Reinforcement Flashcards

Positive Reinforcement

Reinforcement

The addition of a stimulus following a behavior that increases the likelihood of that behavior occurring again.

I love how neat that is!

Example:

• Antecedent: Writing time.

• Behavior: Student keeps their letters on the line.

• Consequence (added): Teacher gives a thumbs-up and specific praise: “I love how neat that is!”

• Effect: Neat handwriting increases during future writing times.

• Reinforcer: Teacher praise/attention (social, conditioned).

Reinforcement

Negative Reinforcement

(Escape)

The removal, reduction, postponement, or avoidance of an aversive stimulus following a behavior that increases the likelihood of that behavior occurring again.

Example:

• Antecedent: Parent is nagging, “Do your homework!"

• Behavior: Alex starts doing homework.

• Consequence (removed): Parent stops nagging.

• Effect: Alex is more likely to do homework right away in the future.

Automatic Reinforcement

Reinforcement

A behavior is strengthened because it directly produces a sensory consequence (no other person delivers it).

Example:

• Antecedent/EO: Itchy mosquito bite.

• Behavior: Child scratches.

• Consequence (automatic SR–): Itch relief (no one else involved).

• Effect: Scratching is more likely next time in a similar situation.

Socially mediated reinforcement

Reinforcement

A behavior is strengthened because another person delivers the consequence after the behavior (either adds something desired or removes an aversive). Not automatic.

Example:

• Antecedent/EO: Sun glare makes relief valuable.

• Behavior: Student asks "can you close the curtain please?"

• Consequence: Teacher closes the curtain (glare removed).

• Reinforcer: Removal of glare, delivered by the teacher.

Reinforcement

Reinforcers

A reinforcer is the stimulus change delivered after a response that makes future responses in that class more likely.

Reinforcement = the process that increases future behavior Reinforcer = the stimulus delivered after the response that does the increasing.

• Edible: preferred food, candy, etc

• Sensory: sounds, lights, vibration, tickles.

• Tangible: toys, stickers, trinkets.

• Activity: access to preferred activities/privileges.

Formal Properties of Reinforcers

• Social: attention, proximity, praise, high-fives.

Unconditioned Reinforcers

Reinforcement

(Primary / Unlearned)

An unconditioned reinforcer works as reinforcement without prior learning (due to biology/phylogeny).

• Food when hungry
• Water when thirsty
• Warmth
• Oxygen
• Touch, sexual stimulation

Positive Unconditioned Reinforcers (SR+)

• Relief from pain/pressure
• Loud noise ending
• Intense light ending
• Extreme temperature ending
• Itch relief

Negative Unconditioned Reinforcers (SR–)

Stimuli whose presentation strengthens behavior:

Stimuli whose removal strengthens behavior:

oxygen

water

food

warmth

touch

CONDITIONED REINFORCER

Reinforcement

(Secondary / Learned)

A previously neutral stimulus that becomes reinforcing through pairing with unconditioned or other conditioned reinforcers. Its value comes from an individual’s learning history, not biology.

Paired with reinforcer

Neutral Stimulus

Conditioned reinforcer

now strengthens behavior

praise

sticker

Conditioned reinforcers gain power through pairing & learning history, not biology.

Generalized CONDITIONED REINFORCERs

Reinforcement

(GCRs)

A conditioned reinforcer that has been paired with many different reinforcers, so it stays effective across conditions and is not tied to one single EO.

Praise is powerful because it’s paired with many reinforcers → effective across conditions.

Social Praise / Attention

Reinforcement

Contingency-Shaped Behavior

Behavior acquired and maintained by direct contact with reinforcement; shaped by immediate consequences, not by verbal rules.

• Learned through direct contact with reinforcement.

• Immediate consequences (seconds) matter.
• Trial-and-error learning.
• Arbitrary behaviors may be reinforced

Example:

• Child touches hot stove → gets burned → stops touching stove.

• Contrast with rule-governed = Dad says don't touch it's hot and child doesn't touch

Rule-Governed Behavior

Reinforcement

Behavior controlled by verbal statements of contingencies rather than direct contact with immediate consequences.

