Behavioural Economics

Introduction

Behavioural economics studies how psychological, social, cognitive, and contextual factors shape economic decisions. It keeps the analytical discipline of economics while relaxing some benchmark assumptions about perfectly consistent preferences, unlimited attention, flawless probability judgement, and complete self-control. The central question is not whether people are simply "rational" or "irrational". It is when, why, and by how much observed behaviour departs from a useful benchmark, and whether a better model can explain and predict those departures.

For university students, the field is especially valuable because it connects microeconomics, psychology, decision theory, experimental economics, public policy, finance, marketing, health, and organisational behaviour. You will learn to compare standard and behavioural models, interpret evidence, design simple experiments, and evaluate interventions without assuming that one behavioural finding applies everywhere.

Daniel Kahneman's work with Amos Tversky helped establish the psychological study of judgement and choice as a major influence on economics. Kahneman received the 2002 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel for integrating insights from psychological research into economic science, especially in judgement and decision-making under uncertainty. Richard H. Thaler received the 2017 prize for contributions to behavioural economics. Herbert A. Simon's earlier work on bounded rationality also challenged the idea that decision-makers always optimise with unlimited information and computational capacity.

This CrashCourse overview introduces risk, perception, nudges, and the ultimatum game. Use it as an orientation, then test its claims against the more formal models and evidence in the course.

Learning Goals

By the end of this aiMOOC, you should be able to explain how behavioural economics modifies benchmark economic models; distinguish heuristics, biases, preferences, and constraints; apply prospect theory to risky choice; analyse present bias and self-control problems; explain social preferences and behavioural game theory; evaluate nudges and other forms of choice architecture; design a basic behavioural experiment; interpret causal evidence critically; and discuss ethical questions raised by behavioural public policy.

Foundations of Behavioural Economics

Benchmark Models and Their Role

Behavioural economics usually starts from a benchmark rather than rejecting standard economics. A benchmark gives you a clear prediction to compare with observed behaviour.

In standard consumer theory, a decision-maker is often modelled as having stable preferences and choosing the most preferred affordable option. Under uncertainty, expected utility theory models choice as the probability-weighted average of utilities over possible outcomes. In intertemporal choice, exponential discounting represents a stable rate of trade-off between consumption at different dates. In many elementary game-theoretic models, players maximise their own material payoff.

These assumptions can be powerful even when they are not literally true in every setting. Behavioural economics asks whether alternative assumptions improve explanation, prediction, welfare analysis, or policy design. The most useful comparison is therefore not "standard economics versus reality" but "one model versus another, evaluated with evidence".

Benchmark idea	Behavioural question	Example alternative
Stable preferences	Do preferences depend on reference points, framing, or context?	Reference-dependent preferences
Correct use of probabilities	Do people systematically overweight, underweight, or neglect some probabilities?	Probability weighting and heuristics
Consistent intertemporal choice	Do immediate rewards receive extra weight?	Present bias
Pure self-interest	Do fairness, reciprocity, identity, or norms affect choices?	Social preferences
Full attention and optimisation	Do limited attention and cognitive costs change decisions?	Bounded rationality

Bounded Rationality

Herbert Simon argued that real decision-makers face limits of information, time, and computation. Instead of solving every problem globally, people may use rules, routines, and aspiration levels. The idea of satisficing describes selecting an option that meets an acceptable threshold rather than searching indefinitely for the theoretical optimum.

Bounded rationality is not the same as stupidity. A shortcut can be adaptive when information is costly and time is scarce. The scientific task is to identify the environment in which a rule performs well and the environment in which it creates systematic error.

Methods and Evidence

Behavioural economists use multiple methods. Laboratory experiments allow tight control over incentives and information. Field experiments test behaviour in natural settings. Randomised controlled trials can estimate causal effects when treatment assignment is genuinely random and implementation is sound. Natural experiments, administrative data, surveys, qualitative research, structural models, and observational studies can answer questions that experiments cannot.

