I also use serial position effects as a demo that almost invariably works. I know some people have had trouble making it work for them and have learned that it can be difficult if you don't keep some parameters just right. Here are my guidelines:
1. make sure you have a long enough list (15 items is a minimum) 2. use 2-syllable words (the extra syllable helps produce the load on working memory that is needed) 3. use common words (I use words with moderately high frequency - in the 40-60 words per whatever on the Thorndike Lorge norms. The two-syllable words from Paivio's imagery norms are a good source of words in this frequency range.) 4. present the words at a rate of about 1 word every 2 seconds - slower rates provide opportunities for the use of mnemonic devices that will undermine the effect 5. if you have time, present more than one list and average across lists to dilute the effects of unusually distinctive and memorable words on recall in any one serial position (also controls for the isolation effects produced by coughs, dropped books, noise from late-arriving students, etc.) Hope this helps! Claudia Stanny At 07:38 AM 11/17/2004 -0500, you wrote: >Hi Annette. I was surprised to hear you say that the serial position curve >demo works for you only 50% of the time. For me, it's one of the most >reliable demos that I can do, and I therefore do it in both my intro class >and my cognitive psych class. For me, the one time it didn't work was when >I had a class size smaller than 20. My classes usually range between 40-50 >in intro (although my cognitive class generally is between 15-20). Out of >curiosity, how many words do you use in your list? I use 15. -- Mark > ________________________________________________________ Claudia J. Stanny, Ph.D. Associate Professor Web Site: http://uwf.edu/cstanny/ General Track Coordinator Department of Psychology Phone: (850) 474 - 3163 University of West Florida FAX: (850) 857 - 6060 Pensacola, FL 32514 - 5751 --- You are currently subscribed to tips as: [email protected] To unsubscribe send a blank email to [EMAIL PROTECTED]
