Textbook Notes (362,735)
Canada (158,032)
Psychology (3,256)
PSYC 2650 (226)

PSYC*2650 Ch 8.doc

12 Pages
Unlock Document

University of Guelph
PSYC 2650
Anneke Olthof

Monday, Feb 18, 2013 Chapter 8: Associative Theories of Long-Term Memory - long-term memory is huge, it has to be to contain all of the memory we remember - how do we ever find anything? - how does memory retrieval work and how is memory set up to support this retrieval> - memory search is aided by connections between the materials to be learned and new things but what are these connections? The Network Notion - memory connections provide much more than retrieval paths, the connections are memories - all of knowledge is represented via a sprawling network of connections, a vast set of associations How Might the Network Work? - the essence of a memory network is straightforward - first, we need some means of representing individual ideals, these representations will be the Nodes within the network - these nodes are tied to each other via connections that we will call Associations or Associative Links - e.g. nodes are cities and associates are highways - not all associations are of equal strength e.g. there is a strong association between the “my birthday” node and the node representing a specific date - how do memory connections get established? - some active intellectual engagement is need to create or strengthen the connections - the nature of this engagement is crucial, creating multiple connections, thinking about the material in distinctive ways Spreading Activation - a node becomes activated when it has received a strong enough input signal - what travels through associative links is akin to energy or fuel - associative links can be thought of as “activation carriers” - once a node has been activated, it can activate other nodes - nodes receive activation from their neighbours, and as more and more activation ar- rives at a particular node, the Activation Level for that node increases Monday, Feb 18, 2013 - the activation level will reach the node’s Response Threshold and the node Fires - the firing has several effects, the node will now itself be a source of activation, sending energy to its nature and it will summon attention to that node (finding the node) - activation levels below the response threshold - Subthreshold Activation is important - activation accumulates so 2 subthreshold inputs may add together or Summate - if a node has been partially activated recently, it is already “warmed up” - very parallel to how neurons work Spreading Activation - as each node becomes activated and fires, it serves as a source for further activation, spreading onward through the network - you do not choose where to go once you are at one node - thanks to frequent use, some associative links are particularly effective - better-establish links will carry activation more effectively and more successful at acti- vating subsequent nodes Evidence Favoring the Network Approach Hints - why do hints help us remember? - mention of South Dakota will activate the nodes in memory that represent your knowl- edge about this province, activation will spread outward reaching nodes that represent the capital city’s name - perhaps you are not very familiar with South Dakota and haven’t thought about the capital for some time so insufficient activation will flow into the Pierre nodes - if a hint is available “South Dakota’s capital is also a man’s name” this will activate the man’s name node and activation will spread from this as well - nodes for Pierre will now receive activation from 2 sources and will be enough to life the nodes’ activation level to threshold levels Context Reinstatement - memory is best if the state you’re in during memory retrieval is the same as the state you were in during learning - if you learn a list of words, including the word “pointer” while underwater - later on if asked what the words were and not enough activation will reach the pointer nodes to activate them - if you are underwater at the time of the test, this will trigger certain thoughts that may be linked to the nodes representing the learned material - the pointer nodes will be receiving a double input Monday, Feb 18, 2013 More Direct Tests of the Network Claims Spread of Activation and Priming - evidence that subthredhold activation can accumulate comes form the Lexical-Deci- sion Task - participants are shown a series of letter sequences on a computer screen and are asked to hit “yes” if the sequence spells a word or “no” if it does not - participants’ speed of response is an index of how quickly they can locate the word - in one study, participants were presented pairs of letter strings (chair, bread = yes, house, fime = no) - summits words were semantically related (nurse, doctor vs lake, show) - of central interest was how the relationship between the words influence performance - in the related pairs, when they search for the first word, the node is activated and spreads outward partially activating the second word’s node - as they turn to the second word, activation is faster Sentence Verification - in an experiment using Sentence Verification Task, participants were shown sentences on a computer screen, some were false some were true, they hit true or false buttons - participants perform this task by traveling through the network, seeking connections - there is a connection between robin and bird so the sentence “a robin is a bird” is true - they would answer that quickly - if the nodes are connected indirectly (robin and animal) it is slower (2 steps) - there is not point in storing in memory the fact that cats have hearts, it is more efficient to store that they are animals and separate that animals have hearts - the property “has a heart” would be associated with the animal node - we would expect a slower response to sentences like “cats have hearts” - we would expect a quicker response to “cats have claws” - response times depend heavily on the number of associative steps that msut be tra- versed to support a response Monday, Feb 18, 2013 - there