Pivotal response treatment

1. Introduction

Version followed: Koegel & Koegel's Pivotal Response Treatment (UC Santa Barbara / Koegel Autism Center lineage; Koegel & Koegel 2006, The PRT Pocket Guide 2012, 2nd ed. of Pivotal Response Treatment for Autism Spectrum Disorders), with session-level components taken from the Stanford PRT package (Hardan et al. 2015; Gengoux et al. 2019), which was adapted from the Koegel et al. (1989) manual How to Teach Pivotal Behaviors to Children with Autism. Classroom adaptations follow CPRT (Stahmer, Suhrheinrich & Rieth). Where these differ, the difference is named.

Pivotal Response Treatment (PRT; earlier "pivotal response training", and before that the "Natural Language Paradigm", Koegel, O'Dell & Koegel 1987) is a naturalistic, play-based form of applied behaviour analysis for autistic children (and, in later work, adolescents and adults). Rather than drilling skills one at a time, it targets a few "pivotal" areas, above all motivation, plus responding to multiple cues, self-initiation, self-management and (in development) empathy, on the theory that change in these areas produces broad collateral gains in untargeted language, social and behavioural skills. Change happens through ordinary operant learning arranged so the child wants to respond: the child chooses the activity, the adult controls access to it, any reasonable attempt is reinforced immediately with the very thing the child was trying to get, and this is delivered in natural settings by parents, teachers and peers so gains generalise. [source: Wikipedia PRT; Koegel PRT Pocket Guide intro; Koegel & Koegel Stanford slides (UNL PRT Part 1); Gengoux et al. 2019]

PRT is a set of procedures, not a fixed-session protocol: it is used as a stand-alone intervention, as a parent-training programme, or as a component inside a broader programme. It is primarily a child intervention; for a psychotherapist, the usual roles are direct delivery in play and, most often, coaching parents and school staff. [source: Gengoux et al. 2019; Koegel empirical support table] [synthesis: the role framing for psychotherapists]

2. Framework

PRT has no officially numbered phases in the Koegel texts. The sequence below is the order of work implied by the Pocket Guide's structure (motivation first, then disruptive behaviour, then self-initiation; then family involvement, natural-environment assessment and data collection) and by the Stanford trial schedules. [source: PRT Pocket Guide table of contents; Gengoux et al. 2019; NCT05948202 protocol] [synthesis: the phase labels and transition criteria are mine unless cited]

Phase 0: Assessment and planning (1–3 sessions). Goals: identify preferred items, activities and interests by watching the child with materials freely available; take a baseline language sample in play (e.g., 10 minutes, counting spontaneous, prompted and imitative utterances, as in the Stanford structured laboratory observation); list easy (maintenance) vs. harder (acquisition) skills; identify disruptive behaviours and their likely function; record family routines where teaching can be embedded. The 2nd edition adds an explicit strengths-based assessment linked to treatment planning. [source: OCALI PRT implementation checklist; Gengoux et al. 2019; Brookes 2nd ed. product page] Move on when you have a preference list, 1–3 initial targets and a baseline. [synthesis]

Phase 1: Motivation (the core; typically the first 6–12 weeks). Goals: establish the motivational package (attention, clear opportunity, child choice/shared control, turn-taking, task variation, maintenance/acquisition interspersal, contingent and natural reinforcement of attempts) in play and daily routines; teach first words or longer utterances depending on level. In the Stanford PRT-P package this was weekly 60-minute parent training plus 10 hours/week of clinician in-home PRT for 12 weeks; in the Stanford group model, 12 weeks of group parent classes; in the NCT05948202 adaptation, 11 group sessions (16 h) plus 4 h individual coaching over 12 weeks, with a weekly 10-minute home video. The Koegels also describe a week-long intensive centre-based parent programme (Koegel, Symon & Koegel 2002). [source: Gengoux et al. 2019; Stanford Medicine news 2014; NCT05948202 protocol; Koegel empirical support table] Transition criterion: the adult implements the components at 80% fidelity or better on a 5–10 minute sample (the criterion used in Stanford trials), and the child is producing communicative attempts in most opportunities. [source: Gengoux et al. 2019; NCT05948202] [synthesis: the child-side criterion]

