By Matt Harrington, BCBA · Behaviorist Book Club · April 2026 · 12 min read

Progressive Discrete Trial Teaching: Moving Beyond the Conventional DTT Model

Overview & Clinical Significance

Discrete trial teaching has been a cornerstone of ABA-based intervention for autism for decades. Its structured format — discriminative stimulus, response opportunity, consequence, inter-trial interval — creates the conditions for high-density, data-monitored skill acquisition that has produced strong outcomes across a wide range of domains. Yet despite this empirical track record, DTT has also been criticized for producing rigid, context-dependent performance and for lacking the naturalness needed to support generalization.

Progressive DTT, as described in this course, represents an evolution of the conventional DTT model that addresses many of these limitations while preserving the structural precision that makes DTT effective. The distinction between conventional and progressive philosophies of DTT is not a superficial rebranding — it reflects a substantive rethinking of how specific components of the teaching trial are conceptualized and implemented. Understanding this distinction is clinically important for any BCBA who supervises DTT programs, because the differences affect skill acquisition rates, generalization, learner independence, and program individualization.

For practitioners trained exclusively in conventional DTT models, this course offers an opportunity to examine the assumptions embedded in their current practice and evaluate whether progressive alternatives might produce better outcomes for specific learners. For those already familiar with progressive approaches, it provides an empirical framework for defending those methods and communicating their rationale to caregivers, administrators, and supervisors.

Background & Context

The conventional DTT model emerged from early applied work building on the experimental analysis of behavior and was formalized in protocols such as the Lovaas Early Intensive Behavioral Intervention model. Its core features — massed trials on a single target, rigid inter-trial intervals, standardized consequence delivery — were designed to maximize experimental control and learning trial density. These features produced strong acquisition data in controlled settings.

Over time, however, practitioners and researchers identified limitations of the strictly conventional approach. Massed practice on isolated targets sometimes produced responding that was highly stimulus-bound, with poor transfer to naturalistic contexts. Rigid protocol adherence across all learners and targets failed to account for the significant individual differences in learning history, motivation, and developmental profile that characterize the autism spectrum. The conventional model's strength — its consistency — could also become a liability when it prevented individualization.

The progressive DTT philosophy, associated with the work of practitioners like Dr. Justin Leaf, proposes that DTT components should be actively adapted based on learner response data and individual learning needs rather than applied uniformly based on protocol. This includes adapting the type and timing of prompts, the nature of the consequence, the distribution of trials, and the sequencing of targets. The progressive approach retains the structural elements of DTT while insisting that each component decision be driven by clinical data and individualized analysis rather than default protocol.

This evolution within the field reflects a broader movement in ABA toward individualized, naturalistic, and flexible programming while maintaining the data-driven core that distinguishes behavior analysis from less rigorous intervention frameworks.

Clinical Implications

The progressive DTT model has several direct clinical implications for program design and supervision. First, the concept of individualized component selection means that supervisors and BCBAs must have a deeper understanding of each DTT component — not just how to implement it but why a specific configuration is chosen for a given learner at a given point in their program. This requires ongoing data review and the conceptual flexibility to adapt components when data indicate that the current approach is not producing expected progress.

Second, progressive DTT's emphasis on naturalistic instructional opportunities means that DTT trials are not confined to a structured tabletop format. Practitioners trained in progressive DTT embed teaching trials within naturally occurring activities, increasing the ecological validity of instruction and reducing the separation between structured and naturalistic teaching that characterizes conventional approaches. This has implications for how sessions are structured, how therapists are trained, and how session data are collected.

Third, the progressive approach changes how reinforcement is conceptualized and implemented. Rather than using a single, program-wide reinforcement approach, progressive DTT advocates for real-time assessment of reinforcer potency and flexibility in consequence delivery that matches the learner's current motivational state. This requires therapists who can read motivating operations as they shift during a session and adjust accordingly — a clinical skill that goes beyond standard protocol adherence.

For BCBAs supervising teams, the progressive DTT model raises the bar for therapist training. Implementing progressive DTT with fidelity requires not just procedural skill but conceptual understanding of why each component works the way it does. Supervision models that invest in conceptual foundations — not just behavioral demonstrations — are better positioned to implement progressive DTT effectively.

