You Get What You Teach For

Being clear on the goal of education is so important for schools. If a school is not clear on what it wants from the education it offers students, then not only will there be a lack of consistency across classrooms, but students will get a mixed message as to what is valued.

One of the things we need to be clear on is the extent to which we value students’ ability to transfer what they have learned to unfamiliar contexts. David Perkins says that many schools’ approach to transfer can be described as the “Bo Peep” method: “Leave them alone and they’ll come home, wagging their tails behind them.” Basically, we kind of hope they will transfer what they have learned, but we don’t systematically put in place the things that are likely to make it happen.

So what are some of the hallmarks of teaching for transfer? And is it possible for even very young children to generalize in ways that enable transfer? Some recent research by Cristine Legare of the University of Texas and Tania Lombrozo of the University of California, Berkeley points to one important component of teaching for transfer and suggests that children even as young as three are capable of transferring abstract reasoning to unfamiliar situations.

The research was conducted over two separate studies, the first involving 95 children between three and five years old and the second involving 87 children between three and six years old.

Study 1

In this study children, were shown a mechanical toy with five interlocking gears. A training task was given to all children where the experimenter modelled how to put the toy together and labelled the parts, after which the child had the opportunity to assemble the toy. During the experimental task which followed, half the children were told to observe the toy for 40 seconds (observation condition) and the other half were asked “Can you tell me how this works?” and were given 40 seconds to give a verbal response (explanation condition).

What were the results of the study?

- Children in the explanation condition performed significantly better on measures of causal learning than those in the observation condition.

- Three-year-olds did not perform as well on measures of causal learning as the four- and five-year-olds, whose performance did not differ significantly.

- Three- and four-year-olds in the explanation condition did not perform as well on a task requiring memory for causally unrelated details such as color as those in the observation condition. There was no difference in performance on this task among five-year-olds in the two conditions.

Study 2

A follow-up study was conducted at least partly to see whether the differences observed in the first study were related to the verbalization required by the explanation condition, which was not required by the observation condition. The set-up for the study was almost identical to that for the first study except that half of the children were told to explain the toy (explain prompt condition) and the other half were told to describe the toy (describe prompt condition). In this study there was also an additional assessment done that measured the children’s ability to generalize and transfer their learning to a task where they were invited to create a new toy out of unseen gears and parts.

What were the results of the study?

- Without the more specific prompt “Can you tell me how this works?” there was no consistency across the two conditions in terms of whether children gave explanations or descriptions. The experimenters believe that the children often did not discriminate between the words “explain” and “describe.” Because of this, the results were coded by the responses the children gave (explainers vs describers) rather than by the prompt conditions.

- Explainers outperformed describers on measures of causal learning.

- On measures of non-causal learning there was no advantage for the explainers.

- Explainers significantly outperformed describers on the task requiring generalization and transfer.

- Older children were more likely to generate explanations as opposed to descriptions and generally outperformed younger children on measures of causal learning. However, those younger children who generated explanations still showed an advantage on measures of causal learning over their age peers who generated descriptions.

What might this mean for our classrooms?

Across the two studies, the generation of explanations brought significant benefits in terms of causal learning, including the likelihood that children would transfer that learning to an unfamiliar situation in the second study. Generating explanations seemed to direct children to notice causal patterns, though for some of the younger learners in the first study, this seemed to come at the expense of noticing details that were causally irrelevant.

The researchers believe that the reason for this may lie in the fact that the younger learners found the generation of explanations too cognitively taxing to also be able to focus on other details. This may be related to working memory capacity. Importantly, however, the benefits of explanation for causal learning were not age-dependent. While younger learners in the second study were not as likely to generate explanations, those who did showed the same benefits in the assessment of causal learning and of transfer as the older children.

Overall, the research strongly suggests that we should be engaging children from a very early age in generating explanations of phenomena. The benefits of explanation also apply to reading comprehension. Other research suggests that the use of self-explanation improves learners’ reading comprehension and problem-solving abilities and that students can be trained to give better explanations and to engage in self-explanation when interacting with text.

References:

Bielaczyc, Pirolli & Brown (1995) “Training in self-explanation and self-regulation strategies.”

Legare, C. & Tombozo, T. (2014) “Selective effects of explanation on learning during early childhood.”

Perkins, D. (2010) Making Learning Whole. SanFrancisco: Jossey-Bass.

---

Please fill out the form below if you would like to post a comment on this article:

Nickname

(this will appear with your comments)

Email

Comments

---

Comments

There are currently no comments posted. Please post one via the form above.

MORE FROM

GORDON ELDRIDGE: LESSONS IN LEARNING