An ABA task interspersal worksheet should reveal the teaching sequence a learner experienced. It should not let a broad label such as “mixed trials” hide how often the acquisition target appeared, whether the interspersed tasks were truly maintained, which prompts and consequences were used, or whether the arrangement helped this learner.

This reusable ABA task interspersal worksheet separates acquisition opportunities, maintenance opportunities, planned ratio, actual order, initial responding, support, preference, and implementation integrity. It does not choose a skill, ratio, task domain, reinforcement schedule, or mastery criterion. A fictional community-garden example shows how to reconcile counts without offering a recipe for treatment.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

Clinical boundary: Task interspersal is an instructional arrangement, not a reason to continue a target that lacks a meaningful, authorized purpose. Keep speech, AAC, gesture, sensory, motor, visual, medical, safety, and relationship supports available throughout teaching. Learners need an accessible way to choose, pause, seek help, decline, and ask for a different order. Use this worksheet only within individualized work overseen by qualified professionals. The BACB ethics page identifies current ethics resources; the form itself cannot establish consent, necessity, effectiveness, payer coverage, school compliance, or legal compliance.

Interspersal is not one uniform procedure

Task interspersal generally means placing one type of trial among another. In skill-acquisition programs, clinicians often place previously acquired tasks among acquisition tasks. Those previously acquired tasks may be described as maintenance tasks. But the label does not reveal whether tasks share a response class or whether a fixed or changing ratio is used. It also does not reveal the consequences, order-selection rule, or actual exposure to the acquisition target.

A survey of task-interspersal implementation practices found substantial variation in task type and domain. Reported ratio and reinforcement practices also varied. The authors noted that empirical best practices for those variations were not settled. A worksheet should therefore document the selected arrangement and its rationale. It should not present the most frequently reported practice as the correct one.

Task interspersal also differs from a high-probability request sequence. A task can be mastered without being easy or preferred in the moment. An interspersed maintenance trial remains part of an instructional series, while a high-probability sequence is defined by response probability and its position before a lower-probability request. Combining the labels can obscure both the teaching dose and the hypothesized relation.

The current BACB certification test-content outlines and the Sixth Edition BCBA outline distinguish measurement, skill acquisition, maintenance, generalization, integrity, and data-based evaluation. They are useful organizing references, not a task-interspersal protocol.

Teaching-purpose and target-definition card

Name the functional reason for teaching the acquisition target. Define the antecedent, response, latency, materials, and expected variations. Record prompts, the correction procedure, and the outcome separately. Then document the evidence that each proposed maintenance task is currently performed under the relevant conditions. If the maintenance response has weakened, it may require its own teaching rather than serving as a stable interspersed item.

FieldCase-specific entryMeaningful teaching purpose and authorized plan referenceAcquisition target and direct measureValid opportunity start, end, materials, and contextInitial correct, incorrect, no-response, and not-scorable definitionsPrompt and error-correction rulesProposed maintenance tasks and current evidenceSame or different task domain and rationaleOrdinary communication, AAC, sensory, motor, and visual supportsChoice, help, break, refusal, and alternative responsesQualified owner, version, effective date, and review date

Avoid targets that are described only as “cooperation,” “work behavior,” or “attending.” The record needs an observable learner response and a valid opportunity. It also needs a place for environmental failure. Missing materials, an unavailable activity, the wrong stimulus array, or an inaccessible instruction should not become an acquisition error.

Maintenance-task qualification table

Task IDOperational responseContext sampledValid opportunitiesInitial correctPrompted or correctedCurrent maintenance evidenceEffort/access notesLearner feedbackInclude, hold, or reteach

The exploratory analysis of task interspersal while teaching object labels compared constant, similar-interspersal, and dissimilar-interspersal arrangements. Five children participated. Acquisition in the constant condition was similar to or better than the interspersal conditions in that study. No consistent difference emerged between the two interspersal forms. That result is a useful warning against assuming interspersal is superior. Its targets, reinforcement arrangements, participants, and mastery criteria do not determine what another learner needs.

Versioned ratio and trial-order map

Write the planned order in a way that can be compared with delivery. A ratio such as 2:1 is ambiguous unless the record says which task type comes first and whether the ratio resets after a prompt or break. It should also explain how partial blocks, re-presented trials, and correction trials count.

