# Behavioral Decision-Making and Cognitive Biases > The descriptive account of judgment and decision-making: how people actually decide, not how they should. Use it to read why a customer, buyer, or stakeholder made a seemingly irrational choice, to et Parent: [Applied Human Psychology](https://llms-explorer.com/tree/applied-human-psychology/) · 17 facets · 70 facts · page: https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/ ## Behavioral Decision-Making & Cognitive Biases - The descriptive account of judgment and decision-making: how people actually decide, not how they should. Use it to read why a customer, buyer, or stakeholder made a seemingly irrational choice, to ethically shape the decision environment, and to catch bias in your own forecasts and recommendations. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#behavioral-decision-making-cognitive-biases) - > Installed as a Claude Code skill with four on-demand reference files (not duplicated here): references/biases-catalog.md (full heuristics-and-biases catalog with canonical experiments + TAM application notes), references/choice-architecture.md (defaults, EAST, MINDSPACE, sludge vs nudge vs boost, choice overload, ethics of influence), references/debiasing-and-application.md (debiasing procedures + worked operator scenarios), and references/replication-status.md (what survived vs what is contested/failed, with citable sources - read before citing any effect externally). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#behavioral-decision-making-cognitive-biases) - Descriptive vs normative: keep them separate. This skill is descriptive (the psychology of real decisions). For the normative side, computing the optimal action under constraints (linear programming, decision trees, EVPI, expected-utility maximization), use da-33-prescriptive-analytics. The gap between the two is the subject matter here: people deviate from the normative optimum in systematic, predictable ways. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#behavioral-decision-making-cognitive-biases) - The honesty rule (read first). Decision/social psychology went through a replication crisis. Several once-famous effects did not survive (power posing, social priming, and ego depletion are contested/failed). The core judgment-and-decision-making findings (anchoring, framing, the disposition effect, present bias, default effects) replicate well; several adjacent social-psych effects do not. Never present a debunked effect as established fact. See references/replication-status.md before citing any effect to a customer or in a written recommendation. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#behavioral-decision-making-cognitive-biases) ## When to reach for this skill - A customer/buyer made a choice that looks irrational -> name the bias, then address the real driver. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - You're setting a price, an opening offer, or a contract renewal -> anchoring and framing. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - A renewal/expansion stalls on "we already invested in X" -> sunk-cost; on "let's keep things as they are" -> status-quo/default bias. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - You're designing a signup, plan-selection, or opt-in/opt-out flow -> choice architecture (and the sludge you should remove). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - You're writing a forecast, capacity plan, or project timeline -> debias your own judgment (overconfidence, planning fallacy). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - Someone cites "power posing" / "priming" / "ego depletion" as fact -> check replication status first. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) - If the task is changing a customer's behavior over time (adoption, habit, enablement), that's behavior-change-psychology, not this skill. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#when-to-reach-for-this-skill) ## 1. Dual-process theory: System 1 / System 2 - Two modes of cognition. System 1 is fast, automatic, associative, affect-laden, and effortless; it produces most snap judgments and most biases. System 2 is slow, deliberate, effortful, and lazy (it endorses System 1 unless prompted). Labels coined by Stanovich & West, popularized by Kahneman (Thinking, Fast and Slow, 2011). Biases are System 1 outputs that System 2 fails to catch. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#1-dual-process-theory-system-1-system-2) - Caveat: treat "two systems" as a useful metaphor, not literal brain architecture. The strict two-box model is contested (better read as a continuum of automaticity). Don't oversell it. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#1-dual-process-theory-system-1-system-2) - Operator use: high-stakes decisions (renewals, escalations, architecture calls) deserve a deliberate System-2 step (a checklist or premortem) precisely because the default is a System-1 gut call. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#1-dual-process-theory-system-1-system-2) ## 2. Heuristics & biases (Tversky & Kahneman, 1974, *Science*) - Mental shortcuts that are "highly economical and usually effective" but produce "systematic and predictable errors." The working set every operator should recognize: anchoring-and-adjustment (the first number dominates; the single most useful effect in negotiation), availability (judging probability by ease of recall; the loud outage feels likelier than the silent risk), representativeness (stereotype/similarity over base rates; the conjunction fallacy), confirmation bias (the engine behind most bad root-cause calls), hindsight bias ("knew it all along"; corrupts postmortems), overconfidence (90%-confident estimates are right far less than 90% of the time; the planning fallacy), status-quo / default bias, and sunk-cost fallacy (honoring unrecoverable past spend instead of deciding on the margin). Detail and operator scripts: references/biases-catalog.md. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#2-heuristics-biases-tversky-kahneman-1974-science) ## 3. Prospect theory (Kahneman & Tversky, 1979, *Econometrica*) - How people choose under risk, the descriptive replacement for expected-utility theory: reference dependence (outcomes judged as gains/losses from a reference point; whoever sets the reference frames the decision), loss aversion (losses loom larger than equivalent gains; classic estimate ~2x, but CONTESTED, do not state "2x" as universal law), diminishing sensitivity (concave for gains, convex for losses), probability weighting (small probabilities overweighted, hence lottery tickets AND insurance; the certainty effect), the fourfold pattern (risk-averse for likely gains and unlikely losses; risk-seeking for unlikely gains and likely losses), and framing effects ("90% uptime" vs "10% downtime" flip the choice). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#3-prospect-theory-kahneman-tversky-1979-econometrica) - > Loss aversion is contested, not debunked. Gal & Rucker (2018), "The Loss of Loss Aversion," argue it is far more context-dependent than the "universal 2x law" implies; gains can loom larger at small magnitudes (Harinck et al., 2007) and predicted pain of loss overstates the actual (Kermer et al., 2006). Real in many settings (especially higher-stakes, endowed goods) but not a context-free constant. Use it as a hypothesis to test for this customer, not a guaranteed lever. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#3-prospect-theory-kahneman-tversky-1979-econometrica) ## 4. Bounded rationality & satisficing (Simon) vs ecological rationality (Gigerenzer) - Bounded rationality (Herbert Simon) - limited information/time/compute, so people satisfice: pick the first option clearing an aspiration threshold rather than optimizing. Buyers rarely run an exhaustive comparison; they stop at "good enough." — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#4-bounded-rationality-satisficing-simon-vs-ecological-rationality-gigerenzer) - Ecological rationality / fast-and-frugal heuristics (Gerd Gigerenzer) - the counterpoint to heuristics-and-biases. Simple heuristics (take-the-best, recognition, 1/N) are adaptive and often more accurate than complex models under scarce/uncertain information (less-is-more). A heuristic is "rational" relative to its environment. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#4-bounded-rationality-satisficing-simon-vs-ecological-rationality-gigerenzer) - Why both matter: one says shortcuts cause errors, the other says shortcuts are often the smart move. The truth is conditional: match the diagnosis to the environment before "fixing" a heuristic. These two programs are the respective foundations of nudging and boosting. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#4-bounded-rationality-satisficing-simon-vs-ecological-rationality-gigerenzer) ## 5. Mental accounting & present bias - Mental accounting (Thaler) - money sorted into non-fungible mental "buckets" (budget categories, "house money," renewal-vs-new-purchase), violating fungibility. A spend framed against the "innovation budget" lands differently than against "BAU/maintenance." — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#5-mental-accounting-present-bias) - Present bias / hyperbolic discounting - near-term costs/rewards discounted far more steeply than distant ones (quasi-hyperbolic beta-delta; Laibson 1997). Produces time-inconsistency: "we'll migrate next quarter" gets reversed when next quarter arrives. Upfront-cost / delayed-benefit work (migrations, upgrades, tech-debt paydown) is chronically under-chosen. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#5-mental-accounting-present-bias) ## 6. Choice architecture & nudges (Thaler & Sunstein, 2008) and boosts (Hertwig) - Choice architecture - every presentation of options (order, defaults, count, framing) influences choice; there is no neutral presentation, so design it deliberately. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - Nudge - alters behavior predictably without forbidding options or changing incentives (libertarian paternalism). The most powerful nudge is the default (opt-out organ donation, 401(k) auto-enrollment, pre-checked tiers). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - EAST (UK Behavioural Insights Team) - make the desired action Easy, Attractive, Social, Timely. The most practical operator checklist. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - MINDSPACE - Messenger, Incentives, Norms, Defaults, Salience, Priming, Affect, Commitments, Ego. (The "Priming" element rests on social-priming research that largely failed to replicate; treat it as the weakest element.) — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - Sludge - friction added against the person's own interest (cancellation mazes, hidden opt-outs). Find and remove sludge in your own onboarding/renewal flows; don't deploy it. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - Boosts (Hertwig & Grüne-Yanoff, 2017) - the contrast to nudges. Instead of steering the chooser, build their competence (teach a decision rule, give a fast-and-frugal tree, present risks as natural frequencies). Boosts preserve agency and persist after the intervention; prefer them for long-term, trust-based relationships, which is most TAM work. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) - Ethics: nudge toward the chooser's own interest, keep it transparent, never sludge. Full applied detail (incl. choice overload): references/choice-architecture.md. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#6-choice-architecture-nudges-thaler-sunstein-2008-and-boosts-hertwig) ## 7. Debiasing - You cannot will a bias away, but structured procedures help: consider-the-opposite (best-evidenced general debiaser, strong against anchoring/overconfidence), premortem (Gary Klein) (imagine the project has failed and explain why; prospective hindsight, cheap and high-yield), reference-class forecasting (Flyvbjerg / Kahneman's "outside view") (estimate from the distribution of comparable past cases; the fix for the planning fallacy), and checklists (force System 2 through a disciplined pass; only work with consistent adherence). Procedures and worked scenarios: references/debiasing-and-application.md. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#7-debiasing) ## Operator quick-map (bias -> tell -> move) - This table is a hypothesis generator, not a verdict. Treat the observed "tell" (what the customer said or did) as data, not a confirmed diagnosis. A tell suggests a candidate effect; confirm it against this specific person/context before acting (several effects are context-dependent). If the tell is ambiguous, gather one more observation or ask a clarifying question first. Never state the bias label to the customer or imply they are irrational; the label is your internal hypothesis, the "move" is what you do. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#operator-quick-map-bias---tell---move) ## Anti-patterns - Calling a customer "irrational." They're predictably boundedly rational. Name the mechanism and design around it. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) - Citing a debunked effect. Power posing, social priming, ego depletion are contested/failed; the "2x loss-aversion constant" is over-stated. Check references/replication-status.md first. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) - Weaponizing nudges (sludge / dark patterns). Steering a customer against their own interest is self-defeating in a TAM relationship. Prefer boosts. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) - Treating System 1/2 as literal neuroanatomy. It's a model. Don't overclaim. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) - One-shot debiasing. Awareness alone barely moves biases; only structured procedures reliably help, and only with disciplined use. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) - Over-applying loss aversion / "fixing" a heuristic that's actually ecologically rational. Diagnose the environment first (Gigerenzer's caution). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#anti-patterns) ## Cross-references - behavior-change-psychology - adjacent and complementary. This skill = the descriptive psychology of a decision (biases, framing, choice architecture). That skill = changing behavior over time (motivation, Fogg B=MAP, stages-of-change, habit loops, adoption). "Design an onboarding nudge to drive adoption" -> that skill; "what default/framing shapes this purchase decision" -> this skill. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#cross-references) - da-33-prescriptive-analytics - the normative counterpart (optimal action under constraints: LP/MILP, decision trees, EVPI, utility theory). Compute the optimum there; understand why humans deviate from it here. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#cross-references) - executive-comms - persuasion and decision-driving communication (board memos, negotiation prep, influence). For the rhetoric/persuasion craft go there; for the underlying decision psychology stay here. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#cross-references) - postmortem-writing - applies hindsight-bias control in incident reviews. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#cross-references) - deep-research-methods - covers confirmation bias / echo chambers as research anti-patterns. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#cross-references) ## Sources - Tversky, A. & Kahneman, D. (1974). "Judgment under Uncertainty: Heuristics and Biases." Science 185(4157), 1124-1131. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Kahneman, D. & Tversky, A. (1979). "Prospect Theory: An Analysis of Decision under Risk." Econometrica 47(2), 263-291. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Tversky, A. & Kahneman, D. (1991). "Loss Aversion in Riskless Choice: A Reference-Dependent Model." QJE 106(4). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Kahneman, D. (2011). Thinking, Fast and Slow. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Simon, H. A. (1955/1956). Bounded rationality and satisficing. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Gigerenzer, G. & ABC Research Group. Fast-and-frugal heuristics / ecological rationality. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Thaler, R. & Sunstein, C. (2008/2021). Nudge (and Nudge: The Final Edition). — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Laibson, D. (1997). "Golden Eggs and Hyperbolic Discounting." QJE. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Hertwig, R. & Grüne-Yanoff, T. (2017). "Nudging and Boosting." Perspectives on Psychological Science 12(6), 973-986. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Dolan, P. et al. (2010). MINDSPACE; Behavioural Insights Team (2014). EAST. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Gal, D. & Rucker, D. (2018). "The Loss of Loss Aversion." Journal of Consumer Psychology. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) - Replication: Open Science Collaboration (2015) Science; Many Labs 2; Ranehill et al. (2015) and Simmons & Simonsohn (2017) on power posing; Hagger et al. (2016) on ego depletion. — [source](https://llms-explorer.com/sources/mdb-context-hub/behavioral-decision-making/#sources) ## Where this helps - A customer, buyer, or stakeholder makes a choice that looks irrational at first glance — naming the specific bias, such as anchoring, sunk cost, or status-quo, turns “they're being unreasonable” into an addressable driver. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Setting a price, an opening offer, or a renewal number, where anchoring and framing effects predictably shape how the other side receives the number. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - A renewal or expansion conversation stalls on “we already invested in X,” a sunk-cost tell, or “let's not change what's working,” a status-quo tell — recognizing the pattern suggests a different move than arguing the merits. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Designing a choice presentation, such as a pricing page, a settings default, or an onboarding flow, where being honest about how order, defaults, and framing influence the decision matters more than pretending the presentation is neutral. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## How to apply this - Before a high-stakes decision — a renewal call, an escalation, an architecture choice — add a deliberate System-2 step such as a checklist or a premortem, rather than trusting the fast, automatic System-1 read. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - When a possible bias appears in a stakeholder's reasoning, treat it as a hypothesis to test against more evidence, not a diagnosis to announce; the bias-to-tell-to-move map is a hypothesis generator, not a verdict. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Use consider-the-opposite as a general-purpose debiasing procedure, especially against anchoring, before finalizing a number or a judgment call. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Keep the descriptive question, why did they actually decide this way, separate from the normative question, what is the mathematically optimal decision — conflating them leads to lecturing about optimality instead of addressing the real driver. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Common mistakes - Labeling a customer or stakeholder irrational instead of naming the specific, predictable mechanism, such as bounded rationality or a named bias, and designing around it. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Citing a contested or since-failed effect as settled science — power posing, social priming, and ego depletion have not replicated, and the commonly cited 2x loss-aversion constant is overstated relative to current evidence. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Using nudge or choice-architecture techniques to steer someone against their own interest, known as sludge or dark patterns, which is self-defeating in any relationship built on trust. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Treating the System 1 / System 2 model as literal brain architecture rather than a useful descriptive metaphor whose strict two-box form is contested. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Limitations - Loss aversion, one of the field's most cited findings, is contested rather than settled — later work argues it is far more context-dependent than the original prospect-theory framing suggested. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - The bias catalog functions as a hypothesis generator: an observed tell, something a customer said or did, is data suggesting a possible bias, not confirmation the bias is actually driving the behavior. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - Ecological rationality is a direct counterpoint to the heuristics-and-biases tradition — the same mental shortcut framed as an error in one lens can be framed as the smart move in the other, depending on the environment. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* - A bias cannot be debiased by simply willing it away; only structured procedures, such as consider-the-opposite or checklists, have real evidence behind them as mitigations. — [source](https://llms-explorer.com/tree/behavioral-decision-making-and-cognitive-biases/) *(AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.)* ## Context files - [Behavioral Decision-Making and Cognitive Biases](https://llms-explorer.com/downloads/sources/mdb-context-hub/behavioral-decision-making.md)