Framing Effect vs. Sunk Cost Fallacy

Human decision-making operates at the intersection of normative rationality and psychological reality. For over two centuries, classical economic doctrine posited the existence of Homo economicus—an idealized agent possessing boundless cognitive capacity, perfect access to market information, and immutable preferences that strictly obey the axioms of expected utility theory. Within this classical paradigm, choices are derived through the dispassionate calculation of marginal costs and marginal benefits, evaluated against an absolute scale of sovereign welfare. Real-world human actors, however, routinely deviate from these mathematical ideals. The pioneering work of behavioral scientists, most notably Amos Tversky and Daniel Kahneman, revealed that human cognition is fundamentally constrained by bounded rationality, relying on computational shortcuts, affective impulses, and contextual cues that produce predictable, systematic departures from normative logic.

Among the catalog of documented cognitive biases, the framing effect and the sunk cost fallacy represent two of the most pervasive, consequential, and analytically fascinating pathologies of judgment. Both phenomena illuminate the mechanisms through which human decision-makers misallocate resources, distort subjective probabilities, and violate standard economic axioms. However, while they frequently intersect within organizational blunders, public policy failures, and personal dilemmas, their psychological mechanics, temporal architectures, and informational triggers are distinct. Conflating the two blurs the diagnostic clarity required to remedy flawed choices.

This treatise provides a definitive, comparative investigation into the framing effect and the sunk cost fallacy. By analyzing their theoretical foundations under Prospect Theory and microeconomic principles, examining their neurological correlates, detailing their operational profiles across diverse sectors, and constructing targeted debiasing protocols, this work equips analysts, executives, and behavioral scholars with the conceptual tools required to identify, decouple, and neutralize these two ubiquitous distortions of human reason.

1. Introduction to Cognitive Biases in Decision-Making

1.1 Conceptual Foundations: Defining Both Phenomena

The framing effect designates a cognitive bias wherein decision-makers reach divergent conclusions when presented with identical objective data, depending solely on the descriptive format, semantic emphasis, or visual architecture of that presentation. Formally, the framing effect represents a systemic violation of the economic axiom of description invariance. Description invariance dictates that an agent’s preferences among alternative prospects must remain stable regardless of the linguistic or symbolic terminology utilized to depict those options. When an individual selects a surgical procedure described as having a “90% one-year survival rate” but rejects the exact same procedure when presented as having a “10% one-year mortality rate,” the decision is not governed by the objective probability distribution of clinical outcomes. Instead, it is governed by the associative emotional valence triggered by the frame itself.

Conversely, the sunk cost fallacy refers to the irrational persistence in an endeavor, strategy, or capital allocation based on the cumulative magnitude of previously expended resources—such as capital, time, physical labor, or emotional effort—which are permanently historical, unrecoverable, and non-refundable. Normative microeconomic theory stipulates the principle of bygones: because historical expenditures cannot be altered by any present or future choice, rational forward-looking decisions must be evaluated strictly upon their anticipated marginal costs and marginal benefits. When an executive allocates an additional $5 million to a failing software project simply because the company has already invested $50 million, the decision maker commits the sunk cost fallacy. The historical capital is treated as active equity to be rescued rather than a closed, irrelevant ledger entry.

Within cognitive psychology, both constructs are situated across the dual-process cognitive continuum delineated by Jonathan Evans, Keith Stanovich, and popularized by Daniel Kahneman. The framing effect operates predominantly within System 1—the autonomous, rapid, associative, and emotionally reactive cognitive engine. Semantic cues, such as “loss” or “gain,” bypass deliberative calculation to trigger immediate visceral heuristics. The sunk cost fallacy, while heavily fueled by System 1 affective dread of waste and ego-protective defense mechanisms, actively enlists System 2—the analytical, deliberate, and effortful cognitive faculty—in an insidious manner. System 2 is co-opted to generate elaborate post-hoc rationalizations, complex financial justifications, and convoluted strategic narratives to justify the continued expenditure of capital on failing endeavors.

1.2 Why These Two Biases Are Frequently Conflated

In executive boardrooms, political analyses, and popular media, the framing effect and the sunk cost fallacy are routinely conflated. This confusion stems largely from their shared theoretical genesis within behavioral economics and their joint reliance on the asymmetric value function of Prospect Theory, formulated by Daniel Kahneman and Amos Tversky in 1979. Both biases demonstrate that human actors do not maximize normative utility across absolute states of wealth. Instead, individuals evaluate alternatives through subjective reference points and show profound loss aversion—the psychological reality that the disutility of a loss is felt approximately twice as intensely as the utility generated by an equivalent gain.

In real-world organizational disasters, the two biases rarely operate in clinical isolation; rather, they present an interconnected empirical profile. When an enterprise software deployment or mega-infrastructure project runs over budget, the historical investment of hundreds of millions of dollars (the sunk cost substrate) is almost invariably presented to board members through manipulative rhetorical frameworks (the framing substrate). A project champion will argue: “If we terminate the initiative now, we are locking in a guaranteed $100 million loss; if we invest an additional $20 million, we preserve the chance to break even.” In such scenarios, the decision-maker experiences a compounding cognitive trap. The sunk capital supplies the emotional pressure to avoid waste, while the strategic narrative reframes abandonment as a certain loss and continuation as a probabilistic gamble.

This dynamic co-occurrence causes analysts to misdiagnose the underlying pathology. A managerial decision to pour resources into an unprofitable enterprise may be categorized as an acute response to negative semantic framing, when in reality it is driven by historical resource entrapment, personal identity, and the inability to close an open mental account. Alternatively, an observer might diagnose a failure as a stubborn devotion to sunk capital, overlooking that the decision-makers were systematically manipulated by a vendor’s selective gain-loss information displays. Disentangling the contextual presentation triggers of framing from the historical resource traps of sunk cost is essential for effective organizational governance and cognitive audit processes.

1.3 Significance in Academic Literature and Applied Strategy

The academic transition from neoclassical equilibrium economics to descriptive behavioral decision science represents one of the most profound paradigm shifts in modern social sciences. The classical model, grounded in the von Neumann-Morgenstern utility axioms and Leonard Savage’s foundations of subjective expected utility, provided mathematically elegant formulations of market efficiency. However, these models proved unable to account for ubiquitous market anomalies, such as asset bubbles, irrational investment persistence, and dramatic preference reversals triggered by simple semantic alterations. The formalization of the framing effect and the sunk cost fallacy provided the empirical scaffolding necessary to explain why competitive markets fail to eliminate cognitive distortions automatically.

