9.3 Discrete Trial Teaching (DTT) & Errorless Learning Procedures

Discrete Trial Teaching (DTT) & Errorless Learning Procedures

Exam Tip: A discrete trial is not simply "tabletop therapy"; it is a scientifically validated, highly structured instructional cycle containing five distinct components: (1) Discriminative Stimulus ($S^D$), (2) Prompt (if needed), (3) Learner Response, (4) Consequence / Reinforcement ($S^{R+}$) or Error Correction, and (5) Inter-Trial Interval (ITI). On the QASP-S exam, pay meticulous attention to the Inter-Trial Interval duration (1–3 seconds), the exact sequence of the 4-Step Error Correction Procedure (Model $\rightarrow$ Prompt $\rightarrow$ Switch $\rightarrow$ Repeat), and the distinction between Cold Probe vs. Trial-by-Trial data collection.

Originally pioneered by Dr. Ivar Lovaas in the 1960s and refined over decades of empirical research, Discrete Trial Teaching (DTT) is one of the most rigorously evaluated pedagogical methods in Applied Behavior Analysis. DTT breaks complex behavioral repertoires down into small, discrete, teachable units. By isolating each component of the three-term contingency ($S^D \rightarrow R \rightarrow S^R$), DTT provides repeated, high-density learning opportunities with immediate, unambiguous feedback. A QASP-S must understand how to engineer discrete trials, prevent error rehearsal, implement structured error correction, and choose appropriate data collection systems.

---

The Five Demarcated Components of a Discrete Trial

Every discrete trial operates as a self-contained learning unit governed by precise behavioral parameters:

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│ 1. ANTECEDENT   │ ──► │ 2. LEARNER      │ ──► │ 3. CONSEQUENCE  │
│    (Sd + Prompt)│     │    RESPONSE     │     │ (SR+ or Correc.)│
└─────────────────┘     └─────────────────┘     └────────┬────────┘
                                                         │
                                                         ▼
                                                ┌─────────────────┐
                                                │ 4. INTER-TRIAL  │
                                                │    INTERVAL     │
                                                │    (1-3 sec)    │
                                                └────────┬────────┘
                                                         │
                                                         ▼
                                                [Next Trial Cycle]

1. Discriminative Stimulus ($S^D$)

The $S^D$ is the clear, concise antecedent directive, question, or environmental presentation that signals reinforcement is available contingent upon the emission of the target response.

• Clarity & Brevity: The $S^D$ must be stripped of extraneous conversational language. Saying "Touch shoe" is clear; saying "Okay, sweetie, can you look down at the table and show me where the shoe is?" introduces auditory clutter that impairs stimulus control for autistic learners.
• Consistency: The phrasing must remain uniform across all therapists during the acquisition phase. Variations are introduced only after the learner has demonstrated mastery, to promote stimulus generalization.
• Attending Prerequisite: The technician must never deliver the $S^D$ until the learner's attention is secured (e.g., eye orientation toward the materials, hands quiet on table, absence of active motor stereotypy).

2. The Prompt (Antecedent Assistance)

When a skill is in acquisition or when following an errorless learning protocol, the prompt is delivered concurrently with the $S^D$ or immediately following it (e.g., within 0–1 seconds), before the learner has an opportunity to emit an incorrect response.

3. Learner Response

The operationalized behavior emitted by the learner following the $S^D$ within a predetermined latency window (typically 3 to 5 seconds):

• Correct Independent: The target response is emitted accurately without adult assistance.
• Correct Prompted: The target response is emitted accurately following adult prompting.
• Incorrect (Error): The learner emits an inaccurate motor or vocal response (e.g., touches the wrong card, speaks an incorrect word).
• No Response (Omission): The learner fails to emit any response within the specified latency window.

4. Consequence ($S^R$ or Error Correction)

The consequence is delivered immediately (ideally within 0.5 to 1 second) following the response to preserve temporal contiguity:

• Contingent on Correct Response: Delivery of immediate positive reinforcement ($S^{R+}$)—either primary (edible, high-value toy) or secondary (token, tickles, enthusiastic descriptive praise).
• Contingent on Incorrect or Non-Response: Delivery of neutral feedback (e.g., a calm "Let's try again," clearing materials without emotional reaction) and immediate transition into the prescribed error correction protocol. No reinforcement is delivered.

5. Inter-Trial Interval (ITI)

The Inter-Trial Interval (ITI) is the brief temporal pause that separates the delivery of the consequence of one trial from the presentation of the $S^D$ for the subsequent trial.

• Optimal Duration: Typically 1 to 3 seconds.
• Clinical Functions: The ITI allows the learner to consume the reinforcer, gives the technician time to record the trial outcome on the data sheet, and permits the resetting or repositioning of tabletop stimuli.
• The Pacing Danger: If the ITI exceeds 5 to 10 seconds, instructional momentum is lost, the learner's attention drifts, rates of off-task and problem behaviors skyrocket, and the density of learning opportunities is severely compromised.

---

Errorless Learning Procedures

Traditional trial-and-error learning permits learners to guess, make mistakes, and self-correct. In Applied Behavior Analysis, Errorless Learning is an instructional strategy designed to prevent errors from occurring during the acquisition of new skills.

The Behavioral Rationale for Errorless Learning

1. Elimination of Faulty Stimulus Control: When learners make errors, the incorrect response may accidentally become paired with reinforcement or become an entrenched part of a superstitious behavioral chain.
2. Prevention of Emotional Responding: For many autistic learners, repeated errors elicit frustration, escape-motivated aggression, elopement, or self-injurious behavior.
3. Maximized Reinforcement Density: Errorless learning ensures that every trial results in a successful response and immediate access to reinforcement, establishing high behavioral momentum.

