How Do Consequences Shape Action?

Reinforcement, choice, punishment, and the evidence for goal-directed actions and habits

Level: Intermediate
Reading load: 9 core readings in 5 units
Prerequisites: How Does Experience Change Behavior? is useful preparation, especially its distinction between learning and performance. Comfort comparing experimental conditions and reading simple graphs. The predictive-learning syllabus provides useful context but is not a prerequisite; technical terms are introduced here.
Last reviewed: September 2026

Introduction

An action acquires a history through what follows it. Yet consequences do not exert one uniform influence: an activity may reinforce one response and leave another unchanged, equal rewards may sustain unequal rates of action, and a routine may outlast the value of its original result. To explain these differences, we must specify the relation between conduct and consequence.

This syllabus begins with an argument about how consequences make some actions more likely to recur and a concrete test of what makes an activity reinforcing. It then examines when and how often consequences follow an action, choice among alternatives, and punishment before asking whether behavior depends on knowledge of an outcome and its current value. The final unit connects laboratory definitions of habit with everyday routines and scrutinizes attempts to induce habitual control in humans.

By the end, you should be able to describe a consequence in terms of its measured effect, distinguish how often a reward occurs from whether receiving it depends on the action, explain why alternatives matter, and evaluate evidence for goal-directed or habitual control. The route combines conceptual arguments, small laboratory experiments, and broader reviews. The final unit is longer and can take several sittings; understanding its comparisons matters more than reproducing every analysis.

Unit 1 of 5

What makes a consequence reinforcing?

Start with an explanatory proposal and then ask how to test it. Skinner treats consequences as selecting behavior over time. Premack makes a more specific prediction from baseline activities. Together they raise the question of how an account goes beyond naming a change after it has occurred.

Reading
1 of 9
Selection by Consequences
Skinner, B. F. (1981). Selection by consequences. Science, 213(4507), 501–504. https://doi.org/10.1126/science.7244649
Foundational argument · Moderate · Full work: 4 pages
Distinguish selection across generations, consequences within an individual’s lifetime, and the extension to cultural practices. Ask what evidence each level would require.
Why you’re reading it. Establish the broad claim that consequences help explain the future distribution of behavior. If you met the article in the foundation, now focus on what would make that claim predictively informative.
A question to consider. An activity becomes more frequent after praise is introduced. What additional comparison would support an effect of the praise, and what would still be needed to explain why it was reinforcing?
Reading
2 of 9
Rate Differential Reinforcement in Monkey Manipulation
Premack, D. (1963). Rate differential reinforcement in monkey manipulation. Journal of the Experimental Analysis of Behavior, 6(1), 81–89. https://doi.org/10.1901/jeab.1963.6-81
Methodological contribution · Moderate · Full work: 9 pages
Trace the baseline ranking, the imposed dependency between activities, and the different clarity of results across monkeys.
Why you’re reading it. Turn the broad selection argument into a prospective prediction about which consequence should reinforce which response.
A question to consider. If activity B is more probable than C but less probable than A, what does Premack predict when access to B follows C versus A? What would a test of both arrangements add to observing just one?

Unit 2 of 5

How do schedules and alternatives shape responding?

The amount of reinforcement does not specify the entire learning environment. Zuriff compares requirements that alternate over time; Herrnstein makes alternatives available together. Keep that design difference visible while asking how response rates and allocation depend on the arranged opportunities.

