Lecture 7 Flashcards
1
Q
What is an analogy?
A
- A relation of similarity/isomorphism between two domains of phenomena
- parts of the world
- cognitive process of transferring conceptual resources from one domain to another
2
Q
How does an analogy help researchers?
A
- Discover similarities and identities between domains
- Generalise over and unify distinct domains
- Create and extend scientific concepts
3
Q
What are the four components of an analogy?
A
- source domain
- target domain
- mapping
- relations
4
Q
What is a source domain?
A
- Domain of phenomena that supplies the terms of the analogy: typically more familiar to us
- The thing you know everything of
5
Q
What is the target domain?
A
- Domain of phenomena to which we apply the analogy: typically less familiar to us
- The thing you don’t know anything of
6
Q
What is mapping?
A
- Set of correspondences between terms in the source and terms in the target domain
- Translation of terms; doesn’t make any claims about the domains
7
Q
What are relations?
A
- Claims that, under the mapping, hold for both the source and target domain
- Claims are found in relations
8
Q
What are three classes of analogies?
A
- Positive analogy: we know already that these relations hold in both the source and target domains
- Negative analogy: we know that these relations hold in one domain but not in the other
- Neutral analogy: We know that these relations hold in the source domain, but we do not know if they hold in the target domain
9
Q
What is a model?
A
- A simplified representation of a domain of phenomena
- An application of a scientific theory to a particular case
- Many scientific models are based on analogies
10
Q
How are models idealisations?
A
- Models are idealisations because they deliberately in order to make reality more tractable
11
Q
What are four types of models?
A
- Abstract models: fictional entities and mathematical equations (atom)
- Material models/concrete models: concrete objects (you can touch them; MONIAC as analogy between machine and economy)
- Computer simulations: equation-based simulations (something goes into the computer, something else comes out)
- Phenomenological models/data models: represent only data or observable properties