is more to this - some connections can be traversed more quickly than others e.g. participants are faster to responding to “a robin is a bird” than “a peacock is a bird” - these are both one-step connections but more typical ones are faster - responses are relatively slow to sentences like “sparrows have feathers” but very fast to “peacocks have feathers” (even though it is a 2-step) - psychologists have moved away from this model Degree of Fan - the amount of things fanning from it, more familiar objects have a larger degree of fan - once a node is activated, the activation will spread form there, flowing through all the links radiating out from the node - the quantity of activation is limited, os each link only gets its “share” of the whole - if a node has a higher degree of fan, each link will receive less activation and carry less to neighboring nodes and will slow the rate at which they are activated - in a study participants memorize a set of sentences about people in locations (the doc- tor is in the bank, the fireman is in the park) - some of the actors appear in only one locations, and some in 2, some locations con- tained only 1 person and others contained 2 - the amount of time the actor or place was mentioned increased their decree of dan - when participants were tested for memory, they had to decide as quickly as possible whether ach of the test sentences had been present - response times were fasted when only one sentence mentioned a specific person or a specific place and slowest when multiple sentences named a specific person or place Retrieving Information From a Network Searching Through the Network via Associative Links Monday, Feb 18, 2013 - when you search something on the internet you get links to get more information e.g. search dog diseases, find specific illness, find symptoms - this depends on having the rights likes and having a lot of links - but the system will collapse is there are too many links - the network that supports your memory probably works the same way - activation can spread from more than one source simultaneously Finding Entry Nodes - there is one problem, on the internet you launch the search process by typing some text into a search engine, how do things get launched in memory? - some nodes within the net are Input Nodes - they are like any others, they receive ac- tivation via associative links and send activation to other nodes once triggered - what is special about these input nodes is that they receive most of their input activa- tion form appropriate Detectors which are connected to the eyes, ears etc - various detectors are actually nodes within a network, functioning like any other nodes, there is no problem with these input nodes (the detectors) sending their activation to the long-term memory nodes - there is really no difference between the feature net and the memory network - detectors send activation to memory nodes which allow cues to trigger events - need more theory to explain how inputs are recognized - the key is that perceiving and identifying an input will involve locating and activating of that input’s node in the network Unpacking the Nodes - a node can contain a lot of complex information so there needs to be a device capable of reading and interpreting this corpus of material - the simpler we keep each node’s informational context, the less we need to rely on this - different models approach the issue in different ways Different Types of Associative Links - one proposal is that nodes represent single concepts, nothing more complicated - e.g. chair, doctor - how do we represent an idea as complicated as “my understanding of Darwin’s theory of evolution” - these complex ideas are presented with more network - associative link may exist between Darwin and evolution, this is too simple though - how do you represent the contract between “Sam has a dog” and “Sam is a dog” Monday, Feb 18, 2013 - introduce different types of associative links, some representing equivalence relations and others representing possessive relations Propositional Networks and ACT - there are limits with these labeled associations - the problem is that we are able to remember and think about a wide range of relation- ships, not just equivalence and possession - another proposal by Anderson is that at the center of this conception is the idea of Propositions - the smaller units of knowledge that can be either true or false e.g. chil- dren love candy, dad likes BMWs - just children and just BMWs are not propositions - the same proposition can be represented in various forms - Anderson’s theory is embodied in a computer program known as ACT where proposi- tions are represented - in this diagram, the ovals identify the propositions themselves, and the associations connect each ellipse to the ideas that are the propositions constituents - associations are identified in terms of their syntactic role within the proposition - diagram shows 4 propositions that together constitute part of the network that repre- sents you knowledge about dogs - to represent each specific episodes or a specific object (rather than a category), ACT makes a distinction between Types Nodes and Token Nodes - “types” refers to a general category, type nodes are embedded in propositions true for the entire category Monday, Feb 18, 2013 - “token” is a specific instance of a category, and token nodes are therefore found in propositions concerned with specific events and individuals - type nodes and token nodes are typically connected to each other - the ACT model distinguishes between timeless truths (“Jacob feeds the pigeons”) and more specific statements (“last spring, Jacob fed the pigeons in Trafalgar Square”) - ACT does this by incorporating time and location nodes as part of propositions - ACT models shares many claims with other network models: nodes are connected by associative links, some of the links are stronger than others, with the strength of length d
More Less

Related notes for PSYC 2650

Log In


Don't have an account?

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.