scholarly journals Troubleshooting Bayesian cognitive models: A tutorial with matstanlib

2021 ◽  
Author(s):  
Beth Baribault ◽  
Anne Collins
Keyword(s):  
Cognitive Processes ◽  
Bayesian Model ◽  
Cognitive Modeling ◽  
Computational Models ◽  
Model Fitting ◽  
Psychological Research ◽  
Cognitive Models ◽  
Model Output ◽  
Mcmc Sampling ◽  
Bayesian Cognitive Modeling

Using Bayesian methods to apply computational models of cognitive processes, or Bayesian cognitive modeling, is an important new trend in psychological research. The rise of Bayesian cognitive modeling has been accelerated by the introduction of software such as Stan and PyMC3 that efficiently automates the Markov chain Monte Carlo (MCMC) sampling used for Bayesian model fitting. Unfortunately, Bayesian cognitive models can struggle to pass the computational checks required of all Bayesian models. If any failures are left undetected, inferences about cognition based on model output may be biased or incorrect. As such, Bayesian cognitive models almost always require troubleshooting before being used for inference. Here, we present a deep treatment of the diagnostic checks and procedures that are critical for effective troubleshooting, but are often left underspecified by tutorial papers. After a conceptual introduction to Bayesian cognitive modeling and MCMC sampling, we outline the diagnostic metrics, procedures, and plots necessary to identify problems in model output with an emphasis on how these requirements have recently been improved. Throughout, we explain how the most commonly encountered problems may be remedied with specific, practical solutions. We also introduce matstanlib, our MATLAB modeling support library, and demonstrate how it facilitates troubleshooting of an example hierarchical Bayesian model of reinforcement learning implemented in Stan. With this comprehensive guide to techniques for detecting, identifying, and overcoming problems in fitting Bayesian cognitive models, psychologists across subfields can more confidently build and use Bayesian cognitive models.All code is freely available from github.com/baribault/matstanlib.

scholarly journals A Tutorial on Cognitive Modeling for Cognitive Aging Researchers

2020 ◽  
Author(s):  
Nathaniel R. Greene ◽  
Stephen Rhodes
Keyword(s):  
Age Differences ◽  
Cognitive Processes ◽  
Cognitive Aging ◽  
Cognitive Modeling ◽  
Hierarchical Modeling ◽  
Model Fitting ◽  
Cognitive Models ◽  
Modeling Software ◽  
Fitting In ◽  
Observed Behavior

Cognitive aging researchers are interested in understanding how cognitive processes change in old age, but the standard analyses used on observed behavior (e.g., ANOVA) are inappropriate for measuring age differences in latent cognitive processes. Cognitive models formalize the relationship between underlying processes and observed behavior and are more suitable for identifying what processes are associated with aging. This article provides a tutorial on how to fit and interpret cognitive models to measure age differences in cognitive processes. We work with an example of a two choice discrimination task and describe how to fit models in the highly flexible modeling software Stan. We describe how to use hierarchical modeling to estimate both group and individual effects simultaneously, and we detail model fitting in a Bayesian statistical framework, which, among other benefits, enables aging researchers to quantify evidence for null effects. We contend that more widespread use of cognitive modeling among cognitive aging researchers may be useful for addressing potential issues of non-replicability in the field, as cognitive modeling is more suitable to addressing questions about what cognitive processes are (or are not) affected by aging.

scholarly journals Cognitive Modeling in Intelligence Research: Advantages and Recommendations for their Application

2018 ◽  
Author(s):  
Gidon T. Frischkorn ◽  
Anna-Lena Schubert
Keyword(s):  
Working Memory ◽  
Cognitive Processes ◽  
Processing Speed ◽  
Cognitive Modeling ◽  
Cognitive Process ◽  
Working Memory Capacity ◽  
Memory Capacity ◽  
Cognitive Models ◽  
Model Fit ◽  
General Intelligence