Phase 2: Expanding pivotal areas (ongoing, overlapping). Add, as the child's level allows: multiple-cue responding (generally once cognitive/language level is roughly 3 years or above, per CPRT guidance); child-initiated questions ("What's that?", "Where is it?", "Whose is it?", "What happened?", plus "Look" and "Help"); self-management for older or more verbal children; social interaction (embedded social reinforcers, play dates, peer-mediated PRT, interest-based group activities); and empathic communication for verbal adolescents and adults. In the Koegel question-asking studies sessions were about 60 minutes twice weekly, with probes about every fifth session. [source: Stahmer, Suhrheinrich & Rieth 2016; UNL PRT Part 2 handouts; Koegel et al. 2010; Koegel et al. 2016 via Hussman summary] Move a target from acquisition to maintenance when it is used without prompts on repeated probes and in a second setting or with a second person. [source: Koegel et al. 2010 probe procedure] [synthesis: generalised rule]

Phase 3: Generalisation, maintenance and fading (Stanford maintenance phase: monthly 60-minute parent sessions plus 5 h/week clinician PRT for 12 weeks). Goals: parents and teachers run PRT across all routines (meals, bath, car, park, homework); fade clinician presence; fade self-management devices; coordinate home and school (priming, shared targets). [source: Gengoux et al. 2019; Stanford Medicine news 2014; Brookes self-management excerpt; UNL priming handout] Ending or stepping down: parents maintain fidelity without coaching, targets maintain at follow-up probes, and new goals can be selected by the family. [synthesis]

Format. Individual play-based sessions in home, clinic or school; parent coaching (individual or group) using didactic teaching, modelling, live or video-based practice with feedback, and homework; classroom delivery (CPRT) for groups of students. Intensity varies widely across sources; no single recommended dose exists. [source: Gengoux et al. 2019; NCT05948202; Stahmer et al. 2016] [uncertain: no Koegel-authored dosage standard was found]

3. Guiding principles

Target pivotal areas, not isolated skills. Choose goals that should unlock many other behaviours (motivation, initiation, multiple cues, self-management, empathy) rather than long lists of discrete skills. [source: Koegel & Koegel slides (UNL Part 1); Wikipedia PRT]
Motivation is the central pivotal area. If the child is not responding, the first question is "what about the set-up is unmotivating?", not "what is wrong with the child?" [source: PRT Pocket Guide intro] [synthesis: the phrasing]
Treatment in the natural environment. Teach where the skill is needed (home, playground, classroom, community) so you do not have to programme generalisation later. [source: Koegel & Koegel slides (UNL Part 1)]
Family involvement; parents as primary agents. Plan sessions so the parent is doing the teaching by the end; the clinician is a coach. [source: Koegel & Koegel slides; Wikipedia PRT; Gengoux et al. 2019]
Child choice and shared control. Follow the child's lead in materials and topics while keeping control of access, so that the child's interest generates the learning opportunity. [source: NCT05948202; Stahmer et al. 2016]
Natural, direct reinforcement. The reward is the thing the response was about (say "ball", get the ball), not an unrelated token or edible. [source: Wikipedia PRT; Williams, Koegel & Egel 1981 via Koegel table]
Reinforce attempts, not just perfection. Any clear, goal-directed attempt earns the reinforcer; this keeps responding high and speeds learning. [source: Koegel, O'Dell & Dunlap 1988 via Koegel table]
Keep success high. Intersperse easy maintenance tasks with new ones and vary tasks so frustration and escape behaviour stay low. [source: Dunlap 1984; Dunlap & Koegel 1980 via UNL Part 1]
Clear, uninterrupted opportunities. Get attention first, then give one short, developmentally appropriate cue. [source: Stahmer et al. 2016; OCALI checklist]
Contingency. Reinforcement follows the target response immediately and is withheld for non-attempts or inappropriate behaviour. [source: NCT05948202]
Disruptive behaviour is reduced mainly by improving motivation and teaching communication, not by adding punishment. [source: Koegel, Koegel & Surratt 1992 and Koegel et al. academic work via UNL Part 2] [synthesis: "not by adding punishment" reflects the PBS orientation of the Koegels' 1996 edited volume; not a quoted rule]
Strengths-based, low-stress delivery for families. Attend to parent stress and affect; strength-based feedback improves parent-child interaction. [source: Pocket Guide ch. 6 title; Mossman-Steiner 2011 via Koegel table]
Measure, but make it easy. Collect brief, regular data on adult fidelity and child responding. [source: Pocket Guide ch. 8 title; Gengoux et al. 2019]
Aim for independence. Fade adult prompts and presence; self-management and child-initiated questions are explicitly about the child not needing the adult. [source: Koegel, Koegel & Parks 1995 via IIDC summary]

