Mathematical Psychology: Computational Modeling of Cognition and Behavior

A comprehensive set of flashcards covering key concepts from the lecture on mathematical psychology, focusing on computational modeling of cognition and behavior.

Mathematical Psychology

The field that uses mathematical models to represent cognitive processes and behavior.

Computational Modeling

The use of mathematical models to simulate and analyze cognitive mechanisms.

Falsifiability

A principle stating that a theory must be structured in a way that allows it to be tested and potentially disproven.

Descriptive Models

Models that describe data patterns without assuming the underlying cognitive processes.

Cognitive Process Models

Models that make theoretical assumptions about cognitive processing rather than just describing data.

Overfitting

A situation in which a model fits the training data too closely, resulting in poor generalization to new data.

Model Comparison

The process of evaluating and comparing different models to determine which one best fits the observed data.

Maximum Likelihood Method

A statistical method used to estimate parameters of a model by maximizing the likelihood of observing the given data.

AIC (Akaike’s Information Criterion)

A criterion for model selection that penalizes models for having more parameters.

BIC (Bayesian Information Criterion)

A criterion for model selection that also considers the model's complexity.

Neural Network Models

Models based on networks of interconnected nodes, similar to neuron connections in the brain.

Simulation

The process of using computational models to replicate human behavior in experiments.

Reinforcement Learning

A type of learning where an agent improves its performance based on feedback from its environment.

Power Law

A mathematical relationship often used to describe phenomena where one quantity varies as a power of another.

Exponential Law

A mathematical relationship where a quantity increases or decreases at a rate proportional to its current value.

Cognitive Aid

A tool or model that helps scientists to communicate and define cognitive processes more precisely.

Latent Psychological Properties

Underlying psychological characteristics that cannot be directly observed but can be inferred from data.

Expected Value

The anticipated value or payoff of a decision, based on probabilities and outcomes.

Sensitivity to Gains

The degree to which individuals respond more strongly to positive outcomes compared to negative ones.

Trial-Level Estimates

Values or predictions made for individual trials in an experiment, reflecting specific conditions.

Critical Question in Falsifiability

What specific data patterns could potentially disprove a given theory?

Parametric Models

Models that quantify relationships using parameters, enabling explicit predictions.

Non-Serious Paper

A publication intended as a joke or satire, rather than a serious scientific contribution.

Mean Squared Deviation

A measure of the average squared deviations between observed and predicted values.

Connectionist Models

Models based on networks that simulate the way neurons interact in the brain.

Backpropagation

An algorithm used in neural networks to adjust weights as the model learns.

Intellectual and Scholarly Judgment

The process of applying critical thinking and expertise to model evaluation beyond quantitative metrics.

Psychological Plausibility

The degree to which a model realistically reflects human cognitive processes.

Trade-off between Model Fit and Generalizability

The balance between a model's accuracy in fitting the training data and its ability to predict new data.

Interference Strength

The degree to which competing information hinders retrieval of the target memory.

Cognitive Models

Mathematical representations that aim to simulate cognitive processes.

Stress Condition

An experimental condition where participants are subjected to stress to observe its effects on behavior.

Control Condition

An experimental baseline condition that allows for comparison against a treatment or stress condition.

Equations as Precise Predictions

Mathematical expressions that provide exact forecasts of outcomes in experiments.

Recency Effect

The tendency to better recall most recently presented information.

Law of Parsimony

The principle that, given competing hypotheses, the simplest explanation is preferred.

Computational Psychiatry

The study of mental disorders through the lens of computational and mathematical modeling.