MCAT content library / Psychology & sociology / Operant learning: consequences and response patterns
A reward label does not determine a consequence’s effect. Follow the action, its consequence, and the next opportunity to respond.
About 17 minutes, including the worked examples and checks.
From measurement to learning · Lesson 4 of 5 · Continue in this path
On this page
- Classify the consequence by what happens next
- Keep two independent dimensions separate
- A schedule specifies which response earns the consequence
- Shaping and extinction change the contingency
- Escape ends an event; avoidance prevents it
- Behavior has a biological history too
- Check your reasoning
- Continue learning
- Sources
Classify the consequence by what happens next
A researcher calls a token a reward. That label does not establish its behavioral effect. If delivering the token after a response makes the response more likely under comparable conditions, it functions as a reinforcer. If the response becomes less likely because of the consequence, that consequence functions as punishment. The classification concerns the future behavior, not the researcher’s intention or whether the consequence sounds pleasant.2
AAMC’s learning outline includes these consequence relationships, reinforcement schedules, shaping, escape and avoidance, and biological influences on learning.1 A useful analysis starts with a concrete action: What did the learner do, what changed because of it, and how did later responding change?
Keep two independent dimensions separate
Positive means a stimulus or event is added; negative means it is removed or prevented. Reinforcement increases the relevant behavior; punishment decreases it. Neither positive nor negative is a judgment of moral value.2
| Consequence of the action | Later action increases | Later action decreases |
|---|---|---|
| An event is added | Positive reinforcement | Positive punishment |
| An event is removed or prevented | Negative reinforcement | Negative punishment |
Consider an original game task in which a button press removes a flashing panel. If removal makes later button presses more frequent, it is negative reinforcement. If a different button press removes earned points and thereby becomes less frequent, it is negative punishment. “Something was taken away” is therefore only half of the classification. The direction of the behavior change supplies the other half.
Measure a defined response in comparable opportunities. A total count of twelve presses in two minutes is not necessarily greater responding than ten presses in one minute. Also distinguish the target action from other behavior: removing points might reduce guessing but increase careful checking. Name the action before saying that a consequence “increased behavior.”
What makes a reinforcer effective can be learned
A primary reinforcer can function without a history of pairing with another reinforcer; food can do so for a hungry animal. A conditioned reinforcer acquires its function through learning, as when exchangeable tokens gain value. Motivation and learning history matter to whether an event actually reinforces a response.2 In the game, a token that cannot be exchanged and has no learned significance should not automatically be treated as equivalent to an established token reward.
A schedule specifies which response earns the consequence
A ratio schedule counts responses. An interval schedule makes reinforcement available after time has passed, but the qualifying response must still occur. Fixed schedules hold the response requirement or waiting interval constant; variable schedules change it across reinforcements, commonly around a specified average.3
| Schedule | What qualifies the next response? |
|---|---|
| Fixed ratio (FR) | Completion of a fixed response count |
| Variable ratio (VR) | Completion of a changing response count |
| Fixed interval (FI) | Being the first response after a fixed interval has elapsed |
| Variable interval (VI) | Being the first response after a changing interval has elapsed |
Do not turn an interval schedule into automatic delivery from a timer. With an FI 20-second rule, a response at 19 seconds is too early; a first response at 23 seconds can earn reinforcement. If no response occurs, the elapsed interval alone does not deliver it. The examples here start the interval at time zero and restart it immediately after each reinforcement.
Worked application: the same response record under two rules
Use a hypothetical record with one press every five seconds. Under FR 3, every third press earns a token. Under FI 20 seconds, the first press at or after twenty seconds since the previous token earns one. All figures and records on this page are original teaching examples.
| Press time (s) | FR 3 token? | FI 20 s token? |
|---|---|---|
| 5 | No | No |
| 10 | No | No |
| 15 | Yes | No |
| 20 | No | Yes |
| 25 | No | No |
| 30 | Yes | No |
| 35 | No | No |
| 40 | No | Yes |
| 45 | Yes | No |
| 50 | No | No |
| 55 | No | No |
| 60 | Yes | Yes |
FR 3 delivers four tokens, at 15, 30, 45, and 60 seconds. FI 20 delivers three, at 20, 40, and 60 seconds. Extra presses before the FI interval has elapsed do not advance the clock; extra presses do advance a ratio count. Changing the schedule changes the action–consequence relation even when the token itself remains identical.
This comparison holds the response record fixed to expose the rules. It does not predict that a real learner will maintain the same press times after a schedule change. Typical patterns differ: fixed schedules can produce pauses after reinforcement, whereas variable schedules can sustain steadier responding. Such tendencies are clues; the stated delivery rule identifies the schedule more securely than a curve’s appearance alone.3
Shaping and extinction change the contingency
Shaping builds a response by reinforcing successive approximations to a target.3 For a novel lever action, a researcher might first reinforce approaching the lever, then touching it, and finally depressing it. Advancing the criterion matters: reinforcing only the fully completed action from the start is not this successive-approximation procedure.
In extinction, a previously reinforced response no longer produces that reinforcer, and responding can decline.2 Withholding an earned token differs from delivering a new consequence that suppresses the response. Read exactly what follows the action rather than labeling every reduction “punishment.”
Escape ends an event; avoidance prevents it
In escape learning, the action reduces or terminates an aversive event already present. In avoidance learning, the action prevents or postpones such an event. Both can involve negative reinforcement when the contingency strengthens the action.4 A press that ends a current buzzer is escape. A press during an earlier warning that prevents the buzzer is avoidance. Hearing a warning does not make a response “escape” from the buzzer if the buzzer has not begun.
The warning can also acquire a predictive relationship with the buzzer. That classical relationship and the operant effect of the press answer different questions: what does the cue predict, and what does the action change? Tracking both avoids forcing a whole experiment into one label.
