Source / pinned edition

Authored exposition of functional communication training and schedule thinning

Source of package aba.fct_thinning · FCT-THINNING-EXPOSITION-1

work
urn:aba:clir:fct-thinning#FCT_EXPOSITION
number
FCT-THINNING-EXPOSITION-1
kind
urn:aba:clir:fct-thinning#doctrine

FCT_EN

edition
urn:aba:clir:fct-thinning#FCT_EN
language
en
officiality
unofficial
materialisation
PINNED_OFFICIAL_BYTES
life cycle
in_force: [object Object]

Publications

URITypedocumentHashRetrieved
urn:aba:fct-thinning:entext/plain; charset=utf-8sha256:aa0c3b34e091d4e493fa1bf630522b85ccf3e1325d1292c1a69e3c8fc433e0712026-09-02T00:00:00Z

Fragments — 13

section/1 · article

en · unofficial · sha256:bf827682aa88a4462b3ad5cd4c01d2901edeb891d3a1946c7fb29b155d479286

This text is an authored exposition of functional communication training (FCT) and of the subsequent thinning of the reinforcement schedule by multiple schedules, within the methodology of Applied Behavior Analysis. It is not a normative legal act, a medical prescription, or a substitute for an individual functional assessment and clinical judgement. The exposition fixes a reproducible form of signalled periods, journaling, procedural-integrity monitoring, the discrimination criterion, and thinning; the choice of communicative response, of reinforcer, and of the initial schedule density must rest on a functional assessment of the individual learner.
section/2 · article · referenced by: 7

en · unofficial · sha256:8a7dd39367bbc22a35c762c5c3716530f05db22e97ca8bd79fec57abcc3f3b04

The learner is taught a functional communication response, the implementer reinforces it and keeps the journal, and the FCT plan belongs to exactly one learner and names one functional communication response together with one problem behavior of the same function, for which that response is the replacement. The functional communication response (FCR) is an observable form of requesting the very reinforcer that maintained the problem behavior.

Referencing norms: FctPlan, Implementer, Learner, MakeRollbackTransitionReadyFromResurgence, ResurgenceByCrossMultiplication, fct_plan, plan_learner

section/3 · article · referenced by: 11

en · unofficial · sha256:d1626a0d3bba70c71b60f00310a693b243886731964d5643d35a2cff887c1300

The session date belongs to the legalTime axis and is the only calendar quantity of this exposition. Time inside a session is measured as a whole number of seconds from the start of the session (the within-session offset) and is not a calendar quantity: offsets are not converted to dates, take no part in deadline computation, and read no calendar. Implementer duties are attached to the closed window of the session date, beginning at the session date and ending at the declared upper bound of that same date.

Referencing norms: DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, ReinforceResponseInReinforcedSignal, ReinforceResponseInReinforcedSignalDuty, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_offset_seconds, period_end_seconds, period_start_seconds, session_date, session_window_end

section/4 · article · referenced by: 12

en · unofficial · sha256:c81683fc2eb0701af5af009fa5a3c12df9e4a5f5abed2dff12eae6b9b175f947

A session is divided into non-overlapping signalled periods, each carrying an ordinal number, a starting and an ending within-session offset, and exactly one signal state: s_plus or s_minus. In an s_plus period the functional communication response is reinforced; in an s_minus period it is not. The signal state is journal data, not an inference: the implementer presents a discriminable signal and records which one. Periods alternate, but the alternation is not inferred from the number — it is recorded. The exposition contains no random quantities: a variable-time schedule is expressed as a pinned list of durations, not as a generator.

Referencing norms: PeriodsOfOneSessionMustNotOverlap, PresentDiscriminableSignalDuty, PresentDiscriminableSignalOfEachPeriod, SignalState, SignalledPeriod, period_end_seconds, period_number, period_record, period_session, period_signal, period_start_seconds, session_periods_overlap

section/5 · article · referenced by: 47

en · unofficial · sha256:18c73d51b6694620ac320601a35b0903501988d1544b00817f4757a56acffe01

Thinning is given by four steps Step1, Step2, Step3, and Step4; each step carries two durations in seconds — of the s_plus period and of the s_minus period — pinned by a set of A3 parameters and selected by legalTime. From step to step the s_plus period shortens and the s_minus period grows: the density of reinforcement falls. A session declares the step at which it was conducted. A period conforms to the step if its duration, that is the difference of its ending and starting offsets, is exactly equal to the duration of its own signal state at that step. A period that does not conform to the step is called non-conforming and is a departure from the procedure.

