Self-Reference Instructor's Page

By:
Gary M. Levine
Department of Psychology
Edinboro University of Pennsylvania
Email: glevine@edinboro.edu

Introduction
Method
Results
Discussion
References

The self-reference experiment is essentially a replication of Rogers, Kuiper, and Kirker (1977), which itself is an extension of the classic Craik and Tulving (1975) depth-of-processing studies. Research participants view a common set of words serially. For some of the words (randomly determined), participants are asked to make an orthographic decision (Does the word contain an "e"?) For others they make a semantic judgment ("How well does the word describe you?"). Words for which the first question is asked are labeled "E-Words." Words for which the second question is asked are labeled "Self-Words." The expectation is that self-words will be recognized more accurately than e-words. The primary difference between this version of the experiment and the original version is the mode of testing.

Method

The words participants judge are drawn from a fixed set of 20 (see list to the right), with half being randomly assigned to be presented along with the "self" question and half with the e-question. This manipulation is within persons. In addition to randomly assigning words to the self- or e-word category, the order is which words appear is random for each participant. Note that half the words have an "e" in them. Note also that half the words are less socially desirable. Because of this, the e-questions and self-questions should produce a relatively equal number of yes and no responses.

The 20 encoding trials are presented serially separated by a 1.5 second "Get Ready For The Next Question" signal. When the signal ends, the question assigned to the upcoming word appears, either

Does this word have an "e?"

Does this word describe you?

One second later, the word appears along with choice buttons marked "Yes" and "No." Although the instructions stress completing the task quickly and that the words will only appear for a second, in actuality, the words remain on screen until the participants answer yes or no.

There are two practice screens (one for each question) that appear before the words are presented. After reading the instructions and answering "yes" or "no" for the twenty words, there is 20-second delay (mainly to reduce recency effects). After the delay, the participants complete a recognition task that intermixes 20 foils with the 20 original words in a new random order. The 20 foils are strong, patient, polite, flexible, calm, logical, frugal, modest, bashful, clean, liberal, quiet, emotional, religious, eccentric, aggressive, immature, materialistic, irritable, somber. Participants are asked to respond "yes" or "no" to indicate whether the word was one of the words from earlier in the experiment.

After the recognition task, there is a screen indicating the results, and a debriefing screen that explains the details of the experiment.
« Return to Top»

Results

The main data consist of the hit rate and false alarm rates for Self Reference words and for E-Words, respectively. To provide a composite measure of discrimination, the hit rate and false alarm rate are combined to create a single discrimination index known as A-prime (A').

One simple analysis is a dependent t-test that compares A' for the two word types. The anticipated result is that A' will be significantly greater for self-words than e-words.

In addition to hit rates and false alarm rates, the program also records the time participants spend answering the e-questions and self-questions. This is called "processing time." These data should be examined as a possible confound because longer answer times alone might produce better discrimination.
« Return to Top»

Discussion

In discussing the results of this experiment, students might want to incorporate information from Bower and Gilligan's (1979) Experiment 2. Also, they might want to comment on the fact that the current experiment did not control for processing time. Thus, it is possible that participants spent more time processing self-reference words. If so, processing time offers an additional reason why self-reference words might produce better discrimination. The discussion could propose a method of controlling for this variable in future research.
« Return to Top»

Data Format for Downloadable csv file

The data obtained via the Download data link on the PsychExperiments homepage are comma delimited. The first entry (up to the first comma) is the affiliation of the research participant. Typically this is the class to which the participant belongs. Non-affiliated participants use the default entry of Interested Person. Next is the randomly assigned ID code for the research participant. The third and fourth entries are the data and time (24-hour format) on the server when the data were received. Next is a text entry for gender. F is for female and M for male. After gender comes the raw data for the experiment. The data concern processing time and recognition memory for Self Reference words and E-Words. Although there are not trials in this experiment, PsychExperiments data require that a trial number precede data. A dummy entry "1" precedes the data for Self Reference words and a "2" precedes the data for E-Words. After this dummy entry, there is a text entry indicating the word type (SelfRef or E-Word). Next is processing time in seconds, the hit rate, the false alarm rate, and then the value of A' (A-prime). The table below gives the data format (excluding user information) The text line below that shows a complete data set for one research participant.

Trial Number

Word Type

Study time

Hit Rate

False Alarm Rate

Trial Number

Word Type

Study Time

Hit Rate

False Alarm Rate

Trial Number

etc.

SelfRef

4.396

15.287,

E-Word

7.364

0.8,

0.95,

etc.

An Interested Person,WIE421,3/25/2002,16:14,F,1,SelfRef,4.396,1,0,1,2,E_Word,7.364,0.8,0,0.95,

« Return to Top»

References

Bower, G. H., & Gilligan, S. G. (1979). Remembering information related to one's self. Journal of Research in Personality, 13, 420-432.

Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology, 104, 268-294.

Rogers, T. B., Kuipers, N. A., & Kirker, W. S. (1977). Self-reference and the encoding of personal information. Journal of Personality and Social Psychology, 35, 677-688.

Symons, C.S. & Johnson, B. T. (1997). The self-reference effect in memory: A meta-analysis. Psychological Bulletin, 121, 371-394.

^{iThe author expresses thanks to Jim Craig, Linda
Smith, and other Indiana University graduate students who helped in the development of the
use of the self-reference paradigm for methodology courses. Additionally, this program
would not be possible without the help and support of the PsychExps team of John Williams,
Ken McGraw, and Mark Tew. Please e-mail the author (glevine@edinboro.edu
) with any comments you have about the program or other ideas / tips you have for other
instructors using this experiment.

«
Return to Top»}