The Learning Scientists

View Original

Expanding Retrieval Practice for Preschoolers

By Megan Sumeracki

Today’s post features a set of experiments conducted by Catherine Fritz and colleagues (1) with preschool children. We have actually cited this paper a few times on our blog, when we covered why the spacing effect has failed to make it into mainstream practice (Part 1, Part 2), and when I wrote about whether retrieval practice is likely to help all of our students learn (see here). However, we have yet to write in detail about the design and results of the experiments in this paper, and we know there is interest in understanding how the most effective learning strategies may work for different types of learners.

Compared to older children, preschool children are easily distracted, are poor at predicting how much they will remember, and tend to forget things quickly (1). Preschool children also do not tend to attempt to adopt effective learning strategies on their own. Given all of this, it is important to understand whether relatively simple strategies, like spacing and retrieval practice, can be implemented in preschool classrooms to help preschool children retain more of what they are learning in school.

Interestingly, the four of us Learning Scientists have observed that when speaking to preschool-aged children, adults often tend to implement retrieval practice naturally (even if they are not intentionally using the strategy). When talking with very young kids, we often say things like “what color is this?” or “what does the cow say?”. Given that, at least anecdotally, many are already doing this with preschool-aged children, it is useful to know whether this truly helps them learn and remember, like it does for older kids. If so, then teachers and parents can use spaced retrieval even more with preschool-aged children.

The goal of this set of experiments was to investigate whether spaced retrieval practice with preschool children would help them learn. Specifically, the authors tested expanding retrieval practice, where the first retrieval attempts are closely spaced, and the amount of space between subsequent retrieval attempts gets larger (i.e., the spacing expands).

Experiment 1: Design and Procedure

Image from Pixabay.

The image shows a plush stuffed pig.

Sixty preschool children (2.5-5 years) learned the names of six plush toy pigs in one of three conditions. Three preschools participated, and each of the schools was randomly assigned to one learning condition. This was done to prevent the children from talking to one another, but of course, does not represent true random assignment. It was, therefore, important to take into account any differences between the groups at the start (and unsurprisingly, there were differences). To do this, a baseline measure of name learning was taken for each child before the experimental procedure began. This measure was used to control for these differences statistically, reducing concerns about this design feature.

The learning phase

The children learned the names of the toys in one of three conditions:

  1. In the control condition, the children were introduced to each toy pig, and the child repeated each toy’s name and was encouraged to say “hello” and shake the toy’s hand or hug it.

  2. In the reward condition, the children were also introduced to each toy pigs in the same way as in the control condition, but they were also told that they would get a sticker if they tried their best.

  3. In the expanding retrieval practice condition, the children were introduced to the toy pigs in the same way as in the control condition, but they were also asked to recall names following an expanded schedule. We can think of each time a toy was introduced or the child was asked to recall the name as a single trial. In expanding retrieval, the number of trials between each repeat of the toy increases. After the first introduction of a given toy, the children were asked to recall the name after one different trial. Then they were asked to recall again after 2 trials, and again after 3 trials, and finally again after 7 trials (i.e., a 1-2-3-7 lag schedule). If the child could not name the pig, then the experimenter said the name and encouraged the child to repeat it.

Visual depiction of the expanding retrieval procedure. Image created by Megan Sumeracki, pig photo from Pixabay.

The assessment phase

The children were asked to recall the names of the toy pigs immediately after the learning phase ended.

Results

After statistically controlling for the children’s baseline ability to remember names, children that practiced expanding retrieval practice remembered significantly more correct names than children in the control and reward conditions. In fact, those in the expanding retrieval condition remembered more than twice the number of names compared to the other two groups. The size of this effect was large, like what is typically found in retrieval practice experiments with older children and adults.

Experiment 2: Design and Procedure

In Experiment 1, the children in the expanding retrieval condition spent more time learning names than did those in the other two conditions. In Experiment 2, time was equated.

The learning phase

The experiment was very similar to the first. Sixty-two preschool children (3 yrs 10 mo - 4 yrs 10 mo) learned the names of 14 different plush toys across two days. The experiment included two new conditions:

  1. In the expanding retrieval practice condition, the procedure was the same as that from Experiment 1.

  2. In the re-presentation condition, the children toys were represented following the same schedule as the expanding retrieval practice condition, except the experimenter said the toy’s name rather than having the children recall them, and the children repeated the names back.

  3. In the massed condition, the total time was spread out evenly among the toys, and the experimenter had an extended discussion of each toy with the children. During this discussion, the experimenter said the toy’s name repeatedly and asked the child to repeat it.

Image from Pixabay.

The image shows three stuffed teddy bears reading a book together.

 The assessment phase

The children were asked to recall the names of the toys one minute after learning, and again one day after learning. (Note, because of the learning phase was separated into two days, some of the toys were also tested after a 2-day delay in addition to 1 minute and 1 day.)

Results

Expanding retrieval practice led to better assessment performance than the other two conditions, both after one-minute and one-day. The children in the re-presentation condition remembered more names than in the massed condition.

Bottom Line

Expanding retrieval practice was found to be a highly successful learning strategy for preschool children. The authors determined that the success in the expanding retrieval practice condition comes from both the expanding spaced schedule and the benefits of recalling the names.

References: 

(1) Fritz, C. O., Morris, P. E., Nolan, D., & Singleton, J. (2007). Expanding retrieval practice: An effective aid to preschool children's learning. The Quarterly Journal of Experimental Psychology, 60(7), 991-1004. https://doi.org/10.1080/17470210600823595