Behavior has a biological history too
Breland and Breland described trained animal actions becoming disrupted by species-typical food-related behavior, which they called instinctive drift. Their observations challenged the assumption that every action is equally easy to establish with the same consequence.5 A changing response is therefore not automatically evidence that the researcher selected the wrong schedule. Examine the consequence rule, response opportunities, motivational conditions, and any supplied biological constraint together.
ACTIVE PRACTICE · COMPARE AND TRANSFER
Apply Reinforcement Schedules to the Same Response Record
The original response record has one lever press every five seconds from t = 5 to 60 s. For fixed ratio 3 (FR3), every third press earns a token; for fixed interval 20 seconds (FI20), the first response at or after 20 seconds since the previous token earns it. Intervals begin at t = 0.
Examine the original evidence
Results are bounded to the original supplied observations. Instructional checks do not establish assessment calibration or clinical effectiveness.
Check your reasoning
Choose an answer before opening its feedback. Each option explains the reasoning behind that choice.
Check 1
In a game, each press of an optional button adds a brief animation. The designer intended the interruption to discourage pressing. With equal session lengths, pressing rises from 12 to 27 times per session when the animation is introduced and returns toward 12 when it is removed. Assuming the change is attributable to that consequence, how did the animation function?
A. Negative punishment
Review this choice. Negative punishment would remove an event after the action and reduce that action. Here an animation is added, and the action increases.
B. Negative reinforcement
Review this choice. The increase supports reinforcement, but the consequence adds an animation. Nothing in the stated contingency is removed or prevented by pressing.
C. Positive punishment
Review this choice. The animation is added, so “positive” fits. Punishment does not: pressing increases rather than decreases. The designer’s intention does not determine the functional label.
D. Positive reinforcement
Correct. Pressing produces an added event, and that consequence increases subsequent pressing under the stated assumption. These are the two features of positive reinforcement.
The principle to carry forward
Identify addition versus removal, then inspect the measured change in the specified action. An intended deterrent can function as a reinforcer.
Check 2
A token task changes from FR 4 to FI 30 seconds. Under the new rule, the first press at or after 30 seconds earns a token, and each token immediately restarts the 30-second interval. The initial interval begins at time zero. For this supplied response record—5, 10, 15, 25, 35, 40, 45, 65, 70, and 90 seconds—at which times are tokens delivered under the new rule?
A. 25 and 65 seconds
Review this choice. These are the fourth and eighth presses, which would satisfy the old FR 4 rule. Under FI 30, the 25-second press is too early.
B. 30 and 60 seconds
Review this choice. No presses occur at these times. An interval becoming eligible does not itself deliver a token; the first qualifying response is still required.
C. 35 and 65 seconds
Correct. The first response after 30 seconds is at 35 seconds. That token restarts the interval, so the next qualifying response is at 65 seconds. The next eligibility time is 95 seconds, beyond the record.
D. 35 and 90 seconds
Review this choice. The first time is correct, but 65 seconds is already 30 seconds after the first token and contains a response. It earns the second token; 90 seconds is then too early for a third.
The principle to carry forward
For an interval rule, track elapsed time from the specified reset and then locate the first eligible response. Counting presses solves a different schedule.
Check 3
In task X, pressing a key terminates an ongoing unpleasant buzz. In task Y, a warning light precedes the scheduled buzz; pressing during the light prevents the buzz but leaves the light unchanged. Key pressing becomes more likely in both tasks because of these consequences. Which interpretation fits?
A. Avoidance in X and escape in Y; both involve negative reinforcement.
Review this choice. The reinforcement classification fits the increase in pressing, but the timing labels are reversed. X ends a present buzz; Y prevents a future buzz.
B. Escape in X and avoidance in Y; both involve negative reinforcement.
Correct. X terminates an aversive event already present. Y prevents its onset. Because these consequences increase pressing, both are negative reinforcement.
C. Escape in X and avoidance in Y; both involve negative punishment.
Review this choice. The timing labels fit, but punishment would decrease the relevant action. Here pressing becomes more likely because it ends or prevents the buzz.
D. Escape in both X and Y; both involve positive reinforcement.
Review this choice. In Y the buzz has not begun, and the light remains unchanged. The stated consequence prevents an event rather than adding one, so neither the common escape label nor positive reinforcement fits both tasks.
The principle to carry forward
Ask whether the aversive event is already occurring, then use the later change in the action to classify its consequence.
Continue learning
From measurement to learning
Lesson 4 of 5.
- Previous in From measurement to learning: Classical conditioning: learn the predictive relationship
- Next in From measurement to learning: Observational learning and reciprocal social influences
- Return to this path’s overview
See every step in From measurement to learning
From measurement to learning
Use behavioral evidence to distinguish adaptation, predictive associations, consequences, and learning from a model.
- Psychological Measurement: What Does the Score Mean?
- Sensory thresholds, adaptation, and habituation
- Classical conditioning: learn the predictive relationship
- Operant learning: consequences and response patterns · You are here
- Observational learning and reciprocal social influences
Related lessons
- Classical conditioning: learn the predictive relationship
- Observational learning and reciprocal social influences
- Sensory thresholds, adaptation, and habituation
- Reading graphs: axes, scales, and the comparison that matters
Browse psychology & sociology · MCAT content library
MCAT preparation and roadmap · Aim High, Score Higher
Sources
1. AAMC: What’s on the MCAT Exam? — 7C Learning
2. OpenStax Psychology 2e: Operant Conditioning
3. University of Minnesota, hosted by Penn State: Learning — Operant Conditioning
4. St. Bonaventure University: Reinforcement and punishment contingencies