In applied strategy and corporate governance, distinguishing between presentation vulnerabilities and historical capital entrapment carries monumental financial implications. Capital allocation constitutes the primary responsibility of executive leadership. When boards of directors fail to recognize framing manipulations, they fall victim to strategic proposals tailored to exploit their loss aversion. They mistake marketing rhetoric for objective risk assessments. When boards fail to insulate their capital budgeting from sunk cost distortions, they facilitate the systemic misallocation of capital, allowing capital expenditure overruns to erode corporate equity and destroy enterprise value.

In public policy and defense procurement, the consequences of these cognitive biases scale to geopolitical dimensions. Public infrastructure projects, such as major transit expansions and specialized aerospace developments, routinely experience cost overruns exceeding 100% of initial projections. Policy architects, constrained by the public visibility of already-spent tax revenues, leverage linguistic framing to convince electorates that discontinuing these initiatives amounts to abandoning national prestige and wasting public sacrifice. The analytical capacity to decouple the present semantic architecture of a policy proposal from its historical resource trail is a baseline prerequisite for the preservation of public wealth and institutional integrity.

2. Comprehensive Overview of the Framing Effect

2.1 Theoretical Mechanism and Prospect Theory

The formal theoretical mechanism of the framing effect is grounded in the foundational tenets of Prospect Theory, articulated by Daniel Kahneman and Amos Tversky (1979). Traditional expected utility theory evaluates choices based on final asset positions, assuming that an individual with $100,000 who faces a $5,000 loss will evaluate the resulting $95,000 position through the same utility curvature as an individual with $90,000 who gains $5,000. Kahneman and Tversky demonstrated empirically that human cognitive apparatuses do not perceive absolute outcomes. Instead, perceptual systems evaluate sensory inputs, cognitive stimuli, and financial returns relative to a subjective reference point, denoted mathematically as an adaptation level or neutral baseline.

The mathematical representation of Prospect Theory’s value function, $v(x)$, possesses three structural characteristics that explain framing susceptibility:

• Reference Point Dependence: The value function is defined on gains and losses relative to an explicit or implicit baseline ($x = 0$), rather than absolute states of cumulative wealth. The presentation format dictates where this reference point is situated within the decision-maker’s mind.
• Asymmetric Curvature: The function is concave in the domain of gains ($v”(x) < 0$ for $x > 0$), which generates risk-averse behavior for favorable prospects. Conversely, the function is convex in the domain of losses ($v”(x) > 0$ for $x < 0$), which mathematically dictates risk-seeking behavior for unfavorable prospects.
• Loss Aversion: The function is significantly steeper for losses than for gains, characterized by the empirical loss aversion parameter $lambda$, where $\lambda \approx 2.0 \text{ to } 2.5$. The psychological sting of losing $1,000 is more than twice as intense as the psychological pleasure of acquiring $1,000 ($v(-x) < -lambda v(x)$).

Because the value function changes curvature precisely at the reference point, shifting the location of that reference point through descriptive phrasing alters the decision-maker’s risk posture. When an outcome is framed as a gain relative to an arbitrary anchor, the decision-maker moves onto the concave segment of the curve, experiencing decreasing marginal utility for additional benefits and exhibiting risk aversion to secure the certain positive outcome. When that exact same reality is framed as a loss relative to a different arbitrary anchor, the decision-maker is pushed onto the convex segment of the curve. Here, the decision-maker faces increasing marginal disutility for compounding losses, triggering a psychological willingness to accept high-variance gambles to avoid a certain loss.

2.2 Taxonomy of Framing Paradigms

To avoid treating the framing effect as a monolithic phenomenon, behavioral scholars rely on the classic taxonomy established by Irwin P. Levin, Sandra L. Schneider, and Gary J. Gaeth in their 1998 methodological synthesis. Levin and colleagues demonstrated that different framing applications manipulate distinct cognitive mechanisms, invoke varying cognitive operations, and generate distinct effect sizes. They categorized framing into three primary paradigms:

1. Risky Choice Framing: This classic paradigm involves presenting choices between a riskless, deterministic option and a probabilistic, high-variance gamble, with the outcomes expressed in positive terms (e.g., lives saved, profits secured) versus negative terms (e.g., lives lost, financial costs incurred). In this paradigm, what changes is the risk posture of the agent: positive framing consistently yields risk aversion, while negative framing yields risk-seeking behavior. The underlying mathematical expectation across options remains identical, making this the purest demonstration of description invariance violations.

2. Attribute Framing: Attribute framing represents a simpler cognitive context where only a single characteristic or dimension of an object, product, or event is systematically manipulated. In their landmark experiment, Levin and Gaeth (1988) asked consumers to evaluate ground beef presented as either “75% lean” or “25% fat.” Because the semantic frame highlights either a desirable attribute (lean, pure, healthy) or an undesirable attribute (fat, greasy, unhealthy), the presentation triggers an associative evaluation in System 1. The 75% lean meat is consistently rated as significantly tastier, higher quality, and less greasy than the 25% fat meat, even when subjects consume the identical product in blind sampling trials.

3. Goal Framing: Goal framing focuses on behavioral persuasion, investigating whether individuals are motivated to adopt a specific action by emphasizing the positive consequences of undertaking the behavior (a gain-framed appeal) or the negative consequences of failing to undertake the behavior (a loss-framed appeal). For example, public health initiatives promoting preventative screening (such as mammograms or skin cancer self-examinations) often find that loss frames—emphasizing the mortality risks and undetectable disease progression of not getting screened—are significantly more persuasive than gain frames highlighting peace of mind. The risk configuration here does not present a mathematical gamble; rather, it highlights the cost of inaction.

2.3 Empirical Evidence and Laboratory Demonstrations

The definitive empirical demonstration of risky choice framing remains the Asian Disease Problem, conducted by Amos Tversky and Daniel Kahneman (1981). The researchers presented participants with a hypothetical scenario involving an anticipated outbreak of an unusual Asian disease expected to kill 600 people, offering two alternative therapeutic intervention programs. When the problem was presented with a positive gain frame ($N = 152$), participants chose between:

• Program A: Exactly 200 people will be saved ($72%$ selected this option).
• Program B: A $1/3$ probability that 600 people will be saved, and a $2/3$ probability that no people will be saved ($28%$ selected this option).