Example:

• A child doesn’t touch the stove because their parent said, “Don’t touch, it’s hot.”

• Contrast with contingency shaped behavior: child touches stove and learns its hot

Immediacy of Reinforcement

Reinforcement

Immediacy refers to how quickly a reinforcer is delivered after a target behavior. The more immediate, the stronger the effect on future behavior.

Immediate reinforcement

Delayed reinforcement

Automaticity of Reinforcement

Reinforcement

(Principle)

Reinforcement works automatically: behavior is strengthened by its consequences whether or not the person is aware of, understands, or verbalizes the contingency.

• Pigeon pecks a key → food appears → pecking increases.

• The pigeon doesn’t need to “know why” for reinforcement to work.

• Reinforcement changes future behavior even if the individual can’t describe the contingency.

• Applies to all operant principles: (reinforcement, punishment, extinction)

Differential reinforcement

Reinforcement

Reinforce one response class while withholding reinforcement (extinction) for another. Core ABA procedure for increasing desired behaviors & decreasing problem behaviors.

Reinforce what you want → withhold from what you don’t.

Differential Reinforcement of Alternative Behavior

Reinforcement

(DRA)

Reinforce a desirable alternative behavior while withholding reinforcement for the problem behavior.

Strengthens the alternative & weakens the problem.

Examples

• Reinforce raising a hand instead of calling out.

• Reinforce asking for a break instead of eloping.

shift Rx here

DRA = Reinforce the alternative , not the problem behavior

Differential Reinforcement of Other Behavior

Reinforcement

(DRO)

Reinforce the absence of the problem behavior for a set time.

Alex throws pencils. Ms. Bloom sets a 20-minute DRO. If Alex doesn’t throw pencils for 20 minutes Ms. Bloom provides reinforcement. If he throws a pencil at 10 minutes → no reinforcement and timer is reset.

Does not tell the student what to do instead.. just reinforcing the absence of the problem behavior.

Differential Reinforcement of Incompatible Behavior

Reinforcement

(DRI)

Reinforce a behavior that cannot happen at the same time as the problem behavior.

Unique Feature: Can’t happen at the same time

Steps:

• Reinforce the incompatible behavior.

• Withhold reinforcement for the problem behavior.

Can’t do both at the same time!

Differential Reinforcement of Lower Rates

Reinforcement

(DRL)

Reinforce a behavior only if it happens at a lower rate than before.

Steps:

• Reinforce if behavior occurs less often than a set limit OR with longer pauses between responses.

• Withhold reinforcement if the behavior happens too quickly or too often.

Not eliminate.. just reduce

Differential Reinforcement of High Rates

Reinforcement

(DRH)

Reinforce a behavior only when it happens at or above a set high rate.

Steps:

• Reinforce if the behavior occurs often enough (meets/exceeds the criterion)

• Withhold reinforcement if the behavior is too slow or too few

Reinforce when behavior happens at a higher rate than before.

Unconditioned reinforcers are biologically built-in (we don’t need to learn to want them), but they are still dynamic. Their effectiveness always depends on motivating operations (MOs):

• Deprivation (EO): makes the reinforcer more powerful.
• Food-deprived → food becomes highly reinforcing.

• Satiation (AO): makes the reinforcer less effective.
• After a big meal → food loses reinforcing value.

Positive reinforcement is used in classrooms, therapy, and workplaces to shape behavior.

• Teacher praises student for raising hand → hand-raising increases.

• Therapist gives token for sitting quietly → sitting quietly increases.

Automatic Positive (adds):

• Humming because the sound feels good.
• Turning on a lamp → light appears.

Automatic Negative (removes):

• Scratching an itch → itch stops.
• Pulling curtain → glare decreases.

Automatic vs Socially mediated:

• Automatic: Behavior → consequence occurs directly (no person delivers it).

• Socially Mediated: Behavior → another person delivers the consequence.

• Example: Scratch itch (automatic) vs. asking parent for lotion (social).

