The Parallel Distributed Processing Approach

Parallel Distributed Processing Approach states that cognitive processing is based on parallel operations in which all attributes are considered at the same time. Each concept is represented as a node and is connected to other nodes. Each time a network connection is made between nodes it strengthens and learning takes places.
Let us look at an example
Using the following clues, guess what is being described
A. It has doors
B. It has windows
C. It has rooms
D. It is on wheels
You were probably thinking about a house; until you came to the sentence, it has wheels. Now you are thinking about either a mobile home or a travel trailer.
Let us look at another example
A. It is green
B. It grows on a vine
C. It is a vegetable
D. It can be string, green, or wax
You should have guessed bean. However, you might have thought about a cucumber until you came to the fourth clue.
Advantages of the Parallel Distributed Processing Approach
1. Explains how our memory helps us when some of the information is missing
2. We make spontaneous generalizations through inferences about general information. Spontaneous generalizations account for some of memory errors and can partially explain stereotyping.
3. Processing is a distributed activity across several locations
4. A network contains nodes which are connected to many other links
5. Connections are weighted. Connection weights determine how much activation a unit can passes to another unit and each can be open at different levels
6. Activation can be either positive or negative
7. Each avtivation changes the stength of connections among related units
8. Graceful degradation is the brain's ability to provide partial memory and explains tip of the tongue phenomenon
Reference
Matlin, M.W. (2005), Cognition (6th ed.) Wiley and Sons
Additonal information obtained from this site