Good evidence requires more than a statistically significant result. You should ask about effect size, uncertainty, sample selection, measurement, multiple testing, preregistration, replication, attrition, treatment compliance, and external validity. Behavioural effects can vary across institutions, cultures, experience levels, stakes, and populations. A famous effect is a starting point for inquiry, not a universal law.

Judgement Under Uncertainty

Heuristics

A heuristic is a simplifying strategy or rule of thumb. Heuristics can save effort, but they can also create predictable judgement errors.

Availability refers to judging frequency or probability partly by how easily examples come to mind. Vivid or recent events may therefore receive disproportionate attention. Representativeness refers to judging probability by similarity to a prototype, which can lead people to neglect base rates or sample size. Anchoring occurs when an initial number or reference value influences later estimates, even when the anchor should receive little weight.

These concepts describe tendencies, not automatic failures. Expertise, feedback, incentives, and task structure can strengthen or weaken them.

Framing and Description Dependence

A framing effect occurs when logically equivalent descriptions lead to different choices. For example, a medical outcome described in terms of survival can produce different responses from the same outcome described in terms of mortality. The economic challenge is that a purely description-invariant preference model predicts the same choice when the underlying outcomes and probabilities are unchanged.

When you study framing, distinguish between a genuine change in preferences and a change in beliefs, attention, comprehension, or inferred meaning. Real communication often contains pragmatic cues, so not every response to wording is evidence of a deep preference reversal.

Overconfidence and Biased Beliefs

Overconfidence can appear as overestimation of ability, excessive precision in confidence intervals, or overplacement relative to others. Confirmation bias describes a tendency to search for, interpret, or remember information in ways that support an existing belief. These mechanisms can affect investment, entrepreneurship, forecasting, hiring, and organisational decisions.

A behavioural diagnosis should specify the mechanism. If an investor trades too often, possible explanations include overconfidence, entertainment value, heterogeneous information, incentives, or mistaken beliefs. Naming a bias without ruling out alternatives is not a complete explanation.

Prospect Theory and Reference Dependence

Why Prospect Theory Matters

Prospect theory, developed by Daniel Kahneman and Amos Tversky, models risky choice around a reference point rather than only around final wealth. Outcomes are evaluated as gains or losses relative to that reference point. The theory was designed to explain patterns that expected utility theory did not capture well in many experimental choices.

The basic intuition has three important parts. First, value is reference-dependent. Second, sensitivity to additional gains or losses often diminishes as outcomes move farther from the reference point. Third, losses can receive greater psychological weight than gains of comparable size, a pattern known as loss aversion. Prospect theory also uses decision weights that can differ from objective probabilities.

The graph illustrates a typical prospect-theory value function. The vertical axis represents subjective value and the horizontal axis represents gains and losses around a reference point. The exact shape and parameter values are empirical questions; they should not be treated as universal constants.

This MIT OpenCourseWare lecture develops reference-dependent preferences and connects them to evidence from housing, finance, and labour-market settings.

A Worked Choice Example

Imagine two gain-framed options. Option A gives you a certain gain of £500. Option B gives you a 50 percent chance of £1,100 and a 50 percent chance of nothing. The expected monetary value of B is £550, but a risk-averse person may still choose A.

Now imagine losses relative to a reference point. Option C gives a certain loss of £500. Option D gives a 50 percent chance of losing £1,100 and a 50 percent chance of losing nothing. Some people become more willing to gamble in the loss domain. This pattern is called the reflection effect. It does not imply that everyone always seeks risk in losses or avoids risk in gains; observed choices depend on probability levels, stakes, reference points, and context.

Probability Weighting

In prospect theory, decision weights need not equal objective probabilities. People may give relatively high decision weight to some small probabilities and relatively low decision weight to some moderate or large probabilities. This helps explain why the same person may buy insurance against a rare loss and also buy a lottery ticket for a rare gain.

Do not confuse probability weighting with simply "getting probabilities wrong". A formal model separates objective probability from the weight a probability receives in choice. Cumulative prospect theory refines the original model and handles risky prospects in a way that preserves useful ordering properties.

