Two Categories of Reinforcers
A reinforcer is a stimulus change that follows a response and increases the future frequency of that response. Reinforcers are classified by how they acquired their function. Unconditioned reinforcers (also called primary reinforcers) function without a history of learning. Conditioned reinforcers (also called secondary reinforcers) acquired their function through a learning history, typically by pairing with another reinforcer.
Unconditioned Reinforcers
Unconditioned reinforcers are biologically relevant stimuli that support survival and comfort. They are effective without prior learning, although their effectiveness depends on motivating operations.
- Food and drink
- Sleep and rest
- Warmth and comfort
- Sensory stimulation, such as vibration or visual patterns
- Relief from pain or extreme temperature
- Sexual stimulation
Their value fluctuates with deprivation and satiation. Food is an effective reinforcer before a meal and a weak one after a large meal.
Clinical use of unconditioned reinforcers
Primary reinforcers are useful early in programs, particularly for learners with limited histories of social reinforcement. Small bites of food, sips of juice, and sensory toys are commonly used. However, they have drawbacks.
- Satiation: The effect weakens quickly with repeated delivery.
- Time: Consuming food or playing with toys interrupts the learning flow.
- Health and dietary concerns: Reinforcers must be safe and consistent with the client's health needs.
- Naturalness: They may not be available or appropriate in typical environments.
Conditioned Reinforcers
A neutral stimulus becomes a conditioned reinforcer when it is repeatedly paired with an established reinforcer. Imagine saying "Good job!" every time you deliver a favorite snack. After many pairings, "Good job!" can function as a reinforcer on its own. Praise starts as a neutral stimulus for many young children, and it gains value through its association with attention, food, and affection.
Examples
- A clicker used in animal training, paired with food.
- A smiley-face sticker paired with access to a preferred activity.
- The sound of the ice cream truck, paired with ice cream.
- A coach's thumbs-up paired with playing time.
Generalized Conditioned Reinforcers
A generalized conditioned reinforcer has been paired with many different backup reinforcers, so it is effective under many motivating operations. This makes it less vulnerable to satiation than a single reinforcer.
- Money: Can be exchanged for countless items and services.
- Tokens: Points, stars, or chips that can be traded for a menu of backup reinforcers.
- Social attention and praise: Paired with many kinds of reinforcement across a person's life.
- Grades: Linked to many outcomes such as admission, approval, and opportunities.
Token economies are a major clinical application. Tokens are delivered immediately after a target behavior, and later exchanged for backup reinforcers. This bridges the delay between behavior and a reinforcer that cannot be delivered right away, such as screen time at the end of a session.
Establishing a Conditioned Reinforcer
- Select a neutral stimulus that is easy to deliver, such as a verbal marker, a sticker, or a token.
- Pair it with a known reinforcer. Present the neutral stimulus immediately before the reinforcer, consistently, over many trials.
- Test its effect by delivering it contingent on a simple response and seeing whether the response increases without the backup reinforcer.
- Maintain the pairing by periodically delivering the backup reinforcer after the conditioned reinforcer. Without this, the conditioned reinforcer can lose value.
- Fade toward natural reinforcers so the learner can function in everyday environments.
Stimulus-stimulus pairing for language
Some programs pair the voices of staff with reinforcers to make speech sounds reinforcing, a procedure called stimulus-stimulus pairing. Research on its effectiveness is mixed, so clinicians should combine it with other evidence-based strategies and monitor data closely.
Pairing Procedures and Their Mechanism
Conditioned reinforcement is built on respondent conditioning: the neutral stimulus is paired with a reinforcer, so that it acquires reinforcing properties. The result is an operant effect, because the stimulus then increases behavior it follows. This illustrates how operant and respondent processes interact. The pairing is respondent, and the function is operant.
Conditioned Punishers
The same logic applies to punishers. A stimulus paired with a punisher, such as the word "No," can become a conditioned punisher. Generalized conditioned punishers, such as reprimands or fines, are paired with many kinds of punishment. Programs should rely first on reinforcement-based approaches and follow ethical guidelines on restrictive procedures.
Strengths and Challenges
Advantages of conditioned reinforcers
- Can be delivered immediately and with minimal disruption.
- Resist satiation, especially if generalized.
- Can bridge delays to backup reinforcers.
- More natural and portable than most primary reinforcers.
Challenges
- Require a history of pairing. A new learner may not respond to praise yet.
- Value depends on the backup reinforcers. If the backup is not available, the token loses value.
- Can lose effectiveness if not occasionally paired with the reinforcer.
- Token economies require careful design to avoid hoarding or ratio strain.
Practical Tips
- Pair praise with tangible reinforcement for new clients before expecting praise alone to work.
- Use a variety of backup reinforcers and update a menu with preference assessments.
- Set a clear token exchange schedule and keep the ratio of work to reward reasonable.
- Teach learners to respond to natural conditioned reinforcers such as smiles and social approval.
- Avoid taking tokens away unless the plan has been carefully designed with response cost and ethical review.
Exam Strategies
- If the stimulus works without learning, it is unconditioned. Food, water, warmth, and sleep are typical examples.
- If the stimulus works because of pairing, it is conditioned. Praise, tokens, and the sound of a clicker are typical examples.
- If the stimulus works across many motivating operations, it is a generalized conditioned reinforcer. Money and tokens are common examples.
- Remember that the backup reinforcer maintains the value of a token.
Key Takeaways
Unconditioned reinforcers work without learning but are vulnerable to satiation. Conditioned reinforcers gain value through pairing, and generalized conditioned reinforcers such as tokens and money are especially flexible. Build and maintain conditioned reinforcers carefully, and fade toward natural reinforcement.