4. Intervention inventory

Grouped by pivotal area. A1–A9 are the core motivational components; they run together in every learning opportunity (attention, clear opportunity, child response, contingent natural consequence) and fidelity checklists score them together. Sources group them differently: Stanford/NCT05948202 uses 5 techniques / 10 components (shared control, opportunities to respond, interspersing tasks, contingent reinforcement, attempts); CPRT uses "maximises motivation", "facilitates responding", "provides appropriate consequences". Content is the same. [source: NCT05948202; Stahmer et al. 2016; OCALI checklist]

A1. Gain attention before the opportunity

Purpose / when: Every opportunity; a cue given while the child is elsewhere teaches nothing. [source: Stahmer et al. 2016; NCT05948202]
Steps: (1) Get to the child's level, near the wanted item, or hold the item near your face. (2) Wait for attention to you or the item (looking, reaching, pausing). (3) Then give the cue (A2). Attention to the item is enough; eye contact is not required. [source: NCT05948202, "attend to the child's focus"] [synthesis: positioning detail]
Example language (illustrative): (Holding the bubbles up, waiting for the child to look toward them, then) "Bubbles?"
Watch for: Cues delivered into the air; insisting on eye contact. Working: the child orients quickly when you hold preferred items.

A2. Clear, appropriate opportunity (instruction)

Purpose / when: Every opportunity; the child must know what response is expected. [source: Stahmer et al. 2016; OCALI checklist]
Steps: (1) Pitch the cue at or slightly above the child's current level (single word for a non-verbal or one-word child; short phrase for a phrase-level child). (2) Keep it short, one cue at a time, no stacked questions. (3) Make it relevant to what the child is doing now. (4) Give it once and wait (see A9 time delay). [source: NCT05948202; OCALI checklist] [synthesis: "one-word-up" pitching]
Example language (illustrative): Child at one-word level reaching for a ball: "Ball." Child at two-word level: "Roll ball."
Watch for: Long sentences, rapid repeated questions, cues far above level. Working: rising proportion of opportunities that get any attempt.

A3. Child choice and shared control

Purpose / when: Throughout; the child's choice supplies motivation, the adult's control of access creates the opportunity. [source: NCT05948202; Koegel, Dyer & Bell 1987 via Koegel table]
Steps: (1) Offer several preferred options (from Phase 0) or let the child pick freely. (2) Follow the child's lead; if interest drops, switch activities rather than pushing on. (3) Identify and control the natural reinforcer: hold the item, keep the container closed, keep pieces in your hand, do the next step of the routine only after a response. (4) Balance adult- and child-selected materials so the adult can also bring in target items. (5) Decide in routines which steps you do and which the child completes. [source: OCALI checklist; NCT05948202]
Example language (illustrative): "Trains or blocks?" (child picks trains) — keep the track pieces, hand them one at a time contingent on attempts.
Watch for: Adult taking over (low child engagement) or adult giving free access (no opportunities). Working: the child stays engaged and keeps coming back to you for access.

A4. Turn-taking with modelling

Purpose / when: In shared play; gives the adult natural control of the material and a chance to model play and language. [source: Wikipedia PRT; Stahmer et al. 2016]
Steps: (1) Take a short turn with the material. (2) During your turn, model an age-appropriate play action and a short phrase. (3) Give the turn back, often contingent on a request ("my turn"/"your turn" or the item name). CPRT varies turn-taking by language level and also sets up turns between students. [source: Stahmer et al. 2016]
Example language (illustrative): "My turn... car goes fast!" (pause, hold car) — child: "Car!" — "Your turn, car!"
Watch for: Turns so long the child disengages; distress at losing the item (shorten your turns). Working: the child tolerates and anticipates turns and imitates modelled play or words.

A5. Task and stimulus variation

Purpose / when: Whenever responding flattens or the child shows boredom. [source: Dunlap & Koegel 1980 via Koegel table]
Steps: (1) Change materials, activities, instructions and settings frequently rather than repeating the same trial. (2) Teach the same target with several exemplars (different balls, books, rooms). [source: OCALI checklist; Stahmer et al. 2016 on multiple exemplars]
Example language (illustrative): After three rounds of "ball" with a red ball, switch to a big bouncy ball, then to rolling it down a ramp.
Watch for: Drilling; changing so fast the child never practises. Working: stable or rising engagement across a session.

