Mem and Cog Ch.8 Cases

Sentance Verification Task

Verify if a sentence is true or false.

Hierarchical Semantic Network Model

A net-like organization of concepts where concepts are nodes that are connected by links.

Superordinate

Name of a category. e.g. Bird

Subordinate

A member of a category. e.g. Canary

Cognitive Economy

Properties and facts are stored at the highest level possible.

Category Size Effect

Definition: the level within the hierarchy

Procedure: Sentence Verification Task

Prediction: Reaction time will increase as the semantic distance increases

Results: The closer the stored fact is to the node, the less time it takes to verify the task.

Collins and Quillian

Problem with Cognitive Economy

Procedure; Sentace verification task, with superordinate and subordinate subjects

Prediction: Reaction time will increase for superordinate categories

Results: Reaction was the same for all sentences

Problem with Hierarcy

Procedure: Sentence Verification Task with common and uncommon categories

Prediction: Reaction time will be better for the categories that are stored higher

Results: Reaction time was better for more commonly used categories

Typicality Effect

Typical examples are easier to recognize.

Spread Activation Theory: Collins and Loftus

Related concepts are connected, and activation spreads to related nodes. Not hierarchical; no cognitive economy.

Category Membership Judgment Task

The participant determines whether an item belongs to a certain category. Longer reaction time when the category boundaries are unclear.

Classical View

A list of necessary and sufficient features. Concepts have precise definitions and features.

Prototype View

The idealized representation of some class of objects from their mental averages.

Bruner Et Al Classical View Study

Procedure: Identify the rule or concept the experimenter predetermined.

Strategies: Simultaneous Scanning: Testing several hypotheses at once

Successive Scanning: Testing hypotheses one at a time

Conservative Focus: Start with one and slowly tweak until correct. (Most common)

Problems: It’s hard to generate a list of features that are necessary and sufficient to specify membership for each individual category.

Family Resemblence

Member share features. Explains typicality.

Rosch and Mervis Prototype View Study

Procedure: Provide a feature for a category → Provide feature for a example of said category → calculate the overlap

Results: Objects sharing many features with the superordinate were better examples of the category. The superordinate feature list didn’t work for all objects ( the classical view didn’t work).