Understanding Theories, Hypotheses, and Scientific Testing

Posted on Mar 27, 2025 in Philosophy and ethics

Theories, Hypotheses, and Contrast

The goal of science is to obtain as much information as possible about the phenomena of reality and their interconnections. A phenomenon is defined as anything we can grasp or perceive. Science uses theories, which are complex sets of definitions.

Theories, like definitions, are complex statements explaining how to use or apply a term in relation to other terms. They do not inherently describe reality and are neither true nor false in themselves, but rather trivially true by definition.

Understanding Theories and Systems

Theories define types of natural or real systems. A system is a set of objects or properties considered inter-related or as a unit. Systems respond to their conditions (the specification of system characteristics at any given time) in two main ways:

  • Deterministic systems: The system’s state completely determines its past and future states.
  • Probabilistic or stochastic systems: The system’s state only determines the probability of possible future states.

The general formula of a theory can be stated as: Theory X: A given real system is of the type defined by Theory X if and only if (iff) it satisfies the laws specified by X (i.e., it behaves as Theory X predicts).

The Role of Hypotheses

It might seem contradictory that science uses theories (which are definitions and don’t directly describe reality) to gain information about reality. This apparent contradiction is resolved through the use of hypotheses.

Theoretical hypotheses are statements asserting that a specific real or natural system conforms to a particular theory (i.e., it behaves as the theory predicts). Their general formula is: Hypothesis X: A specific real system is of the type described by Theory X.

The process of obtaining information to support or refute (negate) a hypothesis is called experimental contrasting or testing.

Other Key Scientific Concepts

Science also employs other important terms:

  1. Laws of Nature: These are hypotheses tested so extensively that nature appears to follow them consistently. They often become part of the statements within theories.
  2. Uses of the word “Model”:
    • Scale Model: System A is a scale model of system B means A is a physical reproduction of B at a specific scale. (This usage is less common in theoretical science.)
    • Model as Analogy: A is a model of B means A is a well-understood system, B is poorly understood, and it’s proposed that A and B are similar in important ways. (This type of model can inspire new theories.)
    • Model as Theory (Approximate): A is a model of B means A is a theory, and B is a real system that approximates the type described by A. This approximation might only be vaguely true. (This is common in psychology and social sciences, used when detailed theories are lacking but some information is still valuable.)

Experimental Contrasting of Hypotheses

This is the process for obtaining information to support or refute a hypothesis. It involves understanding how manipulations, experiments, observations, or findings yield the necessary information.

Elements of Hypothesis Testing

  • Prediction (P): A statement describing what should happen to a system if the hypothesis is true.
  • Initial Conditions (IC): A description of the system’s state or properties at the beginning of the experiment. These must be distinct from the prediction.
  • Auxiliary Assumptions (SA): Statements describing necessary background conditions or assumptions required for the predicted event to occur if the hypothesis is true.

Rationale for Hypothesis Testing

Hypothesis testing is possible because the prediction (P) should logically follow from the hypothesis (H) combined with the initial conditions (IC) and auxiliary assumptions (SA). Conversely, if the hypothesis (¬H) is false, given the same IC and SA, the prediction (P) is unlikely to occur.

In summary:

  • Condition 1: If H (and IC and SA) are true, then P is expected. (H & IC & SA) → P
  • Condition 2: If H is false (¬H) (and IC and SA are true), then P is unlikely. (¬H & IC & SA) → ¬P (likely)