Plan elementCurrent versionOperational descriptionRationale/evidenceExceptionOwnerAcquisition task setMaintenance task setPlanned acquisition:maintenance ratioPlanned order within a blockTask-selection or randomization rulePrompt and correction placementConsequence after acquisition responseConsequence after maintenance responseBreak, choice, refusal, and restart ruleHold, revise, compare, or stop condition

The brief review of task interspersal for neurodevelopmental disorders highlights limitations in the existing literature. It also identifies questions for additional study. The review should temper confident claims about an optimal ratio or universal learning advantage. A case record can support an individualized decision only when the implemented sequence and direct outcomes are visible.

Block ledger preserves exposure and order

Use one row for a planned instructional block. Link it to trial-level records when a block contains several targets. Record invalid blocks even though they leave the learner-performance denominator. Repeated invalidity may identify a materials, scheduling, training, or access problem.

Date/timeVersionPlanned order/ratioActual order/ratioValid? reasonAcquisition opportunity/resultMaintenance opportunities/resultsPrompts/correctionsConsequencesChoice/help/dissentUnwanted effectObserveryes / noyes / noyes / no

Trial-level detail when needed

Block IDTrial positionTask typeTask IDValid opportunity?Initial response/latencyPrompt or correctionFinal supported responseConsequenceAccess or context noteacquisition / maintenanceacquisition / maintenanceacquisition / maintenance

Score the initial response before adding a prompted or corrected outcome. A correct response after a prompt may show successful support without being independent acquisition. Re-presented trials need their own label so the team does not inflate exposure or learning by silently counting the same opportunity twice.

Keep reinforcement, preference, and task status separate

A maintenance task may contact a familiar success history, but that does not prove it functions as reinforcement or is preferred. The study comparing contingent and noncontingent access to preferred academic tasks found different results across five preschool participants. Its findings support measuring case-specific effects and keeping contingency details visible. They do not support assuming a preferred task will work the same way for every learner.

Learner choice can be recorded directly. A study of preference between massed and alternating trial sequences used a computerized choice procedure. Two children selected an alternating arrangement. The small, specific study does not establish a population preference. It does show why the learner's selection and experience can be treated as data rather than guessed from adult impressions.

DateOptions offered and communication formChoice or responseTime/effortAffect or burdenHelp/break honored?Would choose again?Requested change

The ASHA AAC Practice Portal describes communication systems as individualized and flexible. They are often multimodal. The learner's established communication system remains available during both acquisition and maintenance trials. Access to it is not contingent on completing the sequence.

Measurement layers need different denominators

Review questionNumeratorDenominatorResultDoes not showValid block coveragevalid instructional blocksplanned instructional blockslearning or client effortAcquisition initial accuracyinitial correct acquisition responsesvalid acquisition opportunitiesmaintenance performance or causationMaintenance initial accuracyinitial correct maintenance responsesvalid maintenance opportunitiespreference or acquisitionPlanned acquisition yieldinitial correct acquisition responsesplanned blocks expected to contain acquisitionwhy blocks were invalidRatio/order fidelityvalid blocks delivered in the planned ratio and ordervalid blockslearner benefitPrompted completionsupported acquisition responsesacquisition opportunities receiving supportindependent respondingOverall implementation integritycorrectly implemented componentsobservable component opportunitiesmastery, social validity, or generalization

These rows should retain raw counts, not only percentages. A 90% total score can conceal excellent maintenance responding and weak acquisition responding. Reporting only correct trials can also hide reduced exposure to the target. A learning claim requires a pattern across relevant observations, not a single mixed-session average.

Fictional community-garden example

This fictional program concerns learning labels used in a chosen community-garden sorting activity. Eighteen instructional blocks were planned. One block lacked the planned seed packets. Another was cancelled when the garden activity changed. Those two blocks were invalid, leaving 16 valid blocks.

Each valid block scheduled one acquisition opportunity and two maintenance opportunities. The record therefore contains 16 valid acquisition opportunities. It also contains 32 valid maintenance opportunities. Initial correct responses occurred in 10 acquisition opportunities and 28 maintenance opportunities.

