Have you ever noticed what you remember? In our most vivid memories it seems that we remember the poignant, the unique, and the reason that the memory is vivid. But what about everyday life? How do we remember things that simply pass us by, that we see briefly and that do not make much of a difference in our lives? What did I have for dinner last night?
There are two ways to look at the former question. One is that we will remember things that stand out—we will process it deeper into our memory structure, increasing the depth of processing. The other is that we have a plan, or schema, for how and what things happen, and we will tend to remember things happening according to this schema, whether they did or not. For example, What did I have for dinner last night? Well, yesterday was Sunday, and I always have pizza at Robert Purcell Union on Sundays. So, I must have had pizza. That’s an example of schema theory, whether I had pizza or not. But what if, last night when I went to the junk food area, I noticed the hot dogs with sauerkraut, realized it’s been a long time since I’ve had that, and ate that instead. By depth of processing, this event would stand out, being different from the norm, and I would remember it.
We have built up a large structure of schema (or stereotypes) based on sex. (Bem & Bem) We expect men to conform to one norm, and women to conform to another. Bem has devised a test, the Bem Sex Role Inventory (BSRI) that measures how much people view the wold according to the sex norms. Using this test, she has defined two types of people—those who view the world according to sex (sex-typed) and those who do not view the world according to sex (non sex-typed). Traditional people are sex-typed people who also apply the sex norms appropriately to themselves. A traditional male views himself as masculine, and a traditional female views herself as feminine. Androgynous people do not view themselves according to their sex, at least to the extent that traditionals do.
Magazines are among the many things that can by organized according to sex-type. Car and Driver, for example, is seen as male-oriented, as is Field and Stream. Family Circle, and Parents, on the other hand, are seen as female-oriented. According to schema theory, traditional sex-typed people should, on the off chance they saw someone reading Car and Driver, remember that reader as being male. On the other hand, if the reader were female, the Depth of Processing view holds that this cross-sex activity should stand out from the norm, and be remembered easily.
This is where our experiment comes in. We decided to test, using magazines, whether or not abnormal, in this case cross-sex type, activity is remembered as abnormal, or forgotten and ‘remembered’ as having been normal.
First, of course, we had to determine what magazines are viewed as male-oriented, female-oriented, and neutral. For this, we conducted a pre-test, in which 30 questionnaires were distributed, asking the subject to rate 40 magazines according to a scale from 1 to 5, from 1: read predominantly by males, to 5: read predominantly by females. From these 40 magazines we chose ten male-oriented magazines: Car and Driver, Computers and Electronics, Esquire, Field and Stream, Forbes, Harvard Business Review, High Fidelity, Sky and Telescope, Sports Illustrated, and Wall Street Journal; nine female oriented magazines: Bon Appetit, Cosmopolitan, Family Circle, Glamour, Ladies Home Journal, Parents, Redbook, Self, and Vanity Fair; and ten neutral magazines: Architecture Digest, Consumer Reports, Newsweek, New Yorker, Psychology Today, Reader’s Digest, Rolling Stone, Ski, Tennis, and USA Today.
Slides were then taken of various readers reading these magazines. Half of the readers in each group were male, and the other half female.
For the main part of the experiment, 42 subjects were run. The subject first received the Bem Sex Role Inventory and a fake test which asked various questions about magazines, designed to make sure subjects did not guess that they should specifically look at the sex of the readers. The subjects then looked at a slide show, each slide being shown approximately 3 seconds. After the slide show, the subjects were given the memory test, which asked whether or not the reader was standing, wearing glasses, and the reader’s sex, along with whether or not there was food in the picture, and windows in the picture.
Of the 42 subjects, 10 were androgynous and 16 were traditional. Of the 10 androgynous, 6 were female, and of the 16 traditional, 6 were also female. The final results were run through ANOVA on the Cornell C IBM mainframe to determine what, if any, of the results were significant.
There were three results, two with borderline significance, and 1 designated very significant by ANOVA. Significant was defined as P < .05.
The first result of borderline significance (P = .0984) is that for neutral magazines, androgynous subjects had better recall than traditional subjects (see Table 1)
| Table 1 | ||||
|---|---|---|---|---|
| Male | Female | |||
| Neutral Magazines: | Andro | Trad | Andro | Trad |
| Female Reader | 75% | 57.5% | 79.2% | 67% |
| Male Reader | 83.3% | 72.5% | 83.3% | 75% |
| Overall %Correct | 79.2% | 67.9% | 83.3% | 72.2% |
Overall, the androgynous subjects had better recall also, but this probably reflects the fact that traditional subjects did incredibly worse in certain cells, reducing their overall percentage correct.