Reference Points and the Endowment Effect

Reference points can arise from the status quo, expectations, goals, recent outcomes, or social comparison. The endowment effect describes cases in which ownership is associated with a higher valuation of an object than non-ownership. Reference dependence and loss aversion can provide one explanation, but market experience, transaction costs, strategic behaviour, and experimental design can also matter.

The lesson is methodological: when a behavioural explanation is plausible, compare it with competing mechanisms and ask what additional observation would discriminate among them.

Intertemporal Choice and Self-Control

Exponential Discounting and Present Bias

Economic choices often involve trade-offs between sooner and later outcomes: saving, borrowing, studying, exercising, energy conservation, or preventive health. Standard exponential discounting represents a constant proportional discount between adjacent future periods. Under that benchmark, preference orderings should remain dynamically consistent when all options move closer in time by the same amount.

Present bias gives extra weight to immediate utility. A person may prefer £110 in 31 days over £100 in 30 days, yet prefer £100 today over £110 tomorrow. The preference reversal can be represented by quasi-hyperbolic discounting, in which future utility receives both a long-run discount factor and an additional present-bias factor.

The figure illustrates how subjective value can decline with delay. Different functional forms imply different patterns of impatience and dynamic consistency.

This MIT OpenCourseWare lecture introduces exponential discounting, quasi-hyperbolic discounting, and the distinction between sophisticated and naive decision-makers.

Self-Control and Commitment Devices

A self-control problem arises when your current plan conflicts with the preference of your future self. A sophisticated present-biased person anticipates future temptation and may demand a commitment device. A naive person underestimates future self-control problems.

Examples include automatic saving, deadlines, website blockers, voluntary deposit contracts, and removing tempting options from the immediate environment. Commitment can increase welfare when it helps people implement their own long-term goals, but forced commitment can also be harmful when preferences or circumstances change. Welfare analysis therefore requires evidence about what people actually want, not merely evidence that behaviour changed.

Mental Accounting, Attention, and Context

Mental Accounting

Mental accounting describes the tendency to organise money and decisions into separate mental categories rather than integrating every choice into total lifetime wealth. You might treat a tax refund as "bonus money", refuse to use savings while carrying expensive credit-card debt, or maintain separate budgets for food, travel, and entertainment.

Mental accounts can create inconsistencies, but they can also serve useful self-control and budgeting functions. Behavioural analysis should therefore examine both the costs and the benefits of a rule.

Sunk Costs and Narrow Bracketing

The sunk-cost principle in standard economics says that irrecoverable past costs should not affect a forward-looking decision. Yet people sometimes continue a project, subscription, or investment because they have already spent money or effort. This may reflect emotional attachment, reputation concerns, learning, signalling, or a desire not to "waste" the past investment.

Narrow bracketing means evaluating decisions separately rather than as part of a wider portfolio. A person can be highly risk-averse over a single small gamble while accepting a diversified set of risks over time. How choices are grouped can therefore affect observed preferences.

Limited Attention and Salience

Not all departures from benchmark choice require unusual preferences. People may simply fail to notice a tax, deadline, fee, future consequence, or low-probability event. Salience and limited attention can change what information enters the decision process.

This distinction matters for policy. If a problem is caused by missing information, better disclosure may help. If information is available but difficult to process, simplification may help. If incentives are misaligned, a nudge may be too weak and a price, rule, or institutional reform may be more appropriate.

Social Preferences and Behavioural Game Theory

Fairness, Reciprocity, and Cooperation

Standard game theory often uses material self-interest as a benchmark, but experiments show that choices can also reflect fairness, reciprocity, altruism, spite, identity, and social norms. These motives are grouped under social preferences.

In the ultimatum game, one player proposes how to divide a sum of money and the second player accepts or rejects. If the offer is rejected, both receive nothing. A purely money-maximising responder with no other concerns should accept any positive amount, while a proposer who knows this should offer the smallest feasible positive amount. In many experiments, responders reject some low offers and proposers often offer more than the minimum. The pattern varies with stakes, anonymity, institutions, and culture, so the game should be interpreted as evidence about context-dependent social preferences rather than a single universal fairness rule.

