Certainty Effect

Within decision making, the certainty effect is used to describe the impact of certainty on the decision maker. People are drawn to certainty, giving higher preference to options that have high levels [Page 122]of certainty. An option with high certainty (close to 0% or 100%) is more appealing to people than a complex or ambiguous probability. This causes many decision makers to choose options that go against the expected utility of the problem. A reduction in probability has a greater impact on the decision maker if the initial outcome is certain. For example, a reduction in survivability from 100% to 90% would have a greater impact than a reduction in survivability from 70% to 60%.

The underlying reason for the certainty effect falls on a person's preference for certain or absolute values. People will bear psychological effects from feelings both of certainty and of uncertainty. They prefer certainty, rather than complexity and ambiguity. Most decision makers cannot clearly define the difference between two probabilities, especially if they are ambiguous. Rather than consider exact probabilities, people often lump outcomes into categories such as “likely” and “unlikely.” This makes comparison between two “likely” probabilities difficult. For example, if a healthcare provider explains two courses of treatment to a patient, he or she may present some probability of full recovery. If both options presented a midrange probability, it would be difficult for the patient to decipher the true difference between them. Consider the case where the first course of treatment presents a 70% chance of full recovery, whereas the second presents a 60% chance of full recovery. Most people would be unable to differentiate between these two probabilities but would rather refer to them as “good chances,” “likely,” or “better than average.” If one course of treatment had extreme certainty (close to 100% in this example), the decision maker would put a higher weight on the certain treatment. This is due to the fact that decision makers tend to eliminate uncertainty altogether by overweighting the certain outcomes.

Consider the following case, originally presented by Amos Tversky and Daniel Kahneman. Treatment A leads to a 20% chance of imminent death and an 80% chance of normal life, with a longevity of 30 years. Treatment B leads to a 100% chance of normal life with a longevity of 18 years. According to expected utility theory, rational decision makers would choose Treatment A as it provides a higher utility in terms of lifespan (24 years compared with 18 years). However, the majority of decision makers choose Treatment B. This is a prime example of the certainty effect in practice. Decision makers, be they physicians or patients, have a high preference for certain outcomes, regardless of the comparative utilities associated with them.

Decision makers are confident when handling extreme probabilities (near 0 or 1.0). When the probabilities are not as certain, however, the weighting of alternatives becomes disproportionate. Decreasing a risk from 5% to 0% should have the same utility as decreasing that risk from 20% to 15%. However, decision makers greatly prefer the first.

...