Differential reinforcement of other behavior

DRO reinforces an absence criterion

A complete description states:

• the observable target response
• the interval or observation schedule
• whether one target response resets or fails the interval
• what consequence follows a successful interval
• how the initial interval was chosen and later changed
• who monitors, delivers reinforcement, and records exceptions

The term can sound as though a specific “other” response earns reinforcement. The actual requirement is target-response absence. A methodological review by Thompson and Iwata describes DRO as presenting a stimulus change contingent on the absence of the target response and discusses how that negative contingency differs from response-independent reinforcement.

The ABAI basic-principles page lists differential reinforcement in higher-education content. The BACB BCBA Test Content Outline, 6th edition covers differential-reinforcement procedures with and without extinction. These documents identify training and examination content rather than prescribing an individual plan.

Whole-interval and momentary DRO differ

Under whole-interval DRO, reinforcement follows when the target response is absent throughout the entire interval. A response during the interval makes that interval unsuccessful; some arrangements restart the timer, while others wait for the next scheduled interval.

Under momentary DRO, reinforcement depends on whether the target response is absent at the precise observation when the interval ends. The response may have occurred earlier in that interval. Continuous observation is unnecessary for scoring the moment, which can ease implementation and can also make the measure less representative of the full interval.

Intervals may be fixed or variable. These choices affect predictability, reinforcement density, monitoring, and what the resulting percentage means. A study of fixed momentary DRO describes procedural variations and implementation considerations (Sullivan and colleagues). An earlier four-participant comparison found whole-interval DRO more effective than momentary DRO in those cases, with order and maintenance differences across participants (Repp, Barton, and Brulle). These small studies do not create a universal preference.

Choose intervals from observed data

An initial interval should provide realistic contact with reinforcement. A long interval can yield few successful intervals, frustration, and weak learning. An extremely short interval can be impractical or produce reinforcement near an unscored target response.

Record the baseline interresponse times or a comparable opportunity measure, the rule used to select the starting interval, and the criterion for changing it. Increase duration gradually when results, implementation accuracy, and the person's experience support the change. Report completed, successful, failed, reset, missed, and invalid intervals separately.

Reinforcement delivered at an interval boundary can accidentally follow a different response that happened moments earlier. Define a brief delivery hold or safety rule when immediate delivery would reinforce an unsafe or conflicting response. Preserve the earned outcome and deliver it once the safe condition returns rather than silently canceling it.

Target absence does not teach a replacement skill

DRO can reduce a response while leaving the person without an efficient way to obtain help, a break, attention, information, sensory support, or access. Pair the schedule with direct communication, environmental changes, medical or interdisciplinary care, and a functional alternative when those are relevant.

DRA reinforces a specified alternative. DRI reinforces a response physically incompatible with the target. Functional communication training teaches a recognizable communication response that produces the relevant outcome. These procedures can be combined deliberately, with each contingency and measure kept clear.

Relapse planning also matters. A review of recurrence research notes that DRO and DRA can differ in resurgence patterns and that findings vary by arrangement (Kimball and colleagues). This literature supports testing schedule changes and disruptions rather than promising durable response reduction.

A fictional DRO example

Jordan is a fictional fourteen-year-old who wants online coding-club waits to feel more predictable. The defined target is sending the same queue-status message again within two minutes of a prior message. A whole-interval DRO schedule uses four-minute intervals during waits. An interval succeeds when no repeated status message occurs throughout it; one occurrence ends that interval without restarting the session.

Across ten baseline intervals with the queue hidden, the target occurs in 8 of 10. Two intervals meet the absence criterion. Staff provide useful status information after 5 of 8 repeated messages, showing that the environment often responds to the target.

Jordan and the team add a visible queue, a one-tap update request, and DRO. Each successful interval earns one minute of a chosen mentor code review, while the update request receives an immediate honest status. Across twelve later intervals, the target is absent in 9 of 12, staff deliver the earned review after 8 of 9 successful intervals, and Jordan uses the update request in 7 of 8 defined opportunities.

Several components changed together, so the counts describe the package rather than a DRO-only effect. Jordan rates 10 of 12 intervals predictable. The team reports target absence, update use, staff follow-through, and Jordan's experience as separate outcomes.

Protect access, assent, and dignity

ASHA's AAC practice portal says AAC users should always have access to their communication tools or devices. Keep speech, AAC, gesture, writing, and other effective forms available. Honor clear help, dissent, discomfort, and safety communication even when it resets or fails a DRO interval.

Food, water, bathroom access, mobility, prescribed care, pain care, rest, relationships, and emergency help remain available regardless of interval performance. A target definition should describe an actual concern rather than harmless self-regulation, disability expression, culture, or a style adults find inconvenient.

The current BACB Ethics Code applies to BCBA and BCaBA certificants and people who completed an application. It addresses client involvement, consent and assent when applicable, assessment, medical needs, risk, positive reinforcement, data, and continual evaluation.