Original scientific paper
Conditional Reasoning and Success of Deduction
Pavle Valerjev - Department of Linguistics, University of Zadar, Zadar
Igor Bajšanski - Faculty of Humanities and Social Sciences, University of Rijeka, Rijeka
Tanja Gulan - Department of Linguistics, University of Zadar, Zadar
Fulltext (croatian, pages 49-62).pdf
Abstracts
Human reasoning dealing with conditionals in a form of If P, then Q (where P is the antecedent and Q is the consequent) is one of the most prominent problems in the psychology of deductive reasoning. The reason for this lies in the numerous biases that reasoners apply during the process of inference. This resulted in several theories that aim to explain conditional reasoning. However, the mental model theory turned out to be the most dominant. This theory claims that differences in efficiency between modus ponens (MP) and modus tollens (MT) inferences are caused by varying numbers of mental models that need to be constructed. Modus ponens inference needs only one model, while modus tollens needs three models (or at least two). Hence, MP is the easier and quicker type of inference. However, there is another factor which can affect the efficiency of the conditionals – the directionality of conditional. Valerjev (2006) used MP and MT inferences as well as negation of antecedent (NA) and affirmation of consequent (AC) and found that the direction of conditional (from antecedent to consequent or vice versa) affected the response time. In this study the effect of two types of directionality concerning conditionals were investigated. The two types of directionality are syntactic directionality – that depend on the order in which antecedent and consequent are arranged in a conditional sentence – and semantic directionality that depends on usage of words “left” and “right” in the left or the right part of the conditional sentence. Three-factor experiment (2 × 2 × 2) was conducted. Participants in the experiment were psychology students. The efficiency parameters were response time and correctness of the response. The type of conclusion effect showed significant advantage of MP conclusion compared to MT conclusion on both efficiency parameters, as expected. Syntactic directionality showed effects on both MP and MT conclusions. In MP situations standard conditionals were verified faster, while in MT situations reversed pattern was obtained – the reversed conditionals were verified faster. It seems that the way the conditional is represented affects the reasoning process differently for MP and MT conclusions, respectively.
Keywords
deduction, reasoning, conditionals, mental models, directionality of conditionals