Mathematical models of cognition measure individual differences in cognitive processes, such as processing speed, working memory capacity, and executive functions, that may underlie general intelligence. As such, cognitive models allow identifying associations between specific cognitive processes and tracking the effect of experimental interventions aimed at the enhancement of intelligence on mediating process parameters. Moreover, cognitive models provide an explicit theoretical formalization of theories regarding specific cognitive process that may help overcoming ambiguities in the interpretation of fuzzy verbal theories. In this paper, we give an overview of the advantages of cognitive modeling in intelligence research and present models in the domains of processing speed, working memory, and selective attention that may be of particular interest for intelligence research. Moreover, we provide guidelines for the application of cognitive models in intelligence research, including data collection, the evaluation of model fit, and statistical analyses.

scholarly journals Beyond the “Conceptual Nervous System”: Can Computational Cognitive Neuroscience Transform Learning Theory?

2019 ◽  
Author(s):  
Fabian Soto
Keyword(s):  
Nervous System ◽  
Learning Theory ◽  
Cognitive Processes ◽  
Cognitive Neuroscience ◽  
Cognitive Modeling ◽  
Computational Models ◽  
Levels Of Analysis ◽  
Neural Network Approach ◽  
Behavioral Predictions ◽  
Behavior And Cognition

In the last century, learning theory has been dominated by an approach assuming that associations between hypothetical representational nodes can support the acquisition of knowledge about the environment. The similarities between this approach and connectionism did not go unnoticed to learning theorists, with many of them explicitly adopting a neural network approach in the modeling of learning phenomena. Skinner famously criticized such use of hypothetical neural structures for the explanation of behavior (the “Conceptual Nervous System”), and one aspect of his criticism has proven to be correct: theory underdetermination is a pervasive problem in cognitive modeling in general, and in associationist and connectionist models in particular. That is, models implementing two very different cognitive processes often make the exact same behavioral predictions, meaning that important theoretical questions posed by contrasting the two models remain unanswered. We show through several examples that theory underdetermination is common in the learning theory literature, affecting the solvability of some of the most important theoretical problems that have been posed in the last decades. Computational cognitive neuroscience (CCN) offers a solution to this problem, by including neurobiological constraints in computational models of behavior and cognition. Rather than simply being inspired by neural computation, CCN models are built to reflect as much as possible about the actual neural structures thought to underlie a particular behavior. They go beyond the “Conceptual Nervous System” and offer a true integration of behavioral and neural levels of analysis.

scholarly journals Reduced positive evidence within activated self-schema may underpin increased sensitivity to negative evaluation in socially anxious individuals

2019 ◽  
Author(s):  
Alexandra Kathryn Hopkins ◽  
Raymond J Dolan ◽  
Katherine Susan Button ◽  
Michael Moutoussis
Keyword(s):  
Computational Models ◽  
Model Fitting ◽  
Well Being ◽  
Psychological Research ◽  
Negative Evaluation ◽  
Social Evaluation ◽  
Socially Anxious ◽  
Increased Sensitivity ◽  
The One ◽  
Associative Models

Positive self-beliefs are important for well-being, and are influenced by how others evaluate us during social interactions. Mechanistic accounts of self-beliefs have mostly relied on associative learning models. These account for choice behaviour but not for the explicit beliefs that trouble socially anxious patients. Neither do they speak to self-schemas, which underpin vulnerability according to psychological research. Here, we compared belief-based and associative computational models of social-evaluation, in individuals that varied in fear of negative evaluation (FNE). Using a novel analytic approach, ‘clinically informed model-fitting’, we replicated the finding that high-FNE participants learn faster from negative feedback about themselves. Crucially, this could be explained through reduced activation of positive self-schemas. The overall population could be characterized equally well by belief-based or associative models, but many individuals used either the one or the other perspective. Our findings have therapeutic importance, as belief activation may be used to specifically modulate learning

scholarly journals The Predictive Power of Spatial Relational Reasoning Models: A New Evaluation Approach