Other Experimental Games

The dictator game removes the responder's veto and can reveal willingness to give, although demand effects and beliefs about the experiment can influence behaviour. Public-goods games study contribution to a shared resource and the problem of free-riding. Trust games examine trust and reciprocity by allowing one player to transfer resources that are multiplied before a second player decides whether to return anything.

Behavioural game theory does not replace strategic reasoning. Instead, it enriches payoff functions, beliefs, learning models, and assumptions about how players reason.

Choice Architecture, Nudges, and Public Policy

What Is a Nudge?

A nudge changes the choice environment in a way that predictably influences behaviour while preserving options and, in the classic definition, without significantly changing economic incentives. Examples include defaults, reminders, simplification, ordering, salience, and carefully designed social-norm information.

A default is what happens if a person does nothing. Defaults can be powerful because of inertia, procrastination, attention costs, implied recommendations, or switching costs. Automatic enrolment in a pension plan, for example, can increase participation without eliminating the option to opt out.

This staircase uses immediate information about energy expenditure as a salience intervention. It is an example of choice architecture: information is placed directly at the point where a choice is made.

In this University of Chicago video, Richard Thaler gives an overview of nudges and explains why small changes in choice architecture can influence decisions.

This MIT OpenCourseWare lecture examines defaults, nudges, and frames in a university course on psychology and economics.

Nudges, Boosts, Incentives, and Regulation

A nudge is only one policy instrument. A boost aims to strengthen people's decision-making capabilities, for example through statistical literacy or decision aids. Economic incentives change the costs or benefits of an action. Regulation can restrict or require behaviour. Information policy changes what people know, while institutional reform can change market structure or administrative processes.

Choose the instrument that matches the diagnosed problem. A reminder may help with forgetfulness but not with unaffordable prices. Financial education may improve knowledge but not remove conflicts of interest. A default may increase participation but can also anchor people on an unsuitable contribution rate. Policy design should compare alternatives rather than assume that a behavioural tool is always cheaper, softer, or better.

Ethics and Welfare

Behavioural policy raises ethical questions because choice architects have objectives, values, and biases of their own. You should evaluate at least five dimensions: autonomy, transparency, welfare, fairness, and accountability. Ask whether an intervention is easy to resist, whether its purpose is understandable, whose welfare standard is being used, whether burdens and benefits are distributed fairly, and how errors will be detected and corrected.

"Sludge" is a useful contrasting idea: excessive friction, paperwork, waiting, or complexity can make beneficial actions unnecessarily difficult. Reducing sludge can improve access, but simplification should not remove safeguards that protect privacy, informed consent, or due process.

Applications

Consumer and Financial Decisions

Behavioural economics can explain why consumers respond to reference prices, partitioned prices, default options, free trials, payment timing, or complex product menus. In finance, behavioural models examine overtrading, disposition effects, attention, belief extrapolation, and household saving.

An ethical application distinguishes between helping consumers and exploiting predictable mistakes. A commercial interface that makes cancellation deliberately difficult may use behavioural knowledge against a user's interests. A transparent reminder before a subscription renews uses similar knowledge to improve informed choice.

Health and Education

Health decisions involve present bias, uncertainty, identity, habit, social norms, and limited attention. Behaviourally informed interventions can include appointment reminders, simplified forms, commitment mechanisms, or salient feedback. In education, timely prompts, planning tools, peer information, and redesigned administrative processes may affect enrolment, attendance, and completion.

Effects are usually context-specific. A successful intervention in one institution should be tested rather than copied blindly to another.

Work, Organisations, and Management

Organisations are environments of incentives, norms, routines, and limited attention. Behavioural insights can inform performance feedback, meeting design, hiring processes, safety, ethical decision-making, and teamwork. However, employee behaviour may reflect structural problems rather than individual bias. If workloads are impossible, a reminder is not a substitute for adequate staffing.

Development and Public Administration

Behavioural approaches have been applied to savings, tax compliance, public-service take-up, health, agriculture, energy, and development. The World Bank's 2015 report Mind, Society, and Behavior emphasised that people think automatically, socially, and through mental models. The OECD has also developed guidance for applying behavioural science ethically in public policy.