Key Points:

• Direct control: Reinforcement must follow the behavior within seconds for maximum effect.

• Delays weaken reinforcement: Even 1–2 seconds can reduce effectiveness.

• Accidental reinforcement: The behavior occurring immediately before the reinforcer (not always the target) will be strengthened.

• Temporal contiguity: Behavior and consequence must be closely linked in time.

• Operant response class: Immediacy strengthens the class of responses that precede the reinforcer.

Not all conditioned reinforcers are generalized.

• Sticker paired with praise = conditioned, but not generalized.

• Money/tokens/praise = generalized.

• Generalized does not mean “just really strong.” It means broad pairings.

• Primary Reinforcers: naturally reinforcing (food, water, warmth).

• Secondary Reinforcers: learned (tokens, money, praise).

• Reinforcement is individualized → what works for one person may not work for another.

• Exam Tip: Look for the function (does the behavior increase?), not whether you would find the reinforcer motivating.

Key Points:

• No awareness needed: Reinforcement changes future behavior even if the individual can’t describe why.

• Operates across species: Works for humans, animals, infants.

• Different from automatic reinforcement

Not always “sensory behaviors only.”

• Opening a jar to access food, flipping a light switch for light = still automatic.

• People can be present and it’s still automatic if they didn’t deliver the consequence.

• Don’t assume all stereotypy is automatic... verify function.

People nearby doesn’t mean social.. it’s only social if they deliver the consequence.

• Prompts are not reinforcement: Prompts happen before behavior; reinforcement is the consequence after.

GCRs are powerful because:

• Effective even when one EO isn’t in place (money works even if you’re not hungry).

• Versatile → can reinforce a wide range of behaviors.

• Don’t lose value as quickly as simple conditioned reinforcers.

• Called generalized because they work across EO conditions, not because they reinforce more behaviors.

Watch Out For These!

• Don’t confuse automaticity (principle) with automatic reinforcement (type).

• Automaticity = Principle (no awareness required)

• Automatic Reinforcement = Type of reinforcement (behavior produces its own reinforcer, no social mediation)

Negative” means something is removed after a behavior. If the behavior increases, it’s actually reinforcement.

• In this example, starting homework makes the nagging stop (removal of an aversive), so doing homework next time is more likely.

Don’t confuse with rule-governed behavior.

• Contingency-shaped = immediate, direct reinforcement.

• Rule-governed = controlled by verbal rules, even with long delays.

Positive vs. Negative Reinforcers

• Positive reinforcer (SR+) = a stimulus whose presentation after a response makes that response class more likely.
• Ex: praise, tokens, iPad time, snacks.

• Negative reinforcer (SR–) = an aversive stimulus whose termination/reduction after a response makes that response class more likely.
• Ex: loud noise, bright glare, itch, pressure, the relief when these end is the reinforcer.

Both are reinforcers → both increase behavior. “Positive/Negative” = added vs. removed, not good vs. bad.

Socially mediated = another person delivers the reinforcer.

• Works as SR+ (adds item/attention) or SR– (another person removes/reduces an aversive).

• Contrast with: Automatic = the behavior itself produces the consequence (no one else involved).

Neutral stimulus gains power through stimulus-stimulus pairing.

• Conditioned reinforcers require learning history → they start as neutral and only gain value through pairing.

• Can be automatic (behavior reinforcing itself, like babbling) or socially mediated (delivered by another person, like stickers)

• If not re-paired occasionally with other reinforcers, they may lose effectiveness (extinction)

Key Points:

• Guided by rules (verbal “if–then” statements).

• Delayed consequences alone do not reinforce behavior, rules allow behavior to persist across long delays.

• Rules bridge the gap when consequences are too delayed or weak to directly influence behavior.

Automatic = no social mediation. The behavior itself produces the reinforcer.

Social Positive (SR+):

• Raises hand → teacher attention added.
• Turning on a lamp → light appears.
• "Push me!” → adult pushes → swing moves

Social Negative (SR–):

• Says “too loud” → adult turns music down.
• Hands over break card → teacher pauses work.
• Asking for water → teacher delivers water

Don’t confuse with contingency-shaped behavior.

