(Originally a class paper written in 2003, with the full paper, including references, found here.  Related thinking on this topic can be found here.)

When a farmer plants a field of corn, he must provide several environmental conditions to ensure it grows. There must be ample sunlight, water, and fertilizer. The temperature should neither be too hot nor too cold, and the soil can neither be too acidic nor too alkaline. Given time, with these influences set at optimum levels, the seed will mature into a successful plant.

It is not a far stretch to apply this same cultivation process to understanding how a child develops into an adolescent, then into full maturity as an adult. An individual will go through a series of cognitive developmental steps over her life, and at each step there are numerous sources of influence that can impact the step’s progress, either helping or hindering the outcome. One of these possible sources, known as socializing agents, is the mass media, with the understanding that socialization is a developmental process by which children learn and internalize a society’s norms. Since the inception of the mass media, researchers have sought to understand the possible socializing effects due to media exposure. One such possible effect is the impetus behind this paper; what is the role the media plays in an adolescent’s development of her identity?

One theory used to explore and explain media effects is George Gerbner’s cultivation theory. Cultivation theory posits that television, as the prevalent medium in society, can foster certain perceptions about the nature of society by repeatedly portraying this society in a certain manner (Gerbner, Gross, Morgan, Signorielli & Shanahan, 2002). That is to say, the messages on television, such as criminals can strike anywhere and anytime, are pervasive, not varying from program to program. Television viewers, just through their cumulative exposure to this portrayal of reality, and especially if they are not exposed to another depiction of reality, will come to adopt this portrayal as truth. These “heavy viewers” will in turn see the real world in a way that reflects the world “as seen on TV”. Cultivation scholars investigate the prevalence of certain messages in the media, and analyze how these messages interact with the audience to form “an environment within which people live, define themselves and others, and develop and maintain their beliefs and assumptions about social reality.” (Morgan & Signorielli, 1990, p. 18).

According to Gerbner, then, television acts as the conduit through which individuals learn about the values of the culture to which they belong. Thus, his theory is an attempt to explain how culture imparts its norms to its members so that they can function within it. “Culture provides the overall framework in which we imagine that we do not encounter directly, and interpret what we do encounter directly.” (Gerbner, 1990, p. 251). This paper started with an agricultural metaphor because cultivation theory has its roots in agricultural theories. “Thus, cultivation is an agro-aquatic metaphor for the function of television in the construction and maintenance of cultural meaning, and for the way culture works generally.” (Shanahan & Morgan, 1999, p. 12). Television acts as the conduit through which an adolescent, in the process of negotiating a position in said culture and society, can learn and incorporate those norms that would help her development.

A particular developmental challenge adolescents face is the formation of a stable identity. Collapsing across the several distinctions of identity, the term can be defined as a continuous sense of one’s location within a society, wherein this position is shaped and/or defined by one’s concrete presentation of one’s attitudes, values and beliefs (Côté & Levine, 2002). According to Swindler (1986, as cited in Côté & Levine, 2002), a person’s culture aids in this process by providing a ‘tool kit’ of various resources, along with various socializing agents, that one uses to negotiate an identity. An adolescent can rely on the culture’s and society’s various institutions to tell them how to behave in a given situation. These norms can be accepted passively (Côté & Levine, 2002), or an adolescent could engage in active process of exploration and experimentation with the goal of constructing a stable construct of identity (Moshman, 1999). What is in consideration here is the role the media plays in this formation when it acts as a source for socialization. Do adolescents use the media to explore their world, adjusting themselves to find a way to fit into the world?

As a summation of the scope of discussion on this subject, it can be said that there has been much research and speculation as to how the media acts as a socializing agent. Arnett (1995), in a discussion of numerous studies on this topic, reasoned that as adolescents engage in identity formation, and as childhood socialization sources diminish while adult sources are not yet present, adolescents may be more inclined to use the media to understand who they should be (p. 520). Kelly and Donohew (1999), in postulating the primary socialization theory, attempted to incorporate the role of media into the network of socializing agents, calling it a secondary socializer that could exert influence should the primary socializers (e.g., family, school, peers) falter in some way. But this theory was not directly applied to the prospect of identity formation.

