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

One-to-One vs. Group Discrete Trial Teaching: Evidence, Efficiency, and Clinical Decision-Making

Overview & Clinical Significance

Discrete trial teaching (DTT) is one of the most extensively researched and widely implemented instructional formats in applied behavior analysis. It provides a structured, controlled environment for teaching skills across developmental domains — language, cognition, self-care, social behavior — by breaking complex behavioral chains into discrete components, presenting clear antecedent stimuli, prompting correct responses, delivering contingent reinforcement, and recording trial-by-trial data. The decision about whether to implement DTT in a one-to-one format or a group instructional format is among the most consequential programming choices a BCBA makes.

This course, presented by Dr. Leaf, describes a study that directly compared one-to-one DTT to group DTT across six participants, evaluating effectiveness, efficiency, and observational learning. These three outcome variables represent distinct but interrelated dimensions of instructional value: effectiveness captures whether the skill was acquired, efficiency captures how many instructional trials were required, and observational learning captures whether participants acquired skills targeted for other group members — a dimension of learning that one-to-one instruction cannot produce.

For BCBAs, the clinical significance of this comparison extends well beyond the six-participant study. The question of instructional format affects staffing models, program design, cost, and — most importantly — the quality and generalization of skills acquired. A BCBA who can make this decision based on empirical evidence rather than convention or resource availability is practicing at a higher level of clinical sophistication than one who defaults to either format without examination.

Background & Context

Discrete trial teaching as a systematic instructional format is associated with the early intensive behavioral intervention literature, where one-to-one instruction was the dominant model. The rationale for one-to-one DTT includes maximizing stimulus control by minimizing competing stimuli, ensuring that antecedent-behavior-consequence sequences are precisely managed, and allowing for immediate and unambiguous reinforcement delivery. For learners with limited attending, significant prompt dependence, or emerging prerequisite skills, one-to-one formats offer a level of instructional control that group formats cannot easily replicate.

Group DTT emerged from recognition that natural learning environments are inherently social and that one-to-one instruction, while valuable, does not prepare learners for the group contexts they will inhabit — classrooms, social groups, recreational settings. Group DTT preserves the structured trial format while adding social complexity, including the opportunity for observational learning, turn-taking, social reinforcement, and the generalization demands that arise from having multiple participants in the instructional context.

The Autism Partnership Foundation has a history of examining the group instruction question empirically. Earlier APF-associated work investigated observational learning in autistic individuals — a topic with direct clinical relevance because autistic learners have historically been characterized as learning less readily through observation than neurotypical peers. Research demonstrating that autistic children can and do acquire skills through observation when appropriate conditions are established has significant implications for how BCBAs design group-based instruction.

The six-participant study described in this course adds to a literature that has been growing but remains less extensive than the one-to-one DTT research base. Understanding the specific procedural features of this comparison — how groups were structured, how sessions were conducted, how data were collected — is necessary for applying the findings clinically with appropriate generalizability in mind.

Clinical Implications

The procedural differences between one-to-one and group DTT have direct implications for how BCBAs design instructional programs. In one-to-one DTT, the instructor presents every trial to the same learner, controls all antecedent and consequent events, and has unambiguous attribution of each response. In group DTT, each trial is presented to one member of the group while others are expected to attend and may acquire target or non-target skills through observation. This difference in instructional mechanics changes the role of the instructor, the distribution of reinforcement, the learning opportunities per unit time per student, and the social demands on each participant.

Effectiveness outcomes from the study — whether one-to-one and group formats produced comparable skill acquisition — have direct implications for programming decisions. If group DTT achieves comparable mastery criteria to one-to-one DTT for specific skill types and learner profiles, it represents a clinical option that can expand instructional options without sacrificing outcomes. The key question is for which skills and which learners this equivalence holds — a determination that requires BCBAs to understand not just the headline finding but the participant characteristics and skill targets of the comparison.

Efficiency outcomes speak to the resources required for skill acquisition. If group DTT produces equivalent mastery in fewer total instructional trials or fewer hours of staff time, it offers a meaningful advantage in settings where resources constrain service intensity. Conversely, if certain skills require substantially more group trials to achieve mastery, the efficiency advantage of one-to-one instruction may justify the format even when group instruction is feasible.

