Authored exposition of the discrete trial training protocol

en · unofficial · sha256:05025abf9ba3c9742cf76ba3a0fd97f2930f214950ba16e95ac9bc5fa9bb41d2

This text is an authored exposition of a discrete trial training (DTT) protocol within Applied Behavior Analysis. It is not a legal instrument, a medical prescription, or a substitute for individualized clinical assessment. The protocol defines a reproducible form for trials, records, aggregation, prompt fading, mastery, maintenance, and procedural-integrity control; the selection of targets and reinforcers must account for the individual learner.

en · unofficial · sha256:c6c6f1af84fb2a39819a44fae8f4ff2e975b4d4430ea6c22bc05db0d32d4eabb

The Learner acquires a target skill, the Therapist conducts and records trials, and the Target identifies one observable and operationally defined behavior. Each Target belongs to one Learner. Before instruction begins, the Therapist records the discriminative stimulus or instruction, expected response, applicable prompts, reinforcement, error-correction procedure, and criteria. Smith (2001) provides a canonical description of discrete trial training and its boundaries.

Referencing norms: Learner, Target, Therapist

en · unofficial · sha256:12406affe32e8b3182f3c00e2338849c64ebf2836812a5b59516706896ca2b9c

A trial consists of presentation of a discriminative stimulus or instruction, a response-wait interval, a prompt when required, an observed response, a consequence, and an intertrial interval. The session date belongs to the legalTime axis. Response latency and the intertrial interval are Quantity values in seconds and are not deadlines or calendar windows; the protocol does not convert them into legalTime.

Referencing norms: intertrial_interval, response_latency

en · unofficial · sha256:0b02750c6bb9739a5c1a70333fb946351110e4fede3c92427132cdbd94a64cc8

This package establishes a protocol-specific linear response-prompt scale in fading order: full_physical, partial_physical, model, gestural, verbal, independent. This is a convention of this protocol, not a claim that all prompt modalities have a universal order of intrusiveness. When a step does not apply to a Target, the Therapist documents the skip and moves to the next applicable step. The baseline fading direction is most-to-least; least-to-most is a different strategy compared with most-to-least by Libby, Weiss, Bancroft, and Ahearn (2008) and does not replace the order adopted here.

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, PromptLevel, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

en · unofficial · sha256:2c59843b387e334b856bf8b8a33e0766d36115294b9f3455615519d68d5dd7b8

Every trial has exactly one outcome: correct_independent, correct_prompted, incorrect, or no_response. correct_independent means a correct response before a prompt is delivered; correct_prompted means a correct response after a prompt. incorrect means an observed incorrect response, and no_response means that the target response did not occur within the response-wait interval defined for the Target. Prompt level and outcome are recorded separately; a target-matching response after a prompt does not become correct_independent.

Referencing norms: ClassifyIncorrectOutcome, ClassifyNoResponseOutcome, CorrectIndependentAggregateOutcome, CorrectPromptedAggregateOutcome, CountIncorrectTrials, CountIndependentCorrectTrials, CountNoResponseTrials, CountPromptedCorrectAttempts, CreditIndependentOutcome, IncorrectAggregateOutcome, NoResponseAggregateOutcome, Outcome, PreservePromptedAttemptWithoutMasteryEffect

en · unofficial · sha256:28a2a8bb4d2755d1fee129e0ff26b5a9c73ce6d52694bd6ab4cdd0c82ac24f02

Every journal entry has its own identity and records the Target, session date, trial number within the session, PromptLevel, Outcome, and Therapist. Entries in one session for one Target share one date and one Therapist; trial numbers are positive and do not repeat. The journal preserves an attempt even when it produces no mastery or procedural effect. The Therapist is responsible for completeness and accuracy of the record.

Referencing norms: ObserveNonEmptySession, ObserveSessionTherapist, TrialRecord

en · unofficial · sha256:bad94de808b6f39815b44b37d32300d1bb967797625aa0967158bf41f7b66900

A correct independent response receives the reinforcement selected by the plan. A correct prompted response may receive differentiated reinforcement but does not count as an independent response. An incorrect response must not be reinforced. Following incorrect or no_response, the Therapist must conduct the error-correction procedure selected for the Target no later than the end of the same dated session; the duty window closes at the end of the session date. The journal explicitly records both compliance through nonreinforcement of the incorrect response and the opposite fact of reinforcement; silence substitutes for neither fact. Comparative evidence does not establish a single universally best correction procedure: the procedure is selected for the Learner and recorded before trials; Carroll, Joachim, St. Peter, and Robinson (2015) report an experimental comparison of correction procedures.

