Lecture 9: Representation of Knowledge Flashcards
1
Q
What is procedural knowledge?
A
- How to do things
- Type of knowledge required through practice, usually extensive practice
- Hard to extinguish or change it (e.g., playing the piano. Chris’ father in law playing golf example)
2
Q
What is declarative knowledge?
A
- Facts, events (things you know and have experienced)
- Acquire with organization (not just randomly stored)
- Easily forgotten or changed (relative to procedural knowledge. Recall eyewitness studies)
3
Q
What are the two subtypes of declarative knowledge?
A
- Episodic: autobiographical (subjective; idiosyncratic). Temporal tags (e.g., when is the first time you learned to ride a bike, first day of university etc.)
- Semantic: factual, words, concepts. Not referenced to own experiences (e.g., we know what the word house means but it doesn’t mean that we remember the day we learned the word house)
4
Q
How did Cohen and Corkin (1982) examine explicit episodic learning vs. implicit procedural learning?
A
- Compared normals vs. amnesics (like HM)
- ## Task: “Tower of Hanoi puzzle” (Objective: move entire set of disks to another rod. But only one disk can be moved at a time, disks must be move onto another rod, no disk may be placed on top of a smaller disk)
5
Q
What were the results of the study by Cohen and Corkin that examined explicit episodic learning vs. implicit procedural learning?
A
- Normals: learns set of procedures. Transfer to other tasks and have episodic memory of solving the puzzle.
- Amnesics: can learn task! also show transfer (meaning that knowledge can be applied to other types of puzzles)! but no episodic memory
- Shows: a dissociation between explicit (episodic) memory & implicit (procedural) knowledge
6
Q
Who was Henry Molaison?
A
- Retrieval of information from LTM intact
- STM (working memory) intact
- But, could not transfer new declarative info to LTM
- i.e., explicit storage of new info a problem
- this included episodic information (experiences)
7
Q
What was the evidence that supported that HM’s working memory was intact?
A
- HM had no recollection of doing this task from one
day to the next - However, HM improved in this motor learning task!
- Evidence for implicit learning and memory
8
Q
What is semantic memory?
A
- Permanent memory of general world knowledge
- Semantic memory is highly organized
- Eye witness studies (lecture 08)
- Psychoanalysis relies on organization
- Aim is to examine and formalize this organization
9
Q
What different brain processes does semantic memory use?
A
Semantic memory uses different brain processes than episodic memory:
- episodic memory: hippocampus and anterior prefrontal cortex
- semantic knowledge: lateral temporal lobes, posterior prefrontal cortex
10
Q
How did Tulving examine the organization of semantic memory?
A
- Subjective organization
- Present a list of unrelated words, several times & in different orders
- Not semantically related; Do people impose their own organization on these items? Wanted to get an index of this grouping because it shows that organization is subjective. Does this organization have a role in overall memory?
- If two words were grouped together on one trial and the then the same two words later you got points this was a measure subjective organization
- From list to list, people started to impose subjective organization on pairs of words. Which drove an increase in recall. Creating a semantic memory network.
11
Q
What are the 3 models of semantic memory?
A
Feature Model: - Semantic Feature-Comparison Model Two Network Models: - Hierarchical Network Model - Spreading Activation Model
12
Q
What is the structure of the semantic feature comparison model?
A
- knowledge consists of sets of features (We know a robin is a bird because we are comparing features. The features of a robin to the category bird.)
- features weighted on n-dimensional space
- small ——————————————- large
- defining: - important (critical core features that define that category)
- characteristic: - less important (often seen within this category, but they don’t necessarily have to be there)
13
Q
What is the decision process involved in the semantic feature comparison model?
A
- Sentence to Verify (e.g., Is a robin a bird?)
- Stage 1: Global Feature Comparison (features activated for both robin and bird)
- Low Overlap Fast “No”, High Overlap Fast “Yes”, Intermediate Score Go to Stage 2
- Stage 2: Compare Defining Features
- Features Mismatch slow “no”, Features Match Slow“Yes”
14
Q
What evidence exists for the semantic feature comparison model?
A
- Typicality ratings (robin and sparrow rated as most typical, then pigeon, hawk, chicken)
- All birds have most defining features, so assume that typicality reflects difference in characteristic features.
15
Q
What is the problem with the semantic feature comparison model?
A
- Disconfirming sentences
- Responses should be faster to collies are birds STAGE 1 than to Collies are poodles STAGE 1 + 2
- Results are opposite! People should be faster to say that collies are not birds because they don’t share a lot of features, but they aren’t (people say collies are not poodles faster)
