Many previous studies have assumed that the domain of the explanandum determines which type of explanation,mechanistic or teleological, people prefer (domain theory). In this study, I proposed that the explanandum’s thematic relation,which is mostly determined by the predicate, i.e., action verb or state verb, is crucial for the explanation type preference(thematic relation theory). To compare the two theories, participants were asked to read a sentence describing the explanandum,and then judge the appropriateness of two explanations for the explanandum, mechanistic and teleological, one after the other.Order of the two explanations were counterbalanced over participants. The domain and thematic relations were manipulated byvarying the subject and the predicate of the explanandum. Mechanistic explanations were preferred when the predicate was astate verb, whereas teleological explanations were preferred for an action verb. Results of the experiment gave support for thethematic relation theory.

The plausability of the retracting information, which can affect whether the retracted information be replaced bythe retracting information, had not been studied much in previous studies on continued influence effect. An experiment wasconducted to investigate how the plausability of a retracting information affects the understanding of an event in which someof the information had been retracted. In the experiment, the plausibility of the retracting information (low versus high) andthe manner of correcting information (simple retraction versus supplying an alternative) were manipulated, and the participantswere asked to answer memory questions and inference questions. Retracted information was better remembered for memoryquestions, but was used less frequently for inference questions when the retracting information was less plausible comparedto when the retracting information was quite plausible. The results were interpreted to support the ‘memory-based’ theory ofmemory updating.