OBJECTIVE. The purpose of this work was to conduct a systematic review of experimental trials for the effects of television viewing by infants and preschoolers.
METHODS. Searches were conducted as of January 2005 on several online bibliographic databases for the disciplines of medicine, psychology, education, and communications. Relevant bibliographies were also reviewed. The article contents were critically evaluated to determine whether they met inclusion criteria. Data were extracted from each included article regarding study methods and results.
RESULTS. Twelve controlled trials met all of the criteria for inclusion in this review. All of the studies focused on the effect of television content rather than viewing time. The findings suggest that educational television programs are successful in broadening young children's knowledge, affecting their racial attitudes, and increasing their imaginativeness. There is insufficient experimental evidence for effects of viewing these programs on either children's prosocial behavior or their aggressive behavior. Finally, there is some evidence that viewing cartoons has a negative effect on children's attentional abilities.
CONCLUSIONS. A number of gaps remain in the literature. Although the amount of time children spend viewing television raises concern, the studies presented here focus only on content. Despite the fact that infants are the fastest growing television market segment, the controlled trials only include children aged ≥3 years. Finally, the clinical utility of the findings are questionable given that most studies included small group sizes, all took place in nonnaturalistic settings, and all but 1 study only evaluated short-term effects of television viewing.
Young children in the United States watch an astonishing amount of television, spending more time in front of a screen than any other single activity except sleeping.1,2 The American Academy of Pediatrics recommends that children ≥2 years of age have <2 hours of screen time per day and that children <2 be discouraged from TV watching.3 Despite growing concerns, ample evidence indicates these recommendations are ignored.1,2,4,5
The American Academy of Pediatrics guidelines refer to a number of adverse effects of television viewing including aggressive behavior, obesity, and decreased school performance.3 Although the bulk of research focuses on adolescents and school-aged children, a growing body of literature focuses on infants and preschoolers. Observational studies suggest positive effects on learning and behavior from viewing educational programs specifically developed for younger children.6,7 Other studies have correlated viewing noneducational programs with negative outcomes, such as aggressive behavior, obesity, risk-taking behavior, and attentional problems.5,8–10 Although these findings are often cited in both academic literature and the popular media, a causal connection cannot be determined from observational studies: such a connection requires experimental designs. The current state of the science with respect to the effects of commercially available television on young children is not widely known. Accordingly, we performed a systematic review of experimental trials of the effects of television viewing by infants and preschoolers. Our goal was to summarize and synthesize the current state of experimental television studies involving preschool children.
For inclusion, a study needed to be a controlled trial involving prekindergarten subjects, as demonstrated by recruiting from preschools or having an average age <6 years. In addition, the experimental group needed to watch television programs that were broadcast for public viewing; studies were excluded if they involved only commercials or only video programs that were never commercially broadcast. Studies were still included if broadcast programming was edited to meet the needs of the researchers, as long as no new material was introduced. Finally, the measured outcomes needed to be relevant to the effects of viewing rather than only measuring comprehension or attention to the program. We restricted our analysis to studies published in the English language.
Searches were conducted as of January 25, 2005, on several bibliographic databases covering the disciplines of medicine, psychology, education, and communication. The databases included Medline, the Cochrane Library, PsycINFO, ERIC, and Communication Abstracts. Search terms used included “televis* OR TV” AND “child*” AND “experiment* OR random* OR control*”. In addition, relevant bibliographies were reviewed for any studies not identified by the electronic search. The contents of abstracts or full-text articles were then reviewed independently by 3 individuals to determine whether they met the criteria for inclusion in this review, and all of the disagreements were resolved via consensus. Authors were contacted for additional information when necessary. Data were extracted from each article regarding study methods and results. This approach was intended to capture all of the published literature on the subject. Given the paucity of rigorous research, a comprehensive strategy seemed appropriate.
