Modeling Intelligence, Cognitive Models, and Reverse Engineering

Vocabulary flashcards focusing on definitions of intelligence, types of scientific models, Allen Newell's critique, and reverse engineering in cognitive science.

General Definition of Intelligence

A property measuring an agent's ability to achieve goals in a wide range of environments.

Intelligence (S. S. Colvin)

The degree to which a person has learned, or can learn, to adjust himself to his environment.

Intelligence (W. V. Bingham)

The ability of an organism to solve new problems.

Intelligence (American Psychological Association)

Individual differences in the ability to understand complex ideas, adapt effectively to the environment, learn from experience, engage in various forms of reasoning, and overcome obstacles by taking thought.

Intelligence (W. F. Dearborn)

The capacity to learn or to profit by experience.

Intelligence (R. Pinter)

The ability to adapt oneself adequately to relatively new situations in life.

Successful Intelligence (R. J. Sternberg)

An individual's skill in achieving whatever it is they want to attain in life within their sociocultural context.

Scientific Models

Representations of objects, systems, or events used as tools for understanding the natural world, often using familiar objects to represent unfamiliar things.

Good Theoretical Model (Yakov Frenkel)

A theoretical model of a complex system that acts like a good caricature by emphasizing the most important features and downplaying inessential details.

Closed-Loop Science

An iterative scientific cycle where knowledge is integrated to build a model, prediction accuracy is tested, experimental results refine the model, and new knowledge is generated.

Simple Quantitative Models

Models featuring easily understandable parameters that formalize underlying principles, generate quantitative predictions, and are simple to communicate.

Allen Newell's Critique (1973)

A foundational critique arguing that psychology's strategy of collecting lists of phenomena and asking binary questions ('playing 20 questions with nature') fails to lead to clarity or unified theoretical understanding.

Reverse Engineering

The methodology of building candidate runnable brain-body systems—defining architecture, objective functions, learning rules, training data, and morphology—to predict brain mechanisms and behavior.

Runnable Models

Executable computational models in mature sciences that can generate observed phenomena from first principles and integrate results across many tasks.