When the problem was presented to an equivalent cohort with an explicit negative loss frame ($N = 155$), the identical outcomes were phrased in terms of mortality:

• Program C: Exactly 400 people will die ($22%$ selected this option).
• Program D: A $1/3$ probability that nobody will die, and a $2/3$ probability that 600 people will die ($78%$ selected this option).

Normative decision theory dictates that Program A is mathematically identical to Program C, and Program B is identical to Program D. Yet, the choice proportion inverted from a 72% preference for certainty under the gain frame to a 78% preference for the gamble under the loss frame. This experiment demonstrates the dramatic fragility of human preferences in the face of linguistic manipulation.

Advances in cognitive neuroscience have mapped the physiological underpinnings of these framing shifts. In a functional magnetic resonance imaging (fMRI) study, Benedetto De Martino and colleagues (2006) observed subjects engaging in financial decisions framed as either “keeps” (gains) or “losses.” The neuroimaging data revealed that susceptibility to framing was directly correlated with heightened activation in the bilateral amygdala, a subcortical structure critical for processing visceral emotional responses. When participants chose the sure option in the gain frame or the gamble in the loss frame, their amygdala showed profound metabolic engagement. Conversely, participants who demonstrated resilience against the framing effect—maintaining consistent, frame-invariant preferences—exhibited elevated activation in the orbital and medial prefrontal cortex (OMPFC), an area associated with analytical integration, cognitive control, and the suppression of automated emotional impulses.

Cross-cultural and demographic empirical investigations demonstrate that while the framing effect is a near-universal cognitive artifact across human societies, its magnitude varies systematically with contextual and cognitive variables. Research assessing individuals with high numeracy scores, measured via the Lipkus Numeracy Scale, consistently documents reduced susceptibility to attribute and risky choice framing. Numerically literate individuals spontaneously convert relative percentages and framed assertions into absolute figures, insulating their System 2 deliberative apparatus from semantic manipulations. Similarly, professional training in actuarial science, financial trading, and emergency medicine moderates, but does not eradicate, susceptibility to framing traps.

3. Comprehensive Overview of the Sunk Cost Fallacy

3.1 Theoretical Mechanism and Economic Pathology

The sunk cost fallacy represents a fundamental pathology in dynamic resource allocation, characterized by an inability to honor the classical economic principle of bygones. In microeconomic analysis, cost is defined strictly as opportunity cost: the value of the next best alternative foregone to pursue a given course of action. When capital, time, or labor has been expended and cannot be retrieved through resale, refund, or repurposing, its opportunity cost drops to exactly zero. Consequently, optimal capital allocation requires that forward-looking actors treat historical expenditures as economically dead. Marginal decision rules mandate that an enterprise invest additional capital if, and only if, the prospective marginal revenue exceeds the anticipated marginal cost:

$$\text{If } E[\Delta \text{Revenue}] > E[\Delta \text{Cost}], \text{ invest; otherwise, abandon.}$$

The sunk cost fallacy occurs when an agent introduces the historical capital, $C_{\text{sunk}}$, directly into the prospective marginal calculation:

$$\text{Utility} = f(E[\Delta \text{Revenue}] – E[\Delta \text{Cost}], C_{\text{sunk}})$$

The underlying psychological mechanisms driving this pathology were first systematically articulated by behavioral economist Richard Thaler (1980) through his formulation of mental accounting. Thaler observed that human actors do not integrate their financial realities into a single, fungible balance sheet. Instead, they organize their activities into segregated mental accounts established for specific transactions, projects, or categories of expenditure. Once a mental account is opened through an initial financial commitment, closing that account at a loss creates acute psychological pain.

Terminating a project prior to financial fruition requires the decision-maker to permanently close the mental account and book a psychological write-down. This forces the individual to acknowledge that the resources allocated were wasted. To postpone or avoid the painful realization of that loss, individuals persist in unproductive ventures, holding the mental account open in the hope that an eventual breakthrough will restore the account to psychological equilibrium.

This dynamic is intensified by psychological entrapment, self-justification theory, and Leon Festinger’s cognitive dissonance model. Hal Arkes and Catherine Blumer (1985) demonstrated in their landmark series of experiments that the drive to avoid the appearance of waste is deeply ingrained in human social conditioning. When an individual expends non-refundable resources on an endeavor that reveals itself to be substandard, a severe cognitive dissonance erupts between the individual’s self-concept as an intelligent, competent resource allocator and the objective failure of the project. To alleviate this dissonance without admitting an error in judgment, the individual doubles down, rationalizing that the historical investment was an essential preliminary phase of an ultimate success.

3.2 Escalation of Commitment and the Concorde Effect

In organizational psychology and management scholarship, the organizational manifestation of the sunk cost fallacy is analyzed through Barry M. Staw’s (1976) paradigm of the escalation of commitment. Staw established that when managers experience negative feedback regarding an initial resource commitment, they rarely retreat. Instead, if they were personally responsible for the initial capital allocation, they systematically increase their commitment to the failing trajectory. Staw demonstrated that escalation of commitment is driven by a web of structural incentives, including:

• Internal Self-Justification: The psychological need to convince oneself that the original forecast and analytical logic were fundamentally sound.
• External Impression Management: The imperative to signal unwavering resolve, consistency, and strategic leadership to subordinates, superiors, and market analysts, as corporate cultures frequently penalize perceived equivocation or public admissions of error.
• Institutional Inertia: Organizational routines, dedicated business units, and political coalitions that form around large capital allocations, making strategic pivots administratively and politically expensive.

The quintessential modern historical manifestation of this pathology is the Concorde Effect, an eponym coined by behavioral scientists following the development of the Anglo-French supersonic passenger airliner. In the late 1950s and early 1960s, the governments of the United Kingdom and France entered into an international treaty to design, build, and deploy a commercial supersonic transport fleet. Long before the aircraft entered active service, detailed economic and engineering audits confirmed that the platform was commercially unviable due to astronomical development costs, prohibitive fuel consumption, limited passenger capacity, and noise regulations that barred supersonic travel over populated land masses.

Despite conclusive evidence that the project could never achieve a positive return on investment, the British and French governments continued allocating billions of pounds and francs to the enterprise over a multi-decade development cycle. The primary political justification consistently reverted to the enormous volume of taxpayer funds already expended: political leaders insisted that abandoning the Concorde would render previous public sacrifices utterly meaningless and compromise national technological prestige. Sunk costs were treated as an equity asset that required protection through ongoing subsidies, ultimately resulting in an economically catastrophic operational life that ended with the aircraft’s retirement in 2003.

