Concepts & Categorisation (Week 5) Flashcards
Knowledge
Concepts: mental representation useful for cognitive efficiency.
- logical concepts.
- natural concepts.
Categorisation: the process by which things are placed into groups.
Why categorise?
Reduces complexity of the environment (cognitive economy)- when we see new things we haven’t seen before, if we can assign them to a concept/category it allows us to infer a number of things similar to them.
Allows us to recognise novel patterns.
Allows us to establish hierarchies of objects.
Logical concepts: the strict rule based version.
Concept identification: defined by logical rules:
- conjunctive rule: use the logical relation ‘and’ (striped and square).
- disjunctive rule: used the logical relation ‘or’ (striped or square).
- conditional rule: used the logical ‘if’, ‘then’ relation (if striped it must be a square).
- biconditional rule: used the logical ‘if’, ‘then’ in both directions (if striped then square; if square then striped).
Logical concepts: attributes + rule
Attributes= distinct features of objects. Rule= logical relationship between attributes. Exemplar= an object that satisfies a concept.
Conjunctive rule: AND (for two attributes)
Both relevant attributes must be present to be a representative of the concept.
Disjunctive rule: OR (for two attributes)
Either of the relevant attributes must be present to be a representative of the concept. An object may have both attributes.
Conditional rule: IF 1 THEN 2
If the first attribute is present, the other must be present. Any object that doesn’t include the first attribute also fits the category.
Biconditional rule: IF 1 THEN 2; IF 2, THEN 1
The conditional rule applies both ways. Both attributes must be present, or both absent.
This strict rule based version
One task in concept identification involved rule learning: figure out the rule given attributes.
Bourne (1970) tested this using a rule learning procedure.
Findings:
- across successive problems, people get better at this task.
- on the very first trial- it takes quite a bit of time, especially biconditional relationships.
Another task: give the rule and let them discover the attributes.
- studies show attribute learning is also affected by the rule.
- frequency theory explains these differences as a function of exposure to the attributes.
- different rules provide different amounts of exposure to the attributes.
- more exposure= quicker learning.
Criticisms of logical concept identification
Highly artificial and not like natural categories.
Natural categories are not as clear cut:
- natural categories are characterised by typicality gradients (eg. a robin is more bird like than a penguin; RTs support this).
Natural categories are organised hierarchically.
Many natural categories are continuous.
Some categories have fuzzy boundaries.
Natural categories: as a hierarchy
Superordinate level:
- highest level of abstraction.
- consists of general categories.
- consists of only a few defining attributes.
- eg. furniture, bird etc).
Subordinate level:
- lowest level of abstraction.
- consists of specific types of objects.
- consists of many attributes.
- eg. end table, song-sparrow etc.
Basic level: exists in between the two extremes.
- is a balance between informativeness (number of attributes the concept conveys) and economy (the summary of the important attributes).
- usually acquired first by children.
- recognised more quickly by non experts (Rosch et al, 76).
- eg. table, sparrow etc
Evidence that basic-level is special
People almost exclusively use basic-level names in free-naming tasks.
Quicker to identify basic-level category member as a member of a category.
Children learn basic-level concepts sooner than other levels.
Basic-level is much more common in adult discourse than names for superordinate categories.
Different cultures tend to use the same basic-level categories, at least for living things.
Characteristics within a category
Family resemblance: proposed by Wittgenstein (1953).
- continuous variable.
- measure of the overlap between members with a category.
- is measured by number of shared attributes.
Typicality:
- refers to the differences in how well members relate to their category (eg. collie vs. dachshund).
- the higher the family resemblance the more typical the item is for common taxonomic categories.
- Barsalou (1985) argues that family resemblance and typicality are not correlated for goal-derived categories (eg. make people happy when you give a birthday present).
The defining-attribute view
Frege (1952) said a concept can be characterised by a set of defining attributes.
- intension: the set of attributes that define what it is to be a member of the concept (eg. bachelor- male, single, adult).
- extension: the set of entitites that are members of the concept (eg. bachelor- every bachelor).
All of the attributes must be present.
Assumes categories are clearly defined and rigid.
Defining attributes vs. characteristic attributes.
Definitional approach to categorisation
Determine category membership based on whether the object meets the definition of the category.
Does not work well.
Not all members of everyday categories have the same defining features (eg. chairs all look very different- some have legs and arms and some don’t).
