> >Question 1: would knowledge of abstract concepts, like 'justice' be >considered semantic (declarative) memory, or nondeclarative memory?? >Whereas a concrete concept, like "hat" would be clearly declarative >(semantic). > Declarative memories are memories that we can talk about (hence, the label "declarative") and for which retrieval is typically an explicit memory act (that is, we are aware that we are remembering something). Thus, declarative memories include both episodic and semantic memories. Non-declarative memories tap skilled behavior that is manifested by performance rather than recollection (like riding a bicycle). Notice that although I can't convince any one that I really have knowledge about how to ride a bicycle except by getting on one and riding, I can have episodic memories of riding a bicycle and I have semantic knowledge about what riding a bicycle means (I can see some one riding a bicycle and know what they are doing and label their behavior correctly). It makes more sense to talk about whether a given test of memory taps declarative or non-declarative memory. The question about "how or where knowldege gets stored" gets slippery when we start playing around with what we mean by "remembering or having knowledge" -- what do you mean by this? If you mean that you can give a definition of the "justice" or recall an event that typifies the concept of justice, you are making use of declarative memory, relying on either semantic knowledge and/or episodic memories. If you mean that you can _demonstrate_ that you have a sense of justice (to use Rob's example) because you advocate solutions that would be deemed "just" but cannot articulate or explain why these solutions are appropriate, we might argue that you have only an implicit knowledge of justice. (For examle, suppose you choose one solution and can defend your choice only by saying that this solution "feels right" and other solutions don't. For this to really work as an example of _only_ implicit knowldege, you would have to be incapable of explaining _why_ the solution is superior to non-just solutions.) I am trying to draw an analogy here between the implicit knowledge of the rules of grammar of a 6 year old who is incapable of saying why one sentence is acceptable and another is not but _demonstrates_ knowledge of grammatical rules by producing only grammatical sentences. An adult might have both declarative and non-declarative knowledge of the rules of grammar and _could_ explain why "The dogs barks at the cat." is not grammatically correct. We might similarly have both declarative and non-declarative knowledge about justice. The distinctions between implict and explicit memory and declarative/non-declarative memory get further muddied by the fact that we construct tests for all of these memories using verbal materials. But the materials tap different memory systems based on the testing procedures used rather than through differences in the test materials themselves. Thus, when we fail to recall that the word "justice" appeared on a to-be-remembered list of words, we experience a failure of episodic memory and not a failure of semantic memory (we are not suddenly at a loss to provide a definition for "justice"). Similarly, if we do an implicit test of memory and use the word "justice," we don't ask participants to tell us whether they've seen the word recently in the experimental task or to define it. Instead, we ask them to decide quickly whether each stimulus presented is a word or a non-word and we find that reaction time is faster for stimuli that were presented recently than for new stimuli (i.e., priming occurs, which is a type of implicit memory test) or we find that people are more likely to respond with "justice" (rather than "juice") when asked to name the first word they can think of that begins with "ju" (another test of implicit memory). One of the difficulties with implicit memory tests is that responses can be contaminated by explicit recall, but this doesn't mean that there are not two different types of memory performance. > >Question 2: Are repressed memories of childhood abuse part of the >implicit memory system since they may affect our behavior in novel >ways from the direct explicit retrieval of the information? > Clearly, memories that we cannot explicitly retrieve can influence our behavior implicitly so it is possible that forgotten childhoon abuse could create implicit effects on current behavior. The problem with the use of "implicit memory phenomena" as evidence of repressed memories of childhood abuse is that we have no idea about which events did and did not take place. The problem for the repressed memory advocates is that they fail to take into consideration the fact that memories that are unrelated to childhood abuse can also influence behavior implicitly. Do I behave strangely toward this man because he smells of the stinky cigars that Uncle Marvin liked and Uncle Marvin abused me? Do I behave strangely toward this man because he favors a juniper-scented cologne and I had an experience with gin that gave me both a taste aversion and an aversion to the aroma of gin? Or does this man have little habits that resemble those of my fifth grade teacher who I disliked intently and have subsenquently forgotten all about? All of these are implicit memory effects manifested as strange responses toward a specific person - but not all are due to childhood abuse. We can speak confidently about the source of implicit memory effects in a laboratory task because we know which words were presented and which ones were not and can predict the specific implict effects created. We don't have this exclusive knowledge in the clinical cases. Claudia ________________________________________________________ Claudia J. Stanny, Ph.D. e-mail: [EMAIL PROTECTED] Department of Psychology Phone: (850) 474 - 3163 University of West Florida FAX: (850) 857 - 6060 Pensacola, FL 32514 - 5751 Web: http://www.uwf.edu/psych/stanny.html