The systematic literature search identified 62 articles from Medline, 26 from the Cochrane Library, 112 from PscyINFO, 7 from ERIC, and 166 from Communication Abstracts; hand searches of relevant bibliographies yielded an additional 3 articles (numbers are not mutually exclusive). Articles were most commonly excluded because they were not controlled designs, the subjects exceeded our age criterion, or the outcomes measured did not include the effects of television viewing. Another reason for exclusion was that subjects viewed something other than broadcast television programs.11–17 Twelve controlled trials met all of the criteria.
Of note, we did not require that trials use strict random assignment in allocating subjects to groups. Several included trials randomly assigned subjects but then reallocated individuals to ensure an equal distribution of characteristics, such as age or baseline familiarity with the television program. Given the small number of subjects, the experimenters deemed these adjustments necessary to prevent confounding.
The characteristics of the included trials are shown in Table 1. All of the studies but 118 looked only at short-term effects of television viewing, and each study focused on the effect of television content rather than total amount of viewing. In terms of content, 4 studies used noneducational television programming19–23; the others used educational programs targeted to subject age group. Some studies used no television viewing as a control, others used “neutral” television viewing, and some compared across different types of television content. The outcomes evaluated fall into the following 6 domains: learning,18,19 racial preference,24 prosocial behavior,22,25–27 imaginative play,22,25,26,28 aggression,20–23,25,27 and self-regulation (Table 2).20,25–27,29
There is evidence that television viewing can aid acquisition of general knowledge plus improve overall cognition. A study that looked at long-term impact was that by Diaz-Guerrero and Holtzman,18 which studied the effects on cognition and learning of viewing Plaza Sesamo, a Spanish-language version of Sesame Street. During the show's first telecast season in Mexico City, Mexico, 173 children aged 3 to 5 years were randomly assigned to watch either Plaza Sesamo or cartoons and other noneducational TV programs in a day care setting, watching 50-minute segments 5 days per week for 6 months. Academic achievement tests were individually administered before exposure to the programs, 7 weeks later, and 6 months later (at the end of the experiment). Highly significant differences were found for content-achievement tests measuring skills specifically taught by Plaza Sesamo, including general knowledge (P < .001), numbers (P < .001), and letters and words (P < .001). The experimental group also significantly outperformed the control group on cognitive and oral comprehension tests only indirectly related to Plaza Sesamo (P < .01 and P < .001, respectively). Similar to the previous study, the older children showed greater gains than the 3-year-olds. The findings suggest that children not only learn content but also make general cognitive gains by viewing educational programs. Furthermore, unlike the other trials included in this review, this trial shows an effect over time rather than simply an immediate effect of television viewing.
There is evidence that educational television shows that emphasize diversity can change children's racial attitudes. Gorn et al24 found that multiracial inserts into Sesame Street changed the short-term intergroup attitude of preschool children in Canada. In a 3-arm study, white preschoolers watched Sesame Street episodes featuring inserts with either white and nonwhite children playing together in a familiar setting or only nonwhite children playing in an ethnic setting. Children in the control group watched the Sesame Street episodes without the inserts. After watching the program several times in one sitting, the subjects were asked to identify from photographs groups of children whom they would like to play with. Children who saw the inserts were more likely to select a photograph of nonwhite children, 71.4% of those watching the integrated insert and 70.7% of those watching the nonintegrated insert, compared with 33.3% of the children in the control group (P < .01).
There is insufficient experimental evidence on whether television viewing impacts children's display of prosocial behaviors. Friedrich-Cofer et al25 observed 141 children (ages 2.3–5.3 years) in Head Start centers before and during their viewing of neutral films or specially selected episodes of Mister Rogers' Neighborhood that represented prosocial themes. The subjects in each condition saw 20 films in 8 weeks. They were observed and scored for prosocial behaviors at random times during a 3-month baseline period and again during the 8-week experimental period. No significant difference was observed between groups in the number of positive social interactions with and prosocial behavior toward peers and teachers.
