January 22-23, 2004
Renaissance Austin
Austin, TX
This meeting was held in conjunction with the National Children’s Study, which is led by a consortium of federal agency partners: the U.S. Department of Health and Human Services (including the National Institute of Child Health and Human Development [NICHD] and the National Institute of Environmental Health Sciences [NIEHS], two parts of the National Institutes of Health, and the Centers for Disease Control and Prevention [CDC]) and the U.S. Environmental Protection Agency (EPA).
Co-Chairs: Christine A. Bachrach, Ph.D., NICHD, NIH, DHHS, and Ellen Wartella, Ph.D., University of Texas at Austin
Welcome
Dr. Bachrach welcomed participants and described her role. The key challenge of the workshop is to establish how media effects can contribute to the Study and how the Study can contribute to the study of media effects on child health and development.
Introduction to the National Children’s Study and Questions
Peter C. Scheidt, M.D., M.P.H., NICHD, NIH, DHHS
Children have increased vulnerability to environmental exposures. They have critical windows of vulnerability, immature mechanisms for detoxification and protection, and metabolism and behavior differences that compound risk for children. Childhood exposure considerations include greater surface area to body mass ratio, higher respiratory ventilation rate, greater daily water intake, greater daily soil ingestion, and higher rate of lead absorption. The Study was created to develop strategies to reduce the risk of environmental exposures to children. Its longitudinal design could infer causality with multiple exposures and multiple outcomes.
The Study must address generalizability to the U.S. population, high-risk populations, and factors such as agricultural and industrial environments and the effects they may have on fertility and pregnancy in women of childbearing age. Sampling strategies could include household, office-based, and medical center models. Issues for determining the sample include national probability, dispersed clustering, concentrated clustering, and center-based convenience. Design and development will involve debate that will lead to consensus. Examples of hypotheses for the five priority outcomes were given and pilot studies, literature reviews, white papers, workshops, and the timeline for the Study were discussed.
Review of Meeting Purpose and Objectives
Ellen Wartella, Ph.D., University of Texas at Austin
Dr. Wartella described the following goals for the Media Effects Workshop:
- Discuss hypotheses for media effects on children
- Identify broad hypothetical issues that implicate media effects on health and development
- Develop testable hypotheses that link exposures to outcomes
- Provide justifications/rationales for the hypotheses and set priorities based on significance of public health outcomes
- Discuss measurement challenges and issues.
Hypotheses must provide a framework for the Study population, sampling schemes, data collection, and follow-up schedule, as well as justify the large sample size and prolonged follow-up. Hypotheses must be amenable to scientific inquiry, have compelling scientific merit for child health and development, and focus on topics not already thoroughly studied.
Establish Priority Media Hypotheses
Hypotheses were then nominated based on discussions of exposure, age-related impact, amount and cumulative effect of exposure, nature of exposure, complexity of media and developmental outcomes, subgroup vulnerability, short- and long-term outcomes, moderating factors, critical public health outcomes, risk-taking behavior and accidental injury, research on developmental risk over time, outcome trajectories set early in life, technology and defining media, portability of media, physical effects of media, gene-environment interaction, genetic predisposition, effect of family environment on exposure outcomes, effects at various developmental stages, and variations in media format and their effects in different subgroups.
Discussions concerning sample size explored subgroups, rare exposures, rare outcomes, and interactions of types of exposures. Subsequent deliberations and refinements identified nine topics for hypothesis development including obesity, neurological development, cognitive development and academic achievement, sexual behavior and orientation, violent behavior and aggression, substance use, prosocial behavior, physical development, and unintended injuries.
Obesity
Discussion on obesity focused on media exposure and its effects on caloric intake and physical activity, and whether media exposure to sports increases physical activity. Subgroup vulnerability was addressed. Participants also discussed effects of media on activity level, measurement and differences in basal metabolic rate and genetic predisposition to obesity, amount/time of media exposure, interaction of predisposition and media effects, causality, advertising, nutrition, snacking, satiety, cultural aspects of eating disorders, and at-risk subpopulations. A hypothesis for media effects and obesity was formulated:
- Children who are genetically predisposed to obesity are at risk of having a relationship between media use and weight status.
