1.3 Reinforcement Schedules and Stimulus Control

Reinforcement Schedules and Stimulus Control

Reinforcement is the single most fundamental principle of operant behavior. For a Qualified Behavior Analyst, understanding how reinforcement operates across basic schedules, choice environments, and antecedent stimulus arrangements is necessary for designing effective skill acquisition and behavior reduction plans.

---

Principles of Positive and Negative Reinforcement

In behavior analysis, reinforcement is defined strictly by its functional outcome on behavior, never by formal intentions or subjective pleasantness. Reinforcement has occurred only if a stimulus change immediately follows a response and results in an increase or maintenance of the future frequency, rate, or intensity of that response.

1. Positive Reinforcement ($S^{R+}$)

Occurs when a response is followed immediately by the presentation or addition of a stimulus, resulting in an increase in the future frequency of that response under similar conditions.

2. Negative Reinforcement ($S^{R-}$)

Occurs when a response is followed immediately by the termination, removal, reduction, or postponement of a stimulus, resulting in an increase in the future frequency of that response. Negative reinforcement is divided into two clinical mechanisms:

• Escape Behavior: The response terminates an ongoing, currently present aversive stimulus (e.g., opening an umbrella while standing in active rain).
• Avoidance Behavior: The response prevents or postpones the delivery of an impending aversive stimulus before it occurs (e.g., checking weather radar and staying indoors before rain begins).

---

Primary vs. Secondary Reinforcers & Generalized Conditioned Reinforcers

Reinforcers are classified by how they acquire their reinforcing properties:

1. Unconditioned (Primary) Reinforcers: Stimuli that function as reinforcers without any prior learning history (phylogenic origin). Examples include food, water, warmth, oxygen, and sexual stimulation.
2. Conditioned (Secondary) Reinforcers: Neutral stimuli that acquire reinforcing properties through repeated pairing with primary or established secondary reinforcers (ontogenic origin). Examples include sticker charts, points, and verbal praise.
3. Generalized Conditioned Reinforcers (GCSRs): A type of conditioned reinforcer that has been paired with many unconditioned and conditioned reinforcers across diverse contexts. GCSRs do not depend on a specific MO state to be effective. The classic clinical example is money or tokens in a token economy, which can be exchanged for a wide array of back-up reinforcers.

---

Intermittent Schedules of Reinforcement

A schedule of reinforcement is a rule specifying which occurrences of a target behavior will be reinforced. While continuous reinforcement (CRF / FR1) is optimal for initial skill acquisition, intermittent schedules are used to maintain established behaviors and build resistance to extinction.

The four basic intermittent schedules are defined along two axes: Ratio (based on number of responses) vs. Interval (based on time elapsed before a response is reinforced), and Fixed (predictable requirement) vs. Variable (unpredictable average requirement):

---

Herrnstein's Matching Law and Concurrent Schedules

In natural clinical environments, individuals rarely encounter isolated single schedules. Instead, they choose between multiple simultaneously available options—a concurrent schedule of reinforcement.

Formulated by Richard Herrnstein (1961), The Matching Law states that on concurrent schedules of reinforcement, the relative rate of responding matches the relative rate of reinforcement provided by each schedule:

$\frac{R_1}{R_1 + R_2} = \frac{r_1}{r_1 + r_2}$

Where $R_1$ and $R_2$ represent the rates of response on options 1 and 2, and $r_1$ and $r_2$ represent the rates of reinforcement delivered by options 1 and 2.

Clinical Implication for QBAs

If a student can obtain reinforcement on a VI 30-second schedule for challenging behavior (screaming) and a VI 120-second schedule for functional communication (manding), the student will allocate approximately 80% of their responding to screaming. To reduce challenging behavior, the QBA must alter reinforcer parameters (rate, quality, delay, effort) so that alternative appropriate behavior receives a significantly higher density of reinforcement.

---

Stimulus Control, Generalization, and Discrimination

Stimulus control occurs when the rate, latency, duration, or amplitude of a response is altered in the presence of an antecedent stimulus. Stimulus control is established through differential reinforcement:

• Discriminative Stimulus ($S^D$): Reinforcement is delivered for the response in its presence.
• Stimulus Delta ($S^\Delta$): Reinforcement is withheld (extinction) or punished when the response occurs in its presence.

Discrimination vs. Generalization

• Stimulus Discrimination: The organism responds only in the presence of the $S^D$ and not in the presence of $S^\Delta$ (tight stimulus control).
• Stimulus Generalization: The response occurs in the presence of novel antecedent stimuli that share similar physical properties with the original $S^D$. A stimulus generalization gradient shows that as physical similarity to the $S^D$ decreases, response rate drops.

---

Prompting Strategies and Transfer of Stimulus Control

When a learner does not yet respond to the natural $S^D$, clinicians use prompts—supplemental antecedent stimuli used to evoke the correct response.

Prompt Types

1. Response Prompts: Operate directly on the response (Vocal, Modeling, Physical guidance).
2. Stimulus Prompts: Operate directly on the antecedent stimuli (Movement, Position, Redundancy/exaggeration).

Transfer of Stimulus Control (Prompt Fading)

To achieve true independence, stimulus control must be transferred from the prompt to the natural $S^D$. Common fading procedures include:

• Most-to-Least Prompting: Begins with maximum physical guidance and systematically fades to lesser prompts as mastery is achieved (prevents errors during acquisition).
• Least-to-Most Prompting: Gives the learner an opportunity to respond to the natural $S^D$ independently, delivering progressively intrusive prompts only if errors or no responses occur.
• Graduated Guidance: Provides physical prompts as needed moment-by-moment and immediately shadows/fades within a trial.
• Time Delay (Constant or Progressive): Inserts a fixed or gradually increasing time interval between the natural $S^D$ and the delivery of the prompt, allowing independent responding to emerge.