The description-experience gap is a significant phenomenon observed in behavioral studies of decision-making, highlighting a fundamental difference in how people behave based on whether they receive information about choices through clear descriptions or through direct experience. This gap reveals that our decision-making processes are not uniform; they adapt based on the mode of information acquisition. Understanding this distinction is crucial for comprehending
why individuals might make seemingly irrational choices, especially when dealing with probabilities and risks, and how cognitive biases like the recency effect play a role.
Described Versus Experienced Choices
In experimental tasks, participants typically choose between two options that lead to various outcomes, which can be gains or losses, with varying probabilities. In description-based studies, participants are provided with explicit information about the potential outcomes and their probabilities for each choice. For example, they might be told, "Option A has a 70% chance of gaining $10 and a 30% chance of gaining $0." Feedback on the consequences of their choices is usually not provided. Prospect theory largely guides our understanding of these described choices, suggesting that people weigh probabilities differently based on whether outcomes are framed as gains or losses, and whether they are sure or merely probable.
Conversely, in experience-based studies, participants are not initially given any explicit information about outcomes or probabilities. Instead, they must sample from the choices, learning about the potential outcomes and their likelihoods through feedback after each selection. They estimate probabilities based on their direct experiences. This mirrors real-world situations where individuals often make decisions without a clear statistical breakdown, relying instead on their past interactions with similar scenarios. The contrast between these two approaches reveals significant differences in decision-making patterns.
Underweighting Rare Events in Experience
One of the most striking findings from experience-based studies is that people tend to underweight the probabilities of rare outcomes. This means they generally choose the more probable outcome much more often, behaving as if rare outcomes are even less likely than they truly are. This behavior stands in contrast to predictions from prospect theory, which often suggests that people might overweight extreme outcomes in described prospects. The underweighting of rare events in decisions made from experience has been observed in studies involving repeated, small samples of choices.
This tendency can lead to different risk-taking behaviors. For instance, when tasks are framed in terms of gains, people tend to choose the riskier option when deciding from experience, which again contrasts with decisions made from description. This reversal of typical risk preferences, such as the reflection effect (where people are risk-averse for gains and risk-seeking for losses in described prospects), highlights the profound impact of how information is presented and acquired. In experience-based scenarios, people might become risk-seeking for gains and risk-averse for losses, effectively reversing the patterns seen in description-based tasks.
The Role of the Recency Effect and Other Factors
Several factors contribute to the description-experience gap. The recency effect is a key explanation for the underweighting of rare events in experience-based decisions. The recency effect dictates that more recent events are given greater weight or value. Since rare events are uncommon, more common events are more likely to have occurred recently and thus receive greater weight in decision-making. This makes rare events seem even less likely than they are, as they are less likely to be recent experiences. In description-based studies, where probabilities are explicitly known, the recency effect has little to no influence.
Other factors also play a role. The nature of the sampling task itself can influence outcomes; if participants only experience a small number of prospects, they might not even encounter low-probability events, leading them to underestimate their likelihood. Additionally, a basic tendency to avoid delayed outcomes can contribute, as alternatives with positive rare events are often advantageous only in the long term. Furthermore, memory biases can influence decisions from experience, as participants rely on memory to learn about outcomes. More improbable but greater rewards might produce more salient memories, leading to greater risk-seeking in gain choices in experience-based studies. These combined factors illustrate the complexity of human decision-making when faced with uncertainty and varying information formats.