The second result with borderline significance (P = .0772) is that the gender of the person reading the magazine had an effect on recall. (See Table 2) In the neutral magazines, all subjects recall males reading the magazine with better accuracy then if a female is reading the magazine.
| Table 2: Neutral Magazine (Percent Correct) | ||
|---|---|---|
| Male Reader | Female Reader | |
| Androgynous | ||
| Male | 83.3% | 75% |
| Female | 83.3% | 79.2% |
| Traditional | ||
| Male | 72.5% | 57.5% |
| Female | 75% | 66.7% |
| Overall Score | 77% | 67% |
The same pattern holds for male oriented magazines, (Table 3) across the board, although much stronger, and the opposite holds for female magazines, although a bit weaker (Table 4).
| Table 3: Male Magazine (Percent Correct) | ||
|---|---|---|
| Male Reader | Female Reader | |
| Androgynous | ||
| Male | 91.7% | 66.7% |
| Female | 95.8% | 75% |
| Traditional | ||
| Male | 92.5% | 70% |
| Female | 91.7% | 45.8% |
| Overall Score | 93% | 65% |
| Table 4: Female Magazine (Percent Correct) | ||
|---|---|---|
| Male Reader | Female Reader | |
| Androgynous | ||
| Male | 75% | 83.3% |
| Female | 62.5% | 91.7% |
| Traditional | ||
| Male | 45% | 70% |
| Female | 79.2% | 75% |
| Overall Score | 61% | 78% |
By far the largest effect, though (P = .0002), was for interaction between the orientation of the magazine and the sex of the reader, especially when the subjects saw someone of their own sex reading a cross-sex magazine. In all cases except one, subjects (androgynous and traditional) were more accurate recalling same-sex activity than cross-sex activity. (See Table 5). In the case of traditionals, subjects were far less accurate than other subjects when recalling someone of their own sex reading a magazine oriented towards the other sex.
| Table 5 | ||||
|---|---|---|---|---|
| Magazine: | Male Oriented | Female Oriented | ||
| Reader: | Female | Male | Male | Female |
| Subjects | ||||
| Androgynous | ||||
| Male | 66.7% | 91.7% | 75% | 83.3% |
| Female | 75% | 95.8% | 62.5% | 91.7% |
| Traditional | ||||
| Male | 70% | 92.5% | 45% | 70% |
| Female | 45.8% | 91.7% | 79.2% | 75% |
| Overall Score | 65% | 93% | 61% | 78% |
The only spot where cross-sex activity was recalled better than normative activity was for traditional females recalling who read a female oriented magazine, and the difference is very small (4.2%).
The second two of these results can be explained in terms of schema theory. The first is somewhat strange, however. Since it has the least mathematical significance of the three results, I’ll let it be.
The second result, that for neutral magazines and male magazines the male readers are remembered with better accuracy is evidence that it is probably inaccurate to say that while androgynous people do not have to view the world through the ‘schema-colored glasses’ of sex-typing, they can choose to do so if they want to, and it is very hard to escape the lessons of society. As Bem has said, traditional sex-typed people view the world through schema-colored glasses, while androgynous people can see that the glasses are there. (Psychology 277, 1985) This does not preclude using the glasses on occasion.
The fact that even for neutral magazines the subjects recall male readers better shows that nothing is neutral. When given the choice between male and female, without ‘Don’t Know’, subjects chose male more often than female.
The largest result, though, and most important, is found considering the sex of the subject and the sex of the magazine. Cross-sex activity was recalled less accurately across the board than sex-typed activity, exactly as schema theory predicts. And by far the largest effect within this result was that traditional males do NOT remember a male reading a female magazine, and females do NOT remember a female reading a male magazine. The percentage correct in each case was less than fifty percent. Traditional people do not seem to want to remember people of their own sex doing abnormal activities. The effect disappears for traditionals recalling the other sex reading cross-sex material, and for androgynous people recalling anything.
So, what do we remember? We seem to remember whatever fits our preconceived views, stereotypes, or schemas, of the world. And what did I eat for dinner last night? Well, actually, I was hungry, and had both pizza and hot dogs with sauerkraut.