Neurological Development
Current theory postulates that in the first 2 years of life, a child’s attention to screen media establishes a trajectory for later media use, and some believe this exposure affects neurological and cognitive outcomes. Environmental stimuli affect brain development from birth to 2 years of age, and audiovisual media exposure may alter neuronal development during this critical time. The amount of cognitive interaction a child has with the media affects neurological development, and the American Academy of Pediatrics (AAP) recommends that children under 2 years not be exposed to screen media. Several participants noted that there is little research to substantiate this conclusion.
Hypotheses formulated for media effects on neurological development were:
- Mature brain function and architecture (neuronal growth and development) are altered by exposure to audiovisual media during the first 2 years of life.
- Complex brain function and cognitive development that are necessary for today’s society are fostered by audiovisual media use.
Cognitive Development and Academic Achievement
There is little evidence that children under 2½ years of age learn from television exposure. This may be due to the lack of learning assessment tools for preverbal children. Changes in visual perception at about 15 months allow children to interpret two-dimensional images in screen media as representations of three-dimensional objects. Children may then be able to learn from screen media. After 2½ years of age, cognitive development and academic achievement depend on media type, content and amount of exposure.
Participants discussed media displacing other activities, effects of educational media on children with less cognitive stimulation in the home, defining educational content, the relationship of violent content and academic achievement, media effects on children from non-English-speaking households, media use in schools, foreground versus background media exposure, media effects on attention span at age 6, educational/non-educational content effects, and content analysis.
Hypotheses for cognitive development and academic achievement were:
- Media have strong, complex effects on cognitive development; effects are determined by media type and content.
- Media exposure from birth to 2 years of age is influential in setting a trajectory for media use.
A participant subgroup suggested an experiment be embedded within the Study to provide incentives to a randomized subsample of parents to (1) have no television in the house in the first 2 years of life or (2) view educational programming only. Volunteers will develop this option.
Sexual Behavior and Orientation
Discussions of media effects on sexual behavior and orientation covered socioeconomic status and reliance on media for information on sex, sexual stereotypes in the media, voluntary versus involuntary exposure to media with graphic sexual content, media content and frequency of exposure, privacy issues, determining trajectories of sexual behavior, and the impact of graphic sexual content on sexual orientation. Participants acknowledged that there is scant information on media effects across subgroups and agreed maintaining consonance in coding media content across cultures would be difficult. There may be media access confounds based on socioeconomic status. The U.S. Surgeon General Richard Carmona and members of Congress expressed interest in media effects on violence and sexual behavior. Amount of sexual content in various media formats differs, and thus, assessing exposure to specific media formats may more accurately quantify exposure to sexual content. There is a relationship between sexual orientation and suicide. Media exposure regarding sexuality-but not necessarily sexual behavior-may affect subgroup behavior.
Preliminary hypotheses for media effects on sexual behavior and orientation were:
- Exposure to media with sexual content leads to risky sexual behaviors.
- Children and youths exposed to media messages about the risks and responsibilities of sexual activity are more likely to develop healthier sexual behaviors.
Violent Behavior and Aggression
Participants discussed media effects on violent behavior and aggression. Topics included boys’ increased susceptibility to violent content, vulnerability among subgroups, effect of social controls, effects of "real" violence versus "virtual" violence, cumulative effects, psychological response to violence, cultural acceptance of violence and aggression, use of weapons, perceived magnitude of violence in America, and resilience versus vulnerability to violent content.
A multi-national study on fighting shows rates were similar across countries, but weapons use and carrying was more prevalent in America. Another study contained data on the magnitude of violence in the United States, indicating that violence has increased. Social, familial, and biological factors of vulnerability or resistance to the effects of violent media content were discussed.
A null hypothesis on this topic is acceptable and such an approach circumvents implications that researchers do not know which subgroups are resilient or vulnerable to violent content.
Hypotheses for media effects and violent behavior and aggression were:
- Exposure to violent media content increases the propensity for violent behaviors and some subgroups are more vulnerable to violent content.