3.3 Manifestation Profiles Across Human Domains

The sunk cost fallacy manifests across every stratum of human activity, from private life to enterprise software implementations and protracted international conflicts. In personal development, the bias traps individuals in toxic romantic partnerships and dead-end career paths. A professional who has dedicated a decade to securing an advanced credential or climbing a corporate ladder will routinely persist in that discipline long after it ceases to provide intellectual fulfillment or economic utility, rationalizing: “I cannot walk away from this field after putting ten years of my life into it.” In this scenario, the ten years are non-refundable whether the individual stays or departs; the only true decision variable is the utility profile of the decades that remain.

In enterprise governance and information technology, the sunk cost trap appears in multi-year Enterprise Resource Planning (ERP) integrations and core infrastructure migrations. Mega-technology implementations that exceed initial budgets by three to four times their projected capital requirements are regularly sustained because executive boards refuse to cancel projects that have absorbed hundreds of millions of dollars. The initial budget allocation serves as a psychological and political anchor. External systems integrators routinely exploit this vulnerability through the “hostage capital strategy,” continuously assuring steering committees that complete functional utility is just one more phase, release, or budget supplement away.

At the geopolitical level, the sunk cost fallacy generates devastating human consequences in protracted military engagements. Historians and political scientists note that once a nation sustains casualties in a military conflict that encounters strategic deadlock, political rhetoric shifts from strategic objectives to honoring prior sacrifices. The rallying cry becomes: “We must see this mission through so that our fallen soldiers shall not have died in vain.” While honoring military sacrifice is a powerful emotional imperative, integrating accumulated casualties into the forward-looking strategic calculation guarantees additional casualties without altering the underlying geopolitical probabilities of the theater. The sunk cost of human life is weaponized to perpetuate unsustainable military interventions.

4. Key Differences Between Framing Effect and Sunk Cost Fallacy

4.1 Temporal Orientation: Present Architecture vs. Historical Investment

The primary structural divergence between the framing effect and the sunk cost fallacy lies in their respective temporal orientations. The framing effect is strictly synchronic: it operates entirely within the immediate informational architecture of the present moment. It requires no historical commitment, no longitudinal track record, and zero past resource expenditures. An individual encountering a choice between two retirement accounts, two surgical options, or two software packages for the first time can be immediately swayed toward risk aversion or risk seeking based entirely on whether the prospectus describes survival probabilities versus mortality rates, or gross returns versus down-market variances. The bias emerges instantaneously from the contextual presentation of contemporaneous alternatives.

In marked contrast, the sunk cost fallacy is explicitly diachronic and historical. It cannot exist in a vacuum of the present; it demands an accumulated timeline of past resource allocations. An agent cannot commit the sunk cost fallacy regarding a project they encountered five seconds ago; the pathology requires the historical deployment of capital, labor, or emotional energy that is now irrevocably gone. Consequently, while framing manipulates how an individual perceives an identical set of current and future prospective states, the sunk cost fallacy forces the individual to distort prospective realities to justify past actions. The sunk cost fallacy is an investment trap built on historical entanglement, whereas the framing effect is an informational lens applied to current data.

4.2 Cognitive Mechanism: Linguistic Interpretation vs. Mental Accounting

The underlying cognitive architectures of the two biases activate fundamentally distinct psychological operations. The framing effect relies on linguistic and visual interpretation mediated by associative priming. When semantic formulations highlight positive words (“gain,” “profit,” “survival”), they activate System 1 heuristics that assign positive valence to the deterministic option, prompting immediate risk aversion. When negative vocabulary is deployed (“loss,” “cost,” “death”), the brain experiences an aversive threat reaction that induces risk-seeking gambles to avoid that certainty. The framing effect modifies human risk posture without requiring the subject to rationalize prior behavior or calculate internal accounting balances; it alters the perception of reality through the lens of language.

The sunk cost fallacy, however, operates via mental accounting and ego-defense mechanisms. It does not depend on whether a forward-looking proposal is written using positive or negative descriptive vocabulary; it depends on the painful ledger entry required to reconcile an open mental account. The decision-maker feels compelled to avoid the definitive realization of an accounting write-down. The psychological driver is not an automated response to a descriptive cue, but an active internal struggle against the cognitive dissonance of acknowledging that previous decisions were flawed. Framing alters risk preferences instantaneously and passively; the sunk cost fallacy progressively escalates risk tolerance through an active, self-reinforcing psychological commitment.

4.3 Causation, Reversibility, and Diagnostic Measurement

Because their cognitive architectures diverge, the framing effect and the sunk cost fallacy exhibit radically different operational profiles regarding diagnostic identification, psychological reversibility, and experimental measurement. The table below delineates these critical differences across core analytical dimensions:

Analytical Dimension	Framing Effect	Sunk Cost Fallacy
Core Operational Definition	Divergence of choices generated by the linguistic, semantic, or presentation architecture of identical objective facts.	Irrational persistence in an endeavor driven by the magnitude of unrecoverable, non-refundable historical expenditures.
Violated Normative Axiom	Description Invariance (Rational Choice Theory).	The Principle of Bygones / Marginalism (Microeconomics).
Temporal Architecture	Synchronic (Present-focused; operates across immediate informational frames).	Diachronic (Past-dependent; requires historical resource accumulation).
Primary Cognitive Driver	Semantic and associative priming via System 1 heuristics.	Mental accounting, self-justification, and cognitive dissonance mitigation.
Role of Personal Ego	Minimal; driven by automatic neurological responses to presentation.	Extreme; intimately tied to professional reputation and internal self-concept.
Intervention Reversibility	High; easily neutralized by standardizing language or utilizing dual-frame displays.	Low; highly resistant to purely semantic reframing; often requires institutional governance reforms.
Experimental Methodology	Between-subjects randomized trials using alternate phrasing of identical mathematical prospects.	Longitudinal tracking of resource allocation following negative performance feedback.
Organizational Remediation	Tabular data standardization, blind analytical proposals, removal of emotive language.	Zero-based budgeting, executive separation of evaluation from origination, strict stop-loss rules.

