Demonstrating the Serial Position Effect

One of the most stable of all experimental results is the serial position effect: When a sufficiently long list of stimuli is presented, subjects can remember the beginning and end of the list better than they can recall the middle. This effect is often used to support the concepts of short-term memory and long-term memory, two of the three major components of the human memory system. The first few words on the list are remembered better because they are rehearsed more and are thus more likely to be transferred to long-term memory. The last few words have the recall advantage of being held in short-term memory and thus being immediately available for recall. The words in the middle of the list have neither advantage, so their chance of being recalled suffers in comparison with the beginning and the end words.

Experimental Procedure

To do this experiment, you will need four cooperative subjects and copies of the answer sheet provided below.

STEP 1 Ask the first subject to sit down at a desk in a comfortable writing position. Read the following instructions: "You are about to participate in a psychology experiment testing your ability to remember lists of words. I will read a list of fourteen words to you. You should listen carefully to the list and, when I say the word 'go,' you should write down as many words as you can remember. You must, however, get the words written down in their correct order. That is, the first word must be written in the blank beside number 1 and the last word must be in the blank by number 14. You may write the words down in any order you wish; for example, you may write down the last words on the list first. Do you have any questions?

STEP 2 Answer any questions the subject has. You may repeat the instructions if necessary.

STEP 3 When the subject is ready, slowly read the words in word list 1, say "go," and let the subject write down as many words as he or she can remember.

STEP 4 Repeat the procedure for word lists 2 and 3

STEP 5 Repeat steps 1-4 with the three other subjects.

Analyzing the Results

STEP 1 Check each subject's answers against the original word lists. For a response to be counted correct, the subject must recall the words on the original list in the order in which they were read. Thus, if the subject writes the fourth word in the fifth blank, both the fourth and fifth words must be counted wrong. Mark an X through any errors or blanks.

STEP 2 Record in the right-hand column of each subject's answer sheet the total number of correct responses in the three lists for each of the fourteen serial positions. For example, if a subject got the first word correct on lists 1 and 2 but missed the first word on list 3, you would record a "2" in the blank numbered 1 at the right of that subject's answer sheet.

STEP 3 Insert the number of correct responses for each serial position in the table below. (Note: Numbers will range from 0 to 3).

STEP 4 Total the number of correct responses for each serial position for all four subjects. (The maximum possible number of correct responses for each word is 12.)

STEP 5 Plot the data from the "Total" column in Figure 7.1. Do the data from your subjects resemble the standard serial position curve that appears in Figure 7.1? If your curve differs, where is the difference? Can you explain any differences that exist? What do you think the curve would look like if the word lists were twice as long?