Reading
3 of 9
A Comparison of Variable-Ratio and Variable-Interval Schedules of Reinforcement
Zuriff, G. E. (1970). A comparison of variable-ratio and variable-interval schedules of reinforcement. Journal of the Experimental Analysis of Behavior, 13(3), 369–374. https://doi.org/10.1901/jeab.1970.13-369
Methodological contribution · Moderate · Full work: 6 pages
Compare response and reinforcement rates in the alternating ratio and interval components. Keep schedule requirements separate from rewards obtained.
Why you’re reading it. Show why equal reinforcement rates need not produce equal behavior, and why an observed rate should not automatically be read as preference.
A question to consider. Suppose the ratio component produces twice the responding at a comparable reward rate. Why would this not yet establish that the animal would choose it when both schedules are available together?
Reading
4 of 9
Relative and Absolute Strength of Response as a Function of Frequency of Reinforcement
Herrnstein, R. J. (1961). Relative and absolute strength of response as a function of frequency of reinforcement. Journal of the Experimental Analysis of Behavior, 4(3), 267–272. https://doi.org/10.1901/jeab.1961.4-267
Landmark empirical study · Methodological contribution · Moderate · Full work: 6 pages
Follow response allocation across two simultaneously available options and examine the role of the changeover delay.
Why you’re reading it. Introduce a quantitative relation between behavior and its alternatives, while keeping its procedural conditions in view.
A question to consider. Compare this experiment with Zuriff’s. Which change in the availability of alternatives prevents you from assuming that their response proportions should obey the same relation?

Unit 3 of 5

What changes when consequences are adverse?

Calling an event unpleasant does not tell us what is learned. An action may cause it, prevent it, or have no effect on it. This unit distinguishes punishment, negative reinforcement, and Pavlovian suppression before considering what a neural finding could explain.

Reading
5 of 9
Behavioral and Neurobiological Mechanisms of Punishment: Implications for Psychiatric Disorders
Jean-Richard-Dit-Bressel, P., Killcross, S., & McNally, G. P. (2018). Behavioral and neurobiological mechanisms of punishment: Implications for psychiatric disorders. Neuropsychopharmacology, 43(8), 1639–1650. https://doi.org/10.1038/s41386-018-0047-3
Major synthesis · Methodological contribution · Challenging · Full work: 12 pages
Use the response–outcome comparisons to interpret the review’s behavioral and neural evidence. Notice when an experiment leaves punishment and fear as competing explanations.
Why you’re reading it. Extend the earlier contingency reasoning to aversive events and avoid treating every decline in responding as punishment learning.
A question to consider. In one task a press causes a noise; in another it prevents the noise; in a third the noise occurs independently. Which response changes would distinguish punishment, negative reinforcement, and cue-related suppression, and what comparison would you include?

Unit 4 of 5

Does the action depend on its outcome?

A response may be reinforced without its later performance requiring a representation of the outcome’s current value. Dickinson develops criteria for goal-directed action; Hammond supplies a complementary manipulation of whether the outcome depends on acting. Changing value and changing contingency test different parts of the explanation.

Reading
6 of 9
Actions and Habits: The Development of Behavioural Autonomy
Dickinson, A. (1985). Actions and habits: The development of behavioural autonomy. Philosophical Transactions of the Royal Society of London. B, Biological Sciences, 308(1135), 67–78. https://doi.org/10.1098/rstb.1985.0010
Foundational argument · Methodological contribution · Moderate · Full work: 12 pages
Trace the logic of devaluing an outcome after training and testing without fresh reinforcement. Compare the evidence needed for outcome knowledge with the evidence needed for sensitivity to its current value.
Why you’re reading it. Introduce operational criteria for goal-directed action and behavioral autonomy. This placement connects the foundation’s distinction to the schedule and contingency questions developed here.
A question to consider. An animal still performs a well-practiced response after its outcome is devalued. What checks would you require before interpreting that persistence as habitual control?
Reading
7 of 9
The Effect of Contingency Upon the Appetitive Conditioning of Free-Operant Behavior
Hammond, L. J. (1980). The effect of contingency upon the appetitive conditioning of free-operant behavior. Journal of the Experimental Analysis of Behavior, 34(3), 297–304. https://doi.org/10.1901/jeab.1980.34-297
Methodological contribution · Landmark empirical study · Moderate · Full work: 8 pages
Reconstruct the response and nonresponse probabilities in the water-reinforcement experiments, especially the move to zero contingency.
Why you’re reading it. Complement outcome devaluation with a manipulation of action–outcome dependence, while distinguishing contingency from close temporal pairing.
A question to consider. Compare making an outcome less valuable with making it equally available without the action. What does each manipulation change, and why could the same decline in responding have different explanations?