A6. Intersperse maintenance and acquisition tasks

Purpose / when: Every session; keeps success high and limits escape behaviour. [source: Dunlap 1984 via UNL Part 1; NCT05948202]
Steps: (1) From Phase 0, list mastered (maintenance) responses and current targets (acquisition). (2) Mix them; OCALI's guidance is several quick easy requests followed by one or two harder ones. (3) Raise the share of acquisition tasks as the child succeeds. [source: OCALI checklist]
Example language (illustrative): "Car" (mastered), "Car" (mastered), "Blue car" (new).
Watch for: Mostly-new-task sessions with rising frustration. Working: high overall success with steady acquisition.

A7. Contingent reinforcement

Purpose / when: After every response; makes the response-consequence link clear. [source: NCT05948202; Koegel & Egel 1979 via Koegel table]
Steps: (1) Deliver the reinforcer immediately (within a second or two) after an acceptable response. (2) Reinforce only the target or acceptable attempt, not grabbing, crying or unrelated behaviour. (3) If no response, prompt (A9), then reinforce the prompted response. [source: NCT05948202] [synthesis: sequencing]
Example language (illustrative): Child: "Buh!" — adult instantly blows bubbles: "Bubbles!"
Watch for: Delayed reinforcement; giving the item after problem behaviour. Working: responses become faster and more consistent.

A8. Natural (direct) reinforcement

Purpose / when: Always preferred; reinforcer is directly related to the response, which speeds learning and generalisation. [source: Koegel & Williams 1980; Williams, Koegel & Egel 1981 via Koegel table; Wikipedia PRT]
Steps: (1) Choose targets that are part of the chain leading to the reinforcer (the word for the toy, the action word for the action). (2) Give that item or action, with social praise added, not an unrelated edible or token. (3) In classrooms where direct reinforcement is hard, CPRT allows a token system as a bridge. [source: Stahmer et al. 2016]
Example language (illustrative): Child says "open" — you open the box. Child says "push" — you push the swing.
Watch for: "Say ball, get a sweet". Working: the child uses the word spontaneously to get the item outside session.

A9. Reinforce reasonable attempts; prompting and time delay

Purpose / when: Every opportunity; reinforcing attempts produces faster progress than reinforcing only correct responses. [source: Koegel, O'Dell & Dunlap 1988 via Koegel table]
Steps (attempts): Reinforce any clear, goal-directed verbal attempt (approximation, or correct word with attention to item or adult) unless a planned shaping programme specifies a stricter criterion. Do not reinforce responses that are clearly off-task, self-stimulatory or less than the child's demonstrated best. [source: NCT05948202] [synthesis: "demonstrated best" rule]
Steps (prompts, Stanford materials): Prompt types are model prompt (adult says the target), time delay (expectant pause), choice ("X or Y?"), and question ("What do you want?"); a carrier phrase is also listed but not defined. If a time delay draws no response within roughly 3–5 seconds, give a model prompt. Keep the model matched to the object. Keep labels/recasts distinguishable from prompts. Use "say ..." only to model a replacement behaviour or when the child is stuck in echolalia or unsure. Vary prompt types. [source: NCT05948202]
Example language (illustrative): (Holding juice, waiting 3–5 s) — no response — "Juice." — child: "Juh" — give juice: "Juice!" A choice prompt: "Juice or milk?"
Watch for: Over-prompting (no waiting), only one prompt type, the child becoming prompt-dependent; in the Stanford trial gains were mainly in nonverbally prompted utterances, so plan explicit fading toward spontaneous requests. Working: attempts become clearer, and more occur on time delay alone. [source: Gengoux et al. 2019]

A10. Target selection and expanding utterances (Natural Language Paradigm lineage)

Purpose / when: At start and whenever a target is mastered. The 1987 Natural Language Paradigm used child-selected toys, varied often, adult modelling of play and words, and reinforcement of attempts with the toy; it out-performed structured teaching on spontaneous speech and generalisation. [source: Koegel, O'Dell & Koegel 1987 via UNL Part 1 and Koegel table]
Steps: (1) Start from the child's current level (sounds, single words, word combinations, sentences). (2) Pick targets that name things the child actually wants or actions they enjoy. (3) Model one step above current level. (4) Once a target is reliably used, require it (or a longer form) and move it into maintenance. (5) Add new forms (verbs, adjectives, prepositions, pronouns, grammatical morphemes) partly through the question-asking procedures in C. [source: Koegel et al. 2010; Koegel, Carter & Koegel 2003 via Koegel table] [synthesis: the step ladder]
Example language (illustrative): Child says "car" reliably — model "go car" or "blue car" next.
Watch for: Raising criteria too fast (attempts drop). Working: mean length of utterance and spontaneous use rise.

