Risk seeking

Risk seeking is not a general appetite for danger but a context-dependent preference for uncertainty, most clearly visible when all available options are bad. A person may reject a gamble in the domain of gains yet choose one in the domain of losses: the same decision maker who prefers a sure gain to a probable larger gain may prefer a probable larger loss to a smaller loss that is certain. Prospect theory treats this reversal as a central fact of judgment, because people evaluate outcomes relative to a reference point and experience losses as psychologically different from gains.

Definition and Core Contrast

Risk seeking appears in the classic comparison between a sure loss and a gamble with a higher possible loss. In one problem, people choose between receiving $900 for sure and a 90% chance of receiving $1,000; most are risk averse and prefer the sure gain. In the parallel problem, they choose between losing $900 for sure and a 90% chance of losing $1,000. Most choose the gamble. The sure loss feels exceptionally aversive, while the chance of losing nothing offers an escape, even though the gamble has a negative expected value.

The psychological explanation is not simply that people calculate the monetary expectation incorrectly. The negative value of losing $900 is more than 90% of the negative value of losing $1,000, so the certain loss carries disproportionate emotional weight. Diminishing sensitivity makes the difference between smaller losses feel larger than an equivalent difference between larger losses: the subjective gap between losing $900 and losing $1,000 is smaller than the gap between losing $100 and losing $200, but the position of the sure loss makes it especially painful. Risk seeking therefore emerges when a definite unfavorable outcome is set against a worse outcome that retains a possibility of being avoided.

Reference Points and Prospect Theory

The contrast between Anthony and Betty shows why total wealth cannot explain this behavior. Anthony owns $1 million and must choose between a sure outcome that raises his wealth to $2 million and a gamble offering equal chances of reaching $4 million or gaining nothing. Betty owns $4 million and faces the same final wealth options: the sure outcome leaves her with $2 million, while the gamble offers equal chances of losing three-quarters of her wealth or losing nothing. Anthony sees a gain and is likely to prefer the sure thing; Betty sees a loss and is more likely to gamble. Their possible states of wealth are identical, but their reference points make the psychological outcomes opposite.

This is the weakness prospect theory was designed to address in Bernoulli’s model. States of wealth are not the only carriers of value; changes relative to an earlier state are psychologically decisive. The reference point may be the status quo, an expected result, or an outcome regarded as deserved, such as a raise received by colleagues. Above it, an outcome is experienced as a gain; below it, as a loss. The value function is S-shaped and changes slope abruptly at the reference point: sensitivity diminishes within both domains, but losses have a steeper psychological impact than corresponding gains.

The fourfold pattern extends the same logic across probability and outcome domains. For high-probability gains, people commonly prefer certainty; for high-probability losses, they commonly seek risk. The loss-side pattern is strengthened by probability weighting: a 90% probability may receive a decision weight of only about 71, making the unfavorable gamble feel less threatening than its stated probability suggests. In the domain of gains, the certainty effect makes the sure outcome especially attractive; in the domain of losses, it makes the certain loss especially intolerable.

From Individual Choices to Desperate Gambles

Risk seeking becomes most consequential when people face very bad alternatives. Individuals may accept a high probability of making their situation worse in exchange for a small possibility of avoiding a large loss. The desire for complete relief can overpower the sensible decision to cut losses, turning a manageable failure into a disaster. The pattern is therefore not merely a laboratory curiosity: it describes the emotional structure of desperate gambling.

The same mechanism can operate in businesses losing ground to superior technology. Rather than accept a definite decline and close a failing project, managers may wager the organization’s remaining resources on an unfavorable recovery attempt. A company that has already spent $50 million on a delayed project may invest another $60 million in it even when a new project promises higher returns. From the firm’s perspective, this escalation is a sunk-cost error; from the executive’s perspective, continuing may postpone the humiliation of admitting failure or preserve the hope of recovering the original investment. Risk seeking thus reflects both the decision maker’s reference point and the conflict between personal incentives and organizational welfare.

The pattern also appears in collective conflict. When defeat is difficult to accept, a losing side in a war may continue fighting long after the opponent’s victory is effectively certain. The possibility of avoiding a devastating loss can make continued risk seem preferable to acknowledging defeat. In these situations, the gamble is not chosen because the probable outcome is attractive. It is chosen because the sure loss is emotionally unbearable.

Mental Accounts, Framing, and Policy

Mental accounting gives risk seeking a practical structure. People organize money and outcomes into separate accounts for spending, savings, education, emergencies, self-control, or particular transactions. These accounts make experience manageable, but they also narrow the frame of evaluation. A basketball fan who paid for a ticket is more likely to drive forty miles through a blizzard than a fan who received a ticket free from a friend, because staying home closes the paid ticket account with a more painful negative balance. A comprehensive decision maker would treat the ticket cost as sunk; ordinary decision making converts it into an emotional loss that intensifies the urge to proceed.

The same narrow framing helps explain why people continue investing in losing accounts or resist actions that would make failure explicit. Mental accounts keep score in emotional terms, rewarding gains and punishing losses, while regret and responsibility make doing nothing feel different from acting. These influences can produce choices that are irrational for wealth maximization but intelligible as attempts to repair a local account or avoid its humiliating closure.

Framing can also create or reduce risk seeking at the level of public policy. Organ-donation rates differed dramatically between otherwise similar European countries: donation was close to 100% in Austria but only 12% in Germany, 86% in Sweden but 4% in Denmark. The decisive difference was the default. In opt-out systems, people became donors unless they checked a box; in opt-in systems, they had to check a box to become donors. The effect was attributed less to an automatic preference than to the laziness of System 2: when people had not already decided, the default shaped the result.

This observation supports the policy approach associated with Richard Thaler and Cass Sunstein in Nudge. Their idea of libertarian paternalism preserves the right to opt out while arranging defaults, disclosures, and other choice architecture so that people are less likely to act against their long-term interests. Automatic enrollment in pension plans, clear presentation of fuel costs, and simpler food and health information are applications of the same principle. Humans need protection not only from their own risk-seeking reactions but also from institutions that deliberately exploit the quirks of System 1 and the laziness of System 2.

Significance in Decision Theory

Risk seeking exposes the limits of a theory that treats utility as a function of total wealth alone. Bernoulli’s framework could explain why a person prefers a sure gain, but it could not represent why a person with unfavorable options might prefer a gamble. The missing variable was the reference point, together with the different psychological treatment of gains and losses. Once outcomes are encoded as changes from that point, the apparent inconsistency becomes a systematic pattern.

The pattern also clarifies why emotional reactions often precede calculation. In paired decisions involving a 25% chance to win $1,000 or receive $240 for sure, and a 75% chance to lose $1,000 or receive $750 for sure, large majorities preferred the sure gain in the first problem and the gamble in the second; in the original experiment, 73% selected that combination, while only 3% chose the opposite pairing. The choices feel separately compelling even though combining them can reveal a logically inferior overall prospect. Risk seeking is therefore both a preference pattern and a warning about the limits of intuitive coherence: the mind reacts to the emotional character of certainty, loss, and possible relief before it constructs a unified account of consequences.