Decisions from experience

To what extent do people adapt their information search policies and subsequent decisions to the long-and short-run consequences of choice environments? To address this question, we investigated exploration and exploitation policies in choice environments that involved single or multiple plays. We further compared behavior in these environments with behavior in the standard sampling paradigm. Frequently used in research on decision from experience, this paradigm does not explicitly implement the choice in terms of the short or long run. Results showed that people searched more in the multi-play environment than in the single-play environment. Moreover, the substantial search effort in the multi-play environment was conducive to choices consistent with expected value maximization, whereas the lesser search effort in the single-play environment was compatible with the goal of maximizing the chance of winning something. Furthermore, choice and search behaviors in the sampling paradigm predominantly echoed those observed in the single-play environment. This suggests that, when not instructed otherwise, participants in the sampling paradigm appear to favor search and choice strategies that embody short-run aspirations. Finally, the present findings challenge the revealed preference approach in decisions from experience, while also suggesting that information search may be an important and potentially even better signal of preference or aspirations than choice.

This paper examines the effects of different cultural backgrounds on decisions from experience. In Experiment 1, participants from Denmark, Israel, and Taiwan faced each of six binary choice problems for 200 trials. The participants did not receive prior description of the payoff distributions, but obtained complete feedback after each choice. Comparison of choice behavior across cultural groups reveals similar overall choice rates, and similar indications of underweighting of rare events and of the payoff variability effect. In addition, subjects from Taiwan exhibited a stronger tendency to chase recent outcomes. That is, subjects from East Asia behaved “as if” they expected less change in the environment than subjects from West Asia and West Europe. Experiment 2 shows that an increase in the complexity of the choice tasks (i.e., adding slight variability to the safe option, and increasing the number of replicas for each option) does not break the similarity of choice rates across cultural groups, but reverses the observed chasing pattern: In Experiment 2, Israeli participants tended to chase recent outcomes more than did the Taiwanese. These results can be summarized with the assumption that the tendency to rely of small samples of past experiences (a sufficient condition for underweighting of rare events and the payoff variability effect) is robust to cultural differences, but the exact sampling process is culture- and framing-specific. An increase in the number of possible outcomes increases the probability of sampling the most recent trial in the West, but not in the East. Thus, behavior in the East appears less sensitive to task complexity.

When people make risky choices, two kinds of information are crucial: outcome values and outcome probabilities. Here, we demonstrate that the juncture at which value and probability information is provided has a fundamental effect on choice. Across four experiments involving 489 participants, we compared two decision-making scenarios: one in which value information was revealed during sampling (standard) and one in which value information was revealed after sampling (value ignorance). On average, participants made riskier choices when value information was provided after sampling. Moreover, parameter estimates from a hierarchical Bayesian implementation of cumulative-prospect theory suggested that participants overweighted rare events when value information was absent during sampling but did not overweight such events in the standard condition. This suggests that the impact of rare events on choice relies crucially on the timing of probability and value integration. We provide paths toward mechanistic explanations of our results based on frameworks that assume different underlying cognitive architectures.

For over 50 years expected value and expected utility theory has been challenged by behavioral findings in repeated and single plays of risky gambles. The inherent long-term nature of these models has been found to be at odds with preferences indicating short-term maximization in single play situations. With the present study we provide further evidence on the distinction between long-term and short-term oriented behavior. Evaluating experiencedbased decisions over repeated and single play situations we demonstrate that both choice preferences and search behavior change in response to long and short-term framing. This suggests different cognitive approaches for single and repeated play situations, with single decisions often favoring risk-aversion and therefore the underweighting of rare events. These findings are in line with alternative models of risky choice as for example proposed by Lopes (1996) and also the literature on statedependent foraging.