A strong behavioural policy process begins with diagnosis, tests an intervention against a credible comparison, measures intended and unintended outcomes, checks heterogeneous effects, and plans for replication or scaling.

Interpreting Behavioural Evidence Critically

Identification and Causality

Suppose students who receive weekly reminders submit more assignments. This correlation alone does not prove that reminders caused the improvement. Students who opt into reminders may already be more motivated. Random assignment can help identify a causal effect, but even a randomised experiment can fail through attrition, contamination, non-compliance, poor measurement, or selective reporting.

You should distinguish internal validity from external validity. Internal validity asks whether the study identifies the effect in the study setting. External validity asks whether the result is likely to generalise to other people, places, times, and implementation systems.

Effect Size and Practical Significance

A tiny effect can be statistically significant in a very large sample, while an important effect can be estimated imprecisely in a small sample. Report uncertainty, not just whether a threshold was crossed. For policy, compare the effect with implementation costs, opportunity costs, distributional consequences, and alternative interventions.

Behavioural effects should also be assessed over time. A reminder may have a short-term impact that fades. A default may create persistent behaviour. A training programme may initially slow performance but improve long-term competence. Time horizon is part of the causal question.

Replication, Heterogeneity, and Theory

Replication is not mechanical repetition. A useful replication tests whether an effect survives changes in sample, context, operationalisation, and analytic choices. Heterogeneous treatment effects can be substantively important: an intervention may help one group, do nothing for another, and harm a third.

Theory helps organise these patterns. If a default works because of procrastination, reducing switching costs should weaken the effect. If it works because people interpret the default as expert advice, changing the perceived expertise of the choice architect should matter. Predictions like these make behavioural explanations testable.

Mini Case Study: Designing a Savings Intervention

Imagine that a university wants to help employees increase emergency savings. A behavioural team first maps the decision process. It discovers that employees intend to save but must complete a long form, choose among many account options, and remember to act after payday.

A diagnosis based on present bias and friction suggests several treatments: a simplified form, a default contribution with easy opt-out, a payday reminder, or a commitment option. A rigorous design could randomly assign eligible employees to different versions and compare account opening, contribution rates, opt-outs, financial stress, and persistence.

The welfare question cannot be answered by participation alone. Employees with expensive debt may benefit more from debt repayment than from additional saving. A responsible evaluation therefore measures outcomes that matter to participants, offers transparent choices, protects personal data, and checks whether effects differ by income or financial situation.

Key Analytical Questions

When you encounter a behavioural claim, ask: What is the benchmark model? What is the behavioural mechanism? What evidence distinguishes that mechanism from alternatives? How large and robust is the effect? For whom does it occur? What happens over time? Does the intervention change beliefs, preferences, attention, incentives, or constraints? What are the welfare and ethical implications? What would falsify the explanation?

These questions prevent behavioural economics from becoming a catalogue of named biases. At university level, the goal is to build and test explanations.

Interactive Tasks

Quiz: Test Your Knowledge

What is the central purpose of a benchmark model in behavioural economics? (To provide a clear prediction against which observed behaviour can be compared) (!To prove that all people are perfectly rational) (!To replace empirical evidence with assumptions) (!To show that psychology is irrelevant to economics)

Which statement best describes bounded rationality? (Decision-making is constrained by information time and computational capacity) (!People always choose the option with the lowest monetary value) (!People never learn from experience) (!Economic choices are determined only by emotion)

What is a defining feature of prospect theory? (Outcomes are evaluated relative to a reference point) (!All risky choices depend only on final wealth) (!People always prefer risky options) (!Probabilities never influence choice)

What does loss aversion mean in its standard behavioural interpretation? (Losses can receive greater weight than comparable gains) (!Every loss is valued exactly like every gain) (!People refuse all gambles involving losses) (!Losses are ignored when probabilities are small)

What is present bias? (Extra weight placed on outcomes that are immediate) (!A preference for outcomes that occurred in the past) (!Perfect consistency across all future dates) (!A belief that future prices must fall)