As illustrated by the matrix above, the two biases differ profoundly in their reversibility. Framing biases can be neutralized instantaneously by presenting the decision-maker with a balanced, standardized informational display. If an executive rejects an initiative framed around a “20% market failure risk,” providing a concurrent tabular display that explicitly shows both the “80% success probability” and the “20% failure probability” typically eliminates the presentation distortion. The cognitive dissonance is resolved because System 2 is prompted to recognize that both expressions describe the identical state of reality.

The sunk cost fallacy resists such straightforward cognitive adjustments. Presenting a project manager with an objective, neutral statement that their multi-million-dollar program is economically unviable does not eliminate their commitment to the endeavor. The manager’s resistance is not caused by a semantic misunderstanding of probabilities; it is rooted in career vulnerability, organizational identity, and an aversion to closing a mental account at a loss. Consequently, while the framing effect can be debiased via superficial adjustments to the presentation format, neutralizing the sunk cost fallacy requires deep structural governance interventions, such as shifting capital allocation authority away from original project sponsors to independent review committees.

5. Similarities and Intersecting Dynamics

5.1 Rootedness in Loss Aversion and Asymmetric Valuation

Despite their distinct temporal orientations and cognitive mechanics, the framing effect and the sunk cost fallacy share a common theoretical anchor: both are fueled by the fundamental asymmetry of loss aversion within Prospect Theory. Kahneman and Tversky established that human beings experience losses with an intensity approximately double that of equivalent gains. This core insight explains the primary deviation from normative utility in both contexts: human actors will systematically accept irrational, variance-laden gambles if those gambles offer even a marginal probability of avoiding a certain, undeniable loss.

In the framing effect, loss aversion is triggered artificially through linguistic cues. By manipulating whether an outcome is categorized as a “saved asset” or a “lost life,” the architect shifts the reference point, placing the decision-maker on either the concave or convex section of the value function. In the sunk cost fallacy, loss aversion is activated by the prospective closure of a mental account. Abandoning a project forces the manager to book an irrevocable historical loss ($-$10text{ million}$), a prospect that causes acute psychological discomfort. To avoid accepting that certain loss, the manager becomes radically risk-seeking, embracing high-variance gambles—such as deploying an additional $$5text{ million}$ into a failing enterprise with an 80% failure rate—in a desperate bid to reach a breakeven state where no loss is formally registered.

5.2 Cognitive Biases That Co-Occur and Reinforce Both Effects

In dynamic corporate and institutional environments, the framing effect and the sunk cost fallacy are reinforced by a constellation of adjacent cognitive biases, creating an interlocking psychological web that resists rational correctives:

• Confirmation Bias: Once a project has absorbed substantial sunk capital, or once an executive has adopted a strategic course under the influence of a compelling frame, confirmation bias begins filtering all incoming data. Decision-makers selectively search for, interpret, and retain operational evidence that validates their chosen course, while actively discounting or dismissing critical metrics that demonstrate continuing failure.
• Status Quo Bias: Originally formulated by William Samuelson and Richard Zeckhauser (1988), the status quo bias describes the irrational preference for the current state of affairs over alternatives. In sunk cost traps, the status quo is the active continuation of the project. Abandoning the trajectory requires deliberate, difficult intervention, whereas allowing resource allocations to persist via incremental budgetary momentum represents the path of least administrative resistance.
• Optimism Bias and the Planning Fallacy: Decision-makers trapped in sunk cost spirals or influenced by positive gain frames routinely exhibit unrealistic optimism regarding future trajectories. When justifying supplementary capital injections to save historical investments, project champions construct wildly inaccurate forecasts that systematically underestimate prospective costs, overstate future revenues, and compress deployment timelines.

5.3 Compounding Interplay: How Framing Supercharges Sunk Cost Entrapment

The most dangerous manifestation of these phenomena occurs when they compound, where strategic actors leverage the framing effect to preserve, mask, or accelerate sunk cost entrapment. Because human beings are naturally vulnerable to both biases, sophisticated project champions routinely protect failing, sunk-cost-laden initiatives by wrapping them in loss-framed narratives. When an enterprise software deployment or capital asset investment faces corporate review, proponents rarely frame the decision around forward-looking marginal utilities. Instead, they frame the choices around guaranteed losses versus probabilistic salvation:

Frame A (Objective Marginal Reality): “We have spent $50 million, which is gone. We must evaluate whether spending an additional $20 million yields a higher net present value than reallocating that capital to our core logistics network, which guarantees a 12% return.”

Frame B (Manipulative Compounded Frame): “If we stop funding now, we are locking in a catastrophic, guaranteed $50 million loss that completely wastes our previous hard work. If we approve this modest $20 million budget supplement, we preserve the operational opportunity to recover our original investment and bring the system to life.”

By substituting Frame B for Frame A, the project champion supercharges the sunk cost fallacy with the emotional power of loss framing. The board of directors is not simply evaluating historical resource commitments; their System 1 is bombarded with the visceral prospect of finalizing a massive, public loss. The sunk cost provides the raw emotional and financial fuel, while the loss frame shapes that fuel into a vehicle for risk-seeking behavior. As a result, the board routinely approves the supplementary capital, falling victim to a compounded cognitive trap that leads to massive value destruction.

Furthermore, organizations frequently engage in semantic reframing to alter the mental accounting categorization of sunk costs. By shifting past project expenditures out of the “Operational Failure” category and into the “R&D and Organizational Learning” category, leaders reduce the cognitive sting of project abandonment. This dynamic demonstrates how framing can also be deployed constructively: by reframing sunk expenditures as the necessary price of strategic learning, leadership can provide the psychological safety required for teams to terminate dead-end projects without feeling that prior sacrifices were wasted.

6. Practical Contexts and Sectoral Applications

6.1 Financial Markets and Portfolio Allocation

In retail and institutional asset management, the framing effect and the sunk cost fallacy distort capital deployment, generate systematic anomalies in asset pricing, and degrade long-term portfolio yields. The sunk cost fallacy is the primary cognitive engine behind the disposition effect, originally documented by Terrance Odean (1998). The disposition effect describes the well-documented tendency of investors to sell winning equities prematurely to secure a gain, while holding losing positions for extended periods in a desperate attempt to break even.

