Scientific Models

Question 1

define a scientific model

Answer 1

A scientific model is a machine for predicting how part of reality will behave

Question 2

give the Key principles of modelling

Answer 2

Generation of testable predictions: A model which makes no testable predictions is useless

Elegance or simplicity: Occam’s Razor ‘Entities should not be multiplied without cause’.

Question 3

What can be deduced by the fact a model is logico-deductive?

Answer 3

Because a model is logico-deductive, you
can deduce that situations could exist in
the world which have never been observed.

Question 4

Give 3 Types of research

Answer 4

● Develop and test new theories about reality
● Develop logical extensions of theories
● Use logical extensions to develop new
technology

Question 5

What is Shannon entropy?

Answer 5

The Shannon entropy equation provides a way to estimate the average minimum number of bits needed to encode a string of symbols, based on the frequency of the symbols.

Question 6

What are the criteria for a model becoming a law?

Answer 6

Sciences designate as laws those models that:

1. Have a simple, elegant formulation with few parameters
2. Make excellent predictions in an apparently unlimited number of cases

Question 7

Summary points of scientific models are:

Answer 7

● A model must make testable predictions to be scientifically meaningful.
● Information theory gives us a uniform means to measure both models and the
predictions of models in order to evaluate their adequacy.
● One should not be afraid to make use of it.