Referencing norms: CorrectIncorrectResponseDuty, CorrectIncorrectResponseWithinSession, CorrectNoResponseDuty, CorrectNoResponseWithinSession, DoNotReinforceIncorrectResponseDuty, DoNotReinforceIncorrectResponseWithinSession

en · unofficial · sha256:cb1297eed9e4e553fb25a6e09d7a2ccd1ea06fc17ce0faa8258f21c7b31028ae

For each Target and session date, the journal produces aggregates for total trials, correct_independent, correct_prompted, incorrect, no_response, the set of Therapists, and the greatest trial number. The projection of a trial number onto the Target-by-date pair is materialized with the journal and retains the source-record reference; it enables the maximum calculation and does not create an additional trial. No percentage is computed by division. Accuracy is tested by cross multiplication: correct_independent × 100 is at least threshold × total. correct_prompted contributes to total trials but not to the mastery numerator. If a declared session has no journal entries, the greatest-trial-number aggregate is undefined and evaluation must report EMPTY_AGGREGATE rather than zero or a fictional trial.

Referencing norms: CountAllSessionTrials, CountIncorrectTrials, CountIndependentCorrectTrials, CountNoResponseTrials, CountPromptedCorrectAttempts, FindLastTrialNumber, ObserveNonEmptySession, ObserveSessionTherapist, SessionFallsBelowAccuracyByCrossMultiplication, SessionMeetsAccuracyByCrossMultiplication

en · unofficial · sha256:fe10e555881544f05ae4864a77158ac189ccbbb4f3fb567befcd2247ea71d97a

The package baseline parameters are an independent-accuracy threshold of 100 percent, at least 3 consecutive observed sessions, and at least 2 distinct Therapists. These values are pinned in an A3 parameter dataset and selected by legalTime; changing any value creates a visible draft contour. The 100-percent and 3-session values are adopted as a strict baseline: Richling, Williams, and Carr (2019) found the most reliable maintenance at 100 percent across three sessions, and Longino, Richling, McDougale, and Palmier (2022) replicated high maintenance at 100 percent with a most-to-least procedure. The 2-Therapist requirement is a transparent authored minimum for testing generalization across people, not a universal constant established by the literature; published evidence treats across-Therapist assessment as relevant but does not prescribe a single mandatory count. A consecutive run is an interval of observed Target sessions in which every session meets the threshold and no observed session between the first and last observed sessions fails it. Cross-session counts of observed, threshold-meeting, below-threshold sessions, and distinct Therapists enter the criterion as materialized aggregates of the same journal: this is an executable-surface boundary because v1 cannot aggregate over an already derived predicate without the §111 completeness barrier. A materialized aggregate is not an independent clinical assertion and must be reproducible from the pinned trial records. Mastery and maintenance apply the same three-part test. Rollback occurs for a run of the same length and Therapist coverage in which every session falls below the threshold. Empirical findings show that maintenance depends on both criterion level and measurement arrangement and do not justify a hidden universal literal; see Fuller and Fienup (2018) and Richling, Williams, and Carr (2019).

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, ConsecutiveSessionCount/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Aconsecutive-session-count%23r-1/definition, CriterionPurpose, CriterionWindow, EstablishMaintenanceCriterion, EstablishMasteryCriterion, EstablishRollbackCriterion, MakeMaintenanceTransitionReady, MakeMasteryTransitionReady, MakeRollbackTransitionReady, MasteryThresholdPercent/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Amastery-threshold-percent%23r-1/definition, RequiredTherapistCount/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Arequired-therapist-count%23r-1/definition, SessionFallsBelowAccuracyByCrossMultiplication, SessionMeetsAccuracyByCrossMultiplication, StartMaintenanceFromDatedWindow, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

en · unofficial · sha256:1541a80ceffe4ba0683f132f7b2ce760d5615c22a5f691e09b1f642530b9fa5e

