Human Reasoning in the Suppression Task: The Relevance of Commonsense Knowledge

Abstract

A new model for human reasoning during logical tasks is proposed, which possesses the intuitive quality of recognizability in the way it describes the process of reasoning. Experiments have shown that humans do not reason according to classical logic during logical tasks. To solve the question of how humans do reason, alternative logical models have been created to replicate the reasoning during these logical tasks. The logical task focussed on here is the Suppression Task. The most relevant model for the human reasoning in this task is based on a notion of abnormality predicates, and lacks intuitive recognizability both in it’s set-up and the conclusions it yields. The proposed new model is based on a notion of commonsense knowledge. The experiment shows that participants performing the Suppression Task reason in line with this commonsense knowledge based logical model, in stead of the abnormality predicates that have been proposed in previous models.