  • valid block coverage: 16 / 18 × 100 = 88.9%;
  • acquisition initial accuracy: 10 / 16 × 100 = 62.5%;
  • maintenance initial accuracy: 28 / 32 × 100 = 87.5%;
  • planned acquisition yield: 10 / 18 × 100 = 55.6%; and
  • planned ratio and order delivered: 13 / 16 × 100 = 81.3%.

The acquisition and maintenance percentages use separate denominators. Combining 38 correct initial responses across all 48 valid task opportunities would produce 79.2%, but that pooled figure would blur the difference between the new and maintained targets. The 13-of-16 sequence result describes plan delivery, not learning.

Fifteen observations were also scored against eight implementation components, producing 120 observable opportunities. One hundred twelve were correct and eight were mismatches. Overall integrity was 112 / 120 × 100 = 93.3%.

Evidence layerFictional resultBounded interpretationPlanned coverage16 valid of 18 plannedTwo blocks could not test learner performanceAcquisition response10 of 16 opportunitiesNo mastery or causal claim followsMaintenance response28 of 32 opportunitiesMaintenance was not perfect or necessarily preferredPlanned acquisition yield10 of 18 planned blocksPreserves system loss in the recordRatio/order delivery13 of 16 valid blocksThree sequences differed from the planImplementation integrity112 of 120 componentsEight mismatches require reviewPreference and burdenseparate direct/descriptive evidence neededAccuracy cannot answer acceptabilityGeneralization or later maintenancenot establishedThe example covers this period and setting only

The record raises focused questions: why two blocks were unavailable, why three valid blocks departed from the planned order, which acquisition errors or prompts occurred, whether maintenance tasks remained useful, what the learner preferred, and where eight integrity mismatches arose. Review those questions before changing the arrangement. The worksheet does not pick the next ratio or criterion.

Learning review across time

Review window/versionValid acquisition opportunitiesInitial correctPrompted/correctedLatency or fluency measureMaintenance opportunities/correctRatio/order fidelityLearner feedback/burdenGeneralization or retention probeInterpretation limits

The receptive-language program design article describes similar and dissimilar interspersal and expansion arrangements. It repeatedly identifies areas where further comparison is needed. Use such descriptions to define a candidate plan, not to guarantee an outcome. A later probe should be labeled for its actual context and elapsed time rather than folded into training accuracy.

Integrity, decision, and exception record

ObservationCurrent versionMaterials and valid opportunity correctTask type/order correctPrompt/correction correctConsequence correctAAC/access preservedChoice/help/dissent honoredInitial response recordedCorrect componentsOpportunitiesNotes

Decision fieldReview entryVersions, settings, people, and dates reviewedAcquisition and maintenance definitions and current evidencePlanned versus actual ratio/order and exposureInitial, prompted, corrected, latency, and integrity resultsLearner choice, assent/dissent, burden, and unwanted effectsEnvironmental barriers and helpful supportsGeneralization, retention, or comparison evidence still missingHold / revise / compare / retain / fade / stop questionQualified decision owner, rationale, and next review date

The Ethics Code for Behavior Analysts places client involvement, informed consent, assent when applicable, competence, confidentiality, documentation, and data use within professional responsibility. The CASP ABA Practice Guidelines access page offers current context for individualized planning and evaluation. Neither source endorses this worksheet, a particular interspersal ratio, or a decision for an individual learner.

Pause or reconsider when the learner reports burden, maintenance responses deteriorate, acquisition exposure becomes too sparse to answer the review question, the task order creates confusion, communication access is reduced, fidelity remains unstable, or the purpose is no longer meaningful. A simpler, concentrated, differently ordered, embedded, distributed, or non-interspersed arrangement may deserve comparison. The responsible team needs direct evidence before choosing among them.

Related resources

Sources

The cited studies differ in task type, participant, sequence, consequence, measurement, setting, and design, and some report mixed or limited findings. Together they justify transparent planning, direct measurement, learner feedback, and cautious comparison. They do not provide a standard ratio, trial order, task set, consequence, prompting rule, mastery criterion, or expected effect for a particular learner.