When an investor purchases a security at $100 and it declines to $60, standard economic logic dictates that the historical purchase price is a sunk cost with zero bearing on future performance. The investor should ask: “Given the current price of $60, is this the highest-yielding asset available in the market?” If the answer is negative, the asset must be liquidated immediately. However, closing the position forces the investor to realize the $40 loss within their mental account. To evade this psychological blow, the investor holds the asset, transforming an acute financial decision into a high-variance gamble that frequently culminates in deeper capital erosion.

Concurrently, the framing effect dictates how investors respond to broader market dynamics and portfolio performance metrics. Institutional asset managers regularly leverage attribute framing when communicating volatility to wealth management clients. A quarterly portfolio drop is framed as a “routine healthy pullback within a secular bull cycle” rather than a “severe capital impairment.” By manipulating baseline reference points—such as calculating returns against a cherry-picked, depressed index bottom rather than the preceding market peak—fund managers frame portfolio performance as a gain relative to an arbitrary low point, maintaining client retention despite objective underperformance.

6.2 Corporate Governance and Product Development

In corporate boardrooms, the framing effect and the sunk cost fallacy directly impact product portfolio management, resource allocation, and mergers and acquisitions. In research and development (R&D) divisions, the product pivot represents the ultimate organizational battleground between rational marginal analysis and cognitive entrapment. When a consumer tech startup or pharmaceutical company expends tens of millions of dollars developing a proprietary architecture that market signals reveal to be obsolete, terminating the initiative is excruciatingly difficult for leadership. The sunk capital in patents, engineering hours, and corporate identity serves as a powerful anchor. Leaders regularly double down on dying product lines, framing the necessary strategic pivot not as an opportunity to secure high-growth markets, but as an intolerable abandonment of company assets.

In mergers and acquisitions (M&A), the two biases frequently drive disastrous corporate takeovers. When an executive team initiates a public acquisition bid, they incur massive sunk costs in legal retainers, investment banking advisory fees, due diligence analyses, and public reputational commitments. When the target firm demands a higher price or unexpected liabilities emerge during due diligence, standard corporate governance mandates walking away. However, executive teams, trapped by both the sunk capital and the public loss frame of an aborted acquisition, engage in escalating bidding wars. They frame walking away as an intolerable loss of strategic market dominance, ultimately overpaying for assets that destroy shareholder value.

6.3 Public Policy, Healthcare, and Geopolitical Strategy

Public policy design, healthcare communications, and statecraft provide high-stakes environments where these cognitive biases carry life-and-death consequences. In public health, the framing effect serves as an indispensable tool for behavioral persuasion. During national vaccination drives and preventative screening campaigns, behavioral scientists carefully deploy gain versus loss frames to optimize public compliance:

• Preventative Health (Gain Framing): Behaviors perceived as low-risk that prevent illness—such as applying sunscreen or receiving a prophylactic vaccine—are most effectively encouraged through gain-framed messaging that emphasizes personal vitality, protection of community elders, and long-term wellness.
• Detection Health (Loss Framing): Behaviors perceived as carrying immediate risk, uncertainty, or psychological discomfort—such as diagnostic mammograms, colonoscopies, or HIV screenings—respond more strongly to loss-framed messaging that highlights the severe risks of late detection, untreated disease progression, and preventable mortality.

In public sector infrastructure and national procurement, the sunk cost fallacy runs rampant, free from the disciplining forces of private capital market balance sheets. Mega-infrastructure programs—such as high-speed rail lines, civic convention centers, and specialized naval defense platforms—routinely exceed original budget baselines by multiples of their initial projections. Public officials, acutely aware that electorates will hold them accountable for “wasting” millions in taxpayer funds, leverage the sunk cost fallacy to secure ongoing funding. They assert that the project is “too far along to cancel,” successfully convincing voters that canceling an economically unviable, half-built transit corridor wastes the spent capital, while allocating additional billions somehow restores that capital to life.

Similarly, environmental policy and climate change mitigation strategies encounter severe framing and sunk cost friction. Fossil-fuel-dependent regional economies resist green transitions because communities have built extensive identities, infrastructure, and capital allocations around traditional extraction models. The sunk capital in refineries, transport infrastructure, and technical skillsets acts as an anchor against modernization. Transition advocates who frame green energy transitions purely as the elimination of traditional industries trigger loss aversion among workers. Conversely, modern climate communications increasingly frame these initiatives through economic opportunity, highlighting job creation, industrial modernization, and energy independence.

7. Debiasing Methodologies and Strategic Interventions

7.1 Countermeasures Against the Framing Effect

Neutralizing the framing effect requires deliberate interventions designed to disengage the automated, associative responses of System 1 and activate the analytical, deliberative capabilities of System 2. Because the framing effect is driven by the semantic architecture of incoming data, debiasing protocols must focus on informational normalization and structural reframing:

1. Dialectical Frame Inversion: Organizations should establish an analytical protocol where no high-stakes proposal is evaluated in isolation under its original phrasing. If an initiative is presented to an investment committee with a gain-framed narrative—such as an “80% probability of retaining a tier-one client accounts portfolio”—the committee is mandated to formally invert the statement into its mathematical complement: a “20% probability of losing the tier-one client accounts portfolio.” By forcing decision-makers to evaluate the identical strategic prospect simultaneously through both gain and loss frames, the emotional valence of each is neutralized, prompting the brain to assess the actual expected value rather than reacting to linguistic packaging.

2. Standardized Tabular Data Architecture: Corporate and institutional reporting systems should be stripped of emotive adjectives, persuasive vocabulary, and selectively highlighted metrics. Dashboards must present data in standardized, frame-neutral tabular formats that display absolute numbers, comparative percentages, and confidence intervals side-by-side. Replacing narrative presentations with standardized data sheets minimizes System 1 associative priming, anchoring decision-makers to objective probabilities rather than curated linguistic frames.

3. Cognitive Nudges and Absolute Risk Displays: Decision-makers are frequently misled by relative risk statistics, which exaggerate marginal effects. For example, describing an intervention as providing a “50% reduction in system crash frequency” sounds monumental, whereas stating that the intervention “reduces the absolute system crash probability from 2 in 10,000 to 1 in 10,000” puts the cost-benefit trade-off into its proper context. Mandating the inclusion of absolute risk figures across all risk-management proposals prevents analysts from exploiting the relative-risk framing heuristic.