Observational learning is perhaps the most clinically distinctive advantage of group DTT. Skills acquired through observation — without direct reinforcement to the observing learner — represent efficiency gains that have no counterpart in one-to-one instruction. BCBAs who systematically assess whether their clients demonstrate observational learning can design group instruction that maximizes this advantage, selecting group compositions and target skills that create high-value observation opportunities.

Ethical Considerations

Code 2.01 requires behavior analysts to implement only evidence-based procedures, and the growing literature comparing instructional formats provides an increasingly clear basis for format selection. However, the existing research has limitations — small samples, primarily young autistic participants, specific skill domains — and BCBAs should apply Code 6.01's requirement to use scientifically supported procedures with appropriate recognition of where evidence is strong and where it remains preliminary.

Code 2.09 requires least restrictive procedures, and instructional format is one dimension of restrictiveness. One-to-one DTT, while not aversive in the traditional sense, imposes a level of structure and control that may exceed what is necessary for skill acquisition in some learners. When group DTT can produce equivalent outcomes, it may represent a less restrictive option precisely because it more closely approximates the naturally occurring learning contexts into which skills must generalize.

The social validity of group instruction is an ethical consideration independent of its effectiveness. Autistic learners will spend their lives learning in social contexts alongside other people; instruction that develops the social-learning skills — turn-taking, attending while others are instructed, managing arousal during waiting periods — that these contexts require is not merely more efficient but more socially valid. This consideration should weigh in treatment planning discussions alongside effectiveness and efficiency data.

Code 2.11 requires that families understand and participate in treatment format decisions. BCBAs who can explain the evidence for both formats, describe how the decision between them will be made for their specific child, and build in data-based review of the selected format are meeting their informed consent obligations while ensuring that format decisions remain responsive to individual client data.

Assessment & Decision-Making

The decision between one-to-one and group DTT formats requires a multi-dimensional assessment of the learner's current behavioral repertoire, the skill targets under consideration, and the available instructional context. Learner dimensions relevant to format selection include attending repertoire, prerequisite social skills (turn-taking, waiting, tolerating others' responses), level of distraction in the presence of peers, history of observational learning, and current prompt dependence. BCBAs should assess each of these dimensions rather than making format decisions based on global impressions of the learner's functioning level.

Skill target dimensions also guide format selection. Skills requiring high levels of individualized antecedent control — early mand training, initial tact acquisition with novel stimuli, complex discriminations with many exemplars — may be more efficiently established in one-to-one contexts. Skills involving social content, perspective-taking components, or responses that naturally occur in the presence of peers may be better targeted in group formats from the outset. For many skill domains, one-to-one instruction establishes the skill initially while group instruction builds generalization and fluency.

Observational learning assessment is a specific prerequisite for productive group instruction. BCBAs should evaluate whether a client demonstrates observational learning using systematic probes: present instruction to another individual in the client's presence, then probe for the skill in the client without direct instruction. Documenting observational learning history establishes whether the client can benefit from the observational opportunities in group DTT, or whether prerequisite training in observational learning is needed first.

Data-based decision-making requires the BCBA to specify in advance what outcome data will guide format decisions: mastery criteria comparison, trials-to-criterion comparison, observational learning probe outcomes, and social behavior data during group sessions. Building this review process into program plans — with specified decision rules for switching between formats — prevents format decisions from becoming static habits rather than ongoing clinical judgments.

What This Means for Your Practice

The study described in this course provides specific empirical grounding for a clinical decision that BCBAs make continuously but often without adequate individualized data. Three practice-level changes follow directly from this content.

First, if your current practice defaults to one-to-one DTT without systematic evaluation of whether group formats would be appropriate, conduct a format audit for your caseload. For each client, identify which current skill targets might be addressable in a group format and whether the prerequisites for productive group instruction are present. This audit is not an argument for switching formats wholesale — it is a prompt for making format decisions deliberately rather than by default.

Second, build observational learning assessment into your intake and quarterly review process. Knowing whether a client can acquire skills through observation is clinically essential information that too few programs track systematically. A brief observational learning probe — presenting instruction to another person while the target client is present, then probing for the skill — requires minimal additional time and produces information that changes programming decisions.

Third, when group DTT is feasible and appropriate, design group compositions thoughtfully. Matching group members by skill level for specific targets, selecting target skills with natural social relevance, and pairing learners who can serve as high-quality models for each other maximizes the instructional value of the group format. Group DTT is not simply one-to-one DTT with more people in the room — it is a distinct instructional arrangement that requires its own design logic.