What does mental accounting describe? (Organising money and decisions into separate mental categories) (!Calculating every decision from total lifetime wealth) (!Ignoring all budgets) (!Using only formal accounting standards)

What does the ultimatum game help researchers study? (Fairness reciprocity and strategic responses to proposed divisions) (!Only production costs in competitive firms) (!How central banks set interest rates) (!How consumers calculate compound interest)

Which intervention is most clearly a default? (Automatic pension enrolment with an easy option to opt out) (!A legal ban on joining a pension scheme) (!A large tax penalty for not saving) (!A compulsory contribution with no exit option)

What does internal validity primarily concern? (Whether a study identifies the causal effect in the study setting) (!Whether every country will show the same effect) (!Whether the intervention is politically popular) (!Whether the sample is very large)

Which is the strongest way to interpret a famous behavioural effect? (As a testable pattern whose size and generality depend on evidence and context) (!As a universal law that applies equally to everyone) (!As proof that standard economics has no useful models) (!As evidence that incentives never matter)

Memory Game

Reference point	Benchmark relative to which an outcome is experienced as a gain or loss
Anchoring	Influence of an initial value on a later judgement
Present bias	Extra weight placed on immediate outcomes
Mental accounting	Separation of money and decisions into distinct psychological budgets
Default	Outcome that occurs when no active choice is made
Reciprocity	Tendency to respond to kind or unkind actions in kind
Salience	Prominence that draws attention to particular information
Commitment device	Arrangement that helps a person follow a prior long-term plan

Drag and Drop

Match the correct terms.	Topic
Reference dependence	Evaluating outcomes as gains or losses around a benchmark
Probability weighting	Giving choice weights to probabilities that differ from objective probabilities
Present bias	Giving extra importance to immediate utility
Social preferences	Caring about fairness reciprocity or other people's outcomes
Choice architecture	Structuring the environment in which decisions are made

Match each behavioural concept to the explanation that best captures its mechanism. Then create one new real-world example for each pair and discuss what alternative explanation could fit the same behaviour.

Crossword Puzzle

Prospect	Which word completes the name of the theory that evaluates risky outcomes around a reference point?
Anchoring	Which effect occurs when an initial value influences a later estimate?
Default	What is the preselected outcome when no active choice is made?
Discounting	What process reduces the present subjective value of delayed outcomes?
Reciprocity	What social motive involves responding to another person's action in kind?
Heuristic	What is a simplifying rule of thumb used in judgement?

LearningApps

Cloze Text

Behavioural economics compares observed decisions with a useful

rather than assuming that every deviation is irrational. Prospect theory evaluates outcomes relative to a

. The tendency for comparable losses to carry greater weight than gains is called

. An initial value that influences a later estimate can act as an

. Extra weight on immediate outcomes is known as

. Organising money into separate psychological budgets is called

. In the ultimatum game a responder can reject an offer because

may enter preferences. The outcome that applies when no active choice is made is a

. A nudge changes the

while preserving options. A randomised experiment can strengthen claims about

. Evidence that applies outside the original study setting has stronger

. Ethical behavioural policy should consider autonomy transparency welfare fairness and

.

Open-Ended Tasks

Easy

Choice Diary: Keep a three-day decision diary, identify four choices that may involve attention, framing, defaults, or present bias, and explain one competing non-behavioural explanation for each.
Framing Poster: Design a one-page visual that presents the same statistical outcome in two logically equivalent frames and explain why the wording might influence attention or emotion.
Mini Survey: Create a two-version survey with one framing manipulation, collect a small convenience sample, compare response proportions, and state clearly why the result cannot establish a universal effect.
Explainer Video: Produce a two-minute video that teaches reference points, loss aversion, and one real-world application in clear language without describing people as simply irrational.

Standard

Behavioural Interview: Interview two people about a repeated decision such as saving, studying, commuting, or shopping, code the barriers they describe, and distinguish preference-based explanations from attention or constraint explanations.
Choice Architecture Audit: Analyse the sign-up or cancellation process of a real service, map defaults and friction points, and propose a transparent redesign that improves informed choice.
Ultimatum Experiment: Run a small classroom ultimatum-game activity with voluntary participation, record anonymous offers and responses, compare the results with a money-maximising benchmark, and discuss ethical and sampling limitations.
Field Observation: Visit a supermarket, cafeteria, library, transport hub, or university service point and document how ordering, salience, labels, or defaults structure choices; produce an annotated photo essay without recording identifiable people.