Ethical Considerations

The ethical dimensions of progressive versus conventional DTT center primarily on effectiveness and individualization. BACB Code 2.01 requires practitioners to use effective, evidence-based procedures and to individualize treatment. If conventional DTT is producing suboptimal outcomes for a specific learner — slow acquisition, limited generalization, high error rates under naturalistic conditions — the practitioner is ethically obligated to consider modifications. The progressive DTT philosophy provides a framework for making those modifications systematically.

Code 2.09 (Treatment Modification and Interruption) applies when existing DTT protocols are not producing expected progress. A BCBA who continues implementing a conventional DTT protocol that is not producing acquisition because it is the agency's standard protocol is not meeting this standard. The obligation is to the client's outcomes, not to protocol consistency for its own sake.

Code 2.15 (Least Restrictive Procedures) is also relevant. To the extent that progressive DTT produces equivalent or better outcomes with less structured, more naturalistic instructional formats, it may represent a less restrictive alternative to highly structured conventional DTT for some learners. Practitioners should consider this in program design, particularly for learners who show distress or avoidance responses to highly structured instructional contexts.

Transparency with caregivers is obligated under Code 2.03. When BCBAs make decisions about DTT philosophy — whether to use conventional or progressive approaches, and why — those decisions should be communicated clearly to caregivers with an explanation of the clinical rationale. Caregivers who understand the reasons for instructional format decisions are better equipped to implement consistency at home and to advocate for their child's programming needs.

Assessment & Decision-Making

Deciding whether and how to implement progressive DTT for a given learner begins with a review of current programming outcomes. Examine acquisition data across current targets: are learners meeting mastery criteria within expected timeframes? Are skills generalizing to naturalistic contexts? Are there specific components of existing DTT protocols — prompt types, reinforcement schedules, inter-trial intervals — that appear to be limiting progress? These data questions frame the clinical case for progressive modifications.

A component-level analysis is the next step. Review each DTT component — discriminative stimulus presentation, prompt type and timing, response definition, consequence delivery, inter-trial interval, and trial distribution — and assess whether the current implementation for each component is optimally matched to the learner's profile. For example, a learner with a history of prompt dependency may benefit from a shift in the prompt fading strategy; a learner who is showing extinction-like responding during massed practice may benefit from a distributed trial format.

Learner motivation assessment should also guide component decisions. A thorough preference assessment, including both formal assessment and in-session observation of reinforcer potency, should inform consequence delivery decisions. Progressive DTT advocates for continuously updated understanding of the motivating operations that are active during sessions, which requires therapists to observe and respond to shifts in learner engagement and reinforcer effectiveness.

When introducing progressive modifications, implement changes one component at a time where feasible and continue collecting data to evaluate the effect of each change. This systematic approach allows the clinical team to identify which modifications are contributing to improved outcomes and which are not, producing a progressively refined instructional protocol that is increasingly well-matched to the individual learner.

What This Means for Your Practice

Progressive DTT challenges practitioners to move beyond a check-the-box approach to structured instruction and toward a genuinely dynamic, data-responsive model of teaching. For BCBAs who have been implementing the same DTT protocols across all learners for years, this course offers a framework for examining those practices critically and identifying where individualized modifications might produce meaningfully better outcomes.

The most actionable shift is to begin treating DTT components as hypotheses rather than fixed procedures. Each component decision — this prompt hierarchy, this reinforcement schedule, this trial distribution — should be understood as a current best practice based on available data, subject to revision when data indicate it is not working. This epistemological shift transforms the clinical role from protocol implementer to true scientist-practitioner.

For supervisors, progressive DTT has implications for how training is structured. Therapist competency in DTT should include not just behavioral demonstration of correct implementation but verbal demonstration of why each component is designed the way it is for a specific learner. Supervision conversations should address the reasoning behind program decisions, not just their execution.

For administrators and clinical directors, progressive DTT may require updating agency-wide program templates to build in flexibility for individualized component adaptation. Standard protocols can still provide structure, but they should be understood as defaults to be adapted rather than requirements to be uniformly applied. This shift positions agencies to provide more individualized, outcomes-driven care — a competitive and ethical advantage in a field where stakeholders are increasingly sophisticated about what effective ABA looks like.