stimulus-control

name	stimulus-control
description	Use when designing discrimination training, prompt fading, or stimulus transfer procedures — covers stimulus control, generalization, equivalence, faulty stimulus control, and clinical remediation strategies.

Stimulus Control and Transfer Procedures

Definition of Stimulus Control

A behavior is under stimulus control when it occurs at a high rate in the presence of a specific antecedent stimulus (the discriminative stimulus, SD) and at a low rate in the absence of that stimulus or in the presence of other stimuli (S-delta, SΔ). Stimulus control is established through a history of differential reinforcement: responses in the presence of the SD are reinforced, while responses in the presence of the SΔ are not reinforced (or are placed on extinction).

Stimulus control is not an all-or-nothing phenomenon — it exists on a continuum. Tight stimulus control means the behavior occurs almost exclusively in the presence of the SD. Loose stimulus control means the behavior occurs across a broader range of stimuli.

Establishing Stimulus Control

Discrimination Training

Present the SD and reinforce the correct response. Present the SΔ and withhold reinforcement (extinction) for the response. Across repeated trials, the individual learns to respond differentially.

Simple Discrimination

One stimulus occasion and one behavior. Example: When the teacher holds up a red card (SD), saying "red" is reinforced. When the teacher holds up any other color card (SΔ), saying "red" is not reinforced.

Conditional Discrimination

The correct response depends on the combination of two or more stimuli. Example: Given the auditory instruction "Touch red" (conditional stimulus), selecting the red card from an array is reinforced. Given "Touch blue," selecting blue is reinforced. The correct response is conditional on both the instruction and the visual array.

Go/No-Go Discrimination

The individual responds in the presence of the SD and withholds responding in the presence of the SΔ. Example: During a matching task, the individual places a card when the sample and comparison match (go) and does not place a card when they do not match (no-go).

Stimulus Generalization and Discrimination

• Stimulus generalization: A behavior reinforced in the presence of one stimulus occurs in the presence of other, similar stimuli without direct training. Generalization is a gradient — the more similar the novel stimulus is to the training stimulus, the more likely the behavior is to occur.
• Stimulus discrimination: The process by which an organism responds differentially to stimuli that differ along one or more dimensions.

Clinically, both are essential. We need behavior to generalize across relevant exemplars (e.g., the child tacts "dog" for all dogs, not just their pet) but also to be discriminated where relevant (e.g., the child does not tact "dog" for cats).

Stimulus Control Transfer Procedures

The goal of many teaching procedures is to transfer stimulus control from prompts (which are supplementary, artificial stimuli) to the natural SD. All prompting procedures must include a plan for transfer.

Prompt Fading

Gradually reduce the intensity, salience, or intrusiveness of a prompt across trials until the natural SD alone controls responding.

Most-to-Least Prompting

Begin with the most intrusive prompt (e.g., full physical guidance) and systematically fade to less intrusive prompts (partial physical → model → gestural → independent).

• Advantage: Reduces errors during acquisition.
• Disadvantage: May build prompt dependency if fading is too slow.
• Use when: The learner has a history of failure with the task, or errors would be significantly aversive.

Least-to-Most Prompting

Begin with the least intrusive prompt (e.g., a brief time delay or gestural cue) and escalate to more intrusive prompts only if the learner does not respond correctly.

• Advantage: Provides the learner maximum opportunity for independent responding.
• Disadvantage: The learner contacts errors before prompts are delivered, which can be aversive or reinforce error patterns.
• Use when: The learner has some existing skill or the behavior is within their current repertoire.

Time Delay

Constant Time Delay

Initially present the SD and the prompt simultaneously (0-second delay). After a set number of trials, introduce a fixed delay (e.g., 4 seconds) between the SD and the prompt. The learner has the opportunity to respond before the prompt is delivered.

Progressive Time Delay

The delay between the SD and the prompt increases gradually across trials (e.g., 0s → 1s → 2s → 3s → 4s → 5s). This provides a more gradual transfer of control.

Time delay is one of the most empirically supported and efficient transfer procedures. It minimizes errors while systematically transferring control to the natural SD.

