PSICentre for Psychology Athabascau UniversityAthabasca University
 
myAU| Login
 

Experiment Description: Free Recall and the Serial Position Effect

The serial position effect is described in chapter 4 and refers to the U-shaped pattern of performance on a free recall task when recall is plotted as a function of word position. In a free recall task, participants are given a list of 20 or more words, one at a time, and then asked to recall the entire list after the last word has been presented. It is called 'free recall' because the participant can recall the words in any order. The serial position curve is the curve that is produced when performance is plotted as a function of word position (see figure 4.2, p. 104 in Matlin, 2005). The first part of the U-shaped curve is called the primacy effect and refers to the relatively good recall of the first items or primary items on the list. The primacy effect is hypothesized to occur because the items have been successfully stored in long-term memory. The last part of the U-shaped curve is called the recency effect and refers to the relatively good recall of the last items or most recent items on the list. The recency effect is hypothesized to occur because these items are easily retrieved from short-term store because they are still available.

To test these hypotheses, Glanzer and Cunitz (1966) combined the free recall task with two variables, one thought to interfere with long-term memory and one thought to interfere with short-term memory. They reasoned that presentation rate should affect the long-term memory such that the longer the time to study a word, the more likely it will end up in LTM. Thus, increasing the presentation rate (thus shortening the studying time) should disrupt the primacy effect. Alternately, delaying recall should affect short-term memory as a longer delay will lead to less recall from short-term memory. Thus, introducing a delay before recall should disrupt the recency effect. This is indeed, what they found.

In this experiment, you will complete the free recall task with twenty words under three conditions. In the control condition, words will be presented at a standard rate followed by immediate recall. In the Speeded Presentation condition, words will be presented at a faster rate than in the control and followed by immediate recall. In the Delayed condition, words will be presented at a standard rate but recall will be delayed by having you complete a distractor task prior to recall. You will be testing whether the primacy effect and/or the recency effect (i.e. the dependent variables) are affected by the presentation rate or by delaying recall. (i.e. the independent variables).

Participant Instructions: Free Recall and the Serial Position Effect

On each trial in this experiment, a list of 20 common nouns will be presented one at a time in succession in the centre of the screen. After the list has been presented, you will have three minutes to recall as many of the words as you can by typing your words in the space provided. When typing your words at recall, be sure that the spelling is correct. Words can be typed on separate lines or with spaces between them. The case of the letter does not matter nor does the order. If you cannot recall any more words, click on the next trial button to advance to the next list.

There are three conditions in this experiment. In the Control condition, the words will be displayed at a rate of one word every two seconds. After the last word has been presented, you will be asked to recall as many words as you can. In the Speeded-presentation condition, the words will be displayed at a rate of one word every .5 seconds. After the last word has been presented, you will be asked to recall as many words as you can. In the Delayed-recall condition, the words displayed at a rate of one word every two seconds but after the last word has been presented, you will be presented with a series of simple addition and subtraction problems with multiple choice answers. For example, you might be presented with 13+5=? with the following alternatives:

A - 17
B - 18
C - 19
D - 20

Your task is to answer each problem as quickly as you can by using the mouse to click on the correct answer. After 30 seconds of this activity, you will be given the opportunity to recall the words as in the other conditions. The 9 lists (3 for each condition) will be presented in a random order. You will have an opportunity to take a short break between lists.

The raw data from the experiment will be placed in a file called DRECALL.DAT on your computer and can be examined by using the View Raw Data button.

To summarize this data, click on the Analyze Raw Data button to calculate the average correct recall at each position across the 3 lists in each condition. Because there is only one participant in this experiment, the data are further averaged across four positions to provide a smoother curve.

The summarized data will be placed in a file called ARECALL.ANL on your computer. To view the results, click on the View Summarized Data button. This summary data is important and will be necessary to prepare an APA style figure to include with your paper. The file ARECALL.ANL is a simple text file and can also be read and printed with any word processor. For each condition, you will be presented with 5 averages;

Average correct recall in position 1, 2, 3, & 4 across all three lists (max 12)
Average correct recall in position 5, 6, 7, & 8 across all three lists (max 12)
Average correct recall in position 9, 10, 11, & 12 across all three lists (max 12)
Average correct recall in position 13, 14, 15, & 16 across all three lists (max 12)
Average correct recall in position 17, 18, 19, & 20 across all three lists (max 12)

Use this summary data to prepare an APA style figure to illustrate the serial position effects. In your graph, plot the five positions on the abscissa (X axis) with average recall on the ordinate (Y axis). Use separate lines for three conditions.

To view the results of previous participants, click on the Display Group Data button. This button will display the average reaction time for the three conditions. You can use this information to compare with your own performance.

Run Experiment

If you have successfully installed Java Web Start but you are still being taken to the Java installation page, please click on this direct link to the experiment installer.




Athabasca University, Canada Open University

© Copyright 1997-2014, Athabasca University. All rights reserved.
Page created by webmaster@psych.athabascau.ca
Last Modified: Monday, 27-Jun-2005 23:34:14 MDT