Baby sign language is all the rage these days. Upscale day-care centers and nanny services promote it as a better way of understanding what babies want. Babies have been known to reliably produce signs as young as 5.5 months, and studies have shown that they reliably produce signs significantly earlier than spoken words. As we’ve reported here, there is no evidence that teaching sign language delays spoken language development.
But is formal sign training effective? Some studies about baby sign language have been quite informal, with parents and caregivers inventing makeshift signs to “talk” to their babies about bodily functions, favorite toys, desires for comfort, and so on. Other studies have focused exclusively on laboratory settings, with little follow-up to see if the signs acquired in the lab have any practical use.
A team led by Rachel Thompson has now combined these two approaches to explore whether formal laboratory training can have a positive impact on real babies’ lives — and their parents and caregivers.
The researchers systematically taught a modified ASL sign to Heather, a 10-month-old with Down syndrome. Heather learned to say “please” to ask for a toy. A 6-month-old normally-developing child, Betty, was taught a different sign, “more,” to ask for more food.
The training for each baby was similar, but I’ll take you step-by-step through Betty’s training. Remember, a six-month-old is just barely able to sit up on her own, much like Nora in this photo:
In fact, I think Nora’s a little older than six months here — but this photo was still taken several months before she uttered her first word.
Betty’s training was done quite systematically. First she was given a spoonful of baby food. Then the experimenter demonstrated the gesture “more” (bringing both hands together at the body’s midline). If Betty did not duplicate the gesture within five seconds, the experimenter showed her how to do the gesture by gently moving her hands as required. Then Betty was given another bite. The time between the experimenter’s gesture and Betty’s response was gradually extended, giving her more time to duplicate the gesture on her own without being shown how to do it. This graph shows Betty’s progress learning the gesture:
The open circles represent prompted signing — when the experimenter showed Betty how to make the sign. The filled circles are independent signs, made without any prompt from the experimenter. The numbers at the top of the chart represent the number of seconds between the experimenter making the sign and modeling the prompt. If Betty independently produced the sign, then no model prompt was made. The numbers at the bottom are the total number of five-minute learning sessions, which occurred several times each day. The vertical axis of the graph represents the number of each type of sign made per minute. As you can see, after about 30 sessions, fewer than 10 days of training, Betty began producing the signs independently. After 45 sessions, training was stopped and the graph shows how often Betty independently produced the sign with no prompting of any kind.
Finally, after session 50, Betty was moved to real-world settings where the experiment was repeated using different experimenters: a classroom teacher in the classroom, and her father his office. As you can see, Betty continued to produce the sign independently with very little help from the experimenters.
Heather, who had Down syndrome, showed nearly the identical pattern.
In a second study, the real-world benefits of signing were explored more deliberately. Can signing take the place of crying? Before they can talk, most babies cry a lot ,since this is effectively the only way they can communicate to their parents that something is wrong. As Chad Orzel will tell you, this isn’t a pleasant experience for the parents, and it’s unclear that the child gets much out of crying either.
Two boys who cried frequently were trained to give signs instead of crying: Geoffrey, 10 months old, generally cried when he didn’t get enough attention in the classroom. Lyle, 9 months old, cried when he wanted his mother to pick him up.
Geoffrey was trained to give the sign for “please”, and Lyle was shown the sign for “up”, just as Heather and Betty had been trained before. Here’s Lyle’s data:
Lyle followed a similar pattern to Betty and Heather. As you can see, once Lyle learned the sign, his episodes of crying and whining decreased substantially — so signing may actually make life easier for these children and their parents. Geoffrey’s results matched this pattern too.
One big potential weakness of this second experiment: In addition to being trained to give signs to signal what they wanted, the experimenters used the behaviorist strategy of extinction to discourage crying. Extinction is sort of like training in reverse: you give the child what they want only when they don’t exhibit the behavior you’re trying to discourage. So the babies were never rewarded for crying, only for producing the desired signs. Lyle’s and Geoffrey’s crying could very well have been eliminated through extinction alone.
Regardless, parents and children universally seem to appreciate the ability to communicate at an earlier age than would otherwise be possible. This study demonstrates that a rigorous laboratory procedure can yield these benefits in the real world.
Rachel H Thompson, Nicole M Cotnoir-Bichelman, Paige M McKerchar, Trista L Tate, Kelly A Dancho (2007). Enhancing Early Communication through Infant Sign Training Journal of Applied Behavior Analysis, 40 (1), 15-23 DOI: 10.1901/jaba.2007.23-06