选择反转是交通行为中的一种悖论现象，未能通过实验的谨慎误差控制得到良 好解释． 本研究从决策心理学的角度，基于齐当别理论建立起选择支属性的判断维度体 系，在确定、不确定以及风险条件下进行了 3 种公交出行决策场景的心理实验． 实验结果 中匹配任务的效应分析说明，齐当别理论的侦测性占优原则可对出行方式选择进行解释; 选择一致性的重测信度分析说明，出行者对交通方式属性维度的判断是可变过程;选择和 匹配一致性分析说明，选择变异随匹配任务中的维度判断的改变而产生． 总体上，3 组实 验选择反转的变化特征趋势符合实验设计框架，齐当别模型的解释在 3 组实验中均表现 出显著一致性，能够对出行方式的选择反转进行预测和解释，即当出行者采取不同的齐同 策略时，重复选择中令人费解的反转悖论现象便会出现．

Choice reversal is a paradox phenomenon in traffic behavior，and it can not be well explained through error cautious control． From the perspective of decision psychology，the judgment dimension system of option at- tributes，and carries out three groups of psychological experiments in different bus travel decision situation was es- tablished based on the equate-to-differentiate theory． Through the matching task，it was confirmed that choice can be explained under the equate-to-differentiate interpretation． Through the retest reliability analysis for choice con- sistency，it was proved that the judgment for dimension is variable． Through the retest reliability analysis for choice and matching consistency，it was proved that choice reversals are caused by the change of the dimension judgment in matching task． The equate-to-differentiate interpretation shows good consistency in the three groups of experiments． The judgment dimension system can predict and explain the travel choice reversal． That is，the in- explicable paradox occurs when diverse equate-to-differentiate strategy is used．