Referencing norms: FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, PeriodConformsToItsStep, PeriodLongerThanItsStep, PeriodShorterThanItsStep, SMinusDurationInForceAtStep1, SMinusDurationInForceAtStep2, SMinusDurationInForceAtStep3, SMinusDurationInForceAtStep4, SPlusDurationInForceAtStep1, SPlusDurationInForceAtStep2, SPlusDurationInForceAtStep3, SPlusDurationInForceAtStep4, Step1SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep1-s-minus-seconds%23r-1/definition, Step1SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep1-s-plus-seconds%23r-1/definition, Step2SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep2-s-minus-seconds%23r-1/definition, Step2SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep2-s-plus-seconds%23r-1/definition, Step3SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep3-s-minus-seconds%23r-1/definition, Step3SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep3-s-plus-seconds%23r-1/definition, Step4SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep4-s-minus-seconds%23r-1/definition, Step4SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep4-s-plus-seconds%23r-1/definition, ThinningStep, period_conforms_to_step, period_does_not_conform_to_step, session_at_step

section/6 · article · referenced by: 21

en · unofficial · sha256:d6f27352d33bfe5354e37bcf3684ea0886d4235dcb882a9e8924b0692e9866e7

A functional communication response record has an identity of its own and carries the session, the period, the within-session offset, and the implementer. A session carries the date, the plan, the step, the implementer, the upper bound of the duty window, the observation length in minutes, the lengths of the signalled states in minutes, and the number of episodes of the problem behavior. Journal completeness is asserted explicitly: a session is declared either complete or incomplete, and journal silence replaces neither assertion. An incomplete session is unfit for criterion checking and enters no window.

Referencing norms: DeclareIncompleteSessionUnreliable, DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, FcrRecord, ObserveCompleteSession, ObserveSessionResponseImplementer, Session, fcr_implementer, fcr_period, fcr_record, fcr_session, session_duration_minutes, session_implementer, session_journal_complete, session_journal_incomplete, session_measurement_unreliable, session_observed, session_plan, session_problem_behavior_count, session_record, session_response_by

section/7 · article · referenced by: 10

en · unofficial · sha256:5a6f65afbb90cfae49588dcb9fb64e534d505cb3c1a067c4d018a8dbfbfb8fc6

A response is credited to a period if its within-session offset lies between that period's starting and ending offsets inclusive. A response emitted in an s_plus period creates a duty of reinforcement. A response emitted in an s_minus period is preserved by the journal as an attempt without legal effect: it creates no duty of reinforcement and cancels none anywhere. Such a record is neither deleted nor rewritten: it is precisely what makes the course of discrimination learning visible.

Referencing norms: ResponseFallsWithinItsPeriod, ResponseInExtinctionSignal, ResponseInExtinctionSignalHasNoLegalEffect, ResponseInReinforcedSignal, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_in_extinction_signal, fcr_in_reinforced_signal, fcr_within_period, fcr_without_effect

section/8 · article · referenced by: 11

en · unofficial · sha256:74a447e7e821a08fd675a4837e7b9535c3d1964e94e9278f54b30b632c37ee54

For each response in an s_plus period an achievement duty of the implementer toward the learner arises to reinforce that response within the closed window of the session date. For each response in an s_minus period an achievement duty arises not to reinforce it within the same window. For each period an achievement duty arises to present the discriminable signal of that period. For each session an achievement duty arises to declare the journal complete. Performance and its absence are equally reported by the journal as explicit facts; after the window closes an explicit non-performance is visible as a violation. The procedural-integrity audit consists of enumerating these positions.

Referencing norms: DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, PresentDiscriminableSignalDuty, PresentDiscriminableSignalOfEachPeriod, ReinforceResponseInReinforcedSignal, ReinforceResponseInReinforcedSignalDuty, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_not_reinforced, fcr_reinforced, period_signal_presented

section/9 · article · referenced by: 15

en · unofficial · sha256:da5ed4d9e3e73557c1f575347faf5e03eb4426dc7b58dd4799e85d9e38a3142a

The number of response records of a session is counted directly over the journal records. The numbers of responses by signal state, and the numbers of observed and discriminating sessions of a window, by contrast, arrive as materialized data of the same journal: v1 does not aggregate over an already derived predicate without the completeness barrier of §111. A materialized aggregate is not an independent clinical assertion and must be reproducible from the pinned journal records. The response level is not computed by division. A session discriminates if the response level in s_plus is not below the lower threshold and the response level in s_minus is not above the upper threshold; both conditions are checked by cross multiplication: the number of responses multiplied by one hundred is compared with the threshold per hundred minutes multiplied by the length of the corresponding state in minutes. A session fails to discriminate on one of two independent grounds: the level in s_plus is below the lower threshold, or the level in s_minus is above the upper one. The grounds are distinguishable and are written separately. A state of zero length checks no threshold.