Recently, Lloyd (2002) directly addressed media’s influence on identity formation. He proposed that as the media “become increasingly more entertaining and relevant to the targeted audience members, it is much more probable that adolescent audiences will adapt and use this information as a tool for understanding of self and others.” (p. 75). In constructing the AIMSS framework, Lloyd reasoned that an adolescent could engage with the media to “work out” interactions with her peers before employing these interactions. Media use can increase social competence, which would allow the adolescent to compare herself against others, thus fostering a sense of identity. But in the end, this theory appears to be an indirect effect. Media use for the extraction and internalization of norms is not addressed.

Lloyd’s theory appears to have some background in uses and gratifications perspectives, in that adolescents have a need and seek to use media to gratify that need. Boehnke, Münch and Hoffmann (2002), approaching their research from a uses and gratifications perspective, would concur with Lloyd. “Adolescents use the media to accelerate the accomplishment of their developmental goals.” (p. 195). However, if Arnett (1995) is correct, then it is possible cultivation could have an impact on adolescents’ development as well. “Adolescents, who have fewer sources of information and real-world experiences than do adults, are often particularly susceptible to these messages.” (Morgan & Rothschild, 1983, p. 35). Could the proposed cultivation process help us understand how adolescents internalize norms from the media, which in turn would serve as the foundation for identity construction?

The first step in answering this question is to examine research that has specifically looked at cultivation as the reason for media effects in adolescents. There needs to be proof that the media can cultivate adolescents. Typically in these researches, mostly conducted as correlation studies, the researchers were interested in how exposure to the media (i.e. television) is related to certain perceptions and attitudes. Cultivation researchers would imply causality more than just relationship, but because most of these studies were simple survey designs, they cannot totally prove causation. The majority of studies (i.e., Morgan, 1982, 1987; Potter, 1990; Signorielli, 1991, 1993; Cheung & Chan, 1996) were interested in overall exposure without specifying genre or programs, as cultivation theory does not differentiate between overall and specific exposure. However, there were two studies, including the only experiment (Rössler & Brossius, 2001), which did differentiate from overall exposure by examining how watching talk shows could cultivate certain beliefs about sexual norms (Davis & Mares, 1998). Rössler and Brossius reasoned that by selectively choosing to watch talk shows, adolescents may show more of an effect. “We conclude that the two prerequisites of cultivation effects – uniform message and nonselective viewing – might exist within rather than between television genres.” (p. 147). This implies that a more active viewing could have a greater impact on the cultivation process. The implications of this deviation from traditional cultivation theory will be discussed in more detail later in this paper.

For now, it is sufficient to state that the independent variable in these studies was commonly measuring some level of television exposure. The next step was to see how these exposure levels related to some attitude or perception. When first developing the theory, Gerbner was interested in the impact of violence on perceptions of the world as a “bad place”. Accordingly, there has been research conducted to see if this Mean World Syndrome is identifiable in adolescents (Gerbner, Gross, Signorielli, Morgan & Jackson-Beeck, 1979; Cheung & Chan, 1996). Gerbner’s colleagues then moved on to investigate other possible repercussions from watching too much television. The role of television in fostering sex-role stereotypes has been conducted by Morgan alone (1982, 1987) and with Rothschild (1983). The standardization of certain sexual attitudes was studied by Davis and Mares (1998) and Rössler and Brosius (2001). Signorielli studied how television consumption can impact marital attitudes (1991) and attitudes towards different occupations (1993). Finally, Potter (1990) was interested in how television consumption across genres could influence perception of cultural values, such as “good wins over evil” and “honesty is the best policy (p. 848).

