-> Template-Matching Theory [TEMPLATES]
Stimuli are compared to a set of templates (specific patterns stored in memory), and matched with the closest. (Some born with but others developed along life)
Problem: Cannot account for recognising stimuli as the same context, even though presentation is different, e.g. upside down.
- Would require too many templates
- Concrete and not flexible
- Unable to comprehend parts of objects
- Unable to account for the complexity of human perception
-> Prototype Model Theory [PROTOTYPES]
We have prototypes (abstract idealised patterns) in memory, and matched but don't have to be exact. If close enough, you can recognise the stimuli, if inadequate then keep searching.
- Prototypes are good for accounting the flexibility of perception.
- Demonstrates creation of capacity and applying it.
Reed's Experiment: People forming own prototypes of the two sets of faces.
- New faces were compared to the prototypes and the classification was consistent.
Distinctive-Features Models Theory [DISTINCTIVE FEATURES]
We discriminate between items on the basis of a small number of characteristics, not the entire stimulus. Mostly used with testing letter recognition, e.g. Eleanor Gibson.
- Recognise letters by distinctive features of letters (matrix of distinctive features).
- More similar = easier to group, more distinct = harder to group.
- Sub components that build up.
Recognition-by-Component Theory [GEONS]
Objects are represented as an arrangement of simple 3D shapes called geons.
- 24 geons that can be combined to create a meaningful object and words.
- Majority are classified with three geons.
No one theory of perception has explained the ability to perceive so much complexity so quickly.