Television has a bad side. According to a report from the University of Michigan, the average American child has seen sixteen thousand murders on TV by age 18. Indeed, programs explicitly designed for kids often contain more violence than adult programming, and that violence is often paired with humor. Every single animated feature film produced by US production houses between 1937 and 1999 contained violence, and the amount of violence increased throughout that time period. Researchers from the University of Michigan found that just being awake and in the room with a TV on more than two hours a day - even if the kids aren't explicitly paying attention to the TV - was a risk factor for being overweight at ages three and four-and-a-half. This may be related to the fact that two thirds of the twenty thousand television commercials the average child sees each year are for food.

The American Academy of Pediatrics, in their wisdom, recommend that children under age two have zero hours of screen time. (Meanwhile, a bevy of DVDs are marketed to parents of children age zero to 2, promising to “teach your child about language and logic, patterns and sequencing, analyzing details and more.”)

Despite the warning, however, many parents of infants age 0 to 2 do allow their children some screen time. In 2007, Frederick J. Zimmerman of the University of Washington (now at UCLA) wondered what the effects of TV watching were on those infants. He collected data from 1008 parents about the infants' TV habits, as well as the amount of time they spent doing things like reading (with parents), playing, and so on. He also administered, for each child, a survey called the MacArthur-Bates Communicative Development Inventory (CDI). The CDI is a standard tool used by developmental psychologists to assess language development in infants and children. He and his team then looked to see if there were statistical relationships between time spent watching TV (and the other activities) and language abilities, as measured by the CDI. Here's the catch: they only included infants whose TV watching consisted entirely of infant-directed programming. That is, TV programs especially designed for infants age 0 to 2. If the infants were shown other sorts of TV programs, they were not included in the study.

They found that reading at least once a day was associated with a seven point increase on the CDI for 8 to 16 month olds, and nearly twelve points for 17 to 24 month olds, compared with those who read with their parents less frequently. If parents told stories to their children at least once per day, as opposed to less frequently, their kids' scores on the CDI were nearly six and a half points higher for younger infants, and more than seven points higher for older infants. That kids who read more often or were told stories more often scored better on a test of language development is probably not surprising. (Interestingly, there was no statistical correlation between music listening and language development).

Here's the kicker: for each hour, on average, that infants between 8 and 16 months old watched infant-directed television (including DVD versions of those programs) per day, they could expect a seventeen point reduction in their scores on the CDI. Let me say that again: each hour per day, on average, that these kids watched TV was associated with a seventeen point decrease on a measure of language acquisition.

You might argue that this correlation could exist simply because parents who show these programs to their kids may also be somehow less motivated to encourage language development in their kids more generally. Perhaps these parents were just less skilled at parenting overall. The researchers attempted to statistically control for this, by factoring in data related to parent-child interaction. As usual, correlations should be taken with a grain of salt, and this is but one study from a very large literature. Still, this study, combined with others, has led the American Academy of Pediatrics to recommend that children between the ages of 0 and 2 years old watch no television at all.

But television isn't all bad. Shows like Sesame Street and Mister Rogers' Neighborhood stand out as beacons of hope, sparkling diamonds among a sea of coals. For children between the ages of two and five years, at least.

In 2001, Daniel R. Anderson published the results of a massive longitudinal study called "The Recontact Study" as a Monograph of the Society of Research in Child Development. First, they assessed television habits among preschoolers (age 2-5). Then they recontacted 570 of the children a decade later, when they were in high school. They assessed their current (adolescent) media use, and also their grades in English, science, and math, their leisure reading habits, creativity, aggression, participation in extracurricular activities, use of alcohol and cigarettes, and self-image.

They found, among other things, that children age 3 to 5 who watched Sesame Street had larger vocabularies in high school than those who watched other television programming, or even no television at all. The effect could not be explained by gender, family size, or parents' education. Preschoolers from lower income neighborhoods, in particular, who watched Sesame Street were more prepared for school than their peers who did not watch Sesame Street. Kids who watched Sesame Street had higher grades in science and English, had higher total GPA, read more books, placed more value on achievement, and were rated as more creative, compared with their peers. Boys who watched Sesame Street in preschool were rated as less aggressive in high school; girls were more likely to participate in extracurricular art classes.

Similar effects were seen for those who watched Mister Rogers as kids, but not for those who watched other non-educational television programs in preschool.

Importantly, it was the educational content of the television that kids watched in preschool that predicted their future success as high schoolers more than the overall amount of television they watched. "The medium of television is not homogeneous or monolithic, and content viewed is more important than raw amount," Anderson says. "The medium is not the message: The message is."

Reed Larson put it plainly in a commentary on Anderson's monograph: "Educational television works: It has sustained, long-term, positive relationships to development and behavior."

Zimmerman FJ, Christakis DA, & Meltzoff AN (2007). Associations between media viewing and language development in children under age 2 years. The Journal of pediatrics, 151 (4), 364-8 PMID: 17889070

Anderson DR, Huston AC, Schmitt KL, Linebarger DL, & Wright JC (2001). Early childhood television viewing and adolescent behavior: the recontact study. Monographs of the Society for Research in Child Development, 66 (1) PMID: 11326591

Image via Flickr/See-ming Lee.