Huston-Stein et al22 observed 66 preschool boys and girls (ages 3.5–5.8 years) in same-gender paired play sessions immediately before and after viewing children's programs falling into 1 of 3 categories: (1) high-action/high-violence, (2) high-action/low-violence, and (3) low-action/low-violence; the control group watched no TV. There were no significant differences across groups on observed positive social interactions. Tower et al26 exposed 3 groups of preschoolers (median age: 4.1 years) to daily viewings of Mister Rogers, Sesame Street, or a control series of nature films over 2 weeks. The 58 children were observed during unstructured play periods on 2 separate days both before and after the intervention. No significant differences in behavior were observed.
Finally, Friedrich and Stein27 showed 92 preschool children (aged 3.8–5.5 years) 1 of 3 types of television programs each day for 4 weeks: aggressive cartoons (Batman and Superman), prosocial programs (Mister Rogers), and neutral films. Behavior observations were conducted during the entire session. Interestingly, during the experimental period, children in the lower half of the socioeconomic status distribution and assigned to the prosocial viewing condition increased considerably in prosocial interpersonal behavior (P < .05) compared with their counterparts in the other viewing conditions. However, these differences did not continue through the postviewing period. Furthermore, the higher socioeconomic status children showed little change in behavior during either time period.
There is evidence that children's imaginative play can be affected by television content. Psychologists describe the imaginativeness of children's play as important for cognitive development, shaping the ways children interact with their environment. Friedrich-Cofer et al25 found no differences in imaginative play between children who watched neutral films and children who watched Mister Rogers. However, 3 other studies did find that television viewing significantly increased children's imaginative play. Huston-Stein et al22 found that children who watched low-action/low-violence programs showed more imaginative play, with considerable decreases seen among children in the high-action/high-violence group. Interestingly, those who saw no television showed a modest increase in imaginative play, and those in the high-action/low-violence group remained stable. All of the differences were statistically significant (P < .05). The authors concluded that it was not the presence or absence of television but the content that was crucial to encouraging fantasy play.
Two other studies specifically chose to show subjects Mister Rogers, because it places considerable emphasis on make-believe. Singer and Singer28 showed 3- to 4.5-year-old children a half-hour Mister Rogers segment daily for 2 weeks. A control group watched no television and was led by a trained adult in open-ended group play activities. The 60 children were observed and scored for imaginative play on 2 occasions each before and after the intervention. The experimental group showed an increase in imaginativeness of play compared with the control group (P < .003). A third group, which watched Mister Rogers with an adult who occasionally called the children's attention to specific events and points made by the program, showed an even greater increase in imaginative play. This suggests that educational television shows are more likely to achieve their desired effects if a concerned adult mediates the viewing.
Tower et al26 found that low baseline imagination was associated with a greater gain in imaginative play after watching Mister Rogers and Sesame Street but only when analyzing each show separately (P = .001, significant for all 3 groups: Mister Rogers, Sesame Street, and neutral films). Furthermore, the low-imagination children who watched Mister Rogers showed significantly greater gains in imaginative play than the low-imagination children who watched Sesame Street (P < .001). They attributed this to the stronger emphasis on pretend play in Mister Rogers and suggest that content should be explicit to achieve an optimal effect.
There is minimal experimental evidence that television viewing increases children's display of aggression. Steuer et al23 showed 5 preschool children (ages 3–5 years) 10-minute aggressive programs daily for 11 days. A control group of 5 children watched nonaggressive programs. Interpersonal aggressive behavior during free play immediately after viewing was recorded and compared with recordings during a previous 10-day baseline period. The children who viewed aggressive programs showed significantly greater increases in aggressive behavior than did the control group (P < .05). However, other trials have found no change in aggressive behavior after television viewing. Potts et al21 showed pairs of preschool boys (ages 3.3–6.3 years) animated and live television programs that varied in violent content (high or low) and action level (high or low). The 73 boys' postviewing social behavior was then observed as they engaged in free play with each other. Neither high violence nor high action had a significant effect on behavior. Geist and Gibson20 showed 62 children aged 4.1 to 5.6 years either an episode of Mister Rogers or an episode of The Mighty Morphin' Power Rangers. A control group watched no television, instead playing with instructional materials. After the treatment period, the children were observed during free play for “rough and tumble” activity; again, no significant differences were observed between groups. Huston-Stein22 also did not observe any statistically significant differences in aggressive behavior after the viewing of high-action cartoons.
