Class Notes (836,580)
Canada (509,856)
Psychology (3,977)
PSYC 2650 (196)

Chapter 8 notes.docx

5 Pages
Unlock Document

PSYC 2650
Anneke Olthof

Cognitive Psychology: Chapter 8 The Network Notion - Long-term memory is enormous in size, and yet we typically extract information from it extremely quickly and effortlessly - For instance, ambiguous sentences like this one are understood by the automatic application of background knowledge  Steve tried to be careful when he put the vase on the table, but it still broke.” - Much of this chapter explores a single idea: that memory connections are much more than retrieval paths - Instead, the connections are our memories In other words, knowledge is represented via a vast network of connections and associations between all of the information you know - How might the network work? • Representations of individual ideas are the nodes (“knots”) within the network • The connections between the individual ideas are associations or associative links • Think of the nodes as cities on a map, and the associations as the routes between the cities - Spreading activation is the process through which activation travels from one node to another, via the associative links - As each node becomes activated, it serves as a source of further activation, spreading onward through the network - Similar to neurons, nodes have activation levels and fire a signal if the input stimulating them summates to reach threshold We have seen this notion of networks and spreading activation earlier in the course in our discussion of feature nets Evidence Favouring the Network Approach - Networks suggest an explanation for why hints help us remember  South Dakota’s capital is easier to remember with the hint that the answer is a man’s name  With the hint, “Pierre” receives activation from two nodes and is more likely to reach threshold - Networks also suggest an explanation for state-dependent learning and context reinstatement - During learning, connections are strengthened between the context and the learned material - If you are in the same context during testing, the learned material will receive pre-activation from these connections - Further evidence for spread of activation and priming within networks comes from the lexical-decision task - When items are presented in pairs, the semantic relationship between words affects the speed of lexical decision - A word like “butter” will be recognized faster after having seen “bread” because its node has already received spreading activation - Other evidence for the knowledge representation in a network comes from the sentence-verification task - Participants must quickly decide whether sentences like the following are true:  Robins are birds  Robins are animals  Cats have hearts  Cats are birds - For instance, to confirm that “cats have hearts,” one must traverse two associative links - The time to answer these questions depends on the length of the associative path between the pieces of information (Collins & Quillian, 1969) - The degree of fan is the number of associative links radiating out from a node - With a higher degree of fan (more links), each of the associated nodes receives a smaller fraction of the central node’s activation. This is sometimes called a fan effect - An experiment supporting the fan effect (Anderson, 1974) presented sentences such as:  The doctor is in the bank  The lawyer is in the church  The lawyer is in the park - Some agents were associated with one place, and some with two. Similarly, some places were associated with one agent, and some with two - Later, when verifying having seen the sentences before, response times were fastest for sentences with only one association between agent and place, or a smaller degree of fan Retrieving Information from a Network - Searching through memory shares some similarities with performing an internet search - Associative links may guide the search through the network just as hyperlinks guide the search through the Internet - One way that a search through memory gets started is via the connections between perception and memory - An input node is a node within a network that receives at least part of its activation from detectors that are sensitive to events in the external world Unpacking the Nodes - Different types of associative models differ in the details of how networks are structured - For instance, what information is captured by the nodes, and what information is captured by the connections? - Different types of associative links - In some proposals, nodes represent single concepts only (e.g., “dog”) and the nodes are connected by different types of associative links (e.g., “isa” and “hasa”)  Sam “isa” dog  Sam “hasa” dog - However, we can think about a m
More Less

Related notes for PSYC 2650

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.