• Contingency-shaped = immediate, direct reinforcement.

• Rule-governed = controlled by verbal rules, even with long delays.

• Socially Mediated Negative Rx: Someone else removes the aversive (parent stops nagging).

• Automatic Negative Rx: The behavior itself removes the aversive (press snooze → alarm stops).

Examples:

• Seatbelt beeper: Buckle up → beeping stops (automatic, escape).

• Bright sun: Pull curtain → glare goes away (automatic, escape).

• Put on jacket before recess: Avoid being cold (avoidance, automatic).

Key Points:

• Direct Experience → Learner discovers what works through trial-and-error.

• Immediacy → Reinforcement must occur within seconds; otherwise other behaviors may get reinforced.

• Automaticity → Behavior strengthens automatically, without the person needing to “understand” the connection.

• Arbitrary Selection → Any behavior occurring right before reinforcement can be strengthened, even if illogical.

Watch Out For These!

• Intermittent is not the same as delayed reinforcement.. CRF or INT still require immediate delivery of the reinforcer.

• Negative reinforcement requires immediacy too.. relief must follow right after the response.

• The behavior closest in time is what’s reinforced.

• Reinforcer strengthens the behavior that occurs immediately before it... Even 1–2s can matter.

• Don’t confuse delayed reinforcement with thinning schedules—two different concepts.

Reinforcers:

Reinforcers can be delivered automatically (produced by the behavior itself) or socially mediated (delivered by a person).

don't confuse with Punishment

• Negative Reinforcement → behavior increases because an aversive is removed.

• Punishment → behavior decreases (by adding an aversive or removing a preferred item).

Positive does not mean good

• In ABA, “positive” means something is added after the behavior.

• Always think added vs. removed.

How it works:

• Increases reinforced responses

• Decreases unreinforced responses

• Produces response differentiation (new response class emerges)

• Extinction may cause variability, which can lead to closer approximations

Key to Success:

• Select potent reinforcers

• Deliver them immediately & consistently

• Place problem behavior on true extinction

• Consider motivating operations (MOs) and individualization

Applications

• Shaping: build new behaviors step by step

• Reducing problem behavior: least intrusive intervention (basis for DRA, DRO, DRI, DRL, DRH)

• Promoting variability: reinforcing different responses (lag schedules)

How it works:

• Choose an alternative behavior that already exists, is easy to do, and can contact natural reinforcement

• Reinforce it immediately & consistently

• Withhold reinforcement for problem behavior completely

Two Types:

• Interval DRO: No problem behavior throughout the interval → reinforce.

• Momentary DRO: No problem behavior at the end of the interval → reinforce.

Types of DRL:

• Spaced-Responding DRL: Reinforce only when responses are spaced out by a minimum time (increase IRT).

• Example: Reinforce taking bites every 10+ sec instead of eating too fast (every 5 seconds)

• Full-Session DRL: Reinforce at the end of a session if the behavior stayed below a set number.

• Example: less than 5 requests for help in 20 min.

• Interval DRL: Break session into chunks; reinforce if behavior stays under the set limit each interval.

When to use:

• Behaviors okay in moderation (asking for help, raising hand, snacking, talking).

• Not for severe/dangerous behaviors (aggression, SIB).

Types of DRH:

• Full-Session DRH: Reinforcement is delivered at the end of the entire session if the total number of responses meets or exceeds a predetermined criterion.

• Example: A student must complete at least 15 math problems in 30 minutes to earn reinforcement.

• Interval DRH: The session is divided into smaller intervals, and reinforcement is delivered at the end of each interval if the number of responses during that interval meets or exceeds the criterion.

• Example: During a 30-min class, the student must complete at least 5 problems every 10 minutes to earn reinforcement.

• Spaced-Responding DRH: Reinforcement is delivered only if each response occurs within a short interresponse time (IRT), meaning the learner must respond quickly and repeatedly.