Poster #240 - Toddlers’ Imitation and Word Learning from Interactive Videos: Does Task-Relevant Interactivity Support Learning?

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Technology, Media & Child Development

Integrative Statement

Some studies indicate that toddlers learn more from interactive than noninteractive screen media (e.g., Lauricella et al., 2010; Troseth et al., 2006). However, the effect of interactivity may vary by age and interactivity type (Choi & Kirkorian, 2016; Kirkorian et al., 2016). The purpose of this study was to determine whether the task-relevance of interactivity affects learning from video. We predicted that interactivity helps learning if it draws attention to task-relevant information on the screen, especially for younger toddlers.

Toddlers (18-37 months, N = 116) watched videos in which an actress pulled each of four novel objects out of a box, provided either a generic (“this one”) or novel label (e.g., “the gazzer"), and demonstrated a 3-step action sequence with the object. Children either watched non-interactive videos (non-interactive), touched the on-screen box to see the object inside (relevant-interactive), or touched a star sticker in the background (irrelevant-interactive). After watching each set of videos, toddlers were given 30 seconds to interact with each real object while an observer coded for imitation. Then toddlers completed a word-learning test for the labeled object (e.g., Which one is the gazzer?). Of particular interest were the number of action steps imitated and the number of word-learning questions answered correctly.

We conducted 3(condition) x 3(age) ANOVAs and post-hoc t-tests with Bonferroni correction on imitation (Figure 1) and word learning (Figure 2). For imitation, there was an effect of age group [F(4, 106) = 15.74, p < .001], with the oldest group imitating more actions than younger groups (both ps < .001). There was also an age-by-condition interaction [F(2, 106) = 15.74, p < .001], with condition effects in the middle age-group only. Contrary to predictions, these toddlers imitated more actions in the irrelevant (versus relevant) condition (p = .018). There was also an age-by-condition interaction for word learning [F(4, 100) = 3.97, p = .005]. The irrelevant-interactive condition (vs. relevant-interactive) produced marginally higher word-learning for the middle age group (p = .083) but significantly lower word-learning for the oldest age group (p = .025).

We predicted that relevant interactivity would draw attention to target information and increase learning, particularly for younger toddlers. On the contrary, younger 2-year-olds learned more from irrelevant interactivity. For older 2-year-olds, this effect was reversed for word learning. It is not clear why irrelevant interactivity increases learning among younger 2-year-olds; however, this finding replicates a different study in our lab using a spatial search task (author citation redacted for blind review). Thus, it seems that this simple type of “irrelevant” interactivity is not always detrimental to learning, and it may provide just enough difficulty to increase toddlers’ engagement without interfering with learning. However, the impact of interactivity differs by age, even within this narrow age range. More research is needed to identify the optimal conditions for learning across early childhood and the correlates of age that moderate learning. Such research has the potential to inform the development of personalized educational materials that optimize individual learning.

Authors

• Heather Kirkorian, University of Wisconsin-Madison
  Presenting Author

• Tiffany Pempek, Hollins University
  Non-Presenting Author

• Roxanne Etta, University of Wisconsin-Madison
  Non-Presenting Author

• Seung Heon Yoo, University of Wisconsin-Madison
  Non-Presenting Author

• Elizabeth Skora Horgan, University of Wisconsin-Madison
  Non-Presenting Author

• Mengguo Jing, University of Wisconsin-Madison
  Non-Presenting Author