Lecture 18 Memory II

Memory Systems II

Goals

  • Explore semantic memory and its neural basis.
  • Discuss the Complementary Learning Systems (CLS) theory of memory.
  • Examine the distinction and interaction between episodic and semantic memory.
  • Highlight the neural basis of procedural memory and its dissociation from declarative memory.
  • Investigate the role of the hippocampus in memory consolidation and replay.
  • Understand the impact of neurogenesis on learning and memory.

Semantic Memory

Semantic memory refers to context-free knowledge about the world, such as the meaning of words, facts, and object use. Unlike episodic memory, it does not rely on recalling where or when the information was learned.

Key Features

  • Neurological Basis:
    • Depends on the anterior temporal lobe (ATL), which acts as a multimodal hub integrating information from distributed cortical networks.
  • Semantic Dementia (SD):
    • A subtype of frontotemporal lobar degeneration (FTLD) characterized by progressive loss of semantic memory.
    • Patients lose the ability to identify and categorize objects or concepts (e.g., referring to a sheep as “those things”).
    • Memory for specific features erodes, leading to generic or incorrect representations (e.g., drawings of animals lose distinguishing traits).

Evidence from SD

  • Case Study of Mr. M:
    • Retained spatial memory but failed to recognize sheep despite frequent exposure to wool and lamb products.
    • Demonstrates the role of ATL in integrating sensory and conceptual knowledge.

Episodic vs. Semantic Memory

Dissociations

  • Developmental Amnesia:
    • Children with hippocampal damage show severe episodic memory deficits but normal semantic learning.
  • Semantic Dementia:
    • Episodic memory remains intact while semantic memory deteriorates.
  • Case of Patient K.C.:
    • Could not recall personal events but retained factual knowledge and acquired new semantic information post-damage.

Interdependence

  • Semantic memory is often derived from decontextualized episodic memories.
  • Episodic memories are scaffolded on existing semantic knowledge (e.g., recognizing tools in a restaurant episode).

Complementary Learning Systems (CLS) Theory

The CLS theory, proposed by Jay McClelland and colleagues, describes two distinct but complementary memory systems:

  1. Hippocampus (Fast Learning System):
    • Rapidly encodes individual experiences and events.
    • Key processes:
      • Pattern Separation: Differentiates similar events by creating distinct codes.
      • Pattern Completion: Reconstructs entire memories from partial cues.
      • Replay: Trains neocortical networks during sleep and rest to consolidate memories.
  2. Neocortex (Slow Learning System):
    • Gradually extracts structured knowledge by integrating information over time.
    • Supports generalization and abstraction.
    • Avoids catastrophic forgetting through interleaved learning with hippocampal input.

Neural Basis of Memory Consolidation

Hippocampal Replay

  • During sleep, hippocampal place cells replay sequences of activity from prior experiences.
  • Replay trains neocortical networks, aiding long-term memory storage and generalization.
  • Experimental disruption of replay impairs memory consolidation.

Declarative Memory

  • Anterograde Amnesia: Inability to form new memories post-damage.
  • Retrograde Amnesia: Loss of recent memories preceding damage, with older memories remaining intact after consolidation.

Procedural Memory

Procedural memory refers to the implicit learning of skills and tasks, distinct from declarative memory.

Dissociations

  • Patient H.M.:
    • Retained skill in mirror tracing despite severe episodic memory deficits.
  • Weather Prediction Task:
    • Parkinson’s disease patients (basal ganglia damage) struggle with probabilistic learning but retain episodic memory.
    • Amnesic patients (hippocampal damage) show the opposite pattern.

Neural Networks and Learning

Artificial neural networks (ANNs) model aspects of brain function, offering insights into learning and memory: – Perceptrons: – Simple classifiers that inspired early neural network designs. – Backpropagation: – Optimizes network performance by adjusting weights based on error signals.

Challenges in Neural Networks

  • Catastrophic Forgetting:
    • Sequential learning leads to loss of previously acquired information.
    • The hippocampus mitigates this issue by interleaving new learning with existing knowledge during replay.

Neurogenesis and Memory

Neurogenesis in the hippocampus plays a critical role in learning and memory: – Hippocampal Trace Conditioning: – Increases survival of newly generated neurons. – Disrupting neurogenesis impairs hippocampal-dependent learning tasks.

Key Terms

  • Semantic Memory: Context-free knowledge about facts and objects.
  • Episodic Memory: Recollection of personal experiences tied to specific times and places.
  • Pattern Separation: Differentiation of similar events in memory encoding.
  • Pattern Completion: Reconstruction of entire memories from partial cues.
  • Catastrophic Forgetting: Loss of prior knowledge in neural networks due to sequential learning.

This lecture integrated foundational theories of memory with neural mechanisms and clinical case studies, bridging psychology and neuroscience. Future discussions will explore the intersection of memory, emotion, and decision-making.

Print Friendly, PDF & Email