Unit 5 of 5

When is repeated behavior a habit?

Daily routines and laboratory habits are often discussed as though they were interchangeable. A broad review connects context, repetition, and goals; the final experiments test whether extended training actually produces the expected change in outcome sensitivity. Keep the theory, the measure, and the result distinct.

Reading
8 of 9
Psychology of Habit
Wood, W., & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology, 67(1), 289–314. https://doi.org/10.1146/annurev-psych-122414-033417
Major synthesis · Challenging · Full work: 26 pages
Compare the review’s account of goal–habit interaction with Dickinson’s experimental criteria. Notice how evidence changes when the measure is a self-report, a daily routine, or an outcome-devaluation test.
Why you’re reading it. Broaden the laboratory framework without assuming that every form of repetition demonstrates the same control process.
A question to consider. A commuter follows a familiar route despite a changed destination. How would you distinguish a context-triggered response from a still-valued goal, incomplete knowledge, or a failure to attend to the new information?
Reading
9 of 9
Shifting the Balance Between Goals and Habits: Five Failures in Experimental Habit Induction
de Wit, S., Kindt, M., Knot, S. L., Verhoeven, A. A. C., Robbins, T. W., Gasull-Camos, J., Evans, M., Mirza, H., & Gillan, C. M. (2018). Shifting the balance between goals and habits: Five failures in experimental habit induction. Journal of Experimental Psychology: General, 147(7), 1043–1065. https://doi.org/10.1037/xge0000402
Replication · Critical reappraisal · Challenging · Full work: 23 pages
Identify the predicted training-by-devaluation contrast in each procedure and examine why the five experiments do not all constitute the same replication.
Why you’re reading it. End by testing the bridge from repetition to habitual control. The optional Pool study extends this question through a preregistered multilaboratory design.
A question to consider. Design one comparison that would distinguish a stronger habit from weaker goal-directed control after extended training. What manipulation check and informative null result would your interpretation need?

Scope and omissions

The focus is instrumental or operant learning: how actions relate to their consequences. Reinforcement increases a response, while punishment decreases it. Positive and negative name the addition or removal of an event; they do not name its moral value. The sequence develops these distinctions through baseline comparisons, reinforcement schedules, choice, aversive learning, outcome devaluation, and contingency degradation.

These are explanatory tools, not a catalogue of reward techniques. Animal studies make carefully controlled relations visible; human habit studies reveal challenges of translation and measurement. Detailed behavioral economics, delay discounting, computational reinforcement-learning algorithms, neural circuitry, and language-mediated rules lie beyond this core. The punishment review introduces biological explanation without making the path a neuroscience course. Social learning, skill acquisition, and the persistence or return of learned behavior warrant further development in their own right. This syllabus does not establish the effectiveness or appropriateness of particular clinical, parenting, or classroom interventions.

Synthesis

Consequences shape action within relations that evidence must make clear: what follows a response, what happens without it, what alternatives are available, and whether the outcome retains its value. Reinforcement, allocation, suppression, and persistence describe different patterns. Their explanation depends on the response being measured and the conditions under which it is tested; no single pattern reveals the whole mechanism.

Consider a person who repeatedly checks a message app. Describe one possible account involving contingent rewards, one involving avoidance of an aversive state, and one involving familiar context cues. For each, specify what would happen if the same outcomes occurred independently of checking, if their value changed, or if the context changed. Propose a comparison that could distinguish two accounts, identify the behavior you would measure, and explain what result would leave the question unresolved. Do not infer habit simply from repetition or reinforcement simply from an event’s appeal.

Return to How Does Experience Change Behavior? to connect this analysis with observation, practice, and persistence. How Do We Learn What Predicts What? develops the cue–outcome relations that can operate alongside action–outcome learning. The optional readings below deepen particular questions without adding another compulsory sequence.

Continue with How Do We Learn From Other People? to examine how observation, demonstrations, instruction, and testimony contribute to learning. It distinguishes what a learner acquires from what they express, and asks when another person provides useful information.