There is a growing concern in the English language about how to properly refer to the gender of an ambiguous person. The APA Publication Manual (APA, 2001) contains an entire section devoted to non-sexist language and "preferred" examples. This inclusion of non-sexist language has been included in the manual for thirty years now; it was first published in 1977. The APA is not the only organization now warning against sexist language, the Modern Language Association, the American Medical Association, the American Marketing Association, the Association of American University Presses, and the Chicago manual of style (Madson & Hessling 1999). Despite the best efforts to reduce the use of sexist language, students are often confused and resistant when they are required to use nonsexist language in their research papers and others fail to understand the reasons for the requirement. Another problem is the difficulty choosing the best nonsexist alternative.

The APA manual recommends that sex designation only be used when pertinent to the material being discussed. For example it is acceptable to say "13 female doctors and 22 male doctors," because the relative proportions of males and females may be important, however, problematic uses of sex designation are "woman doctor, lady lawyer, male nurse, and woman driver." The manual prefers "doctor or physician, lawyer, nurse, driver" (71). This could be explained by the clause at the bottom of the page that states, "woman and lady are nouns; female is the adjective counterpart to male."

Personal pronouns intended to be gender neutral are often interpreted as references to males (Harrigan & Lucic, 1988). Some believe that using “he” is unfavorable to women. “The effect of using he, etc., as a generic pronoun is believed likely to perpetrate the cultural stereotype that men and masculine aspects are superior and more favorable” (Harrigan & Lucic, 1988). Certain personal pronouns intended to be gender neutral, such as "he, his, him, and man" have been found to not always be truly gender neutral (Harrigan & Lucic, 1988). One proposed resolution is the use of more modern "he/she" or "s/he." The majority of existing research has found that using "he/she" will invoke fewer images of males and be more neutral than "he", and that "they" will be more neutral than "he/she" (Gastil, 1990; Harrigan & Lucic, 1988). However, using "they" may cause complications. Using "they" for sex type singular slows down reading ability (Foertsch & Gernsbacher, 1997). Somewhat less popular suggestions include the creation of neologisms, such as "tey, tem, and ter." Research on these neologisms (Todd-Mancillas, 1984) found that it was harder to understand "tey, tem and ter" than the more traditional “they, he/she, or s/he.” However, Buck (1990) proposed that neologisms test the limits of exceptional compostion.

It is also possible that the age of the person to which one is referring can inflence decisions regarding sex associated characteristics. For example, these classifications exceeded the norms for adults from the original work by Sellers, Satcher and Comas (1999) found that high percentages of children tend to be classified as androgynous or undifferentiated. The percentages of children with Bem (1974). Thus, if children are less associated with a particular sex than adults are, words that invoke images of children should be less associated with a particular sex as well.

Indeed, if the movement toward a non-sexist use of the English language is to have a fair and balanced sense of guidance, it is vital that we monitor the pulse of present opinion and perception of gender equality in our use of language. More importantly, it is critical that we determine the exact impact that the use of various sex associated words hason our cognition. One way we can investigate the cognitive impact that gender related words have is to utilize a priming paradigm.

Words that elicit male ideas (e.g. man, men) should prime male traits utilizing the basic properties of spreading activation (Collins & Loftus, 1975; Schvaneveldt & Meyer, 1973). Words that do not elicit ideas related to a particular sex should not show priming effects. Previous studies (Banaji & Hardin, 1996) have utilized a priming paradigm where items from the Bem (1974) Sex Role Inventory were used to prime target pronouns (he/him/his/she/her/hers). Banaji and Hardin found gender information can automatically influence decisions but there are other factors that also have an influence. An interesting twist on this paradigm would be to use different primes to determine the effects on targets from the Bem Sex Role Inventory.

Due to the fact that the use of these words is both a violation of the stylistic requirements of the APA, and may have effects on language or psychological development, "the search for appropriate, etymologically sound gender neutral terms is a worthwhile and intellectually stimulating enterprise" (Buck, 1990).

The purpose of this study is to test the effects of different primes on both choices and reaction times for masculine/feminine/androgynous traits. The effects of whether or not prime words are singular or plural, whether they portray young or old , and whether or not they are associated with a particular sex will be examined.

PARTICIPANTS
Participants were 235 undergraduate students enrolled at Missouri Western State University, a 5,000 student public university in northwest Missouri. Participants received class credit for their participation. 71 participants (30%) were male, and 164 (70%) were female.