7.2 Countermeasures Against the Sunk Cost Fallacy

Mitigating the sunk cost fallacy requires deeper structural, governance, and institutional interventions. Because the bias is driven by mental accounting, professional identity, and the pain of realized losses, altering the presentation format alone is rarely sufficient. Instead, organizations must change the structural incentives that govern capital allocation:

1. Institutionalized Zero-Based Budgeting (ZBB): Standard corporate budgeting frameworks invite sunk cost entrapment by using prior-year expenditure as a baseline to which incremental allocations are added. Zero-based budgeting neutralizes this historical momentum by requiring every department and project manager to justify every dollar of operational and capital expenditure from a baseline of zero for every budget cycle. Historical expenditures are systematically barred from the budget document; every prospective allocation must prove its marginal utility against competing corporate uses of capital.

2. Decoupling Capital Evaluators from Project Originators: The psychological entrapment identified by Barry Staw stems largely from personal responsibility: the executive who championed an initiative cannot objectively terminate it without admitting an error in judgment. To eliminate this conflict, enterprise governance should mandate that phase-gate reviews and milestone capital allocations are decided by an independent capital allocation committee whose compensation and career progression are completely detached from the project’s historical trajectory. When the evaluators have no personal reputation tied to past expenditures, the urge to justify historical capital disappears.

3. Pre-Committed Stop-Loss Triggers and Exit Milestones: Before allocating the initial dollar to a new venture, R&D program, or acquisition, the governing body must establish immutable, quantitative, and time-stamped exit criteria. For example, a software venture may be launched with the condition: “If customer acquisition costs do not fall below $50 per unit with 10,000 active users by Month 12, the initiative is terminated, with zero right of appeal.” Pre-committing to programmatic stop-loss triggers while the mental account is fresh and free from emotional distortion protects the organization from subsequent sunk cost rationalization when negative feedback occurs.

7.3 Systemic Decision Architecture for Enterprise Resilience

To insulate enterprise operations from the combined, compounding effects of both the framing effect and the sunk cost fallacy, leadership must construct a comprehensive institutional decision architecture. This systemic approach relies on formalized methodologies that challenge institutional narratives:

1. Institutionalized Red Teaming: Organizations should establish formal “Red Teams”—independent groups of internal analysts or external experts charged with challenging strategic proposals. The Red Team’s explicit mandate is to expose framing distortions, strip away rhetorical narratives, identify hidden sunk cost rationalizations, and calculate the true opportunity cost of proposed capital allocations. Red Teams normalize critical evaluation within corporate cultures, making the identification of cognitive bias a routine institutional process rather than a confrontational personal challenge.

2. The Prospective Pre-Mortem Protocol: Developed by cognitive psychologist Gary Klein, the pre-mortem requires an executive team to imagine that a proposed strategic initiative has failed catastrophically five years in the future. Each participant is instructed to write an independent narrative detailing the precise systemic causes of this hypothetical collapse. This simple exercise circumvents the planning fallacy, defuses ungrounded optimism, and neutralizes promotional gain framing. By projecting the team into a future failure state, it legitimizes the discussion of risks, vulnerabilities, and exit parameters that corporate optimism bias routinely suppresses.

3. Cultural Normalization of Timely Project Termination: The most enlightened corporate governance architectures align compensation structures and cultural rewards with the intelligent termination of failing assets. When a company celebrates a team for swiftly killing an unviable R&D initiative—rewarding the capital saved and highlighting the organizational learning gained—it eliminates the social stigma of project abandonment. When managers recognize that terminating a dead-end project enhances rather than terminates their career, the psychological drive to persist in sunk-cost-fueled endeavors is systematically dismantled.

8. Frequently Asked Questions

8.1 Can the Sunk Cost Fallacy Be Considered a Special Case of Framing?

While theoretical arguments can be made that the sunk cost fallacy operates through an internal, psychological “frame” regarding historical investments, behavioral decision science maintains a clear boundary between the two constructs. The framing effect strictly involves description invariance violations: an observer is presented with an identical set of contemporaneous outcomes that generate different choices based purely on how those outcomes are described. In contrast, the sunk cost fallacy represents a failure of marginal analysis: an actor integrates irrevocable, non-refundable historical expenditures into forward-looking decisions.

Framing is an external, presentation-level phenomenon driven by language, visual cues, and the positioning of arbitrary baselines. The sunk cost fallacy is an internal, dynamic trap driven by mental accounting, self-justification, and the psychological pain of booking a realized loss. While a sunk cost dilemma can be exacerbated by presenting choices using loss-framed narratives, it remains a distinct, past-dependent pathology. Treating the sunk cost fallacy simply as a subset of the framing effect obscures its unique temporal foundations and structural solutions.

8.2 How Does Loss Aversion Function Differently in Each Bias?

Although loss aversion serves as the psychological engine for both biases, it operates through different mechanisms and temporal horizons across the two phenomena:

• In the Framing Effect: Loss aversion is activated immediately by semantic or visual cues that place an outcome into the loss domain relative to a subjective baseline. When an option is labeled with a negative attribute (“mortality,” “cost,” “downside risk”), it triggers an aversive emotional reaction in the amygdala, driving the decision-maker to accept high-variance gambles to avoid that perceived loss. The loss is contextual, descriptive, and hypothetical.
• In the Sunk Cost Fallacy: Loss aversion is activated by the prospective realization of a real, historical loss. The decision-maker has already expended physical capital, time, or reputation. Abandoning the project requires the formal closure of that mental account and the realization of a write-down. To delay or avoid that psychological pain, the actor becomes risk-seeking, continuing to commit resources to a failing endeavor in a desperate effort to reach a breakeven state where no loss is formally booked.

8.3 Which Bias Is More Difficult to Mitigate in Organizational Contexts?

Within enterprise and public policy governance, the sunk cost fallacy is universally more challenging to mitigate than the framing effect. The framing effect can be neutralized through administrative data standardization: presenting options in neutral tables, displaying complementary gain-loss percentages, and standardizing report formats. These interventions are straightforward to implement because framing distortions rarely involve the decision-maker’s personal identity or professional track record.

The sunk cost fallacy, by contrast, is deeply tied to personal reputation, executive ego, and institutional power dynamics. Terminating a multi-million-dollar program that an executive championed requires a public admission of error, which can threaten career progression, shareholder confidence, and department headcounts. Neutralizing the sunk cost fallacy requires deep governance overhauls: removing project champions from capital allocation committees, implementing zero-based budgeting, and enforcing strict, pre-committed exit milestones. The sunk cost fallacy is not merely a cognitive miscalculation; it is an organizational and political challenge that demands institutional discipline to overcome.