Advanced

Preregistered Behavioural Experiment: Design a simple experiment on anchoring, framing, or present bias, state hypotheses and analysis rules before collecting data, and explain how randomisation and sample size affect inference.
Behavioural Policy Proposal: Develop a policy intervention for savings, energy, health, education, or public-service take-up, compare a nudge with an incentive and a regulatory option, and include an ethics and distributional-impact assessment.
Replication Critique: Select a published behavioural study, reconstruct its causal claim, identify possible threats to validity, and design a replication that tests whether the mechanism generalises to a different population or context.
Behavioural Data Project: Analyse an open dataset or simulated experimental dataset, estimate an effect with uncertainty, test at least one heterogeneous effect, and write a policy memo that separates evidence from speculation.

Learning Assessment

Model Comparison Assessment: For a risky-choice scenario, derive the prediction of an expected-utility benchmark and a prospect-theory explanation, then identify one new observation that would help distinguish them.
Experimental Design Assessment: Design a randomised test of a reminder or default intervention, specifying treatment, control, outcome measures, ethical safeguards, and the main threat to external validity.
Evidence Interpretation Assessment: Interpret a hypothetical study with a statistically significant but small effect, explaining practical significance, uncertainty, possible heterogeneity, and what additional evidence is needed before scaling.
Mechanism Diagnosis Assessment: Analyse a case of low programme take-up and compare explanations based on preferences, beliefs, attention, administrative friction, and financial constraints before recommending an intervention.
Ethics and Welfare Assessment: Evaluate a proposed behavioural policy from the perspectives of autonomy, transparency, welfare, fairness, and accountability, then recommend whether to adopt, modify, or reject it.
Transfer Assessment: Apply behavioural concepts to a new domain such as cybersecurity, climate action, workplace safety, or digital subscriptions, and defend which behavioural mechanism is most plausible against at least two alternatives.

Evidence of Learning

Evidence of learning should demonstrate more than recall. Strong work shows that you can move between theory, evidence, design, and application.

Dimension	Evidence
Knowledge	Accurate explanation of benchmark models, bounded rationality, heuristics, prospect theory, present bias, mental accounting, social preferences, and choice architecture
Analytical skill	Ability to generate competing explanations, derive qualitative predictions, interpret uncertainty, and distinguish correlation from causation
Research skill	A defensible experiment or observational design with clear variables, ethical safeguards, and awareness of internal and external validity
Product	A data analysis, policy memo, visual explanation, presentation, experiment report, or redesigned choice environment supported by evidence
Transfer	Successful application of behavioural reasoning to an unfamiliar problem without assuming that a named bias automatically explains the behaviour
Reflection	Explicit discussion of limitations, heterogeneity, unintended effects, welfare criteria, and ethical trade-offs

Research and Further Study

For deeper study, compare primary and institutional sources rather than relying on summaries alone.

Nobel Prize: Daniel Kahneman: Background on the 2002 prize and Kahneman's work on judgement and decision-making under uncertainty.
Nobel Prize: Richard H. Thaler: Background on the 2017 prize for contributions to behavioural economics.
MIT OpenCourseWare: Psychology and Economics: University-level lectures, slides, and problem sets on behavioural economics.
OECD: Ethical Behavioural Science in Public Policy: Guidance on responsible use of behavioural science in government.
World Development Report 2015: Mind, Society, and Behavior: Applications of behavioural and social insights to development policy.

OERs on the Topic

Linked Learning Areas

Behavioural Economics

Behavioural economics links microeconomic theory with psychology and evidence. The key transferable skill is disciplined diagnosis: define a benchmark, identify a plausible behavioural mechanism, generate alternative explanations, test them with suitable data, and evaluate welfare and ethics before recommending action.