However, at the same time as investigating a possible direct link between exposure and effect, some researchers conceptualized this relationship as subject to moderation or mediation by intervening variables. In a longitudinal study, gender and IQ were found by Morgan (1982) to, respectively, moderate the direction of the effect and mediate for whom the effect was noticed. A similar relationship was found in his second study, whereby acting out behavior that was inline with traditional sex-role stereotypes moderated how the exposure would impact attitudes (Morgan, 1987). Then again with Rothschild (1983), the peer network an adolescent has established mediated the impact of television exposure. While this relationship with peers could be informative, especially when combined with the theory postulated by Lloyd (2002), it is the research by Rössler and Brosius (2001), which incorporated the idea that active selection of exposure could be the mediating factor in the cultivation process, that may prove more useful in answering the question of media use and identity formation. This tantalizing possibility will be discussed after a review of the findings.

Like results from cultivation studies conducted on adult samples, there was a general trend, significantly positive yet weak, that television exposure could predict the fostering of perceptions inline with the televised messages. Gerbner et al. (1979) found that heavy viewers were more likely to describe the real world as dangerous, overestimating the number of people involved in violence and the number of criminals walking the street. Cheung and Chan (1996), in studying Hong Kong adolescents whose bulk of television viewing is of Hollywood product, found results reflecting Gerbner et al.’s findings. Signorielli (1991) found that higher television watching was positively related to the adolescent’s desire to be married when older, yet at the same time they indicated that one sees so few good marriages as to question marriage as a way of life (p. 145), which the researcher argued represented the ambivalent nature of marriage on television. Signorielli (1993) found that adolescents who were over-exposed to programs that over-represented adventurous jobs were more likely to desire having a high paying, prestigious job. After collapsing across genres, Potter (1990) found that greater television viewing could substantiate more recall and perception of cultural values. Again, these are all generally direct relationships.

Studies that also analyzed for moderators and mediators still detected the general trend, but this relationship was conditional on some other factor in the adolescent’s environment. Morgan (1982) found that when an adolescent’s IQ was considered, the relationship was still present for girls with IQ’s higher than average, who were described as being “otherwise” the least sexist (p. 952). Additionally, it was found that time-order predicted the relationship only for girls, not boys; in fact, sexism at time one predicted more viewing at time two for boys, the opposite direction from girls. This was explained by the fact that overall boys held more sexist views than girls, so by being heavy viewers, those girls just converged with the boys on their level of sexist stereotypes. A similar situation was found when considering how engaging in behaviors seen as reflecting traditional sex stereotypes could moderate cultivation (Morgan, 1987). Again, if a boy did not engage in stereotypical behavior, and was a heavy viewer, he was more likely to develop sexist attitudes than a boy already engaging in stereotypical behaviors. This pattern could again be explained by a convergence effect. But for girls, those already engaging in stereotypical behavior, while also being heavy viewers, were more likely to cultivate sexist attitudes. Gerbner would account for these results as a mainstreaming effect occurring for boys, and a reinforcing impact for girls (Gerbner et al. 2002).

Morgan and Rothschild (1983) were interested in how peer networks could impact cultivation. Socialization theory predicts that peers form a way for adolescents to learn about the world. If an adolescent lacks a good peer network, then she would have to turn to other sources, such as the media, to know about the real world. Thus, the argument is that “the effects of television should be more pronounced for those who tend to be less socially integrated.” (p. 37). In the end, the results refuted this claim. As the number of friends went down, and the amount of television consumed increased, then incorporation of sexist attitudes likewise increased. The researchers reasoned that the effects of television could be mitigated and suppressed by alternate information sources (p. 48). Relating this finding to the Lloyd (2002) theory, there is evidence that, at least compared to his conceptualization, cultivation theory is not a process related to identity formation. If Lloyd is correct and adolescents use the media for social competence, than higher exposure and having more friends would be related to more incorporation of sexist attitudes, because one would be integrating these norms to conform to friends. While this could still be happening, the findings of Morgan and Rothschild did not support the claim. However, as mentioned before, the Rössler and Brosius (2001) study could aid the answering of the question.