The Mechanics of Errorless Learning

1. 0-Second Delay Trials: The technician delivers the $S^D$ (e.g., "Touch blue") and simultaneously delivers the controlling prompt (e.g., physically guiding the child's hand to the blue card). The learner cannot fail.
2. Immediate Reinforcement: The child receives reinforcement for the prompted correct response.
3. Systematic Fading: After a set number of consecutive 0-second trials (e.g., 5 successful trials), the technician introduces a time delay (e.g., 2 seconds) or fades prompt intrusiveness (e.g., full physical $\rightarrow$ partial physical $\rightarrow$ gestural).

---

The 4-Step Error Correction Procedure

When a learner is being taught using standard trial formats or after prompt fading has begun, errors will inevitably occur. When an error or non-response happens, the technician must execute a systematic 4-Step Error Correction Procedure:

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│ STEP 1: MODEL   │ ──► │ STEP 2: PROMPT  │ ──► │ STEP 3: SWITCH  │ ──► │ STEP 4: REPEAT  │
│ Clear materials,│     │ Re-present Sd,  │     │ Insert 1-2      │     │ Re-present target│
│ demonstrate the │     │ provide immediate│    │ mastered distrac-│   │ Sd unprompted   │
│ correct response│     │ controlling cue │     │ tor tasks (easy)│     │ to test control │
└─────────────────┘     └─────────────────┘     └─────────────────┘     └─────────────────┘

Step-by-Step Breakdown

1. Step 1: Model (End Initial Trial & Demonstrate): The technician neutrally terminates the error trial, removes the materials briefly, resets the array, and demonstrates the correct response. Example: If the child touches "fork" when asked to "Touch spoon," the technician says, "This is spoon," while touching the spoon.
2. Step 2: Prompt (Evoke with Controlling Support): The technician re-presents the original $S^D$ ("Touch spoon") and immediately provides a controlling prompt (e.g., full physical or gestural prompt) to ensure the learner touches the spoon correctly. Praise delivered here is minimal or neutral.
3. Step 3: Switch / Distractor Trial: To ensure the learner is not simply engaging in immediate echolalia or motor perseveration, the technician inserts 1 to 2 simple, mastered distractor demands (e.g., "Touch nose!" or "Clap hands!"). The learner completes these easily, clearing the immediate motor memory.
4. Step 4: Repeat / Test Probe: The technician re-presents the original target $S^D$ ("Touch spoon") without any prompt to test whether stimulus control has transferred.

• Reinforcement Rule: If the learner responds correctly on the Repeat step, deliver moderate praise or a secondary reinforcer. Never deliver the highest-value, premium reinforcer on a repeat trial, as this would accidentally reinforce the error-correction chain (Learner learns: Make error $\rightarrow$ get prompted $\rightarrow$ get big cookie).

---

Trial Presentation Formats

Discrete trial instruction utilizes three primary trial distribution formats depending on the learner's stage of acquisition:

1. Massed Practice (Massed Trials)

• Operational Structure: Presenting the exact same target $S^D$ repeatedly across consecutive trials without any intervening distractor items (e.g., Trial 1: "Touch car"; Trial 2: "Touch car"; Trial 3: "Touch car").
• Clinical Indications: Restricted strictly to initial skill introduction when a learner has zero baseline competence. It allows the learner to establish the initial motor or echoic pattern.
• Clinical Limitations: Highly unnatural; induces rapid satiation, robotic or perseverative responding, and position biases. Must be faded as quickly as possible.

2. Distributed Practice (Distributed Trials)

• Operational Structure: Target trials are interspersed across time, embedded into naturally occurring daily routines, or distributed across different academic subjects.
• Clinical Indications: Ideal for naturalistic environment teaching (NET), conversational targets, and social skills.
• Clinical Advantage: Promotes generalization and maintenance under natural antecedent conditions.

3. Random Rotation (Interleaved Practice)

• Operational Structure: The target acquisition stimulus is rapidly and randomly alternated with multiple mastered distractor items and varying instructional demands within the same array (e.g., Trial 1: "Touch car" [Target]; Trial 2: "What does a cow say?" [Mastered]; Trial 3: "Touch shoe" [Mastered distractor in array]; Trial 4: "Touch car" [Target]).
• Clinical Value: This is the gold standard for verifying true discriminative stimulus control. It eliminates position bias, prevents rote memory habits, and mirrors real-world cognitive processing.

---

Data Collection Methodologies: Cold Probe vs. Trial-by-Trial

A QASP-S must design data collection systems that balance measurement precision with instructional efficiency. The two primary methods in DTT are:

                               DATA COLLECTION COMPARISON
  ┌───────────────────────────────┬───────────────────────────────────────────────┐
  │ FIRST-TRIAL COLD PROBE        │ TRIAL-BY-TRIAL DATA RECORDING                 │
  ├───────────────────────────────┼───────────────────────────────────────────────┤
  │ • Scores ONLY the first trial │ • Scores EVERY individual trial emitted       │
  │   presented in the session.   │   throughout the entire session.              │
  │ • Measures true retention and │ • Captures within-session learning, prompt    │
  │   generalization from past day│   fading kinetics, and error frequencies.     │
  │ • Extremely fast; zero loss   │ • Slows down ITI pacing; technician must      │
  │   of instructional momentum.  │   look away to record data after every trial. │
  │ • Low sensitivity to small,   │ • High measurement sensitivity; heavy         │
  │   within-session improvements.│   administrative recording burden.            │
  └───────────────────────────────┴───────────────────────────────────────────────┘

---

Comparison of DTT Instructional Formats & Error Handling