Discrete-trial teaching

A discrete trial has defined parts

A clinician might arrange a choice of pictures, say “Show me the bus,” wait for the defined response window, and provide the planned feedback. The record shows whether the response was independent, prompted, incorrect, absent, or interrupted under the plan.

The parts can be summarized as:

1. materials and attention are ready
2. a clear cue or instruction occurs
3. the learner has a defined opportunity to respond
4. the teacher follows the planned feedback or correction procedure
5. the trial ends before another begins

Definitions matter. One instruction repeated three times could be one opportunity or three, depending on the prewritten rule. Stable counting supports useful interpretation.

DTT is a teaching format

The CDC autism treatment page describes DTT as an ABA teaching style that breaks instruction into simple steps. That public overview is useful orientation. It does not prescribe a particular trial structure, prompt sequence, consequence, schedule, or dose for a specific person.

DTT can teach discrimination, imitation, communication, academic, play, self-care, or other skills when assessment supports the goal and format. The skill should matter to the person and work through an accessible response, including speech, AAC, gesture, selection, movement, or another reliable form.

Prompts and feedback require explicit decisions

The plan states which prompts may occur, when they occur, and how they fade. A model, gesture, visual cue, positional support, spoken hint, or physical prompt has different access, learning, and consent implications.

Feedback after a correct response should follow the learner’s preferences and the skill’s natural context. Error correction should be brief, respectful, and defined. Food, water, bathroom access, communication, mobility, prescribed care, and emergency help remain freely available.

Teaching pace should leave room for the learner

A high trial count can increase practice, and it can also create fatigue or reduce meaningful participation. Plan the session around response effort, communication, health, breaks, latency, signs of distress, and the person’s willingness.

When assent applies, provide an accessible way to start, pause, change, or stop. Record withdrawal, equipment problems, medical or safety interruptions, and invalid opportunities separately from learning data.

Structured teaching needs everyday transfer

A person may select the right picture at a table and still need support to use the skill in a kitchen, classroom, store, or conversation. Generalization should be planned and measured with different people, examples, settings, and ordinary cues.

Natural-environment teaching can complement DTT. The team might introduce a clear discrimination in short trials, then practice it during chosen activities. Transfer probes should remain distinct from coached teaching trials.

Procedure selection is clinical work

The BACB BCBA Test Content Outline, 6th edition includes trial-based and free-operant procedures, prompting, prompt fading, observable goals, contextual fit, procedural integrity, and effectiveness review. It is examination content, not a stand-alone protocol or authority to practice.

For covered behavior analysts, the current BACB Ethics Code addresses competence, client and stakeholder involvement, consent and assent when applicable, assessment-based intervention, risk, data, and continual evaluation. A qualified clinician selects and revises the teaching plan within scope.

Research compares formats in narrow settings

A small study by Geiger and colleagues compared traditional and embedded DTT with two four-year-old autistic boys. Both formats produced skill acquisition under the study’s conditions, and the paper measured procedural fidelity. Two participants and a particular teaching context cannot establish a preferred format for everyone.

The study is useful as a reminder to define the trial, measure whether the teacher followed the procedure, and assess learner performance separately.

A fictional DTT record

Micah chooses a goal of identifying three transit symbols before a community trip. During 12 defined teaching trials with AAC available, Micah responds independently and correctly in 8 of 12. The teacher completes every planned procedure component in 10 of 12 trials; two trials miss the required wait time.

Learner performance is 8/12, while integrity is 10/12. Combining them would hide two different questions. The counts do not prove that DTT caused learning or that Micah will identify the symbols at a station. The team schedules uncoached community probes and asks Micah about the teaching format.

Useful measures stay specific

Track independent correct responses divided by eligible opportunities, prompt level, errors by type, latency, teaching duration, procedure components completed, and performance in separate transfer probes. Report exclusions and assent withdrawals with reasons.

Mastery should include a prewritten threshold, relevant conditions, and a plan for maintenance and generalization. A percentage cannot repair an ambiguous response definition or a biased set of opportunities.