Babies "flip" over toys

Why do babies fail to search for hidden objects before nine- months when, in other kinds of tasks, they seem sensitive to hidden objects months earlier? One common explanation is that babies lack the means-end skill for retrieving objects from barriers. The goal or "end" is to get the object, and the means to that end is to first move the barrier out of the way. According to this account, babies as young as four- months know that hidden objects continue to exist, despite being unable to coordinate their reaching into a means-end search.

However, there is some evidence against this account. For example, babies are more likely to retrieve a toy from a transparent barrier than an opaque barrier. But there is a criticism of studies using transparent barriers. The problem is that babies might succeed in retrieving a toy from a transparent barrier without even noticing the barrier, precisely because it is transparent. That is, babies might simply see the toy and reach directly for it, moving the barrier out of the way only by accident. We tried to solve this problem by using a transparent barrier that babies had to notice in order to retrieve the toy.

In this study, we presented seven-month-olds with a transparent screen that they had to pull down (or "flip") like a drawbridge in order to get the toy behind it. If they ignored the screen and tried to reach directly through it, they would fail to get the toy. We presented babies with events in which either a toy or no toy was behind either the transparent screen or an opaque screen. According to the means-end account, babies should search as often for a toy in the opaque event as they do in the transparent event. Likewise, they should not search very often when no toy was presented, regardless of whether the screen is transparent or opaque. In contrast, we predicted that babies would have more success discriminating between toy and no-toy trials with the transparent screen than with the opaque screen.

Sixteen seven-month-olds from the Denver Metro area participated in the final sample. The transparent screen was a piece of Plexiglass with white stripes across it, mounted on a gray wooden platform (see figure). The opaque screen was identical, except that the Plexiglass was completely covered with white Contact paper.

First, we trained babies how to pull the screens down, which they learned easily. Then on toy trials, the experimenter placed the toy on the gray wooden platform, tapped her fingers behind the toy to make sure the baby was looking at it, and raised the screen up in front of the toy. No-toy trials were identical, except that the experimenter did not place a toy on the platform. Babies then had 5 seconds to retrieve the toy.

As we predicted, the results showed that babies discriminated more between toy and no-toy trials with the transparent screen than with the opaque screen. They pulled the transparent screen down on 82% of toy trials but only 44% of no-toy trials. In contrast, they pulled the opaque screen down on 51% of toy trials and 39% of no-toy trials - a similar amount.

We concluded that babies' failure to retrieve hidden toys can not be due solely to a means-end deficit or to an ability to ignore transparent barriers. Our results address the problems of previous studies, and further challenge the means-end account. If problems with means-end reaching do not explain why babies fail to search for hidden objects before 9 months, then what does? We believe that one possibility is that babies' mental representations of hidden objects get gradually stronger with development, and that a fairly strong representation is needed for babies to actively search in the face of a barrier. To test this account, our next set of studies will focus on testing babies' memory for familiar versus novel toys. Babies might search more for a hidden toy if it is familiar rather than if it is novel, because their memory for the familiar toy should be stronger.

Toddlers lost in space -- Beginner's Luck

We've all heard the phrase "beginner's luck". Well, "beginner's luck" captures toddlers' performance in the Orient study, where toddlers were most successful the first time they tried the task.

When we last reported on our continuing studies of how toddlers orient themselves in space, our toddlers were not using featural information such as colored posters or colored walls to reorient after becoming disoriented. In these studies, two-year olds watch a toy being hidden in one of four boxes placed in the corners of a rectangular room containing featural information. After being blindfolded and turned by their parents, toddlers try to find the toy. To find the toy, children need to reorient, or get their bearings in the room. Children are successful if they use the room's featural cues to reorient and remember the corner that holds the toy. However, children have only a 50% chance of being right if they reorient using the room's geometry, thus searching equally in the corner where the toy was hidden, and in the opposite corner, where the short and long wall have the same relationship to the child (for example - short wall on the child's left, long wall on the child's right).

Toddlers got four toy hiding trials in one of several rectangular rooms with 1) posters of colored shapes, 2) one blue wall and three white walls, 3) four different-colored walls, or 4) a panoramic design on the walls. Toddlers did not consistently find the toy across trials in any of these rooms. However, across the room environments, toddlers tended to find the toy on the first trial but then used the room's geometry to search on the later trials.

Since children used features to reorient and find the toy on the first trial but then used geometry to reorient in later trials, what can we conclude how the brain works in the reorienting task? In other researchers' studies, toddlers reoriented using only the room's geometry. These findings resulted in the theory that children have a specific area of the brain, or module, that uses only geometric information to help one reorient. (Adults use more than the module and so are successful in reorienting tasks). Our alternative theory is that the hippocampus, an area of the brain important for certain types of memory, encodes a "snapshot" of the environment. If the featural cues are strong enough they will be included in the "snapshot" and used later to reorient. Our current findings seem to refute the idea of a reorienting module and support the idea that featural cues are included in memory, allowing kids to find the toy on the first trial. However, these cues may not be encoded strongly enough to help kids ilater trials. Future studies will use a circular room to see whether the lack of geometry will affect toddlers' toy-finding success!

Do as you say, not as you do!

