Empirical Justifications for the Universalness of the Mental Logic and Mental Models Paradigm
Selmer Bringsjord, Jiahong Guo, Shier Ju, Yingrui Yang, Jianmin Zheng, Yi Zhao · eScholarship (California Digital Library) · 2005
Empirical Justifications for the Universalness of the Mental Logic and Mental Models Paradigm Yingrui Yang ([email protected]) Yi Zhao ([email protected]) Department of Cognitive Science Rensselaer Polytechnic Institute, Troy NY USA Institute of Logic and Cognition Zhongshan University, Guangzhou, China Jianmin Zeng ([email protected]) Jiahong Guo ([email protected]) Institute of Logic and Cognition Zhongshan University, Guangzhou, China Institute of Logic and Cognition Zhongshan University, Guangzhou, China Shier Ju ([email protected]) Selmer Bringsjord ([email protected]) Dept of Cog Science/Dept of Comp. Science Rensselaer Polytechnic Institute, Troy NY USA Institute of Logic and Cognition Zhongshan University, Guangzhou, China Abstract Though there is a long-standing controversy between mental logic theory and mental model theory in human deductive reasoning, the two theories do share one general hypothesis: that whether people reason by applying inference schemas or by constructing mental models, these mental representations and mechanisms should be universal across homo sapiens, which implies that they should be available to speakers of different languages. We examined this hypothesis by conducting a series of experiments in Chinese, parallel to the empirical research supporting mental logic (Yang, Braine, and O’Brien, 1998), and to that supporting mental models (Yang and Johnson-Laird, 2000). The results show that the Chinese data sets are significantly correlated with English data sets, indicating that performance in Chinese and performance in English are likely to be based on similar mental representations and mechanisms (in tasks of deductive reasoning with quantifiers and predicates). Introduction While logic tells us what reasoning is, psychology of reasoning studies how people reason. There are two major competing approaches in psychology of deductive reasoning: mental logic theory (Braine & O’Brien, 1998; Rips, 1994) claims that people reason by applying inference schemas, and mental model theory (Johnson- Laird & Byrne, 1991) claims that people reason by constructing mental models. Though there is a long- standing controversy between the two (O’Brien, Braine, & Yang, 1994; Johnson-Laird, Byrne, & Schaeken, 1994), they share one general normative hypothesis: viz., that whether people reason by applying inference schemas or by constructing mental models, these mental representations and mechanisms are universal across homo sapiens, which means they should be available to speakers of different languages. However, recently some researchers have argued that there is a cultural difference in systems of thought: a difference between Western analytical cognition versus Eastern holistic cognition (Nisbett, Peng, Choi, & Norenzayan, 2001). There is a great deal of empirical evidence in the psychological literature supporting both mental logic theory (e.g., Yang, Braine, & O’Brien, 1998) and mental model theory (e.g., Yang & Johnson-Laird, 2000). Both sets of experiments in these examples were conducted in English. This paper reports two sets of experiments conducted in Chinese, strictly parallel to the above English counterparts. In the following two sections, we report on the set of experiments in Chinese based on mental logic theory first, and then based on mental model theory. Mental Logic The theory Mental logic theory claims that people reason by applying inference schemas. An inference schema specifies the form of an inference: Given information whose semantic representation has the form specified in the schema, one can infer the conclusion, whose form is also specified. Below is an example: All the children found some red beads. The red beads were either round or square. The round beads were plastic. The square beads were wooden. Did all the girls find either plastic or wooden beads? Mental logic theory predicts that people solve this problem in steps. Step 1 is to infer from the first premise that all the girls got red beads. This step can be seen as a constant of the inference schema called quantified modus ponens. Step 2 is to consider the next three premises and infer that the red beads are either plastic or wooden. This step is an instance of another schema. Step 3 is to apply this schema again on the two intermediate conclusions in Steps 1 and 2, and infer that all the girls found either plastic or wooden beads; and this leads to answer Yes to the question.