8.4 Do Highly Analytical or Quantitatively Trained Individuals Avoid Both Biases?

Empirical research indicates that advanced quantitative training, mathematical literacy, and high cognitive reflection provide substantial protection against framing effects, but offer surprisingly little protection against the sunk cost fallacy. Individuals who score highly on cognitive reflection tests (such as Shane Frederick’s Cognitive Reflection Test) or possess professional training in statistics, economics, and actuarial analysis routinely demonstrate resistance to attribute and risky choice framing. Their analytical faculties instinctively convert framed percentages into absolute numbers, maintaining description invariance across alternate presentations.

However, when confronted with sunk cost dilemmas, highly analytical individuals remain vulnerable. While they understand the theoretical principle of bygones, their cognitive abilities are often co-opted to generate sophisticated post-hoc rationalizations for their commitment. Highly intelligent managers leverage financial models, scenario projections, and market forecasts to justify why an unprofitable capital investment is on the verge of a breakthrough. Because the sunk cost fallacy is driven by emotional self-justification, intellectual sophistication simply provides more advanced cognitive tools to defend flawed choices. True resilience requires systemic governance structures rather than individual intelligence alone.

9. Conclusion and Strategic Synthesis

9.1 Summary of Comparative Dimensions

The framing effect and the sunk cost fallacy illustrate the vulnerabilities of human cognition when navigating complex decision environments. While both biases reveal the descriptive accuracy of Prospect Theory over the classical axioms of expected utility, their foundational mechanisms are fundamentally distinct. The framing effect demonstrates that human rationality is vulnerable to the informational lens through which choices are observed. By manipulating semantic labels, baseline references, and positive or negative framing, choice architects can nudge decision-makers toward risk-averse or risk-seeking postures without altering the underlying objective probabilities.

The sunk cost fallacy reveals that human decision-making is equally vulnerable to an historical trap. It demonstrates the psychological difficulty of divorcing forward-looking marginal choices from irrevocable past expenditures. Trapped by mental accounts, the pain of realized losses, and the need for self-justification, decision-makers persist in failing, resource-intensive endeavors, converting bad situations into catastrophic failures. Framing is an operational vulnerability of the present moment; sunk cost entrapment is an ongoing vulnerability to historical commitments.

9.2 Actionable Takeaways for Rational Strategic Choice

To protect enterprise capital, optimize public policy design, and elevate personal judgment, decision-makers should internalize this rigorous diagnostic checklist when evaluating strategic proposals:

• Perform the “Clean Slate” Interrogation: When evaluating an ongoing, resource-intensive project, isolate the historical investment entirely. Ask: “If we arrived at this organization today with zero historical involvement, would we invest these specific resources into this project, given its current state and future probabilities?” If the answer is negative, the project must be terminated or restructured immediately, regardless of the historical capital expended.
• Enforce Inverted Frame Reviews: Never approve a high-stakes proposal presented in a single, curated frame. Mandate that every positive gain narrative be paired with its mathematically equivalent negative loss narrative. Neutralize the semantic packaging to force the analytical faculties of System 2 to evaluate the objective expected utility of the endeavor.
• Decouple Capital Review from Strategic Origination: Institutionalize governance frameworks that remove the authority to continue funding long-term initiatives from the managers who originally approved them. Empower independent review committees to conduct ruthless, marginal-utility audits that evaluate only future costs and expected returns.
• Pre-Commit to Concrete Exit Thresholds: Never launch an experimental project, capital venture, or portfolio allocation without establishing immutable, time-stamped, and quantitatively verified stop-loss criteria. When the criteria are breached, execute the exit protocol automatically, bypassing subsequent emotional rationalization.
• Shift Organizational Narratives on Project Closure: Redefine the institutional meaning of project termination. Transform project cancellations from personal failures into strategic decisions that preserve capital and yield valuable organizational insights. By dismantling the social penalty for ending dead-end initiatives, organizations liberate their teams from the sunk cost trap.

By implementing these structural guardrails and mastering the distinction between presentation manipulations and historical resource traps, leaders and organizations can insulate their decision architectures from these ubiquitous cognitive distortions, ensuring that capital, time, and human talent remain dedicated strictly to the most promising forward-looking opportunities.

References

• Arkes, H. R., & Ayton, P. (1999). The sunk cost and Concorde effects: Are humans less rational than lower animals? Psychological Bulletin, 125(5), 591–600. https://doi.org/10.1037/0033-2909.125.5.591
• Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140. https://doi.org/10.1016/0749-5978(85)90049-4
• De Martino, B., Kumaran, D., Seymour, B., & Dolan, R. J. (2006). Frames, biases, and rational decision-making in the human brain. Science, 313(5787), 684–687. https://doi.org/10.1126/science.1128356
• Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford University Press.
• Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. https://doi.org/10.2307/1914185
• Levin, I. P., & Gaeth, G. J. (1988). How consumers are affected by the framing of attribute information before and after consuming the product. Journal of Consumer Research, 15(3), 374–378. https://doi.org/10.1086/209174
• Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames are not created equal: A typological analysis of framing effects. Organizational Behavior and Human Decision Processes, 76(2), 149–188. https://doi.org/10.1006/obhd.1998.2804
• Odean, T. (1998). Are investors reluctant to realize their losses? The Journal of Finance, 53(5), 1775–1798. https://doi.org/10.1111/0022-1082.00072
• Samuelson, W., & Zeckhauser, R. (1988). Status quo bias in decision making. Journal of Risk and Uncertainty, 1(1), 7–59. https://doi.org/10.1007/BF00055564
• Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance, 16(1), 27–44. https://doi.org/10.1016/0030-5073(76)90005-2
• Staw, B. M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577–587. https://doi.org/10.5465/amr.1981.4285694
• Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior & Organization, 1(1), 39–60. https://doi.org/10.1016/0167-2681(80)90051-7
• Thaler, R. H. (1999). Mental accounting matters. Journal of Behavioral Decision Making, 12(3), 183–206. https://doi.org/10.1002/(SICI)1099-0771(199909)12:3<183::AID-BDM318>3.0.CO;2-F
• Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 453–458. https://doi.org/10.1126/science.7455683
• Tversky, A., & Kahneman, D. (1984). Choices, values, and frames. American Psychologist, 39(4), 341–350. https://doi.org/10.1037/0003-066X.39.4.341