DUE DATE: Tuesday August 19, 2003

Your lab write-up should be type-written, double-spaced and spell-checked.

I. Purpose

The purpose of this lab is to explore the reliability of long term memory using a relatively simple word list task and to discuss its implications in understanding the reliability of other types of long-term memories. Further, current memory models will be analyzed according to their efficacy in accounting for your findings as well as the other findings presented in text and lecture.

II. Background Reading

Read the Scientific American article Beardsley, T. (1997, May). As time goes by…Scientific American, 276, 24, 28. (Available as a class handout).

This article is a summary and demonstration of Roediger and McDermott’s paradigm as presented in their paper: Roediger, H. L. & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803-814. The full article has been put on reserve.

III. Instructions

In this lab you will be generating and testing at least two different hypotheses about the reliability of long term memory using at least 6 different people (12 people if you are comparing two groups). Your first hypothesis (Part 1) should replicate the methods of Roediger and McDermott’s (1995) free recall manipulation as described in the Scientific American article and as demonstrated in class (read three lists to six people and have them recall on a piece of paper as many items as they can without "guessing"). A straightforward way of doing this is to read the all of categorized words at a pace of about 1 word per second and then have your participants write down as many words they can recall as soon as you finish. You may use the word lists provided in the Scientific American article; additional word lists from the Appendix of Roediger and McDermott’s (1995) journal article are also included here.

In analyzing your data, you can just look at the number of each of the target false alarms per category (e.g., the number of people who produced "cold" for the cold category items).

NOTE: A more complicated method would be to categorize the words as the percentage of items per person that were correct targets (words produced/the 45 words actually presented), target false alarms (incorrect recall of the "target" word/3 potential targets), other semantic false alarms (incorrect recall of semantically-related, non-presented words/number of semantic false alarms actually produced), and other/random false alarms (incorrect recall of "random" non-presented words/number of unrelated items actually produced). Done this way, it would be predicted that of the false alarms, target false alarms should be produced at the highest rate (e.g., out of the 3 possible).

Your second hypothesis (Part 2) can be tested with the same or different people. It might test factors like recognition, speed of presentation, gender or age differences, mode of presentation (auditory vs. visual), confidence, etc. Potential sources for ideas might come from lecture/text, your own observations, further browsing of my weblinks, etc. You may use the same or additional lists or create your own materials. Think carefully about your second hypothesis because it can have an impact on whom you test, how you test your participants, and what word lists you might use. One word of warning: if you do not counterbalance the use of word lists if testing the same people twice (to be explained in class), you run the risk that the results you get may be because of the word lists used rather than your manipulation (e.g., auditory vs. visual).

IV. Your Lab Write-up

Your lab write-up will contain two sections (Part 1: A Replication of Roediger and McDermott (1995) and Part 2: Testing My Own Hypothesis - NOTE: These can be combined if you test two hypotheses within the same "experiment") with 4 parts each: the hypothesis, the method, the results and the discussion. You should use subheadings to organize your paper. Use past tense throughout. All lab reports should be type-written, double-spaced, and spell-checked.

For the Hypothesis section - state your hypothesis as a prediction. Each hypothesis should examine an aspect of false memory. In addition, your hypothesis should have some rationale, relating back to the text and the readings. Develop your rationale in about one paragraph.

For the Method section - in two to three paragraphs, briefly describe the participant characteristics, the materials used, and the basic procedure (what the participant went through - you will have to develop your own set of instructions!). Use the past tense. If you used the same participants and/or procedure in Part 2 as you did in Part 1, just state so.

For the Results section - Look for similarities and differences in what people recalled and how they recalled the information. You do not need any statistical tests; instead concentrate on the pattern of results shown. Does everyone do better in one condition (e.g., false alarms to the targets as opposed to other false alarms) as compared to the other? Do most people? You should summarize your data using charts or graphs which can be on separate pages (all graphs and tables should report group data rather than individual data). In at least one paragraph (per hypothesis), briefly describe what each graph shows, report the means for each condition, and indicate the consistency of your results (how many people showed the predicted pattern of results?). Give specific examples from the memory protocols/people’s self-reports to help illustrate the ideas. Refer to the graphs in your written Results section (e.g., See Figure 1; See Figure 2).

For the first Discussion section (hypothesis 1) - Briefly describe what you found and whether your results support your hypothesis. In what ways do your data support what Roediger and McDermott found? What about what was said in text/lecture? In what ways do your data and/or the readings bring up new ideas or interpretations?

For the second Discussion section (hypothesis 2) - Did you find what you predicted? Why or why not? What do you think that means? List and describe at least two new things that you learned about long term memory with your variation of Roediger and McDermott's (1995) false memory task. Make any other observations that you feel are relevant.

V. General Discussion: DO NOT FORGET THIS NEXT PART!!

1) How well do you think Roediger and McDermott's (1995) list paradigm captures what might be occurring with false memories for items in related lists and under other more complicated conditions (e.g., eyewitness testimony, autobiographical memory, flash bulb memory, etc. - choose at least one of these other types of memories or discuss them as a group more generally. You might want to examine some of the other weblinks for Chapters 6 & 7 to get a sense of some of the controversies and research findings)? Why? Based on what you've learned, how accurate do you think other long term memories are? Are they more or less likely to result in false memories? Why or why not? What ramifications might that have for how these other long term memories are studied?

2) Choose two models of memory (e.g., Atkinson and Shiffrin’s memory store model, Craik and Tulving’s levels of processing model, Tulving's episodic/semantic memory model, McClelland and Rumelhart’s connectionist model) and use them to explain the formation and alteration of long term memories both in this task and in real life. Which one seems to fit the data better? Explain.

3) How well did doing this lab improve your understanding of the workings of LTM? How might you apply one of the ideas you learned in this lab in your own life?

VI. References

Include a separate reference page listing the text, the Beardsley article, and any other sources you used (e.g., Roediger & McDermott, 1995; weblinks from my website).

VII. Grading

Grades will be based on adherence to the basic mechanics of writing, the clarity of presentation, synthesis with class material, and the depth and originality of thinking. See the class syllabus for more detail on grading considerations for the labs.