Every Target follows the Acquisition(t) procedure. Its stages are the product of six PromptLevels and the three statuses acquisition, mastered, and maintenance, for eighteen stages. The initial stage is full_physical × acquisition unless a documented skip sends the Target directly to the next applicable step; independent × maintenance remains a monitored stage rather than a closed terminal because a later error run must permit rollback. Within a level, acquisition → mastered is available only when the mastery criterion is established; mastered → maintenance follows the first dated maintenance session; maintenance → acquisition at the next less intrusive applicable level is available when the maintenance criterion is established. A below-threshold run returns the Target to acquisition at the immediately preceding more intrusive applicable level, or to full_physical × acquisition when already at full_physical. Every guard is calculated from journal entries, aggregates, and pinned A3 parameters; an attempted_transition whose guard is not met remains an attempt without procedural effect.

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, MakeMaintenanceTransitionReady, MakeMasteryTransitionReady, MakeRollbackTransitionReady, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

en · unofficial · sha256:e3a80e9d64732830e9689ecc9430dc0774dbefe3df0cbc93a601b1e2190034c3

For every incorrect or no_response trial, the Therapist has an achievement duty to perform error correction in a closed session-date window beginning and ending on that session date. For every incorrect trial, the Therapist has an achievement duty in the same closed window to ensure and explicitly record nonreinforcement of the incorrect response; this form makes established reinforcement visible as violated after the window closes instead of losing it in a non-executable negative formulation. Correction, nonreinforcement, the opposite fact of reinforcement, and monitoring completeness are supplied by the journal; missing data alone do not invent a violation. After the window closes, explicit nonperformance permits a violated finding. process-run must show these §140 positions together with §164 attempts without effect and the transitions that remain available.

Referencing norms: CorrectIncorrectResponseDuty, CorrectIncorrectResponseWithinSession, CorrectNoResponseDuty, CorrectNoResponseWithinSession, DoNotReinforceIncorrectResponseDuty, DoNotReinforceIncorrectResponseWithinSession

en · unofficial · sha256:fd1fb89661617cff0a630e76cf1ea005a2850c90a4c238f47ae0f2684461450b

The “sessions to criterion” calculation returns the number of observed Target sessions from the first session of the selected uninterrupted run through the session at which the criterion is first established; it counts sessions, not calendar days or trials. Bibliographic foundations of this authored exposition are: Smith, T. (2001), “Discrete Trial Training in the Treatment of Autism”, Focus on Autism and Other Developmental Disabilities 16(2), 86–92, doi:10.1177/108835760101600204; Libby, M. E., Weiss, J. S., Bancroft, S., & Ahearn, W. H. (2008), “A Comparison of Most-to-Least and Least-to-Most Prompting on the Acquisition of Solitary Play Skills”, Behavior Analysis in Practice 1(1), 37–43, doi:10.1007/BF03391719; Carroll, R. A., Joachim, B. T., St. Peter, C. C., & Robinson, N. (2015), “A Comparison of Error-Correction Procedures on Skill Acquisition During Discrete-Trial Instruction”, Journal of Applied Behavior Analysis 48(2), 257–273, doi:10.1002/jaba.205; Fuller, J. L., & Fienup, D. M. (2018), “A Preliminary Analysis of Mastery Criterion Level: Effects on Response Maintenance”, Behavior Analysis in Practice 11(1), 1–8, doi:10.1007/s40617-017-0201-0; Richling, S. M., Williams, W. L., & Carr, J. E. (2019), “The Effects of Different Mastery Criteria on the Skill Maintenance of Children with Developmental Disabilities”, Journal of Applied Behavior Analysis 52(3), 701–717, doi:10.1002/jaba.580; Longino, E., Richling, S. M., McDougale, C. B., & Palmier, J. M. (2022), “The Effects of Mastery Criteria on Maintenance: A Replication With Most-to-Least Prompting”, Behavior Analysis in Practice 15(2), 397–405, doi:10.1007/s40617-021-00562-y. The publications themselves are not reproduced in this package; the citations are bibliographic references.