Rössler and Brosius (2001) studied exposure, but it was mediated by being selective exposure. Preceding them, Davis and Mares (1998) found that exposure to talk shows was related to overestimation, but not attitude endorsements. However, Rössler and Brosius found that exposure to talk shows on homosexuality lead to overestimating the prevalence of homosexuality in society as well as generating less restrictive attitudes towards homosexuals. The difference may lie in how the studies measured exposure. Davis and Mares asked for the frequency of watching talk shows, while Rössler and Brosius showed the adolescents talk shows chosen for specific content, such as homosexuality, and then measured for attitudes based on this selective exposure. Thus, Rössler and Brosius took Davis and Mares concept of audience activity one step further. And although the adolescents were shown the television clips in an experiment setting, the study presupposes that viewers actively select what television they want to be exposed to. This presupposition touches upon one of the most vocal critiques on cultivation theory, that it perceives the audience as passively watching and accepting television’s messages. However, research has shown that adolescents are particularly active consumers of “identity-related content” (Levy & Windahl, 1985, p. 120). If this is true, then there could be a spanner in the works; if cultivation theory presupposes passivity, but adolescents are more active media users than either children or adults, how then could cultivation apply to identity formation?

As it turns out, this activity-passivity dichotomy has received, and is still receiving, much discussion amongst those interested in cultivation research. Signorielli (1986) does not see any need for a distinction, because different genres and programs will portray the same issues, themes and values in the same manner. Thus, no matter what program a person chooses, they will still be exposed to the same messages. “In this sense, television programming is necessarily a relatively non-selective activity.” (p. 75). However, research has been done to show that activity can both decrease and increase the effect of cultivation. Reimer and Rosengren (1990) reasoned that active persons would be more susceptible to cultivation effects because, by selectively choosing which messages to expose themselves to, they would undergo a conscious cultivation. In other words, “active subjects using the media consciously may be more sensitive to a message or a system of messages than a passive viewer/reader would be.” (p. 187).

In support of this assertion, Kim and Rubin (1997) found that an increase in fear of victimization, a cultivation effect, was positively correlated with content involvement and selective perception of program content, but there was no direct link with instrumental viewing methods and selective exposure. Thus, certain aspects of audience activity were shown to increase the amount of the cultivation effect. However, at the same time, Rouner (1984) found a negative relationship between active television watching and the cultivation of a mean world perception. Thus, the debate goes on. What is known for certain is that in the studies conducted on adolescents and included in this report, none directly measured how active an adolescent is in media consumption, and what this activity could do to the cultivation of certain attitudes. But at the same time, it has been discussed that adolescents are typically active media users (Levy & Windahl, 1985) and could be active selectors when it comes to identity formation (Arnett, 1995; Moshman, 1999).

So where does this leave us in our understanding of whether or not cultivation theory could be a viable explanation for how adolescents use the media in developing their identity?

The point has been reached to determine that there is some preliminary support to say that cultivation could be applied to identity formation, but such an application at this time would be tentative for several reasons. First, cultivation has yet to be proven as a causal relationship when it comes to adolescents. Of the studies discussed in this report, only one was an experiment, where cumulative exposure was substituted for prolonged-exposure (Rössler & Brosius, 2001). But is watching a series of television clips over a five-day span truly equitable to weeks, months, years of constant exposure, such as an adolescent performs in her daily life? The longitudinal studies of Morgan (1982, 1987) do better to address this factor, but then they don’t address activity, which is the second reason.

