Cognitive Modeling

by Jerome R. Busemeyer, Adele Diederich

Cognitive Modeling is the first book to provide students with an easy-to understand introduction to the basic methods used to build and test cognitive models. Authors Jerome R. Busemeyer and Adele Diederich answer many of the questions that researchers face when beginning work on cognitive models, such as the What makes a cognitive model different from conceptual or statistical models? How do you develop such a model? How can you derive qualitatively different predictions between two cognitive models? Focusing on a few key representations, the authors introduce a basic problem in each chapter, illustrate the concept with three examples, and end with a summary of general principles, making this book by far the most accessible cognitive modeling book on the market.

Key Features
Cognitive Modeling is ideal for students and researchers across the various domains of cognitive sciences, including perception, learning, decision making, and inference. It is intended for use in upper-level undergraduate and graduate courses such as Cognition/Cognitive Modeling, Cognitive Science, Cognitive Psychology, Quantitative Methods, and Mathematical Modeling in Psychology.

Genres: Psychology, Artificial Intelligence

224 pages, Paperback

First published March 26, 2009

Reviews

[Max · Rating 4 out of 5]

I enjoyed this and would recommend it as a quick crash course into cognitive modeling. It‘s short but still manages to get into the theoretical weeds. It’s easy to read and to follow with little background knowledge needed. Their introduction to Bayes rule stood out to me as the most „basic“ content that they thought might be useful to some readers. It‘s very hands-on and focused on examples form the literature and it motivates the modeling procedures well (paraphrasing: „We assume this distribution because it‘s convenient“ :p).


Broad overview of topics:
- Examples of cognitive models (Signal Detection, Classification, Decision Making, Learning)
- Parameter estimation (Gradient descent, Maximum Likelihood) and objective functions (weighted least squares, Likelihood)
- Qualitative and quantitative model comparison (BIC, AIC, Cross Validation)
- Hierarchical models and numeric methods for parameter estimation (MCMC)

Some minor things:
I would have enjoyed some small excercises here and there. I think they would‘ve been useful for me, especially if they are feasible in a short amount of time and test the key ideas.

Busemeyer and Diederich do sometimes mention what they left out and what approaches could be used additionally, but I have the gut feeling that they must have left out a ton and other cognitive modelers would have written a very different book. I wonder if would be feasible to ask the community for some feedback what is missing and put this as further reading at the end of each chapter, or into the preface.

[Ricardo Acuña · Rating 4 out of 5]

This book explores the cognitive modeling of some typical cognitive process like decision making, memory and perception. Cognitive modeling is an attempt to mathematically represent the internal cognitive process that occurs within our minds. I found that this book requires some knowledge of mathematical and statistical concepts, but you can follow it. For me it was interesting to identify how closely are related the machine learning and artificial intelligence algorithms with the cognitive models and how the comparison methods can be used to test the models against empirical results. But more interesting was to find how psychology and mathematics can be mixed to explain our brain process with a provocative new view (Mathematical Psychology). Highly recommended for people who are interested in cognitive science.

[M. Apple · Rating 2 out of 5]

The style is too dense for anyone but a student of cognitive psychology. The use of the word "model" is confusing to me. I didn't see any "models" per se...at least not in the sense I'm used to (structural equation modelling...) All I saw was pages and pages of mathematical formulae and line graphs. The description of "cognition" made it seem as if all psychologists have to do is mathematically predict the choices people might make based on probability curves. I see no indication or awareness of social influences. The individual mind does not live in isolation. Curiosity made me buy this, but it's not something I can agree with conceptually.