Set membership talk at the London Reasoning Workshop
I will be talking at the London Reasoning Workshop on my work with Phil and Max Lotstein that concerns new studies and the computational modeling of set membership inferences. The talk is titled, “The psychology of set membership”, and the workshop will be held on the 25th and 26th of July, 2013.
Here’s the abstract:
We describe a new theory of set membership inferences from quantified assertions, such as: None of the artists is a bohemian; Amy is an artist therefore, Amy is not a bohemian. The theory is based on mental models and implemented in a computer program, mReasoner. It embodies three principles: 1.) individuals use a representation of the meanings of assertions and their knowledge to construct models of possibilities; 2.) models are iconic insofar that is possible; and 3.) models represent what is true rather than what is false. The theory makes two novel predictions about set membership inferences: first, inferences in which the initial mental model of the premises yields a correct conclusion should be easier than inferences in which an reasoners have to modify their initial model. Second, reasoners should be susceptible to illusory set membership inferences in which they systematically conclude that an individual is a member of a set even though the premises preclude the possibility. Experiment 1 corroborated the two levels of difficulty, and Experiment 2 established illusory set membership inferences. We discuss the results as a step towards the development of a unified theory of monadic reasoning.