B1. Responding to multiple cues (conditional discrimination)

Purpose / when: Addresses stimulus overselectivity (responding to only one feature). CPRT recommends conditional discrimination training only for learners with cognitive age above about 3; below that, use multiple exemplars instead. [source: Stahmer et al. 2016; Schreibman, Charlop & Koegel 1982 via Wikipedia PRT; Koegel & Koegel slides list "multiple cues" as pivotal]
Steps: (1) Arrange items differing on two features, e.g., red and blue cars plus red and blue blocks. (2) Ask for an item defined by both features in a motivating context. (3) Reinforce with the requested item. (4) Have the child make such requests too ("red car"), with you giving only the exact item named. [source: Stahmer et al. 2016 example of colour and shape] [uncertain: step sequence reconstructed; the Koegel manual chapter is not drawn on]
Example language (illustrative): "Can you hand me the big red block?" Or, the child asks for "blue" and you give a blue block while they wanted the car: "Blue block? Do you want the blue car?"
Watch for: Too many features at once; using non-preferred materials. Working: accurate selection on new combinations.

C. Self-initiation (child-initiated queries)

General rules for all queries: use favourite items as the reinforcer; prompt the question; answer with the target structure; give the item; fade the prompt by pausing expectantly; fade favourite items toward neutral ones; probe in new settings. [source: UNL PRT Part 2 handouts; Brookes PBS excerpt; Koegel et al. 2010]

C1. "What's that?"

Purpose / when: Child has some words but few spontaneous labels; increases vocabulary and initiations. [source: Koegel, Camarata, Valdez-Menchaca & Koegel 1998 via Koegel table]
Steps: (1) Put favourite items in an opaque bag. (2) Prompt the question. (3) When the child asks, pull out the item and label it. (4) Have the child repeat the label. (5) Give the item. (6) Gradually add less familiar or neutral items, then remove the bag. [source: UNL PRT Part 2; Brookes PBS excerpt]
Example language (illustrative): (Shaking the bag) "What's that?" — child: "What dat?" — "Dinosaur!" — child: "Dinosaur" — hand it over.
Watch for: Rote asking without interest in the answer. Working: the child asks about new objects in other settings.

C2. "Where is it?"

Purpose / when: Child combines words but lacks prepositions or "where" questions. [source: Koegel et al. 2010]
Steps: (1) Hide a preferred item. (2) Prompt the question. (3) Answer with the target preposition. (4) Child repeats (in the parent version) and retrieves the item. (5) Fade the verbal prompt: hide and pause. (6) Model each structure several times per session (2–8 in the study); probe roughly every fifth session in new locations; keep unmastered structures in rotation. [source: Koegel et al. 2010; Brookes PBS excerpt] Ordinals (first, second...) can be added for older children.
Example language (illustrative): "Where is it?" — "Under the cup!"
Watch for: Adult giving the item without the preposition. Working: unprompted questions and correct preposition use in a home language sample.

C3. "Whose is it?"

Purpose / when: Teaches possessive pronouns (mine/yours) and possessive -s. [source: Brookes PBS excerpt; UNL Part 2]
Steps: (1) Use items associated with family members (and preferred items). (2) Prompt the question. (3) Answer with the possessive. (4) Child repeats and gets the item. [source: Brookes PBS excerpt]
Example language (illustrative): "Whose is it?" — "It's Daddy's." / "It's yours!"
Watch for: Pronoun reversal; model from the child's perspective. Working: correct pronouns in everyday talk. [synthesis: perspective tip]

C4. "What's happening?" / "What happened?"

Purpose / when: Verb forms (-ing, past tense). [source: UNL Part 2]
Steps: (1) Use pop-up or tab books about the child's interests. (2) Let the child pull tabs. (3) Prompt the question. (4) Answer with the verb form ("The dog is jumping" / "The dog jumped"). [source: UNL Part 2] [synthesis: answer wording]
Watch for: Overly long answers. Working: the child uses target verb forms spontaneously.

C5. "Look!"