2021 ◽  
Vol 12 ◽  
Author(s):  
Marco Ragni ◽  
Daniel Brand ◽  
Nicolas Riesterer
Keyword(s):  
Cognitive Modeling ◽  
Computational Models ◽  
Predictive Power ◽  
Model Comparison ◽  
Cognitive Models ◽  
Relational Reasoning ◽  
Response Patterns ◽  
Research Focus ◽  
Evaluation Approach ◽  
The Individual

In the last few decades, cognitive theories for explaining human spatial relational reasoning have increased. Few of these theories have been implemented as computational models, however, even fewer have been compared computationally to each other. A computational model comparison requires, among other things, a still missing quantitative benchmark of core spatial relational reasoning problems. By presenting a new evaluation approach, this paper addresses: (1) developing a benchmark including raw data of participants, (2) reimplementation, adaptation, and extension of existing cognitive models to predict individual responses, and (3) a thorough evaluation of the cognitive models on the benchmark data. The paper shifts the research focus of cognitive modeling from reproducing aggregated response patterns toward assessing the predictive power of models for the individual reasoner. It demonstrate that not all psychological effects can discern theories. We discuss implications for modeling spatial relational reasoning.

scholarly journals Cognitive Models in Intelligence Research: Advantages and Recommendations for Their Application

2018 ◽  
Vol 6 (3) ◽  
pp. 34 ◽  
Author(s):  
Gidon Frischkorn ◽  
Anna-Lena Schubert
Keyword(s):  
Working Memory ◽  
Individual Differences ◽  
Cognitive Processes ◽  
Processing Speed ◽  
Cognitive Modeling ◽  
Working Memory Capacity ◽  
Memory Capacity ◽  
Cognitive Models ◽  
Model Fit ◽  
General Intelligence

Mathematical models of cognition measure individual differences in cognitive processes, such as processing speed, working memory capacity, and executive functions, that may underlie general intelligence. As such, cognitive models allow identifying associations between specific cognitive processes and tracking the effect of experimental interventions aimed at the enhancement of intelligence on mediating process parameters. Moreover, cognitive models provide an explicit theoretical formalization of theories regarding specific cognitive processes that may help in overcoming ambiguities in the interpretation of fuzzy verbal theories. In this paper, we give an overview of the advantages of cognitive modeling in intelligence research and present models in the domains of processing speed, working memory, and selective attention that may be of particular interest for intelligence research. Moreover, we provide guidelines for the application of cognitive models in intelligence research, including data collection, the evaluation of model fit, and statistical analyses.

scholarly journals Opening Burton's Clock: Psychiatric insights from computational cognitive models

2018 ◽  
Author(s):  
Daniel Bennett ◽  
Yael Niv
Keyword(s):  
Psychiatric Illness ◽  
Computational Models ◽  
Model Comparison ◽  
Model Fitting ◽  
Computational Modelling ◽  
Quantitative Model ◽  
Cognitive Models ◽  
Qualitative Comparison ◽  
Computational Psychiatry ◽  
Reinforcement Learning Models

Computational psychiatry is a nascent field that seeks to use computational tools from neuroscience and cognitive science to understand psychiatric illness. In this chapter, we make the case for computational cognitive models as a bridge between the cognitive and affective deficits experienced by those with a psychiatric illness and the neurocomputational dysfunctions that underlie these deficits. We first review the history of computational modelling in psychiatry and conclude that a key moment of maturation in this field occurred with the transition from qualitative comparison between computational models and human behaviour to formal quantitative model fitting and model comparison. We then summarise current research at one of the most exciting frontiers of computational psychiatry: reinforcement learning models of mood disorders. We review state-of-the-art applications of such models to major depression and bipolar disorder, and outline important open questions to be addressed by the coming wave of research in computational psychiatry.

scholarly journals On the measurement and estimation of cognitive processes with electrophysiological recordings and reaction time modeling

2021 ◽  
Author(s):  
Gabriel Weindel
Keyword(s):  
Reaction Time ◽  
Cognitive Processes ◽  
Cognitive Modeling ◽  
Model Fitting ◽  
Generative Model ◽  
Electrophysiological Recordings ◽  
Time Modeling ◽  
Experimental Effect ◽  
Decision Choices ◽  
Present Thesis