- Vulnerable subgroups may include children lacking family and cultural controls, younger children, those with a biological propensity toward violent behavior, and children with violent family environments.
Substance Use
Topics regarding media effects on substance use included age, biological predisposition, thrill seeking, advertising, materialism and related mental health issues, portrayal of substance use, definition of "normative," and use mediators. Recent research suggests advertising has a significant impact on young people, but its effects on preschool-age children remain unclear. A participant noted that the concept of normative behavior is reliant on the amount of media exposure.
The hypothesis for media effects on substance use was:
- Exposure to media content portraying substance use as normative behavior increases propensity for that use.
Prosocial Behavior
Participants discussed the relationship of prosocial behaviors to self-esteem, the definition of prosocial media content, and content analysis and categorization.
The hypothesis for media effects and prosocial behavior was:
- Exposure to prosocial media content increases the likelihood of altruism, empathy, cooperation, volunteer work, community participation, and "helping" behaviors.
Physical Development
With regard to media formats and physical outcomes, participants discussed distorted perceptions of interactive media; repetitive motion injuries; ergonomic issues; eye strain; spatial, iconic skills; cumulative effects; and radiation/chemical exposure. Participants acknowledged that media apparatuses will evolve over time and that it may be difficult to predict future physical outcomes from media use. No hypothesis was formulated for this topic.
Unintended Injuries
The participants discussed combining unintended injuries with physical outcomes, and also talked about portrayals of driving, wrestling, stunts, and risky physical behaviors. No hypothesis was formulated for this topic.
Measurements and Outcome Measures
A new database development project designed to compile research results into a single online source similar to Medline was described. Determining content and performing content analysis is the greatest challenge. One approach is to obtain and record titles of media exposure. Pilot studies may be performed to determine the focus of children’s attention in various media. Automated data collection methods will be used as much as possible. Data will be gathered in telephone and face-to-face interviews, as well as mail-in written surveys, Web-based tools, and medical records. Participants discussed ways to monitor and record media use. Technologies include personal digital assistants, cameras, cell phones, and computer and television monitoring devices. Participants also discussed other data collection tools and reliability problems of automated data collection and developed a list of media use constructs and measurements that included time use diaries/experience sampling/time use analyses, favorites, most used/watched, parent mediation/co-viewing, home environment media inventory, parent attitudes, media content imitation, media focus, and computers and monitoring devices. Measurement discussion covered use of technology for communication by older children, pilot studies to analyze non-television media content, multi-tasking and simultaneous overlapping use of different media, content analysis of complex/overlapping media, and media use during daycare/childcare.
Recommended Hypotheses
Participants crafted an umbrella hypothesis for media effects on child health and development, consolidated several of the preliminary hypotheses, and categorized them in five topic areas.
Overarching hypothesis. The amount, type, content, and context of media exposure from infancy through adolescence critically influence cognitive, social, behavioral, and physical outcomes. Effects and effect sizes will differ with timing of exposure and cumulative amount of exposure and in different populations with differing vulnerabilities.
Physical development. Media exposure contributes to obesity and other nutrition-related disorders (bulimia/anorexia, type II diabetes) through sedentary behavior due to displacement of physical activity; through increased intake of nutritionally poor food due to advertising, as well as food intake during media use; and through distorted body image as a consequence of exposure to idealized human images.
The relationship between media exposure and body mass index is moderated by genetic predispositions as partially reflected in a family history of obesity.
Alternative: The correlation between media use and obesity is the result of the causal effect of obesity on increased sedentary activities, including media use.
Media devices used for extended periods of time lead to increased risk of repetitive strain injuries and musculoskeletal disorders. Effects may be cumulative, especially when the content reinforces repetitive motion and sustained use and when the ergonomics of the device are not appropriate for children.
Children who are heavier users of media involving immersion in virtual environments-and to a lesser extent, videogames-will be at increased risk for problems in visual development.
Children who are heavier users of portable media have higher risk for myopia and other eyesight problems.
Neurodevelopment. Brain function and architecture are altered by exposure to audiovisual media during critical development between birth and 2 years of age. This will be reflected in long-term cognitive, attentional, and behavioral outcomes.
