and television shows to be watched anywhere, any time. Many media experiences, such as giant screens with surround sound, for example, are far more intense than ever before. And software for babies—even devices alleged to promote the development of fetuses in the womb—are bringing new forms of sensory intensity to young nervous systems that are highly malleable and vulnerable to significant developmental changes.
What impact might all this have on people and their behavior? Anderson noted that multiple claims are made of negative effects on social behavior, health, education, and cognition (e.g., increases in aggression and obesity, interference with reading, increase in problems with attention), as well as positive effects and opportunities in the same areas (e.g., teaching or modeling prosocial behavior, providing a vehicle for public health messages that reduce obesity, teaching reading and other skills, enhancing attention skills). Some, he explained, view media in general as a toxin, exposure to which should be controlled in the same way that environmental pollutants are permitted in small doses. The problem with this approach is that it does not address the significant variation in media content and the context in which exposure takes place, nor does it account for variations in the findings about the effects of exposure.
A more promising approach may be to consider the components of a media diet in terms of the different types, amounts, and effects of media influences, Anderson suggested. In this analysis, certain media influences could be viewed as desserts or stimulants rather than toxic agents. As with food, the overall quantity ingested is important, but equally important are quality and balance. Thus, for example, children may benefit from information, educational programming—and, most likely, entertainment, as well—although the research base for developing precise guidelines as to the recommended levels or proportions of each is insufficient. While the harmful effects of media influences such as violent programming are well established, the basis for making concrete recommendations related to the degree or nature of the violence, maximum safe exposure, and so forth, is similarly absent. The alternative concept of a media diet accommodates recommendations that address such qualitative aspects as the nature, content, and context of media exposure, as well as the quantity, allowing for flexibility in the face of almost overwhelming pressure to consume media.
An extensive research literature is now available that examines the health effects of media use, especially in the areas of violent and sexual behavior and, increasingly, in dietary choices and eating behaviors. The workshop participants focused on these initial areas to review how research studies currently measure important dimensions of media interactions with children and youth.
Regardless of which metaphor is most useful for public health recommendations, a strong research base is available for some important conclusions about media exposure. Perhaps strongest is the case for the effects of violent programming, which some researchers have suggested is at least as strong as the evidence base regarding links between exposure to tobacco smoke and lung cancer (Oakes, 2006; Anderson et al., 2003). A meta-analysis of the varied research base on this issue has shown that exposure to media violence has demonstrable long- and short-term effects on the likelihood of