Friedrich-Cofer et al25 found that children who viewed Mister Rogers scored lower on indices of aggression than children who viewed neutral programs, but the difference was only significant when the results were stratified by how structured the preschool classrooms were. Even then the difference was only significant for prosocial aggression (ie, children telling peers how they feel or standing up for their rights), not the usually accepted meaning of aggressive behavior, such as name calling or hitting. Finally, Friedrich and Stein27 found that for children who were above average in initial interpersonal aggression, exposure to aggressive cartoons led to higher levels of interpersonal aggression than similar children showed neutral programs. However, this pattern did not hold for children with low levels of initial aggression. This finding suggests that television violence may encourage the expression of existing impulses, rather than acting equally on all children.
There is disagreement in the literature regarding the effect of educational television viewing on children's ability to self-regulate their behavior. Behavioral outcomes grouped under self-regulation include attention, task persistence, impulse control, and tolerance of delay. Friedrich-Cofer25 found no difference in self-regulatory behavior between children who viewed neutral films and children who viewed prosocial programming (Mister Rogers). Similarly, Tower et al26 found no significant differences in postviewing concentration among children who viewed Mister Rogers, Sesame Street, or nature films.
Experiments comparing children's behavioral self-regulation after viewing educational programs versus entertainment programs did find significant differences. Friedrich and Stein27 observed children in their classroom for persistence in completing assigned tasks and in play and tolerance of delay, which was defined as patiently waiting for materials or adult attention for ≥1 minute. They found an increase in tolerance of delay for children who viewed Mister Rogers or the neutral programs compared with a decrease in delay tolerance among children who viewed the action programs, Batman and Superman (P < .05). The results for task persistence were more complicated, where a significant effect was only found in children who scored above average in baseline intelligence testing. Among those children, the children who viewed Mister Rogers became more persistent, the children who viewed the action programs became less persistent, and the control group showed little effect (P < .05). The authors concluded that aggressive content produces negative effects on self-regulation, whereas prosocial content produces positive effects. They suggest that the effect of intelligence quotient might be mediated through better understanding of the “complex” themes of self-control and persistence presented in Mister Rogers.
Geist and Gibson20 observed children during free play after they viewed an episode of either Mighty Morphin' Power Rangers or Mister Rogers; the control group watched no television. The authors found that children who viewed Mighty Morphin' Power Rangers spent less time on individual tasks (P < .05) and changed tasks more frequently (P < .05) than the children in the other groups, with no significant difference between the Mister Rogers group and the control group. The authors suggest that the viewing of fast-paced entertainment programs can lead to the development of hyperactivity and also caution that children may be done a disservice if classroom lessons or educational programs are made fast paced to keep children's attention.
Anderson et al,29 on the other hand, examined whether the rapid pacing of many Sesame Street segments affects the attentional skills of 72 four-year-old children. One group was shown rapidly paced segments taken from Sesame Street programs, and a second group was shown slowly paced segments taken from the same programs. Each child viewed the segments individually with a parent present; the control group listened to a parent read stories for the same length of time. Immediately after the session, the children completed 2 tests commonly used in psychological research to assess attention, perseverance, and impulsivity and were observed for activity level and task persistence while playing. No differences among the 3 groups were found on any measure. These findings suggest that any effect that television programs may have on self-regulation may be because of their content rather than their pacing.
The majority of television research focuses on adolescents and school-aged children, which is unfortunate given that an increasing number of programs are targeting very young children.30 This review focuses on television research across multiple disciplines with a focus on the first 5 years of life. These years are generally accepted as the critical period of childhood development. Despite the fact that infants are the fastest growing television market segment, this review was unable to find any studies in this age group that met our inclusion criteria with children under the age of 3 years.