The relationship between media use and cognitive development and academic achievement is primarily due to content rather than time spent with media.
Children exposed to educational media content have enhanced development of cognitive skills and improved academic achievement. These effects are maximized in the developmental period between 2 and 5 years of age, where other sources of cognitive support and stimulation are minimal, where English is not the first language, and when children are exposed to same-ethnic models in the media.
Conversely, noneducational entertainment content, especially violence, has a negative influence on cognitive development and academic achievement.
Social development. Children who are consistently exposed to prosocial media content early in their development are more likely to demonstrate prosocial behaviors and to select prosocial media content over the course of development.
Children who are at risk and vulnerable (for example, due to attention deficit hyperactivity disorder) will be more likely to demonstrate favorable developmental outcomes (that is, be more prosocial) when they grow up in stable homes where prosocial media are frequently shown.
Children who are at risk and vulnerable (for example, due to attention deficit hyperactivity disorder) will be more likely to demonstrate unfavorable development outcomes when they grow up in homes where coercive discipline is used and there is more exposure to antisocial media than to prosocial media. These children are at high risk to become juvenile delinquents.
Behavioral development. The heavier the diet of violent media exposure, the greater the risk of aggressive behavior and desensitization (including acceptance of violence). The effect is amplified by earlier age of exposure, developmental delay and/or brain injury, violence in the home, poverty, and peer-group aggression during early adolescence.
Heavy exposure to news and other "realistic" media violence increases the propensity for children to experience chronic fear and anxiety. This effect is increased by living in high-crime areas, predisposition for anxiety, and personal experience with victimization.
Exposure to media portrayals and ads of alcohol, tobacco, and illicit drugs predicts beliefs about substance use norms and practices and early and increased substance use. The relationship between exposure and substance use is moderated by the nature of media portrayal (prosubstance versus antisubstance), genetic predispositions to substance use, individual sensation seeking/risk taking, exposure to substance use in family, exposure to substance use by peers, access to substances in community, and state/local policy governing penalties for selling alcohol or tobacco to minors.
The amount of exposure to sexual content in entertainment media predicts beliefs about sexual norms and practices, early sexual activity, greater number of sexual partners, increased frequency of sexual activities, and less frequent use of condoms and other protection. Through these outcomes, it also predicts higher incidence of unplanned pregnancy and sexually transmitted diseases.
Children and youths exposed to media messages and information about sexual risks and responsibilities are at lower risk for sexually transmitted diseases or experiencing/causing unplanned pregnancy in adolescence.
Subgroup depictions. Positive or negative portrayals of gender, race, ethnicity, sexual orientation, and disabled groups in the media are related to the formation of stereotyped social expectations and moderate media influences on social and behavioral development, depending on the degree to which the viewer can identify with the portrayal.
Participants
Daniel R. Anderson, Ph.D., University of Massachusetts
Christine A. Bachrach, Ph.D., NICHD, NIH, DHHS
Adelaide Barnes, B.A., NICHD, NIH, DHHS
David Bickham, M.A., University of Texas at Austin
Sandra Calvert, Ph.D., Georgetown University
Allison Gilman Caplovitz, M.A., University of Texas at Austin
Hiroshi Hattori, Japan Broadcasting Corporation
Dale Kunkel, Ph.D., University of California, Santa Barbara
June Lee, University of Texas at Austin
Jesse N. Marquez, Coalition for a Safe Environment
Carol J. Pardun, Ph.D., University of North Carolina
Michael Rich, M.D., M.P.H., Harvard University
Vicky Rideout, M.A., Kaiser Family Foundation
Donald F. Roberts, Ph.D., Stanford University
Fuki Hayamizu Saal, Japan Broadcasting Corporation
Peter C. Scheidt, M.D., M.P.H., NICHD, NIH, DHHS
Vic Strasburger, M.D., University of Mexico School of Medicine
Elizabeth A. Vandewater, Ph.D., University of Texas at Austin
Ellen Wartella, Ph.D., University of Texas at Austin
Barbara J. Wilson, Ph.D., University of Illinois