Referencing norms: CalculateSessionsToMasteryCriterion

ru · unofficial · sha256:bc381c8497a9daded8a0ec65e5fcdc37c5f25eaa6f00c97535de90b7a551877cRU

Настоящий текст является авторским изложением учебного протокола дискретных проб (Discrete Trial Training, DTT) в методологии Applied Behavior Analysis. Он не является нормативным правовым актом, медицинским назначением либо заменой индивидуальной клинической оценки. Протокол задаёт воспроизводимую форму проб, журналирования, агрегирования, снятия подсказок, освоения, поддержания и контроля процедурной точности; выбор целей и подкреплений должен учитывать конкретного обучающегося.

ru · unofficial · sha256:47788a90ed167bd13f7881a3e1ed6a0aeb2719708f59b67e3628dcb5fe9d5e50RU

Обучающийся (Learner) приобретает целевой навык, терапевт (Therapist) проводит и регистрирует пробы, а цель (Target) обозначает одно наблюдаемое и операционально определённое поведение. Каждая цель относится к одному обучающемуся. До начала обучения терапевт фиксирует различимый стимул или инструкцию, ожидаемый ответ, допустимые подсказки, подкрепление, способ коррекции ошибки и применимые критерии. Каноническое описание дискретного обучения и его границ приведено у Smith (2001).

Referencing norms: Learner, Target, Therapist

ru · unofficial · sha256:5ace1b2b01324529f1980801ccb7cb92a6e4b1d1441ecda374fd4193c733a5e7RU

Одна проб состоит из предъявления различимого стимула или инструкции, интервала ожидания ответа, при необходимости подсказки, наблюдаемого ответа, последствия и межпробного интервала. Дата сессии относится к оси legalTime. Латентность ответа и межпробный интервал являются величинами Quantity в секундах и не являются сроками или календарными окнами; протокол не преобразует их в legalTime.

Referencing norms: intertrial_interval, response_latency

ru · unofficial · sha256:3b94f9104b61da8dc934f0b2fd8854f3d0cd2a46a4ed65b314e94d2c41f91a72RU

Для этого пакета устанавливается протокол-специфическая линейная шкала response prompts в порядке снятия: full_physical, partial_physical, model, gestural, verbal, independent. Это соглашение данного протокола, а не утверждение об универсальной сравнительной интрузивности всех видов подсказок. Для цели, к которой отдельная ступень неприменима, терапевт документирует её пропуск и переходит к следующей применимой ступени. Базовое направление снятия является most-to-least; least-to-most представляет иную стратегию, сравниваемую с most-to-least у Libby, Weiss, Bancroft и Ahearn (2008), и не подменяет установленный здесь порядок.

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, PromptLevel, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

ru · unofficial · sha256:0673da047c935b60ea2549922434649b8003b1eb6bbcccafe2b86125e0ca6dedRU

Исход каждой пробы имеет ровно одно значение: correct_independent, correct_prompted, incorrect или no_response. correct_independent означает правильный ответ до предоставления подсказки; correct_prompted означает правильный ответ после подсказки. incorrect означает наблюдаемый неверный ответ, no_response — отсутствие целевого ответа в установленном для цели интервале ожидания. Уровень подсказки и исход регистрируются раздельно; совпадение ответа после подсказки с целевым ответом не превращает исход в correct_independent.

Referencing norms: ClassifyIncorrectOutcome, ClassifyNoResponseOutcome, CorrectIndependentAggregateOutcome, CorrectPromptedAggregateOutcome, CountIncorrectTrials, CountIndependentCorrectTrials, CountNoResponseTrials, CountPromptedCorrectAttempts, CreditIndependentOutcome, IncorrectAggregateOutcome, NoResponseAggregateOutcome, Outcome, PreservePromptedAttemptWithoutMasteryEffect

ru · unofficial · sha256:50ec385facc1b6fc111095a8fed60bbee5aad03bb912c65b609819f501508393RU

Каждая запись журнала имеет самостоятельную идентичность и содержит цель, дату сессии, номер пробы в сессии, уровень подсказки, исход и терапевта. Записи одной сессии для одной цели имеют одну дату и одного терапевта; номера проб являются положительными и не повторяются. Журнал сохраняет попытку даже тогда, когда она не породила эффекта освоения или процедурного перехода. Терапевт отвечает за полноту и точность записи.