Although most people focus on screen time, all of the studies in this review focused on the effect of TV content rather than viewing time. Some argue that, especially for young children, television content is more important than viewing time and that programs with age-specific curricula may allow children to learn at a time when they are not receiving formal school instruction, whereas entertainment programming may displace more developmentally valuable activity. Some studies addressed the content question by directly comparing educational programs to entertainment programs, whereas many of the studies that compared educational television to no television viewing were actually testing the use of the medium as an educational tool. None of the studies showed children general-audience programming, although children often watch such programs in the company of adults.
All of the studies in this review focused on developmental and behavioral outcomes of television viewing. Obesity is largely ignored in the television literature involving preschoolers. The findings from the experiments described in this review suggest that educational and prosocial television programs are successful in broadening young children's knowledge, affecting their racial attitudes, and increasing their imaginativeness.
There are reasons to question the generalizability or clinical use of the findings presented here. Because most of the studies had small sample sizes, they may have been underpowered to detect important meaningful effects. All of the studies were completed in nonnaturalistic settings, and it is possible that the effects of viewing television at home are very different from viewing television in a classroom or laboratory setting. All of the studies but 118 looked only at short-term effects of television viewing. Similarly, none of the studies evaluated the cumulative effects of viewing television over a long period of time. The longest study involved 6 months of viewing time, and few studies targeted home-based viewing, meaning that detected effects would only be marginal, because they would be competing with other programming that children viewed.18
As with many systematic reviews, ours found that after an exhaustive review of the published literature, more research is needed. In particular, more population-based, experimental studies of the effects of specific amounts and types of programming on young children are needed. Such studies must use pragmatic strategies, have long-term follow-up, and use meaningful end points. Although controlled experiments in natural settings looking for sustained effects are resource intensive to conduct, they have been successfully conducted in trials of the effects of breastfeeding.31 The typical approach is to try and aggressively modify behavior in an intervention group through a multifaceted approach. Such trials are necessary to provide strong evidence to address the gaps documented by this review.
- Accepted July 24, 2006.
- Address correspondence to Dimitri A. Christakis, MD, MPH, Child Health Institute, University of Washington, 6200 NE 74th St, Suite 210, Seattle WA, 98115. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Rideout VJ, Vandewater EA, Wartella EA. Zero to Six: Electronic Media in the Lives of Infants, Toddlers, and Preschoolers. Menlo Park, CA: Kaiser Family Foundation; 2003
- ↵American Academy of Pediatrics, Committee on Public Education. Children, adolescents, and television. Pediatrics.2001;107 :423– 426
- ↵Certain LK, Kahn RS. Prevalence, correlates, and trajectory of television viewing among infants and toddlers. Pediatrics.2002;109 :634– 642
- ↵Dennison BA, Erb TA, Jenkins PL. Television viewing and television in bedroom associated with overweight risk among low-income preschool children. Pediatrics.2002;109 :1028– 1035
- ↵Anderson DR, Huston AC, Schmitt KL, Linebarger DL, Wright JC. Early childhood television viewing and adolescent behavior: the recontact study. Monogr Soc Res Child Dev.2001;66 :I-VIII, 1–147
- ↵Singer J, Singer D, Rapaczynski W. Children's imagination as predicted by family patterns and television viewing: A longitudinal study. Genet Psychol Monogr.1984;110 :1– 8
- ↵Christakis DA, Zimmerman FJ, DiGiuseppe DL, McCarty CA. Early television exposure and subsequent attentional problems in children. Pediatrics.2004;113 ;708– 713
- ↵Geist EA, Gibson M. The effect of network and public television programs on four and five year olds ability to attend to educational tasks. J Instruct Psychol.2000;27 :250– 261
- ↵Friedrich-Cofer LK, Huston-Stein A, Kipnis DM, Susman EJ, Clewett AS. Environmental enhancement of prosocial television content: Effects on interpersonal behavior, imaginative play, and self-regulation in a natural setting. Environ Psychol.1979;15 :637– 646
- ↵Garrison M, Christakis D. A Teacher in the Living Room? Educational Media for Babies, Toddlers, and Preschoolers. Menlo Park, CA: Kaiser Family Foundation; 2005
- Copyright © 2006 by the American Academy of Pediatrics