Asymmetry Studies Of Predictive Reasoning And Diagnostic Reasoning

Causal relation is the most basic way to understand the relationship between the objective events in life, most human judgments under uncertainty involve reasoning about causal relationships. Causal reasoning has two directions: from the causes to the results (predictive reasoning) and from the results to the causes (diagnostic reasoning).Alternative causes refer to other factors that affect the outcome of the common outcome structure. Alternative causes play a very important role in predictive reasoning and diagnostic reasoning.Causal Bayes net theories posit that judgments of uncertainty are generated from causal beliefs in a way that respects the norms of probability,and the model assumes that the uncertainty of the causal belief is based on the probability of compliance with the probability. Analysis of previous research showed that the diagnostic reasoning conforms to the standard probability model, but the predictive inference is significantly undervalued; people can consider alternative causes in the diagnosis of spontaneous reasoning, but ignored in the prediction of reasoning can be the reason. On this phenomenon, Fernbach et al. (2010, 2011a) put forward "the pragmatic interpretation of the alternative causes of neglect". In this study, we choose the daily causal story as the research material, to study the non symmetric phenomenon which can be replaced by the predictive reasoning and diagnostic reasoning, and to investigate the pragmatic interpretation.A total of three experiments, experiment verified Fernbach et al. (2011a) standard proposed probabilistic model and are tested in response to predict problem when considering what information; using the Bias causal network, established a standard model formula, namely prediction probability and probability judgment should be how to diagnosis as the intensity change alternative causes the strength, cause and effect and the prior probability and change. The results showed that the diagnosis was very consistent with the model prediction, and the prediction ignored the alternative reasons,and the relative model was not consistent with the model.Experiment two examined how to understand the standard predictive reasoning problem, in the experiment, so that the choice of subjects to choose the most able to reflect their understanding of the standard prediction problems and understanding of the options. The standard problem form of predictive reasoning and diagnostic reasoning is given, and then by making the subjects choose one of the three options, as they consider the question to be answered. The results found that 45% of the subjects had to understand the problem of predicting the causal effect of the problem, and another 36%of the subjects had to predict the problem to be understood as a result of the core causes of the results of the judgment.Experiment three examined whether changes in the form of the problem can improve the performance of the correct answer, by directly telling the subjects to be able to replace the cause of a clear or clear does not appear. If you ignore for alternative reasons is because alternative causes of state uncertainty reasoning difficulties caused by the prediction is that judgment should be increased as the alternative state does not appear, and the unknown, and the diagnosis is changed in the opposite direction.Three conclusions are drawn from this study: first, people do not take into account all the factors that should be considered in the prediction of reasoning. Second, it is often used to understand the problem of predictive judgment as a causal effect, and the proposed form of the prediction of the problem will be confused between the two, so there may be a substitute for the cause of the problem. Third, they are more likely to be ignored when the state of alternative causes is not certain.