Referencing norms: ObserveNonEmptySession, ObserveSessionTherapist, TrialRecord

ru · unofficial · sha256:e4506096d5d3a975c454f0274de9aa1febace93d2bf493b830c0bca5049c5801RU

Правильный независимый ответ получает предусмотренное планом подкрепление. Правильный ответ с подсказкой может получить дифференцированное подкрепление, но не засчитывается как независимый ответ. Неверный ответ не подкрепляется. После incorrect или no_response терапевт обязан провести выбранную для цели процедуру коррекции ошибки не позднее окончания той же датированной сессии; окно обязанности закрыто концом даты сессии. Журнал явно фиксирует как соблюдение отсутствия подкрепления неверного ответа, так и противоположный факт подкрепления; молчание не заменяет ни один из этих фактов. Сравнительные данные не устанавливают единственную универсально лучшую коррекцию: конкретная процедура выбирается для обучающегося и фиксируется до проб; обзор экспериментального сравнения процедур дан у Carroll, Joachim, St. Peter и Robinson (2015).

Referencing norms: CorrectIncorrectResponseDuty, CorrectIncorrectResponseWithinSession, CorrectNoResponseDuty, CorrectNoResponseWithinSession, DoNotReinforceIncorrectResponseDuty, DoNotReinforceIncorrectResponseWithinSession

ru · unofficial · sha256:d8c37a34d9500adf42207dad4c51a2a3280468e1eaac28f4b01fb3e65c50fbfdRU

Для каждой цели и даты сессии журнал образует агрегаты: общее число проб, число correct_independent, correct_prompted, incorrect и no_response, множество терапевтов и наибольший номер пробы. Проекция номера пробы на пару «цель × дата» материализуется вместе с журналом и сохраняет ссылку на исходную запись; она нужна для вычисления максимума и не создаёт дополнительной пробы. Процент делением не вычисляется. Условие точности проверяется кросс-умножением: correct_independent × 100 не меньше threshold × total. correct_prompted входит в общее число проб, но не входит в числитель освоения. Если объявленная сессия не содержит ни одной записи, агрегат наибольшего номера не определён и вычисление должно завершиться громкой проблемой EMPTY_AGGREGATE, а не нулём или вымышленной пробой.

Referencing norms: CountAllSessionTrials, CountIncorrectTrials, CountIndependentCorrectTrials, CountNoResponseTrials, CountPromptedCorrectAttempts, FindLastTrialNumber, ObserveNonEmptySession, ObserveSessionTherapist, SessionFallsBelowAccuracyByCrossMultiplication, SessionMeetsAccuracyByCrossMultiplication

ru · unofficial · sha256:7d35ae81a4e4c6513ab5bc4b13cc7e144cd91f97e4b5366eab9c948c7b76b59fRU

Исходные параметры пакета составляют: порог независимой точности 100 процентов, не менее 3 последовательных наблюдаемых сессий и не менее 2 различных терапевтов. Эти значения закрепляются в A3-наборе параметров и выбираются по legalTime; изменение любого значения создаёт новый видимый draft-контур. Значения 100 процентов и 3 сессии выбраны как строгая исходная конфигурация: Richling, Williams и Carr (2019) получили наиболее устойчивое поддержание при 100 процентах в трёх сессиях, а Longino, Richling, McDougale и Palmier (2022) реплицировали высокое поддержание для 100 процентов при most-to-least-процедуре. Требование 2 терапевтов является прозрачным авторским минимумом проверки генерализации между людьми, а не установленной литературой универсальной константой; опубликованные данные признают проверку между терапевтами значимой, но не фиксируют единое обязательное число. Последовательная серия означает интервал наблюдаемых сессий цели, в котором каждая сессия удовлетворяет порогу и между начальной и конечной наблюдаемыми сессиями нет наблюдаемой сессии, не удовлетворяющей порогу. Межсессионные числа наблюдаемых, пороговых и подпороговых сессий, а также различных терапевтов поступают в критерий как материализованные агрегаты того же журнала: это граница исполнимой поверхности, поскольку v1 не агрегирует по уже выводимому предикату без барьера полноты §111. Материализованный агрегат не является самостоятельным клиническим утверждением и должен воспроизводиться из закреплённых записей проб. Освоение и поддержание используют эту же трёхчастную проверку. Откат возникает при серии той же длины и того же охвата терапевтов, в которой каждая сессия находится ниже порога. Эмпирические работы показывают зависимость поддержания от уровня критерия и организации измерения и не оправдывают скрытый универсальный литерал; см. Fuller и Fienup (2018), Richling, Williams и Carr (2019).

