Long-term Memory: Semantic Flashcards
What are the three main theories that explain how semantic memory is organised?
Prototype
Exemplars
Network models
Explain the prototype theory.
Each category is represented by a prototype. Our semantic memory stores one prototype for each category.
A prototype is an average of characteristics of that category acquired through experience.
The prototype shares the highest number of attributes with other examples of the category (Rosch and Mervis,1975). For instance, “car” is generally considered the most prototypical example of “vehicles”. It shares many attributes with “truck”, “bus”, etc (has wheels, uses fuel..).
Categories are recognised by comparing the sensory input with the prototype stored in semantic memory.
Explain examplars theory.
Each category stores as many exemplars as all the instances of that category you have accumulated throughout the experience:
If you have seen 1000 dogs throughout your lifetime, your semantic memory will have stored 1000 dogs.
Categories are recognised by comparing the sensory input with each exemplar stored
in semantic memory. Very time-consuming.
The exemplars theory has often been criticized according to the principle of cognitive economy because we would need to store thousands of exemplars for each concept in semantic memory.
What is semantic categorisation?
Within semantic knowledge, different levels of categorisation exist.
What is the semantic dementia?
Link this to semantic categorisation.
Semantic dementia is a neurodegenerative disease in which the patient progressively loses the ability to remember the meaning of things. For instance, they may forget what an umbrella is used for, or what a dog is, because they have lost their semantic representations.
Dissociations: their episodic memory and short-term memory are relatively unimpaired. Their language problems are not related to impaired abilities to find words.
What lobe was semantic memory found to be from?
Several studies have shown that semantic memory is related to the integrity and activation of the anterior temporal lobe.
Atrophy is more pronounced in the anterior temporal lobes in patients with semantic dementia (La Joie et al., 2014).
Concepts related to different categories are stored in different parts of the anterior temporal lobe.
Explain Huth et al (2012) experiment.
Showed participants two hours of video clips showing 1700 categories of objects and actions while their brains were being scanned with
fMRI. They looked at how each of the 30000 smallest brain units of investigation (voxel) responded to each category. They found that categories were widely distributed in the brain, rather than being located in specific regions (continuous semantic space).
Explain the network model.
Suggest a net-like organization of concepts in memory with numerous interconnections. Each concept represents a node. When you see information related to a node, the information spreads to neighbouring nodes: spreading activation model (Collins and Loftus, 1975).
Explain the Parallel Distributed Processing model.
Use distributed representations, meaning that a concept is represented across multiple nodes. For instance, the concept of ‘apple’ is represented by different nodes with specific properties, e.g., it can be represented by a node in a taste network and by another node in the ‘smell’ network.
Each concept/node may represent a neuron, which is connected to other concepts/nodes through brain connections. In this way, knowledge is distributed across different parts of the brain.