Have you ever told your child that they can pick one candy bar from the candy aisle and they come back with two candy bars? And they do this despite correctly reciting your rule of picking one candy bar. Why is it that children say one thing, but then act differently? Is it because children know one thing but do another? Is it because the question we ask children is easier than the act they do?

To try to understand why toddlers sometimes say one thing and do another, we observed what they said and what they did while they sorted cards into two piles. The cards had either a blue flower or a red truck on them. Some toddlers were first asked to sort the cards based on their colors (red cards in one pile, blue cards in the other pile). Other toddlers were first asked to sort the cards based on their shapes (flowers in one pile, trucks in the other pile). Most toddlers were able to sort the cards by this first rule, whether it was color or shape. Then, toddlers were asked to sort the cards by a second rule. Toddlers that sorted cards by color were asked to sort cards by shape and those that sorted by shape were asked to sort by color. Using the second rule required children to place a card in the opposite pile from the first rule (look at the figure to see an example of how a card fits with each pile depending on the rule of the game). Most toddlers ignored these new instructions (just like the child who disobeys the parent by grabbing more than one candy bar!), and continued to sort cards by the first rule instead of the second rule.

Interestingly, toddlers did one thing and said another (again, like the child who grabs more than one candy bar but can correctly recite the instruction to get only one candy bar). When toddlers were asked about the second rule (for example, "In the shape game, where does the truck go? Where does the flower go?"), they could answer these questions correctly, even though they incorrectly sorted the cards by the first rule of color!

To try to understand these differences between what toddlers say and do, we also asked them harder questions. These questions included information about color and shape (for example, "In the shape game, where does the red truck go? Where does the blue flower go?"). Most toddlers answered these questions incorrectly; answering based on the first rule rather than based on the second rule. Thus, toddlers answered these harder questions in the same way they sorted the cards.

This suggests that toddlers' ability to demonstrate knowledge depends on the difficulty of the task. Conflict may be part of what makes a task difficult. For example, when a child needs to decide what to do with an object that is both red and a truck, this task is difficult when the decision has about what to do with red things conflicts with the decision about what to do with trucks. This task is hard whether the child has to sort cards or answer questions that have conflict. But when the conflict is taken away (as when the question focuses on a single dimension, like shape), the task is easier. Similarly, when children are asked about a previous instruction ("How many candy bars did I say you could have?), they may be able to answer this relatively easily. The more difficult task is resolving conflict between this instruction and their desire for more than one candy bar. This is why children sometimes may say one thing and do another!

Are you listening?

Following instructions is an important part of life, but not something that kids readily do. We often assume that this reflects a problem of motivation, so that if kids somehow just tried harder, listened more carefully, or just thought more about they've been told, they'd be more likely to follow through. However, we are discovering in our research that, at times, following instructions can be quite difficult for kids.

We've been looking at how effectively 6-year-olds can follow some simple rules in a computer-based listening game. Children listen to a "mixed-up" speaker who describes happy situations in a sad tone of voice, and sad situations in a happy tone of voice. At the beginning of the game, children are asked to focus on what the speaker is talking about, so that if the speaker says something happy, they should press a "happy button," but if she says something sad, they should press a "sad button." After a number of successful turns, the rules of the game are switched around, so that now, kids need to ignore what the speaker is talking about, and listen instead to the sound of her voice. Kids get at few turns at the new game, and then are asked to recall the new rules.

As many parents who've watched their 6-year-olds play this game will agree, the results are quite striking. Kids generally have no difficulty following the rules when asked to listen to what the speaker is saying. In fact, this appears to be kids' natural inclination at this age. However, when asked to switch and listen instead to the speaker's tone of voice, over 60% of children continue to listen to what the speaker is saying. Interestingly, most kids remember that they are supposed to listen to the speaker's voice and not to what she is saying, and they can also accurately distinguish the sound of a happy voice from the sound of a sad voice. Evidently, children have difficulty following these very simple instructions. To date, we have been trying to discover why.

In one study, we used the same sentences throughout the game, thinking that as kids became more familiar with the sentences, it would be easier for them to distinguish between what the speaker was saying and how she was speaking. Although most children noticed that the sentences were repeating, they continued to have difficulty following the new rules. In a second study, we explored how well children actually remember the new rules. We found that children who don't follow the instructions will do OK on a simple test of the new rules (i.e., "If the speaker's voice sounds like this **happy humming**, what's the right answer?), but not so well on a more stringent test of the new rules (i.e., "If the speaker says something happy and her voice sounds sad, what's the right answer?"). This suggests that children's memory of the new rules may not be strong enough to guide them through situations of conflict.

In future studies, we plan to look at whether the instructions are difficult because they force kids to do go against their natural inclination, which is to listen to what is being said. If so, then it should be easier for kids to switch and listen to what is said than it is to switch and listen to how the voice sounds. As well, we plan to look at whether switching or just following instructions is more difficult for kids. We will see whether kids can switch in a version of the game that has no instructions. If they can, it would more clearly suggest that following instructions can be difficult for kids, as opposed to simply switching.

While it is true that children are often not terribly motivated to carry out simple requests, our findings show that following instructions can also be difficult for kids.