Stimulus Fading

Gradually change the physical characteristics of the stimuli (not the prompts) to transfer control.

• Example: In teaching letter discrimination, the correct letter is initially presented in bold red while distractors are faint gray. Across trials, the color and boldness of the correct letter are faded until all letters appear identical.
• Stimulus fading manipulates the irrelevant dimension of the stimulus while maintaining the relevant dimension.

Stimulus Shaping

Gradually change the shape or form of a stimulus from one that already controls the behavior to the target stimulus.

• Example: To teach a child to read the word "cat," begin with a picture of a cat superimposed on the word, then gradually fade the picture while the word remains, transferring control from the picture to the text.

Stimulus Equivalence

Sidman (1971) described stimulus equivalence as the emergence of untrained relations among stimuli. If an individual is taught to match A→B and B→C, equivalence is demonstrated when the following untrained relations emerge:

• Reflexivity: A=A, B=B, C=C (generalized identity matching).
• Symmetry: B→A and C→B (if A→B was trained, B→A emerges without training).
• Transitivity: A→C (if A→B and B→C were trained, A→C emerges without training).
• Combined symmetry and transitivity (equivalence): C→A.

Clinical Significance

Stimulus equivalence explains how teaching a small set of relations can produce a much larger set of derived relations without direct training. For example:

1. Teach the child to match the spoken word "dog" to a picture of a dog (A→B).
2. Teach the child to match the picture of a dog to the printed word "DOG" (B→C).
3. Without direct training, the child can now match the spoken word "dog" to the printed word "DOG" (A→C, transitivity) and the printed word "DOG" to the spoken word "dog" (C→A, equivalence).

This is the basis for efficient language and academic instruction — teaching a few relations and testing for the emergence of the rest.

Faulty Stimulus Control

Prompt Dependency

The behavior is controlled by the prompt rather than the natural SD. The individual waits for the prompt before responding, even when the natural SD is present. Prompt dependency results from:

• Prompts delivered too quickly (not allowing latency for independent responding).
• Insufficient or unsystematic fading.
• Inadvertent differential reinforcement of prompted vs unprompted responses.

Remediation: Implement systematic prompt fading with time delay. Differentially reinforce independent (unprompted) responses with higher-quality reinforcement.

Stimulus Overselectivity

The individual attends to only one feature of a compound stimulus and fails to respond to the relevant dimension. Common in learners with autism and intellectual disabilities.

• Example: A child taught to select a picture of a dog may attend only to the color of the background rather than the features of the dog. When the background changes, the child can no longer select correctly.

Remediation:

• Teach across multiple exemplars that vary irrelevant features while maintaining the relevant feature.
• Use conditional discrimination training with varied stimuli.
• Assess which stimulus feature controls responding by systematically varying dimensions.

Restricted Stimulus Control

Behavior is controlled by a narrower range of stimuli than intended. The individual responds correctly in the training context but fails to generalize to novel exemplars or settings.

Remediation:

• Train sufficient exemplars to establish a generalized response class.
• Vary irrelevant stimulus features across training (multiple exemplar training).
• Program common stimuli between training and generalization settings.
• Teach across multiple settings, people, and materials.

Assessing Stimulus Control

• Probe trials: Present the SD without prompts to assess whether the natural stimulus controls responding.
• Transfer tests: Present novel stimuli to assess generalization.
• Error analysis: Categorize errors to identify which stimulus features control responding — errors reveal faulty stimulus control.
• Stimulus control assessment probes: Systematically vary stimulus dimensions to determine which features control the behavior.

Key References

• Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied Behavior Analysis (3rd ed.). Pearson.
• Sidman, M. (1971). Reading and auditory-visual equivalences. Journal of Speech and Hearing Research, 14(1), 5–13.
• Sidman, M., & Tailby, W. (1982). Conditional discrimination vs. matching to sample: An expansion of the testing paradigm. Journal of the Experimental Analysis of Behavior, 37(1), 5–22.
• Terrace, H. S. (1963). Discrimination learning with and without "errors." Journal of the Experimental Analysis of Behavior, 6(1), 1–27.
• Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of Applied Behavior Analysis, 10(2), 349–367.