Theories

What is a psychological theory?

A theory explains why a pattern occurs. In psychology, theories are built from constructs—abstract entities like "working memory," "attachment," or "belief"—and the relationships between constructs (e.g., "higher working-memory capacity → better reasoning under load").

Specific theories vs. general frameworks. A specific theory makes narrow, testable predictions; a general framework (like reinforcement learning) organizes many phenomena but predicts less precisely on its own.
The cost of a bad theory. A compelling but false theory can misdirect a whole literature for decades. Theory choice is consequential, not merely decorative.

How do we test theories?

Falsificationism. A theory earns its keep by surviving attempts to refute it. A claim that cannot be disproven is not scientifically useful. Good experiments specify in advance what result would count against the theory.
A holistic viewpoint. In practice we never test a theory in isolation—auxiliary assumptions (about the sample, the measure, the statistics) ride along. When a prediction fails, something in the bundle is wrong; figuring out which piece is the hard part.

Designing experiments to test theory

The most informative experiments are those where competing theories make different predictions. If two theories predict the same outcome either way, the experiment cannot discriminate between them. Plan the manipulation and measure so that the data can actually pit theories against each other (see Design and Measurement).

Formalizing theories

A theory stated in words is flexible; a theory stated in equations or code is committed. Formalizing a theory—as a computational model or a set of equations—forces precision about constructs and relationships, makes predictions checkable, and reveals hidden assumptions. Even a simple formal model clarifies what a verbal theory actually claims.

Key terms

Construct — an abstract theoretical entity (not directly observable).
Operationalization — translating a construct into a concrete, measurable indicator.
Falsifiability — the capacity to be shown wrong by evidence.