What is the key difference between semantic and episodic memory?
Semantic memory contains general world knowledge with no record of when or where it was learned, whereas episodic memory includes contextual details about personal experiences (time, place, event).
How do reaction time studies provide evidence for the organization of semantic memory?
Sentence verification tasks measure reaction time (RT). Differences in RT (e.g., verifying “a canary is a bird” vs. “a canary is an animal”) indicate how information is stored and accessed in semantic networks.
Describe the basic structure of a semantic network
A semantic network consists of nodes (concepts) connected by pathways (associations).
Explain the concept of spreading activation and how it works in semantic networks.
When a concept is activated, activation spreads to connected concepts, making them easier to retrieve. Activation travels through pathways and weakens over distance/time.
What are the main limitations of the Collins and Quillian hierarchical network model?
- Hierarchy does not always predict reaction time (e.g., “chimpanzee is an animal” is sometimes faster than “chimpanzee is a primate”).
- Typicality effect: typical examples (robin → bird) are verified faster than atypical ones (ostrich → bird), which the model cannot explain.
How does the Smith, Shoben, and Rips feature list model differ from the hierarchical network model?
It replaces networks with feature lists. Concepts are stored as lists of defining and characteristic features, and verification occurs in two stages (global feature overlap → defining features).
What is the typicality effect and how does it support feature comparison models of semantic memory?
Typical category members share more features with the category, producing faster Stage 1 “yes” responses (e.g., robin vs. chicken). This supports the idea that feature overlap, not strict hierarchies, drives verification speed.
What are the advantages of the ACT-R network model over earlier network models?
- Does not require strict hierarchies.
- Link strength varies based on frequency of experience.
- Explains both typicality and reaction time problems of earlier models.
What is priming and how is it studied in the context of semantic memory?
Define connectivity and resonance effects and explain how they influence memory.
What is the DRM effect and how does it demonstrate the reconstructive nature of memory?
How does network size influence memory differently in associative cuing tests versus cued-recall tests?
Explain how schemas can lead to both accurate and inaccurate memories.
Connectivity, Resonance, and the DRM Effect
This section explores the impact of network connectivity and resonance on memory. It explains that items with densely connected and resonant associates are better remembered. The section then introduces the Deese, Roediger, & McDermott (DRM) effect, a false memory paradigm demonstrating how strong existing networks can lead to the recall of unstudied items.
The Influence of Network Size (Set Size Effects)
This section investigates how network size affects memory performance depending on the testing method. It explains the “set-size effect,” where items with smaller pre-existing networks are better remembered in associative cuing tests but not in cued-recall tests. It highlights how the presence of a cue during study can constrain activation spread and mitigate the set-size effect.
Schemas: Organizing Knowledge and Shaping Memory
A. Defining Schemas and Scripts: This section introduces schemas as integrated chunks of knowledge and focuses on “scripts,” which are schemas for events and their sequences. It provides the example of a typical restaurant script to illustrate how schemas organize our understanding of common situations.
C. Schemas in Everyday Life: This section examines Brewer and Treyens’ (1981) study on schema-driven memory errors in an office setting. It highlights how participants were more likely to recall and recognize schema-consistent objects, even those that weren’t present, demonstrating the powerful influence of schemas on retrieval.
Semantic Memory
A type of long-term memory that stores general knowledge about the world, concepts, and language. It is not tied to specific personal experiences.
Episodic Memory
A type of long-term memory that stores specific personal experiences and events. It includes details about what happened, where, and when.
Network Model
A representation of semantic memory where concepts are represented as nodes connected by links. The strength of the links reflects the relationship between the concepts.
Spreading Activation
A process in network models where activation spreads from one node to another along connecting links.
Priming
A phenomenon where exposure to a stimulus makes it easier to process related stimuli.
Typicality Effect
The finding that it is easier and faster to verify typical members of a category than atypical members.
Feature Overlap Model
A model of semantic memory that represents concepts as lists of features. The more features two concepts share, the more closely related they are.
Hierarchical Network Model
A network model where concepts are organized in a hierarchy from general to specific.