MATERIALS
Twelve primes were words that were inherently masculine, feminine, or neutral and which varied in age (young/old) and whether or not they were plural. Thus, man, woman, adult represented the older singular versions of the masculine/feminine/neutral primes. Men, women, adults represented the older plural versions. Boy, girl, child represented the younger singular primes. Boys, Girls, Children represented the younger plural primes.Target words consisted of the 60 items from the Bem (1974) Sex Role Inventory. Seven of the items requried slight changes in order to be grammatically correct when paired with both singular and plural primes. "Defends own beliefs" was changed to "defending of their own beliefs." "Strong personality" was changed to "strong in personality." "Has leadership abilities" was changed to "able to lead." "Makes decisions easily" was changed to "decisive." "Acts like a leader" was changed to "likely to act as a leader." "Does not use harsh language" was changed to "likely to use harsh language." Finally, "loves children" was changed to "loving of children."

PROCEDURE
Participants were given a basic description of the task and asked to press "1" if the answer to the question was "Yes", and to press "2" if the answer was "no." All prime/target pairs were presented on the computer as questions like "Is a [PRIME] [TARGET]?" For example, "Is a man self-reliant?"Reaction times and choices were recorded using E-Prime 1.2 (Psychology Software Tools, 2006). Each participant completed 120 trials encompassing each of the 60 Bem items cycled through twice with a different random prime before each. Randomization and order of presentation was controlled by the E-Prime software.

Trials which had reaction times more than 3 standard deviation above the mean were removed prior to analysis. 484 of 28,200 trials had a reaction time greater than 7190ms and were removed.

A 3 (Bem category of the target) x 2 (whether or not the prime was plural) X 2 (age of the prime) X 3 (sex of the prime) ANOVA was calculated to compare reaction times to the choices participants made. Table 1 represents the ANOVA results. No significant 4-way or 3-way interactions were found, however, several 2-way interactions and several main effects were significant. The significant main effects and interactions are shown in the various figures.

When studying the effects the plural versions primes, we found that the subjects were faster to make decisions about plural primes than singular primes. More specifically, the prime sex x prime plural interaction showed that participants were fastest to make decisions about female primes that were plural. These results suggest that we stereotype women more readily than we stereotype men, and that groups of women are stereotyped even more than individual women. These results may be contrary to Foertsch and Gernsbacher (1997) who found that “they” slowed down reading timesWe also found that participants were faster to make decisions about adults than children. More specifically, it was harder for the subjects to stereotype children as masculine consistent with Sellers, Satchers, and Comas (1999).When studying the effects of the sex of the prime, there was an overall effect showing that subjects were much faster to make decisions about females than both males and neutrals, and that males were faster than neutrals. This finding suggests that we are more likely to stereotype when we encounter words that represent females than we are to stereotype when we encounter words that represent males or are neutral.This experiment sheds light on previous research, such as the works that state how personal pronouns intended to be gender neutral are often interpreted as references to males (Harrigan & Lucic, 1988). Our results show that females tend to be stereotyped more than males which hint that when looking at pronouns a person doesn’t always think of a male image and are more gender neutral. A concern that might be present with this study is the sample of subjects that were obtained. A high proportion of our participants were females which may have had an impact on the results. The subject pool was also all college students so there was a small age range which was a limitation. For further research it’s recommended to conduct the study on a different subject base, perhaps all men or all women, and varied age ranges. A major difference between our research and previous research was the way we referred to the gender neutral and gender specific subjects. Previous research used pronouns for the primes (he, she, they) while we used gender specific or gender neutral nouns (man, woman, child, adult). This seemingly small difference may have had a large impact on our results.

APA (2001). Publication Manual of the American Psychological Association. Washington, DC: American Psychological Association.Banaji, M.R. & Hardin, C.D. (1996). Automatic Stereotyping. Psychological Science, 7, 136-141.Bem, S.L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.Buck, J.L. (1990). Nonsexist language: Successes, neologisms, and barbarisms. Teaching of Psychology, 17, 198-199.Collins, A.M., & Loftus, E.F. (1975). A spreading-activation theory of semantic memory. Psychological Review, 82, 407-428.Foertsch, J. & Gernsbacher, M. A. (1997). In search of gender neutrality: Is singular they a cognitively efficient substitute for generic he? Psychological Science, 8, 106-111. Gastil, J. (1990). Generic pronouns and sexist language: The oxymoronic character of masculine generics. Sex Roles, 23, 629-643.Harrigan, J.A., Lucic, K.S. (1988). Attitudes about gender bias in language: A reevaluation. Sex Roles, 19, 129-140.Madson, L., Hessling R. (1999). Does alternating between masculine and feminine pronouns eliminate percieved gender bias in text? Sex Roles, 41, 559-575.Schvaneveldt, R.W., & Meyer, D.E. (1973). Retrieval & comparison processes in semantic memory. In S. Kornlum (Ed.) Attention & performance IV. New York: Academic Press.Sellers, N., Satcher, J., Comas, R. (1999). Children`s occupational aspirations: Comparisons by gender, gender role identity, and socieconomics status. Professional School Counseling, 3, 314-317. Todd-Mancillas, W.R. (1984). Evaluating alternatives to exclusive he. Communication Research Reports, 1, 38-41.