Purpose / when: Joint-attention initiation. [source: UNL Part 2; Vismara & Lyons 2007 via Koegel table]
Steps: Use a favourite activity; prompt "look" before showing the interesting event; fade the prompt. [source: UNL Part 2]
Example language (illustrative): (Before the jack-in-the-box pops) "Look!" — then pause on subsequent rounds for the child to say it.

C6. "Help!"

Purpose / when: Requesting assistance. [source: UNL Part 2]
Steps: Set up situations needing help (tight jar, toy out of reach); prompt "help"; give help immediately; fade prompt. [source: UNL Part 2]
Watch for: Frustration if the set-up is too hard; keep waits short.

D1. Self-management package

Purpose / when: Older or more verbal children, adolescents and adults; for behaviour that should occur (or stop) when no adult is present, e.g., stereotypy, responsiveness to others, staying seated. [source: Koegel & Koegel 1990; Koegel, Koegel, Hurley & Frea 1992 via Koegel table]
Steps: (1) Operationally define the target behaviour; start simple. (2) Identify functional reinforcers, ideally chosen by the person. (3) Choose a recording method: check marks or stickers, picture cards (for non-verbal children), wrist counter (event recording) or timer/countdown watch (interval recording). (4) Teach use: model accurate recording, reinforce accurate recording, check understanding (e.g., require around 8/10 correct practice discriminations). (5) Build independence: fade adult prompts, lengthen intervals, increase recordings needed per reinforcer, teach self-delivery of reinforcers. (6) Check use in home, school and community. [source: Koegel, Koegel & Parks 1995 via IIDC; Brookes PBS excerpt] The 2nd edition combines functional behavioural assessment with self-management. [source: Brookes 2nd ed. page]
Example language (illustrative): "When the timer beeps, if you were in your seat, put a sticker here. Five stickers and you choose free time."
Watch for: Inaccurate or gamed recording; reinforcer losing value. Working: behaviour stays changed with the adult absent and as intervals lengthen.

E1. Embedding social interaction in the reinforcer

Purpose / when: Child requests well but treats the adult as a vending machine. [source: Koegel, Vernon & Koegel 2009 via Koegel table]
Steps: Make the reinforcer an activity that needs you (tickles, swinging, chase, you making the toy do something) rather than a solitary item; deliver it contingent on the request. [source: Koegel table summary] [synthesis: activity examples]
Example language (illustrative): Child: "Spin!" — you spin them, with a smile and "Spin, spin, spin!"
Watch for: Working: more eye gaze, smiling and child-initiated social bids.

E2. Motivational play dates

Purpose / when: Reciprocal play with a typically developing peer. [source: Koegel, Werner, Vismara & Koegel 2005 via Koegel table]
Steps: Choose activities that build on the autistic child's interest but need both children to play (cooperative arrangements); adult facilitates then fades. [source: Koegel table summary, "motivational strategies during play dates"] [uncertain: cooperative-arrangement detail]
Watch for: Adult-directed play; the peer losing interest. Working: synchronous, reciprocal exchanges increase.

E3. Peer-implemented PRT

Purpose / when: School or playground. [source: Pierce & Schreibman 1995, 1997; Harper, Symon & Frea 2008 via Koegel table]
Steps: Teach peers a simplified set of PRT components (getting attention, giving choices, taking turns, modelling, reinforcing attempts) through explanation, role-play and feedback; then support them in natural play. [synthesis: training format; peers learning the PRT components is documented, the teaching method is not drawn on]
Watch for: Peer fatigue; rotate peers. Working: more initiations and turn-taking by the target child.

E4. Symbolic and sociodramatic play training

Purpose / when: Child plays repetitively. [source: Stahmer 1995; Thorp, Stahmer & Schreibman 1995 via Koegel table]
Steps: Apply the motivational components to pretend play: adult models a pretend action or role with child-chosen toys, takes turns, reinforces attempts with access to the toy and continued play. [source: Koegel table summary] [synthesis: step detail]

E5. Interest-based social activities (adolescents)

Purpose / when: Teenagers with limited peer contact at school. [uncertain: Koegel and colleagues' high-school lunch-club work (Koegel, Kim, Koegel & Schwartzman 2013) is not drawn on]
Steps: Build a regular peer activity around the student's preferred interest, open to peers who share it; adult facilitates briefly then fades. [synthesis]

E6. Empathic communication training (verbal adolescents and adults)

