1. Prospect Theory
Kahneman & Tversky (1979): people evaluate outcomes relative to a reference point, not absolute values. Value function: concave for gains (risk aversion), convex for losses (risk seeking), steeper for losses (loss aversion, λ ≈ 2.25). Probability weighting: overweight small probabilities (lottery effect), underweight large probabilities. Framing effects: "90% survival rate" vs. "10% mortality rate" elicit different choices despite identical information. Endowment effect: ownership increases valuation (WTA > WTP by factor of ~2). Mental accounting: money is not fungible — different "accounts" for different categories. Status quo bias: preference for current state, inertia in decision-making.
2. Cognitive Biases
Anchoring: initial information disproportionately influences judgment (even random anchors). Availability heuristic: judging probability by ease of recall — overestimate vivid/recent events. Representativeness: ignoring base rates, conjunction fallacy (Linda problem). Confirmation bias: seeking evidence supporting existing beliefs. Overconfidence: 90% confidence intervals are correct only 50% of the time. Sunk cost fallacy: continuing losing investments due to prior commitment. Hindsight bias: "I knew it all along" distorts memory. Present bias: hyperbolic discounting — prefer $100 today over $110 tomorrow, but $110 in 31 days over $100 in 30 days. Dunning-Kruger effect: low-competence individuals overestimate their ability.
3. Nudge Theory
Thaler & Sunstein (2008): choice architecture can guide behavior while preserving freedom. Libertarian paternalism: steer toward beneficial choices without mandates. Default effects: opt-out organ donation increases rates from ~15% (opt-in) to ~85%. Automatic enrollment: 401(k) pension participation rises from 49% to 86%. Save More Tomorrow: commit to future saving increases tied to pay raises. Simplification: reducing form complexity increases completion rates. Social norms: "73% of your neighbors recycle" increases recycling. Calorie labeling: 8% reduction in calories purchased. Choice overload: too many options reduces decisions — jam study (24 vs. 6 options). Commitment devices: pre-committing to goals improves follow-through.
4. Experimental Methods
Lab experiments: controlled environment, random assignment, incentive-compatible mechanisms (real monetary stakes). Field experiments: natural settings with randomization — Randomized Controlled Trials (RCTs). Natural experiments: policy changes create quasi-random variation. Behavioral game theory: ultimatum game (responders reject "unfair" offers ≥20% of the time), dictator game, public goods game. Neuroeconomics: fMRI, EEG during decision-making — ventromedial prefrontal cortex for value, anterior insula for risk/loss. Eye tracking: reveals attention and information processing. Survey experiments: contingent valuation, stated preference methods. Replication crisis: many behavioral findings have p-values near 0.05, effect sizes shrink upon replication.
5. Policy Applications
UK Behavioural Insights Team (BIT): government "nudge unit" since 2010. Tax compliance: personalized letters with social norms increased payment rates 5–15%. Health: default healthy options in cafeterias, warning labels, sugar taxes. Finance: auto-enrollment in pensions (UK: 10.6 million new savers), simplified financial disclosures. Environment: home energy reports (Opower: 2% reduction, scaled to 100M+ households), default green energy tariffs. Education: text message reminders reduce summer melt by 7%, simplified financial aid forms. COVID-19: behavioral interventions for compliance — clear messaging, social norms, reduced friction for testing/vaccination. Critique: effect sizes often small (1–5%), ethical concerns about manipulation, paternalism vs. autonomy debate.
Try it live
Everything above runs in your browser — open Prospect Theory: Loss Aversion & Probability Weighting and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Prospect Theory: Loss Aversion & Probability Weighting simulation