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, ConsecutiveSessionCount/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Aconsecutive-session-count%23r-1/definition, CriterionPurpose, CriterionWindow, EstablishMaintenanceCriterion, EstablishMasteryCriterion, EstablishRollbackCriterion, MakeMaintenanceTransitionReady, MakeMasteryTransitionReady, MakeRollbackTransitionReady, MasteryThresholdPercent/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Amastery-threshold-percent%23r-1/definition, RequiredTherapistCount/value/urn%3Aaba%3Aclir%3Adtt-protocol%3Aparam%3Arequired-therapist-count%23r-1/definition, SessionFallsBelowAccuracyByCrossMultiplication, SessionMeetsAccuracyByCrossMultiplication, StartMaintenanceFromDatedWindow, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

ru · unofficial · sha256:5c37eaa023f237e4f033308b350c812acad7530ee66ff70929010829983e82a8RU

Каждая цель проходит процедуру Acquisition(t). Стадии образуют произведение шести уровней подсказки и трёх статусов acquisition, mastered и maintenance, всего восемнадцать стадий. Начальной стадией является full_physical × acquisition, если документированный пропуск не направляет цель сразу на следующую применимую ступень; independent × maintenance остаётся наблюдаемой стадией, а не закрытым terminal, поскольку последующая серия ошибок должна допускать откат. Внутри уровня переход acquisition → mastered допускается только при установленном критерии освоения; mastered → maintenance — после появления первой датированной maintenance-сессии; maintenance → acquisition следующего менее интрузивного применимого уровня — при установленном критерии поддержания. Серия ниже порога переводит цель в acquisition непосредственно предшествующего более интрузивного применимого уровня, а на full_physical возвращает в full_physical × acquisition. Все гарды вычисляются из записей журнала, агрегатов и закреплённых A3-параметров; attempted_transition без выполненного гарда сохраняется как попытка без процедурного эффекта.

Referencing norms: Acquisition, Acquisition/FadeFullToPartial, Acquisition/FadeGesturalToVerbal, Acquisition/FadeModelToGestural, Acquisition/FadePartialToModel, Acquisition/FadeVerbalToIndependent, Acquisition/FullPhysicalAcquisition, Acquisition/FullPhysicalMaintenance, Acquisition/FullPhysicalMastered, Acquisition/GesturalAcquisition, Acquisition/GesturalMaintenance, Acquisition/GesturalMastered, Acquisition/IndependentAcquisition, Acquisition/IndependentMaintenance, Acquisition/IndependentMastered, Acquisition/MaintainFullPhysical, Acquisition/MaintainGestural, Acquisition/MaintainIndependent, Acquisition/MaintainModel, Acquisition/MaintainPartialPhysical, Acquisition/MaintainVerbal, Acquisition/MasterFullPhysical, Acquisition/MasterGestural, Acquisition/MasterIndependent, Acquisition/MasterModel, Acquisition/MasterPartialPhysical, Acquisition/MasterVerbal, Acquisition/ModelAcquisition, Acquisition/ModelMaintenance, Acquisition/ModelMastered, Acquisition/PartialPhysicalAcquisition, Acquisition/PartialPhysicalMaintenance, Acquisition/PartialPhysicalMastered, Acquisition/RollbackFullMaintenance, Acquisition/RollbackFullMastered, Acquisition/RollbackGesturalAcquisition, Acquisition/RollbackGesturalMaintenance, Acquisition/RollbackGesturalMastered, Acquisition/RollbackIndependentAcquisition, Acquisition/RollbackIndependentMaintenance, Acquisition/RollbackIndependentMastered, Acquisition/RollbackModelAcquisition, Acquisition/RollbackModelMaintenance, Acquisition/RollbackModelMastered, Acquisition/RollbackPartialAcquisition, Acquisition/RollbackPartialMaintenance, Acquisition/RollbackPartialMastered, Acquisition/RollbackVerbalAcquisition, Acquisition/RollbackVerbalMaintenance, Acquisition/RollbackVerbalMastered, Acquisition/SkipFullPhysical, Acquisition/SkipGestural, Acquisition/SkipModel, Acquisition/SkipPartialPhysical, Acquisition/SkipVerbal, Acquisition/VerbalAcquisition, Acquisition/VerbalMaintenance, Acquisition/VerbalMastered, AcquisitionState, AcquisitionStep, AcquisitionTransition, MakeMaintenanceTransitionReady, MakeMasteryTransitionReady, MakeRollbackTransitionReady, attempted_transition, current_state, entered_state, initial_state, left_state, procedure_instance, state_before, valid_transition

ru · unofficial · sha256:fc85cde7d9af667a0f0e09b8dfd712551fabe7b6a8d3dfec2f91b3f7e30538a5RU

Для каждой incorrect или no_response пробы возникает achievement-обязанность терапевта провести коррекцию ошибки в закрытом окне даты сессии, начиная с даты записи и заканчивая этой же датой. Для каждой incorrect пробы возникает achievement-обязанность в том же закрытом окне обеспечить и явно зарегистрировать отсутствие подкрепления неверного ответа; такая форма нужна, чтобы после закрытия окна установленное подкрепление было видимо как violated, а не терялось в неисполняемой отрицательной формулировке. Факт коррекции, факт отсутствия подкрепления, противоположный факт подкрепления и полнота наблюдения сообщаются журналом; отсутствие записи само по себе не выдумывает нарушение. После закрытия окна явная неисполненность позволяет вывести violated. Process-run обязан показывать такие позиции по §140 вместе с попытками без эффекта §164 и доступными последующими переходами.

Referencing norms: CorrectIncorrectResponseDuty, CorrectIncorrectResponseWithinSession, CorrectNoResponseDuty, CorrectNoResponseWithinSession, DoNotReinforceIncorrectResponseDuty, DoNotReinforceIncorrectResponseWithinSession

ru · unofficial · sha256:b6df167956abed50a0d5db026af1f8f97388b4d83d555f40e31003aee52b1900RU

Расчёт «сессий до критерия» возвращает число наблюдаемых сессий цели от первой сессии выбранной непрерывной серии до сессии, на которой критерий впервые установлен; он считает сессии, а не календарные дни и не число проб. Библиографические основания настоящего авторского изложения: Smith, T. (2001), “Discrete Trial Training in the Treatment of Autism”, Focus on Autism and Other Developmental Disabilities 16(2), 86–92, doi:10.1177/108835760101600204; Libby, M. E., Weiss, J. S., Bancroft, S., & Ahearn, W. H. (2008), “A Comparison of Most-to-Least and Least-to-Most Prompting on the Acquisition of Solitary Play Skills”, Behavior Analysis in Practice 1(1), 37–43, doi:10.1007/BF03391719; Carroll, R. A., Joachim, B. T., St. Peter, C. C., & Robinson, N. (2015), “A Comparison of Error-Correction Procedures on Skill Acquisition During Discrete-Trial Instruction”, Journal of Applied Behavior Analysis 48(2), 257–273, doi:10.1002/jaba.205; Fuller, J. L., & Fienup, D. M. (2018), “A Preliminary Analysis of Mastery Criterion Level: Effects on Response Maintenance”, Behavior Analysis in Practice 11(1), 1–8, doi:10.1007/s40617-017-0201-0; Richling, S. M., Williams, W. L., & Carr, J. E. (2019), “The Effects of Different Mastery Criteria on the Skill Maintenance of Children with Developmental Disabilities”, Journal of Applied Behavior Analysis 52(3), 701–717, doi:10.1002/jaba.580; Longino, E., Richling, S. M., McDougale, C. B., & Palmier, J. M. (2022), “The Effects of Mastery Criteria on Maintenance: A Replication With Most-to-Least Prompting”, Behavior Analysis in Practice 15(2), 397–405, doi:10.1007/s40617-021-00562-y. Тексты публикаций в пакет не включаются; ссылки являются библиографическими.

Referencing norms: CalculateSessionsToMasteryCriterion