Purpose / when: Conversation partner shares feelings and the person does not respond to them. Empathy is listed by the Koegels as a pivotal area "in progress". [source: UNL Part 1; Koegel, Ashbaugh, Navab & Koegel 2016 via Hussman summary]
Steps: Weekly 40-minute sessions over about 10 weeks: (1) ~30 minutes reviewing video of last week's conversations with clinician feedback on empathic and non-empathic responses; (2) ~10 minutes live conversation practice. Teach a three-step rule: notice a statement containing an emotion; express understanding; ask a related question. [source: Hussman summary of Koegel et al. 2016]
Example language (illustrative): Partner: "My cat is sick." — "Oh no, that sounds worrying. What did the vet say?"
Watch for: Formulaic responses. Working: empathic responses appear in un-coached conversations.

F1. Motivational PRT for academics (writing, maths, homework)

Purpose / when: Disruptive or avoidant behaviour during academic tasks. [source: Koegel, Singh & Koegel 2010 via UNL Part 2]
Steps: Let the child choose topic and materials (write about trains, map their town); intersperse easy and hard items; embed the reinforcer in the task itself (writing a story about the favourite topic, playing hangman, playing teacher). [source: UNL Part 2]
Example language (illustrative): "Want to write a list of all the dinosaurs you know, or make a dinosaur quiz for me?"
Watch for: Working: faster completion, less disruption, self-initiated writing.

F2. Priming

Purpose / when: School difficulties or transitions; preview upcoming activities in a relaxed setting. [source: Wilde, Koegel & Koegel 1992; Koegel, Koegel, Frea & Green-Hopkins 2003 via UNL priming handout]
Steps: (1) Pick a time (day before, start of day, or before the activity). (2) Any adult or peer with rapport can prime. (3) Quiet place. (4) Use the actual materials and a visual schedule. (5) Briefly preview each activity, about 10 minutes, no teaching, correcting or testing. (6) Reinforce attending; log the session. [source: UNL priming handout]
Watch for: Priming turning into tutoring. Working: more academic responding and less problem behaviour in class.

F3. Classroom PRT (CPRT)

Purpose / when: Teachers delivering PRT in small groups or 1:1 activities. [source: Stahmer et al. 2016]
Steps: Same components as A1–A9; tokens permitted for direct reinforcement; turn-taking between students; multiple exemplars below cognitive age 3; data forms and paraprofessional training materials in the CPRT manual. Fidelity was scored on a 1–5 scale per component, 4–5 counting as meeting fidelity. [source: Stahmer et al. 2016]

G1. Reducing disruptive behaviour

Purpose / when: Tantrums, aggression, escape behaviour. PRT itself produced less disruption than discrete-trial teaching; choice reduces disruption; parent-implemented plans reduced sibling aggression. [source: Koegel, Koegel & Surratt 1992; Carter 2001; Koegel, Koegel & Steibel 1998 via Koegel table]
Steps: (1) Identify the likely function (escape, access, attention) through observation and a functional behavioural assessment. (2) Apply motivational components to make tasks less aversive (choice, interspersal, natural reinforcers). (3) Teach a communicative replacement ("help", "break", "my turn") and reinforce it immediately. (4) Do not deliver the reinforcer after the disruptive behaviour. (5) Add self-management for older children. [source: Pocket Guide ch. 3 title; 2nd ed. FBA + self-management chapter; NCT05948202 on "say" for replacement behaviour] [synthesis: step order; the Pocket Guide chapter itself is not drawn on]

H1. Parent education and coaching

Purpose / when: Core delivery mode. [source: Gengoux et al. 2019; Koegel, Symon & Koegel 2002; Hardan et al. 2015]
Steps: (1) Teach one or two components per session using written materials and video examples. (2) Model with the child. (3) Parent practises with live coaching or via a weekly 10-minute home video reviewed in session. (4) Strengths-based feedback first, then one specific suggestion. (5) Assign practice in named routines; review it next session even if not done; track practice frequency. (6) Group formats work; groups add peer learning. [source: NCT05948202; Stanford Medicine news 2014; Mossman-Steiner 2011 via Koegel table] [synthesis: feedback ordering]
Example language (illustrative): "You waited for him to look at the train before you said 'train' — that's exactly it. Next time, try giving the train the second he says 'tuh'."
Watch for: Parent stress; information overload. Working: parent fidelity at or above 80%.