A primary goal in cognitive psychology is to describe the latent information processingunits that operate between the onset of a stimulus and a measured behavior. Mathe-matical models of cognition aim at decomposing behavior into such processing unitsby formalizing an assumed generative model. Unfortunately, a generative model mayexplain the behavioral data while not necessarily reflecting the underlying processes.Obtaining measurements between the stimulus and the responses could provideadditional information that fruitfully constrains the processing assumptions.The present thesis explores this issue by focusing on models of perceptual deci-sion making, a field with a long tradition of cognitive modeling. These models areconstructed to account for decision choices and their durations (reaction time inthe range of a second) on the basis of a decomposition into encoding, decision andresponse execution stages. We used electrophysiological measures (electromyographyand electroencephalography) to decompose each reaction time into different intervals,presumed to contain these stages. Simultaneously, we manipulated time-honoredexperimental factors to compare the cognitive locus of experimental effects inferredfrom both electrophysiological recordings and from model fitting procedures.Throughout four empirical chapters, we show that the inferences drawn from cogni-tive models conflict with the electrophysiological decomposition when: 1) the model’score assumption of independence between decision and non-decision processes isproven to be false; 2) standard modeling strategies are inadequate to capture thelocus of an experimental effect revealed by the electrophysiological decomposition;3) opposite experimental effects are revealed in decision vs. encoding and responseexecution processes.This thorough assessment of a generative model of decision making delineates itsvalidity, merits and limitations to account for the latent cognitive processes. Newinsights are thus provided on the information processes that allow humans to decidebetween alternatives.

ANALYSIS OF THE INFLUENCE OF ELEMENTS OF THE DIESEL COOLING SYSTEM ON ITS THERMAL STABILIZATION BY MEANS OF COMPUTER MODELING

2021 ◽  
Vol 75 (4) ◽  
pp. 3-12
Author(s):  
Zayats Yuriy Aleksandrovich ◽  
◽  
Zayats Tatiana Mikhailovna ◽  
Savelyev Maksim Anatolevich ◽  
◽  
...  
Keyword(s):  
Life Cycle ◽  
Computer Modeling ◽  
Cognitive Modeling ◽  
Cooling System ◽  
Target Function ◽  
Thermal Stabilization ◽  
Cognitive Models ◽  
Operating Conditions ◽  
Practical Significance ◽  
Mechatronic Systems

Logistics support of products at all stages of the life cycle is gaining increasing influence. This is facilitated by the increasing complexity of structures, a large number of elements, the intro-duction of mechatronic systems. Under these conditions, the relevance of developing methods for analyzing the design of samples increases. The developed model for analyzing the diesel cooling system is based on the principles of cognitive modeling. The practical significance of cognitive models is shown, which consists in the possibility of predicting changes in the influence of system elements on the target function in various operating conditions.

Are similes and metaphors interchangeable?

2017 ◽  
Vol 15 (1) ◽  
pp. 1-33 ◽  
Author(s):  
Manuela Romano
Keyword(s):  
Cognitive Processes ◽  
Cognitive Models ◽  
Figures Of Speech ◽  
Conceptual Semantic ◽  
Discourse Type ◽  
Opposite Process ◽  
Usage Patterns ◽  
News Interviews

Abstract Since Aristotle, scholars have regarded similes and metaphors as equivalent figures of speech sharing very similar comprehension, interpretation and usage patterns. By analysing the use of similes in real discourse, the aim of this study is to show that these two analogical figures reflect different cognitive processes, as well as different discursive functions, using as a framework cognitive models. To this end, this work presents, first, the main differentiating features of the two figures existing in the literature. And, second, it analyses 100 natural-occurring similes in English opinion discourse (news, interviews and commentary sections) in order to explain the conceptual-semantic and formal-syntactic factors which explain why similes and metaphors are not interchangeable in the discourse type under study; that is, why metaphors can usually be transformed into similes by adding like, whereas the opposite process seems to depend on specific conditions of structure, use and interpretation.

Sign in / Sign up

Export Citation Format

Share Document