical Psychotherapy. [cited via Wikipedia]