Can D–E Gap Be Completely Explained By Sampling Error?

The description-experience gap(D-E gap) refers to the observation that choices are influenced by whether information about potential alternatives is learned from a summary description or from the experience of sequentially sampling individual outcomes. D-E gap stems from the contrastive study between decision from experience and decision from description. Wild attentions from researchers in the filed of behavioral decision-making was paid to the discovery of D-E gap. Until now the key research is still the investigation of the causes of D-E gap. Although numerous studies have discussed the reasons for the D-E gap, and many valuable conclusions are drawn, such as small samples, recency, information format and cognitive algorithms. But the researchers do not have a radical interpretation of the causes of D-E gap. Take small samples for example, it has been recognized by the majority of researchers that in sampling paradigm small samples lead to the occurrence of D-E gap, but the sampling deviation theory and theory of psychological factors is uncompromising for whether the D-E gap is completely explained by sampling error.Research conclusions and interpretations are subject to the research methods and tools, most of the early studies of D-E gap follow the classic research paradigm and decision task, and the relevant research findings are limited to the monetary gambles in classic research paradigm. There are studies that had improved the previous research methods, and found that the implication of the D-E gap may not be limited to the effect of underweight rare events. Secondly, the scholars still have many disputes and questions about the reasons for D-E gap, such as the he debate between the sampling deviation theory and the theory of psychological factors. Thirdly, from the view of specific research, these factors are put forward for a particular paradigm and tasks, such as recency led to the D-E gap in active sampling paradigm.Therefore, in this study we select Chinese subjects to repeat the previous monetary gambles under the classic sampling paradigm to verify the existence of D-E gap, and through the analysis of sampling frequency and sampling distribution todetermine the existence of the external biases caused by small sample sampling and the internal biases, and these sample error will affect the D-E gap. In addition, the paradigm of decision from experience, which emphasizes the process of sequential information processing is a kind of decision task paradigm. And this process is the core of the statistical probability scenario, which is ignored by the traditional decision from description. Therefore, this research follows the logic of the process of obtaining information in paradigm of decision from experience to improve traditional prisoners dilemma game and chicken game, to explore the difference of the results between decision under experience situation and describe situation.In specific, we though 4 experiments to verify the hypothesis: sampling bias is not enough to fully explain the D-E gap. Experimental 1 follows the previous monetary gambles to verify the existence of D-E gap; there exist internal biases and external biases in decision from experience. Experiment 2 by referring to the design of Camilleri's judgment probes(2009), pseudo-random number(2010) and Rakow's yoked situation to investigate the influence of sampling error, whether the sampling bias can completely explain D-E gap. In experiment 3 and 4, we use prisoner's dilemma game to explore the influence of sampling error on decision from experience and D-E gap. The main conclusions are drawn as follows: 1, There exist D-E gap between the result of decision from description and the result of decision from experience, there exist internal biases and external biases in decision from experience.2, In monetary gambles, there exist internal biases and external biases in information acquisition process before decision from experience will affect the decision from experience and D-E gap. Controlling the external biases and internal biases can reduce the size of the D-E gap, but the D-E gap still exists. In prisoner's dilemma game, the main effect of sampling bias on decision from experience is not significant,the D-E gap can not be completely explained by sampling bias, and there exist other factors that affect the D-E gap.In the end, the whole study was summarized and reflected, the shortcomings and the issues worthy of further research were also outlined, such as improving research methods, the comprehensive comparison of both decision-making paradigms; thestudy also describe the theoretical and practical significance of the research of D-E gap.