PSYC 350: Human Memory (Unit 6)

Generic/Semantic Memory

Generic memory (semantic memory)

Memory for facts, concepts, and meanings that is context free and not associated with a particular point in time

Concept

A general idea/mental representation of something derived or inferred from specific instances or occurrences; make up semantic memory

Semantic dementia

Impaired generic memory due to damage to the anterior temporal lobe

Hierarchical Network Model (Quillian)

Model of generic memory in which concepts are organized from the most general at the top to the most specific at the bottom, with facts about a concept attached to the highest level of the hierarchy to which they apply

Sentence-verification task

Technique primarily associated with the hierarchical network model that presents participants with a statement and asks them to verify as quickly as possible whether it is true or false

Critique of the hierarchical network model

1. Does not explain the speed with which people verify statements as false

2. Frequency is a better predictor of verification speed than level of the hierarchy

3. Some statements that do not correspond to hierarchy level are verified faster than those that do

4. Typicality effect: verification times are faster for more typical instances of a category, even when they are at the same hierarchy level

Spreading Activation Model (Collins and Loftus)

Model of semantic memory that suggests that related concepts are connected in a network, and concepts that share more properties have more links between them; when one concept is activated, activation spreads to nearby concepts

Decision criterion

A threshold of activation that must be met in order for an item to be retrieved from semantic memory

Critique of the spreading activation model

1. Very hard to disprove

2. Cannot explain mediated priming (as far too many concepts would need to be activated than is practical)

3. Predicts priming effects are shorter in duration than they are (as in theory, new concepts should be activated and the primed concept should deactivate)

4. Assumes each concept in a network has a single, fixed representation

5. Suggests localized representations in the brain, when really semantic memories are stored as patterns of activation that are distributed throughout the brain

Compound-Cue Model

Model of generic memory that assumes recognition of an item is based on a measure of familiarity that is determined by the compound events of multiple cues, including context, associations with recently presented items, and associations in memory

Three types of cues (compound-cue model)

1. Context cues: the context within which a memory was learned

2. Inter-item cues: the items that were learned at the same time as a memory

3. Self-self cues: sense of familiarity caused by an item cuing itself

Hub-and-Spoke Model

Model of generic memory that proposes that six types of modality-specific representations (visual, verbal, smells, sounds, motor/praxis, somatosensory) meet at a central hub in the anterior temporal lobe

Brain areas associated with generic memory

• Hippocampus (encoding) and neocortex (storage); semantic memories are initially episodic memories
• Left inferior prefrontal cortex (controlled searches of generic memory)
• Anterior temporal lobe (integration of concepts)
• Sensory/motor areas (concepts related to that modality)
• Semantic memories are distributed across the brain rather than localized

Evidence for concepts as modality-specific

• Damage to brain regions produce impaired conceptual processing that differs demanding on the damaged area
• Sensory/perceptual brain areas are active during completion of conceptual tasks
• Motor areas are active when listening to action-words that correspond to that motor area

Evidence for concepts as distributed

• Concepts are modality-specific, which representations being spread across sensory modalities
• Individuals with neurodegenerative diseases do not lose access to all parts of a concept at once
• Different features of a concept are activated depending on the concept, suggesting distinct features are coded by different groups of cells

Evidence for concepts as learned

• Conceptual features are coded from direct perceptional and action-based experiences, which is why living things tend to rely on perceptual processing and nonliving things tend to be rely in motor processing

• In a study where participants were shown novel objects and either pantomimed interacting with it or pointed at it, only the pantomime group showed activity in motor areas when recalling the object

Evidence for concepts as flexible

• Concepts contain both context-independent (always activated) and context-dependent (sometimes activated) features, and people verify context-dependent, but not context-independent statements faster when they are preceded by a relevant statement
• Different brain areas are activated depending on the relevant feature of a concept

Encoding abstract concepts

Abstract concepts are grounded largely in their relationship to other words, and may be modality-specific and linked to perceptual, motor, or emotional areas of the brain