Activity is important when discussing adolescents because numerous studies have indicated that adolescents seek out the media as a new form of socializing agent (Arnett, 1995). They tend to seek out media that give them greater autonomy from their families and more integration with their peers (Larson, Kubey & Colletti, 1989), which leads them away from television consumption (the “family” room) and towards music consumption (the bedroom). In support, Boehnke, Münch and Hoffmann (2002) found that adolescents seeking autonomy turned to the radio to fulfill this developmental need. In fact, research has indicated that heavy television viewers, those cultivation theorists say are most susceptible to television, spent more time with their family and less time with friends (Larson et al., 1989). Is it possible that the cultivation effects found in the studies discussed were due more to the socialization impact of the families and not the television? Yet another aspect for future investigation, but it touches on a problem of integrating cultivation theory into understanding a process of media’s impact on identity formation.

Cultivation theory predicts that television exposure would cause the internalization of a culture’s norms. But what if an adolescent is not watching as much television as they are listening to the radio, reading magazines or going on the Internet? Can it be said that cultivation effects are occurring when an adolescent chooses to consume media other than television at levels equal to or greater than her consumption of television? This is a question I have not yet seen an answer for in research, although much has been speculated on how newer media technologies, such as the Internet, would impact cultivation effects.

Shanahan and Morgan (1999) argue that history has shown that a new media, when introduced into a culture or society, will adopt the patterns of message content the dominant media had been employing. The impact on cultivation is not so much from the medium, but from the content that medium contains, because cultivation theory is a theory of storytelling. Gerbner et al. (2002) argued that the predominant online sites are all linked with some dominant television producer, such as Disney and Time-Warner. The question of higher consumption of the Internet displacing time spent with television is likewise dismissed. “Rather, the emerging evidence indicates that audience members add online time on to other previously existing media use patterns, extending (not redistributing) the overall amount of time spent with media.” (Morgan & Shanahan, 1990, p. 214). Thus, cultivationists argue the Internet is no treat to television, not only because it is falling under the control of mainstream Hollywood, but also because there is no strong indication it detracts from time spent with television.

While this may be true for adults, how true is it for adolescents? Larson et al. (1989) found that as adolescents grow older, they spend less time watching television. A study conducted by the National Science Foundation (Papadakis, 2001) found that the greatest predictor of Internet use was generational differences, with teens more likely to use the Internet than adults. Also, the NSF report cited studies indicating that households with the Internet tended to watch less television (p. 21). Stranger (1998) found that if children had computers in their home, they would spend less time in front of the television. Woodard and Gridina (2000), in looking specifically at children, found that adolescents spend the most time using the Internet; however, television use remained stable for children. At the same time, though, the study did not differentiate Internet use and displacement per age group, but instead lumped all ages into one group and found no change in television use. This raises a flag in the face of other research, saying that adolescents watch less television than younger children. Is it possible then that adolescents, just as they listen to music more, are going online more than they are watching television?

Ah, yet another question that does not yet have a satisfactory answer. And the answer could be changing as more and more young people are growing up with computer and digital technologies when compared to older generations, especially when considering how online services like Napster, Kazaa and Apple Music could be fulfilling their need for music.

So what is the current tally? Could cultivation be applicable to identity formation? As said before, several conditions need to be meet for Gerbner’s theory to apply. First, cultivation needs to be proven causally. Second, active media use and its relationship with identity formation needs to be explored. “Unfortunately, this model of active development is still missing from much of the research and theory on adolescents and media.” (Arnett, Larson & Offer, 1995, p. 514). If it is proven that adolescents actively seek out media to aid in this developmental challenge, then for cultivation to apply, it has to be shown that active media use is not detrimental to the cultivation process. Third, cultivation effects need to be shown to still have an effect even if television viewing is displaced by other media consumption, such as radio, magazines and the Internet.

All three of these conditioning statements need further study before it can be concluded that an adolescent can internalize a culture’s norms from media consumption, in a manner predicted by cultivation theory, with the goal of forming a sense of identity and a sense of place and purpose within that culture.