H2. Data collection and fidelity scoring

Purpose / when: Weekly. [source: Pocket Guide ch. 8 title; NCT05948202; Gengoux et al. 2019]
Steps: Fidelity: score a 5–10 minute video in 30-second (or 1-minute) intervals, each component present/absent; compute percentage; 80% is the criterion. Child: count spontaneous vs. prompted vs. imitative utterances in a 10-minute sample; probe targets in new settings. [source: NCT05948202; Gengoux et al. 2019; Rezaei et al. 2018]

5. Handling problems

The Koegel sources address low motivation and disruptive behaviour directly; most other guidance below is [synthesis] built from the PRT components and standard practice.

Child will not engage or shows no interest in anything offered. Return to Phase 0: observe with free access to a wider range of items, including sensory and perseverative interests (these can serve as natural reinforcers; Vismara & Lyons 2007). Use E1 social-physical reinforcers. Shorten each opportunity and reinforce any orientation or vocal attempt. [source: Koegel table; OCALI checklist] [synthesis: sequencing]
Child stuck: attempts but no progress beyond approximations. Check that the attempt criterion is not too loose (reinforcing less than the child's best) or too strict; plan a shaping step; vary prompt types; add model prompts after 3–5 s time delays. [source: NCT05948202] [synthesis]
Prompt dependence (only speaks after a model). Increase time-delay opportunities, use choice and open-question prompts before models, fade models systematically, and track spontaneous utterances separately (Stanford gains were mainly prompted). [source: Gengoux et al. 2019; NCT05948202] [synthesis]
Echolalia. Use "say ..." sparingly and only as the stanford materials allow (stuck in echolalia, unsure, or modelling a replacement); model exactly what the child should say from their perspective. [source: NCT05948202] [synthesis: perspective rule]
Resistance, escape, tantrums during teaching ("resistance" in this modality is behavioural, not interpretive). Increase choice and maintenance tasks, check reinforcer value, shorten demands, teach "help"/"break"; do not give the item following the tantrum; FBA if persistent. [source: G1 sources] [synthesis]
Overwhelm or dysregulation (sensory overload, shutdown). Stop placing demands, reduce stimulation, allow the child's regulating activity, restart with maintenance tasks only. Treat distress as a sign to change the set-up, not to push through; respect clear refusals and signs of non-assent. [synthesis]
Parent non-adherence (no home practice). Review practice every session without blame; shrink assignments to one routine and two minutes; use the 10-minute home video; address parent stress directly (Pocket Guide ch. 6); offer strength-based feedback. [source: NCT05948202; Pocket Guide TOC; Mossman-Steiner 2011] [synthesis: shrinking tasks]
Parent fidelity below 80%. Teach one component at a time, model live, have the parent practise immediately with the child while you coach in the moment, re-score. [source: Gengoux et al. 2019 criterion] [synthesis: coaching sequence]
Rupture with parents (feeling judged, disagreement about goals or about ABA). Name the disagreement, invite the parent's goals, show how PRT is child-led and uses the child's interests; adjust targets to family values; repair by reviewing what is going well. [synthesis] Some autistic adults and advocacy groups criticise ABA-based approaches; discuss this openly with families and older clients and prioritise the child's assent and wellbeing. [synthesis]
Generalisation failure (skills only with clinician). Move teaching into home routines and community, train multiple adults and peers, use multiple exemplars (A5), self-management (D1) and priming (F2). [source: Koegel & Koegel slides; Koegel, Koegel, Hurley & Frea 1992] [synthesis: combination]
Classroom constraints (group size, time). Use CPRT: tokens for direct reinforcement, turn-taking between students, short embedded opportunities in existing activities. [source: Stahmer et al. 2016]
Older or very verbal clients for whom play-based PRT feels childish. Use interest-based conversation, self-management, video feedback (E6) and age-appropriate interest activities (E5). [source: Brookes 2nd ed. page on adolescents and adults] [synthesis]
Child with minimal speech after months of PRT. Some children respond better to more structured teaching (Jobin 2019 reported individual differences between PRT and discrete-trial training); consider adding structured teaching or augmentative and alternative communication (AAC), coordinating with a speech-language pathologist. [source: Wikipedia PRT on Jobin 2019] [synthesis: AAC suggestion; PRT sources do not address AAC]
Self-injury or serious aggression. Beyond PRT's scope alone: arrange a full functional assessment and positive behaviour support plan with a qualified behaviour analyst, and medical review where relevant. [synthesis]

Sources

Sources: