You are here: Home › Teaching › Cognitive Modeling (SS 2020)

Cognitive Modeling (SS 2020)

Organizer: Marco Ragni
Assistants: Nicolas Riesterer, Daniel Brand, Hannah Dames, Paulina Friemann

A syllogism typically consists of two quantified premises, from which a conclusion has to be deduced. Consider the following example:

All cognitive scientists are intelligent.
Most intelligent people are successful.

Though the majority of people deduce from these statements that most cognitive scientists are successful, the only logically valid response states that no valid conclusion (NVC) exists.

While a huge body of research investigating syllogistic reasoning with the classical quantifiers all (A), some (I), some not (O), and none (E) exists, only few theoretical approaches provide predictions for generalized quantifiers like most (M) or few (F). Research in the field of generalized quantifiers emerges from the goal of examining a wider range of reasoning and transferring theories from hypothetical conditions to everyday reasoning. Thus it has been criticized that classical quantifiers are not relevant with regard to everyday contexts, because A and E are too strict not allowing any exceptions, whereas I and O are considered too weak requiring only a single individual. Dealing with uncertainty is a common feature of human reasoning when interacting with the world. This leads either to the revision of premises which have been believed to be true (i.e. non-monotonic reasoning) or to enabling exceptions from the beginning, e.g. through generalized quantifiers, which are discussed in the present seminar.

The seminar is a block seminar. Most if not all parts of the seminar will be organized online.

Background Literature

  • Khemlani, S., & Johnson-Laird, P. N. (2012). Theories of the syllogism: A meta-analysis. Psychological Bulletin, 138(3), 427.
  • Ragni, M., Singmann, H., & Steinlein, E. M. (2014). Theory comparison for generalized quantifiers. In Proceedings of the Annual Meeting of the Cognitive Science Society.

Schedule:

  • Organizational meeting (via Zoom) May 13th at 3 pm
    (please mail your name and interest in participation no later than May 11th to the organizer)

  • Midterm presentation (via Zoom) June 26th at 3 pm

  • Seminar presentation (onsite/Zoom) July 10th from 9-17 and July 11th from 9-17

Requirements

  • Presentation of your preliminary & final results
    • Theoretical and computational foundation
    • Predictive performance
    • Ideas for improvement
  • Written report of your work (~6 pages, CogSci format)
    • Introduction/Motivation
    • Theoretical Foundation
    • Method/Model
    • Results
    • Conclusion/Discussion

Materials