Referencing norms: CountSessionResponses, ExtinctionSignalRateMetByCrossMultiplication, FcrSMinusCeilingPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Afcr-s-minus-ceiling-per-hundred-minutes%23r-1/definition, FcrSPlusFloorPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Afcr-s-plus-floor-per-hundred-minutes%23r-1/definition, ReinforcedSignalRateMetByCrossMultiplication, SessionDiscriminatesWhenBothThresholdsAreMet, SessionFailsOnHighExtinctionSignalRate, SessionFailsOnLowReinforcedSignalRate, session_discriminates, session_does_not_discriminate, session_extinction_signal_rate_met, session_fcr_count, session_reinforced_signal_rate_met, session_state_fcr_count, session_state_minutes

section/10 · article · referenced by: 46

en · unofficial · sha256:0d547e2609ce38fe067b304bc07300b9227730e85b1dc4fd83f534db9be5811b

A review window belongs to a plan, a step, and a purpose — advancement or rollback. Advancement is established if the window holds no fewer than the parameterized number of consecutive observed sessions and each of them discriminates. Rollback is established if the window holds no fewer than the same number of consecutive observed sessions and none of them discriminates. Independently of any window, rollback is opened by resurgence: a session in which the level of the problem behavior exceeded the parameterized ceiling returns the plan to the previous step immediately. Resurgence is the return of previously reinforced behavior when reinforcement of the replacement response is weakened; it is measured by the same cross multiplication and the same journal.

Referencing norms: ConsecutiveDiscriminatingSessionCount/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aconsecutive-discriminating-session-count%23r-1/definition, EstablishAdvanceCriterion, EstablishRollbackCriterion, FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, MakeAdvanceTransitionReady, MakeRollbackTransitionReadyFromResurgence, MakeRollbackTransitionReadyFromWindow, ResurgenceByCrossMultiplication, ResurgenceCeilingPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aresurgence-ceiling-per-hundred-minutes%23r-1/definition, ReviewPurpose, ReviewWindow, advance_criterion_met, advance_ready, discriminating_session_count, non_discriminating_session_count, observed_session_count, review_window_plan, review_window_purpose, review_window_record, review_window_step, rollback_criterion_met, rollback_ready, session_resurgence

section/11 · article · referenced by: 29

en · unofficial · sha256:e3935d6557a049727b37a81dbb7b2aeeaa5964f8c5c50921c01ae0e063eef5bd

A plan passes through the stages Step1Active, Step2Active, Step3Active, Step4Active, and Thinned. Step1Active is initial and is the densest pinned schedule; Thinned is terminal and means that the last pinned step has been reached. The forward transition leads from the stage of a step to the stage of the next step, and from Step4Active into Thinned. The backward transition leads from the stage of a step to the stage of the previous step. There is no rollback out of Step1Active: the exposition defines no denser schedule, and a demand for rollback at that stage remains without a transition. There is no exit from Thinned: resuming thinning after it is complete is a new plan. Both boundaries are declared deliberately.

Referencing norms: FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, MakeAdvanceTransitionReady, MakeRollbackTransitionReadyFromResurgence, MakeRollbackTransitionReadyFromWindow, advance_ready, rollback_ready

section/12 · article · referenced by: 52

en · unofficial · sha256:c57663d6b9cdbee8cb434f58e530b4c6738b817cd2ea5e04a7e215e360caedf0

Teaching the learner to tolerate a delay in reinforcement and a denial of reinforcement is given by this exposition as a SEPARATE procedure DelayTolerance of the same plan, rather than as a second axis of the thinning procedure. The delay levels Level1, Level2, and Level3 carry pinned durations in seconds and end at the terminal stage Tolerant. Advancement and rollback along the delay levels are established by their own review window and their own criterion: a session tolerates the delay if the learner's response to a delay of the corresponding length was not accompanied by the problem behavior, and this is reported by the journal as an explicit fact. Keeping the axes apart is a substantive claim, not a convenience of notation: it means that a failure to tolerate a delay does NOT cancel the thinning step already reached, and conversely. To join the axes into one procedure would be to assert the opposite; the exposition makes no such assertion and leaves the choice to whoever runs the program.

Referencing norms: ConsecutiveToleratedSessionCount/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aconsecutive-tolerated-session-count%23r-1/definition, DelayInForceAtLevel1, DelayInForceAtLevel2, DelayInForceAtLevel3, DelayLevel, DelayLevel1Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level1-seconds%23r-1/definition, DelayLevel2Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level2-seconds%23r-1/definition, DelayLevel3Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level3-seconds%23r-1/definition, DelayTolerance, DelayTolerance/AdvanceToLevel2, DelayTolerance/AdvanceToLevel3, DelayTolerance/CompleteDelayTolerance, DelayTolerance/Level1Active, DelayTolerance/Level2Active, DelayTolerance/Level3Active, DelayTolerance/RollbackToLevel1, DelayTolerance/RollbackToLevel2, DelayTolerance/Tolerant, DelayToleranceState, DelayToleranceStep, DelayToleranceTransition, DelayTolerance__attempted_transition, DelayTolerance__current_state, DelayTolerance__entered_state, DelayTolerance__initial_state, DelayTolerance__left_state, DelayTolerance__procedure_instance, DelayTolerance__state_before, DelayTolerance__valid_transition, EstablishDelayAdvanceCriterion, EstablishDelayRollbackCriterion, MakeDelayAdvanceTransitionReady, MakeDelayRollbackTransitionReady, RecordDelayFailureAtSessionLevel, RecordToleratedDelayAtSessionLevel, delay_advance_criterion_met, delay_advance_ready, delay_failed_session_count, delay_failure_recorded, delay_observed_session_count, delay_rollback_criterion_met, delay_rollback_ready, delay_seconds_in_force, delay_tolerance_recorded, delay_tolerated_session_count, delay_window_level, delay_window_plan, delay_window_purpose, delay_window_record, session_at_delay_level, session_delay_not_tolerated, session_delay_tolerated

section/13 · article · referenced by: 3

en · unofficial · sha256:3d65288bda88f9a15fb09a1b3991fc234ef22b0419fcd9873e53062337c11975

The "period duration in force" calculation returns the length of a signalled state at the declared step of a session, in seconds rather than in dates. The boundaries of this version are named explicitly: the exposition models neither competing stimuli that raise tolerance to thinning, nor delayed reinforcement as an independent axis, nor extinction of the problem behavior, nor the functional analysis from which the plan originates; all of these remain outside the model rather than being declared nonexistent. The bibliographic grounds of this authored exposition are: Hanley, G. P., Iwata, B. A., & Thompson, R. H. (2001), “Reinforcement Schedule Thinning Following Treatment with Functional Communication Training”, Journal of Applied Behavior Analysis 34(1), doi:10.1901/jaba.2001.34-17; Fisher, W. W., Kuhn, D. E., & Thompson, R. H. (1998), “Establishing Discriminative Control of Responding Using Functional and Alternative Reinforcers During Functional Communication Training”, Journal of Applied Behavior Analysis 31(4), doi:10.1901/jaba.1998.31-543; Hagopian, L. P., Kuhn, S. A. C., Long, E. S., & Rush, K. S. (2005), “Schedule Thinning Following Communication Training: Using Competing Stimuli to Enhance Tolerance to Decrements in Reinforcer Density”, Journal of Applied Behavior Analysis 38(2), doi:10.1901/jaba.2005.43-04; Tiger, J. H., Hanley, G. P., & Bruzek, J. (2008), “Functional Communication Training: A Review and Practical Guide”, Behavior Analysis in Practice 1(1), doi:10.1007/BF03391716; Hagopian, L. P., Boelter, E. W., & Jarmolowicz, D. P. (2011), “Reinforcement Schedule Thinning Following Functional Communication Training: Review and Recommendations”, Behavior Analysis in Practice 4(1), doi:10.1007/BF03391770; Fuhrman, A. M., Fisher, W. W., & Greer, B. D. (2016), “A Preliminary Investigation on Improving Functional Communication Training by Mitigating Resurgence of Destructive Behavior”, Journal of Applied Behavior Analysis 49(4), doi:10.1002/jaba.338; Briggs, A. M., Fisher, W. W., Greer, B. D., & Kimball, R. T. (2018), “Prevalence of Resurgence of Destructive Behavior When Thinning Reinforcement Schedules During Functional Communication Training”, Journal of Applied Behavior Analysis 51(3), doi:10.1002/jaba.472.

Referencing norms: CalculateSessionsToAdvanceCriterion, sessions_to_criterion, state_duration_in_force

FCT_RU

edition
urn:aba:clir:fct-thinning#FCT_RU
language
ru
officiality
unofficial
materialisation
PINNED_OFFICIAL_BYTES
life cycle
in_force: [object Object]

Publications

URITypedocumentHashRetrieved
urn:aba:fct-thinning:rutext/plain; charset=utf-8sha256:2cfb721f1e2de6a58c01039c3ea8fce005851d256810a35a68cffb74438d02d62026-09-02T00:00:00Z

Fragments — 13

section/1 · article

ru · unofficial · sha256:247613412132bdb187253a018ee11b71810ce8e665e2dc5f1411a86b5cc56b4fRU

Настоящий текст является авторским изложением обучения функциональной коммуникации (functional communication training, FCT) и последующего разрежения расписания подкрепления по множественным расписаниям в методологии Applied Behavior Analysis. Он не является нормативным правовым актом, медицинским назначением либо заменой индивидуальной функциональной оценки и клинического решения. Изложение задаёт воспроизводимую форму сигнализированных периодов, журналирования, контроля процедурной точности, критерия дискриминации и разрежения; выбор коммуникативного ответа, подкрепления и исходной плотности расписания должен опираться на функциональную оценку конкретного обучающегося.
section/2 · article · referenced by: 7

ru · unofficial · sha256:2408d0371590ac22143ce4e0ff75aac2c220b85352fd65486b51e9d1f577113fRU

Обучающийся (Learner) обучается функциональному коммуникативному ответу, исполнитель (Implementer) подкрепляет его и ведёт журнал, план FCT (FctPlan) относится ровно к одному обучающемуся, обозначает один функциональный коммуникативный ответ и одно проблемное поведение той же функции, замещением которого этот ответ является. Функциональный коммуникативный ответ (FCR) есть наблюдаемая форма запроса того же подкрепления, которое поддерживало проблемное поведение.

Referencing norms: FctPlan, Implementer, Learner, MakeRollbackTransitionReadyFromResurgence, ResurgenceByCrossMultiplication, fct_plan, plan_learner

section/3 · article · referenced by: 11

ru · unofficial · sha256:ae2fb7fb705bba44512b53ad14cbb1b0b06d6f95092a2e0ae98f0121b4466922RU

Дата сессии относится к оси legalTime и является единственной календарной величиной изложения. Время внутри сессии измеряется целым числом секунд от начала сессии (внутрисессионный офсет) и календарём не является: офсеты не переводятся в даты, не участвуют в вычислении сроков и не читают календарь. Обязанности исполнителя привязаны к закрытому окну даты сессии, начинающемуся датой сессии и заканчивающемуся объявленной верхней границей той же даты.

Referencing norms: DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, ReinforceResponseInReinforcedSignal, ReinforceResponseInReinforcedSignalDuty, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_offset_seconds, period_end_seconds, period_start_seconds, session_date, session_window_end

section/4 · article · referenced by: 12

ru · unofficial · sha256:184d57bbae7b944be8712dfef408e2923dba1388db62c4de886c5b451d2fc978RU

Сессия разбита на непересекающиеся сигнализированные периоды, каждый из которых имеет порядковый номер, начальный и конечный внутрисессионные офсеты и ровно одно состояние сигнала: s_plus либо s_minus. В периоде s_plus функциональный коммуникативный ответ подкрепляется; в периоде s_minus он не подкрепляется. Состояние сигнала есть данная журнала, а не вывод: исполнитель предъявляет различимый сигнал и записывает, какой именно. Периоды чередуются, но чередование не выводится из номера — оно записано. Изложение не содержит случайных величин: расписание переменного времени выражается закреплённым списком длительностей, а не генератором.

Referencing norms: PeriodsOfOneSessionMustNotOverlap, PresentDiscriminableSignalDuty, PresentDiscriminableSignalOfEachPeriod, SignalState, SignalledPeriod, period_end_seconds, period_number, period_record, period_session, period_signal, period_start_seconds, session_periods_overlap

section/5 · article · referenced by: 47

ru · unofficial · sha256:4c17a726510a6fd5c60d9bd72577525abb19d172f155fd8c9403c92e918cdc65RU

Разрежение задаётся четырьмя шагами Step1, Step2, Step3 и Step4; каждому шагу соответствуют две длительности в секундах — периода s_plus и периода s_minus, закреплённые набором параметров A3 и выбираемые по legalTime. От шага к шагу период s_plus сокращается, а период s_minus растёт: плотность подкрепления падает. Сессия объявляет шаг, на котором она проведена. Период соответствует шагу, если его длительность, то есть разность конечного и начального офсетов, в точности равна длительности своего состояния сигнала на этом шаге. Период, не соответствующий шагу, называется несоответствующим и является отступлением от процедуры.

Referencing norms: FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, PeriodConformsToItsStep, PeriodLongerThanItsStep, PeriodShorterThanItsStep, SMinusDurationInForceAtStep1, SMinusDurationInForceAtStep2, SMinusDurationInForceAtStep3, SMinusDurationInForceAtStep4, SPlusDurationInForceAtStep1, SPlusDurationInForceAtStep2, SPlusDurationInForceAtStep3, SPlusDurationInForceAtStep4, Step1SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep1-s-minus-seconds%23r-1/definition, Step1SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep1-s-plus-seconds%23r-1/definition, Step2SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep2-s-minus-seconds%23r-1/definition, Step2SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep2-s-plus-seconds%23r-1/definition, Step3SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep3-s-minus-seconds%23r-1/definition, Step3SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep3-s-plus-seconds%23r-1/definition, Step4SMinusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep4-s-minus-seconds%23r-1/definition, Step4SPlusSeconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Astep4-s-plus-seconds%23r-1/definition, ThinningStep, period_conforms_to_step, period_does_not_conform_to_step, session_at_step

section/6 · article · referenced by: 21

ru · unofficial · sha256:76befc2231be582f527a2f31a06b50b5cecc71dc6af4dcb35d681c19ecb43165RU

Запись функционального коммуникативного ответа имеет самостоятельную идентичность и содержит сессию, период, внутрисессионный офсет и исполнителя. Сессия содержит дату, план, шаг, исполнителя, верхнюю границу окна обязанностей, длительность наблюдения в минутах, длительности сигнализированных состояний в минутах и число эпизодов проблемного поведения. Полнота журнала утверждается явно: сессия объявляется либо полной, либо неполной, и молчание журнала не заменяет ни одно из этих утверждений. Неполная сессия непригодна для проверки критериев и не входит ни в одно окно.

Referencing norms: DeclareIncompleteSessionUnreliable, DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, FcrRecord, ObserveCompleteSession, ObserveSessionResponseImplementer, Session, fcr_implementer, fcr_period, fcr_record, fcr_session, session_duration_minutes, session_implementer, session_journal_complete, session_journal_incomplete, session_measurement_unreliable, session_observed, session_plan, session_problem_behavior_count, session_record, session_response_by

section/7 · article · referenced by: 10

ru · unofficial · sha256:4fa326e54961aef7866a6c280f8c6e51d48ce288fa405c464fae45f1f65a43feRU

Ответ засчитывается периоду, если его внутрисессионный офсет лежит между начальным и конечным офсетами этого периода включительно. Ответ, поданный в периоде s_plus, порождает обязанность подкрепления. Ответ, поданный в периоде s_minus, сохраняется журналом как попытка без правового эффекта: он не порождает обязанности подкрепления и не отменяет её нигде. Такая запись не удаляется и не переписывается: именно из неё виден ход обучения различению.

Referencing norms: ResponseFallsWithinItsPeriod, ResponseInExtinctionSignal, ResponseInExtinctionSignalHasNoLegalEffect, ResponseInReinforcedSignal, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_in_extinction_signal, fcr_in_reinforced_signal, fcr_within_period, fcr_without_effect

section/8 · article · referenced by: 11

ru · unofficial · sha256:d7d0e3569e4e7d3bde54fdd5e9a6be07f684a27939ef699338efb9ed7fd31a45RU

Для каждого ответа в периоде s_plus возникает обязанность-достижение исполнителя перед обучающимся подкрепить этот ответ в закрытом окне даты сессии. Для каждого ответа в периоде s_minus возникает обязанность-достижение не подкреплять его в том же окне. Для каждого периода возникает обязанность-достижение предъявить различимый сигнал этого периода. Для каждой сессии возникает обязанность-достижение объявить журнал полным. Исполнение и его отсутствие равно сообщаются журналом явными фактами; после закрытия окна явная неисполненность видна как нарушение. Аудит процедурной точности состоит в перечислении этих позиций.

Referencing norms: DeclareSessionJournalCompleteDuty, DeclareSessionJournalCompleteWithinSession, PresentDiscriminableSignalDuty, PresentDiscriminableSignalOfEachPeriod, ReinforceResponseInReinforcedSignal, ReinforceResponseInReinforcedSignalDuty, WithholdReinforcementInExtinctionSignal, WithholdReinforcementInExtinctionSignalDuty, fcr_not_reinforced, fcr_reinforced, period_signal_presented

section/9 · article · referenced by: 15

ru · unofficial · sha256:dcd46fb74ff2b0d40e3c5c2902719f00271b1b93e3988e9898fb52cf04753595RU

Число записей ответов сессии считается непосредственно по записям журнала. Числа ответов по состояниям сигнала, числа наблюдаемых и дискриминирующих сессий окна, напротив, поступают материализованными данными того же журнала: v1 не агрегирует по уже выводимому предикату без барьера полноты §111. Материализованный агрегат не является самостоятельным клиническим утверждением и обязан воспроизводиться из закреплённых записей журнала. Уровень ответов не вычисляется делением. Сессия дискриминирует, если уровень ответов в s_plus не ниже нижнего порога и уровень ответов в s_minus не выше верхнего порога; оба условия проверяются кросс-умножением: число ответов, умноженное на сто, сравнивается с порогом на сто минут, умноженным на длительность соответствующего состояния в минутах. Сессия не дискриминирует по одному из двух самостоятельных оснований: уровень в s_plus ниже нижнего порога либо уровень в s_minus выше верхнего. Основания различимы и записаны отдельно. Состояние нулевой длительности порогов не проверяет.

Referencing norms: CountSessionResponses, ExtinctionSignalRateMetByCrossMultiplication, FcrSMinusCeilingPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Afcr-s-minus-ceiling-per-hundred-minutes%23r-1/definition, FcrSPlusFloorPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Afcr-s-plus-floor-per-hundred-minutes%23r-1/definition, ReinforcedSignalRateMetByCrossMultiplication, SessionDiscriminatesWhenBothThresholdsAreMet, SessionFailsOnHighExtinctionSignalRate, SessionFailsOnLowReinforcedSignalRate, session_discriminates, session_does_not_discriminate, session_extinction_signal_rate_met, session_fcr_count, session_reinforced_signal_rate_met, session_state_fcr_count, session_state_minutes

section/10 · article · referenced by: 46

ru · unofficial · sha256:f0e9e515a1dd7fb73bcadd9a5e48b56ad1de87d51c4af5510ee8d62cb9210e5cRU

Окно проверки относится к плану, шагу и назначению — продвижение либо откат. Продвижение установлено, если окно содержит не менее параметрического числа последовательных наблюдаемых сессий и каждая из них дискриминирует. Откат установлен, если окно содержит не менее того же числа последовательных наблюдаемых сессий и ни одна из них не дискриминирует. Независимо от окна откат открывает резургентность: сессия, в которой уровень проблемного поведения превысил параметрический потолок, возвращает план на предыдущий шаг немедленно. Резургентность есть возвращение ранее подкреплявшегося поведения при ослаблении подкрепления замещающего ответа; она измеряется тем же кросс-умножением и тем же журналом.

Referencing norms: ConsecutiveDiscriminatingSessionCount/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aconsecutive-discriminating-session-count%23r-1/definition, EstablishAdvanceCriterion, EstablishRollbackCriterion, FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, MakeAdvanceTransitionReady, MakeRollbackTransitionReadyFromResurgence, MakeRollbackTransitionReadyFromWindow, ResurgenceByCrossMultiplication, ResurgenceCeilingPerHundredMinutes/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aresurgence-ceiling-per-hundred-minutes%23r-1/definition, ReviewPurpose, ReviewWindow, advance_criterion_met, advance_ready, discriminating_session_count, non_discriminating_session_count, observed_session_count, review_window_plan, review_window_purpose, review_window_record, review_window_step, rollback_criterion_met, rollback_ready, session_resurgence

section/11 · article · referenced by: 29

ru · unofficial · sha256:c5d7572e0b787cd7cf455001b244de816205edd46ca4c5d7a7b1e3714a94f797RU

План проходит стадии Step1Active, Step2Active, Step3Active, Step4Active и Thinned. Начальной является Step1Active — наиболее плотное закреплённое расписание; Thinned является терминальной и означает, что достигнут последний закреплённый шаг. Переход вперёд ведёт от стадии шага к стадии следующего шага, а от Step4Active — в Thinned. Переход назад ведёт от стадии шага к стадии предыдущего шага. Отката из Step1Active не существует: более плотного расписания изложение не задаёт, и требование отката на этой стадии остаётся без перехода. Выхода из Thinned не существует: возобновление разрежения после его завершения является новым планом. Обе границы объявлены намеренно.

Referencing norms: FctThinning, FctThinning/AdvanceToStep2, FctThinning/AdvanceToStep3, FctThinning/AdvanceToStep4, FctThinning/CompleteThinning, FctThinning/RollbackToStep1, FctThinning/RollbackToStep2, FctThinning/RollbackToStep3, FctThinning/Step1Active, FctThinning/Step2Active, FctThinning/Step3Active, FctThinning/Step4Active, FctThinning/Thinned, FctThinningState, FctThinningStep, FctThinningTransition, FctThinning__attempted_transition, FctThinning__current_state, FctThinning__entered_state, FctThinning__initial_state, FctThinning__left_state, FctThinning__procedure_instance, FctThinning__state_before, FctThinning__valid_transition, MakeAdvanceTransitionReady, MakeRollbackTransitionReadyFromResurgence, MakeRollbackTransitionReadyFromWindow, advance_ready, rollback_ready

section/12 · article · referenced by: 52

ru · unofficial · sha256:502b9e6e7b8450d063f8d8a68a922053f441bd84f250e2675603759892bd56c0RU

Обучение переносить задержку подкрепления и отказ в подкреплении настоящее изложение задаёт ОТДЕЛЬНОЙ процедурой DelayTolerance того же плана, а не второй осью процедуры разрежения. Уровни задержки Level1, Level2 и Level3 несут закреплённые длительности в секундах и завершаются терминальной стадией Tolerant. Продвижение и откат по уровням задержки устанавливаются собственным окном проверки и собственным критерием: сессия переносит задержку, если ответ обучающегося на задержку соответствующей длительности не сопровождался проблемным поведением, и это сообщается журналом явным фактом. Разделение осей есть содержательное утверждение, а не удобство записи: оно означает, что неудача в переносе задержки НЕ обнуляет достигнутый шаг разрежения, и наоборот. Соединить оси в одну процедуру означало бы утверждать обратное; изложение такого утверждения не делает и оставляет выбор за тем, кто ведёт программу.

Referencing norms: ConsecutiveToleratedSessionCount/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Aconsecutive-tolerated-session-count%23r-1/definition, DelayInForceAtLevel1, DelayInForceAtLevel2, DelayInForceAtLevel3, DelayLevel, DelayLevel1Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level1-seconds%23r-1/definition, DelayLevel2Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level2-seconds%23r-1/definition, DelayLevel3Seconds/value/urn%3Aaba%3Aclir%3Afct-thinning%3Aparam%3Adelay-level3-seconds%23r-1/definition, DelayTolerance, DelayTolerance/AdvanceToLevel2, DelayTolerance/AdvanceToLevel3, DelayTolerance/CompleteDelayTolerance, DelayTolerance/Level1Active, DelayTolerance/Level2Active, DelayTolerance/Level3Active, DelayTolerance/RollbackToLevel1, DelayTolerance/RollbackToLevel2, DelayTolerance/Tolerant, DelayToleranceState, DelayToleranceStep, DelayToleranceTransition, DelayTolerance__attempted_transition, DelayTolerance__current_state, DelayTolerance__entered_state, DelayTolerance__initial_state, DelayTolerance__left_state, DelayTolerance__procedure_instance, DelayTolerance__state_before, DelayTolerance__valid_transition, EstablishDelayAdvanceCriterion, EstablishDelayRollbackCriterion, MakeDelayAdvanceTransitionReady, MakeDelayRollbackTransitionReady, RecordDelayFailureAtSessionLevel, RecordToleratedDelayAtSessionLevel, delay_advance_criterion_met, delay_advance_ready, delay_failed_session_count, delay_failure_recorded, delay_observed_session_count, delay_rollback_criterion_met, delay_rollback_ready, delay_seconds_in_force, delay_tolerance_recorded, delay_tolerated_session_count, delay_window_level, delay_window_plan, delay_window_purpose, delay_window_record, session_at_delay_level, session_delay_not_tolerated, session_delay_tolerated

section/13 · article · referenced by: 3

ru · unofficial · sha256:e5db73cd85532f58afe821813c1acc020143ab0c6a2eeb46db766feaca0e121bRU

Расчёт «длительность периода в силе» возвращает длительность сигнализированного состояния на объявленном шаге сессии в секундах, а не в датах. Границы настоящей версии названы явно: изложение не моделирует ни конкурирующих стимулов, повышающих переносимость разрежения, ни отсроченного подкрепления как самостоятельной оси, ни угасания проблемного поведения, ни функционального анализа, из которого план происходит; всё перечисленное остаётся вне модели, а не объявляется несуществующим. Библиографические основания настоящего авторского изложения: Hanley, G. P., Iwata, B. A., & Thompson, R. H. (2001), “Reinforcement Schedule Thinning Following Treatment with Functional Communication Training”, Journal of Applied Behavior Analysis 34(1), doi:10.1901/jaba.2001.34-17; Fisher, W. W., Kuhn, D. E., & Thompson, R. H. (1998), “Establishing Discriminative Control of Responding Using Functional and Alternative Reinforcers During Functional Communication Training”, Journal of Applied Behavior Analysis 31(4), doi:10.1901/jaba.1998.31-543; Hagopian, L. P., Kuhn, S. A. C., Long, E. S., & Rush, K. S. (2005), “Schedule Thinning Following Communication Training: Using Competing Stimuli to Enhance Tolerance to Decrements in Reinforcer Density”, Journal of Applied Behavior Analysis 38(2), doi:10.1901/jaba.2005.43-04; Tiger, J. H., Hanley, G. P., & Bruzek, J. (2008), “Functional Communication Training: A Review and Practical Guide”, Behavior Analysis in Practice 1(1), doi:10.1007/BF03391716; Hagopian, L. P., Boelter, E. W., & Jarmolowicz, D. P. (2011), “Reinforcement Schedule Thinning Following Functional Communication Training: Review and Recommendations”, Behavior Analysis in Practice 4(1), doi:10.1007/BF03391770; Fuhrman, A. M., Fisher, W. W., & Greer, B. D. (2016), “A Preliminary Investigation on Improving Functional Communication Training by Mitigating Resurgence of Destructive Behavior”, Journal of Applied Behavior Analysis 49(4), doi:10.1002/jaba.338; Briggs, A. M., Fisher, W. W., Greer, B. D., & Kimball, R. T. (2018), “Prevalence of Resurgence of Destructive Behavior When Thinning Reinforcement Schedules During Functional Communication Training”, Journal of Applied Behavior Analysis 51(3), doi:10.1002/jaba.472.

Referencing norms: CalculateSessionsToAdvanceCriterion, sessions_to_criterion, state_duration_in_force

The hash beside the text is the sha256 of its exact bytes. It is the check that the text shown matches the one pinned in the package: a divergence fails the oracle’s static analysis.