Time delay

Teaching often starts at zero seconds

At zero delay, the prompt follows the natural cue immediately. This arrangement can establish a correct response with few errors. After the learner responds reliably under the prompt, the plan introduces time between cue and prompt.

For reading a transit symbol, the teacher might say the route number immediately after showing it during zero-delay trials. Later, the symbol appears and the teacher waits. A correct response during the interval is independent of the added vocal prompt under the operational definition.

Constant delay uses one fixed interval

In constant time delay, the prompt interval changes from zero to a selected duration and then stays fixed. The duration might be two, four, or five seconds, based on the learner’s latency, response form, task, and evidence.

A review of constant time delay for reading instruction describes the common zero-delay phase followed by a fixed interval that permits independent responding. Its findings concern selected students with intellectual disability or autism and particular reading targets.

Progressive delay increases in steps

Progressive time delay lengthens the interval gradually, such as zero, one, two, and four seconds, according to a prewritten performance rule. Smaller changes may help when response latency develops slowly.

Reichow and Wolery compared progressive prompt delay with and without instructive feedback for three autistic children and one child with developmental delays. The small study suggests that procedural additions can affect efficiency and maintenance. It does not establish one schedule for every learner.

The natural cue must remain clear

Time delay transfers control only when the intended cue is present and distinguishable. Teacher posture, materials, position, repeated wording, or visible preparation to prompt can become unintended cues.

Probe with ordinary people, materials, and settings. If the response occurs only when the familiar teacher leans forward, waiting longer has transferred control to the wrong feature.

Early responses and errors need rules

The plan states what happens when the learner responds before the prompt, after the prompt, incorrectly, or after the response window. It also defines whether correction trials count as new opportunities and how help requests are scored.

A response that occurs during the delay can still be wrong. A prompted correct response is useful learning data and remains distinct from an independent response. Reliable labels protect interpretation.

Delay should respect communication and pace

Some people need more time to visually scan, compose AAC, move, or regulate. A short interval can produce avoidable prompting. A long silent wait can also feel confusing when the cue is unclear.

Select the interval with the person and relevant supporters. Keep AAC available and provide an accessible way to ask for help, pause, or withdraw when that right applies. Delay never justifies withholding ordinary communication support.

Clinical authority and fading stay visible

The BACB BCBA Test Content Outline, 6th edition includes time delay, prompt fading, stimulus control, observable goals, contextual fit, integrity, and evaluation. It is examination content, not a protocol.

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

A fictional delay record

Owen practices recognizing four pantry symbols. After zero-delay teaching, the plan uses a four-second constant delay. Across 12 eligible opportunities, Owen responds independently and correctly during the delay in 7 of 12. The prompt produces a correct response in four more; one response is incorrect.

Independent performance is 7/12. Prompted correct performance is 4/12, and the error is 1/12. These categories cover the same locked cohort and sum to 12. They do not prove mastery, causation, or transfer to a store shelf.

Review transfer, effort, and integrity

Track independent correct, prompted correct, incorrect, and absent responses by eligible opportunity; latency; delay level; integrity; and unprompted transfer probes. Record device failure, unclear cues, interruptions, and assent withdrawal separately.

If progress stalls, review the cue, controlling prompt, interval, correction rule, reinforcement, response effort, and goal fit. Changing time alone may leave the underlying problem untouched.

Generalization checks should preserve the same delay definition while varying one feature at a time, such as the speaker, symbol size, room, or device. This helps the team see whether responding follows the intended cue or a familiar teaching setup. Maintenance probes after teaching pauses answer another question: whether the response persists over time. Label acquisition, generalization, and maintenance data separately because one strong session cannot stand in for all three.