Reviews On The Sets Mapping Model

Four types of proposition were chosen as experiment material in four inference forms whose antecedents and consequents could be binary or non-binary by prepared experiment. The influence of four types of proposition on the performances of the four inferences was investigated in the experiment with university students. There exists the semantic connection of "not this but that" to the antecedents and (or) consequent of the major proposition and there is no the third instance, which is called the binary proposition. If an object has only two sides, we call it binary term, for example, there are only two genders, male and female. On the contrary, if there is a third condition, we call it non-binary, for example, for colors, we have red, white, black, and so on.There was a 4×4 within-subject design. Proposition types and reasoning rule produce two factors within-subjects design. Factor 1 was conditional propositions with different types: Type I , both the antecedent and the consequent are binary (BB). Type II, the antecedent is binary while the consequent is non-binary (BN). Type III, the antecedent is non-binary while the consequent is binary (NB). Type IV, both the antecedent and the consequent are non- binary (NN). One example of the four types of conditionals was given as follow. BB: If someone scores (lose points). then he answers the questions correctly (false). BN: If it is the outside of an overcoat (inside), then there hangs the medal. NB: If someone wears shoes with high-heel, then she is female (male). NN: If someone goes to Beijing, then he takes a train. The experiment material was concrete from type I to type IV. Every type of conditional has three homogenous conditional propositions, so there were twelve conditional propositions in all. All propositions tested pre-experiment conformed to the experiment standard. Dependent measure is the scores participants got in the conditional reasoning task. Each participant gets 1 point if the answer was logical, 0 point while the answer was illogical in each of the four inferences (MP, DA, AC, MT). These data were analyzed using a two-factor within-subject ANOVA. The results showed interaction between them was significant too (F (9,735) =26.96, P=0.000). Specifically, there are four different response patterns. For BB conditionals, it elicits a strong bi-conditional response pattern, that is, participants only failed to suppress the invalidate DA and AC inferences, while generated high rates of validate MP and MT inferences. For BN conditionals, except for MP, participants generated high rates of false answers across the rest DA, AC and MT inferences. For NB conditionals, participants seemed to automatically make the valid inferences and resist the invalid inferences logically to a great extent. For NN conditionals, only the MT inference was strongly suppressed. As for the validate form, typically, a great percentage of participants easily and happily to make the valid MP inference on any of the four conditional types and MT inference on BB and NB conditional types. While on BN, especially NN conditional types, the valid MT inference were much hard to be made. And for the invalidate form, many subjects tended to accept the invalidate inferences on BB and BN conditionals while the fallacious answers were greatly suppressed on NB and NN conditionals. The design in the experiment 2 is the same as the experiment 1. The difference between experiment 1 and experiment 2 was the materials we adopted. One example of the four types of conditionals was given as follow: BB: If someone scores (lose scores), then he makes efforts (no efforts). BN: If it is the outside of an overcoat (inside), then it is yellow. NB: If someone drinks milk, then she is female (male). NN: If someone goes to Beijing, then he takes a train. Unlike the results of experiment 1, here we only get one response pattern over four different conditional types, that is, compared with high scores in invalidate inference forms, participants do quite poor in validate inference forms, especially for MT, a strong suppression effect was observed.Using the similar materials, the two experiments above show divergent results, which makes sure that it is not the quality of the conditional types but the semantic connection activated between the antecedent and the consequent that influences people's conditional reasoning. We think there exits three Stages for conditional reasoning: 1, representing the semantic connection between two sets when comprehending the conditionals. 2, forming a certain mapping pattern for certain conditionals. 3, recollecting the mapping pattern and applying to the reasoning process. These results verified the sets mapping model of conditional reasoning to a great extent.