Active and observational conventional action learning: Relations with toddlers’ skills and behaviors

• In Event: Object-based understanding and learning: Roles of infant active experience and skill

Abstract

Active experience and observation benefit early object learning. Infants are keen imitators of adult actions (Bauer, 1996) and in some cases, viewing demonstrations enhances object engagement beyond independent actions (Fagard & Lockman, 2009). However, infants’ active experience with objects benefits object exploration (Needham, Barrett, & Peterman, 2002) and understanding of others’ goals (Sommerville, Woodward, & Needham, 2005). Yet, few studies have tested both instruction strategies in relation to object learning. During everyday instruction, toddlers learned functions of novel objects (“conventional action”) when caregivers guided their actions; caregiver demonstrations negatively predicted learning (Brezack, Radovanovic, & Woodward, under review). This study suggested that active experience might be more effective than observation for toddlers’ conventional action learning.
Here, we experimentally tested the findings of Brezack et al. (under review) by pitting active and observational experience against one another to examine which boosted toddlers’ learning. 46 22-26-month-olds (mean age = 23.4 months, 22.2-26.0 months; 21 male) were taught to use one novel, multi-step toy through active experience and another through observing an experimenter, after which toddlers were tested on their toy reconstruction abilities (Figure 1). Surprisingly, results did not show a difference in learning from active or observational instruction (B = 0.03, p = 0.291). However, children’s cognitive problem-solving skills (measured by the Bayley Scales, Cognitive subtest) significantly predicted learning across instruction styles (B = 0.01, p = 0.011). Further, children’s problem-solving skills interacted marginally with instruction style such that more capable children performed marginally better on items learned through observational instruction (B = 0.01, p = 0.069; Figure 2). Children’s independent skills had a greater impact on conventional action learning than instruction style.
While learning did not differ between instructions styles, children’s behavior during instruction predicted learning differently by style. When learning from active experience, children with poorer problem-solving skills learned more when they were positively engaged (B = 0.003, p = .022). In contrast, when learning from observation, children with better problem-solving skills learned more when they looked longer (B = 0.762, p = .004; Figure 2). This suggests that different skills supported learning from each type of instruction: Positive engagement benefitted active learning, and visual attention benefitted observational learning.
Further, data collection is ongoing to test whether children’s memory for taught objects differed by teaching style one year after instruction. Preliminary results from 21 3-year-olds suggested that memory for actively-taught items might be greater than memory for observationally-taught items (active memory score was marginally greater than chance, p = .069; observational memory score was not different from chance, p = 1.0; Figure 2). If this result is borne out in the full sample, it would indicate that while immediate conventional action learning from active and observational instructions was similar, long-term retention could be enhanced for actively-learned objects. This study has implications for our understanding of the instruction styles and child skills that support object learning.

Authors

• Natalie Brezack, University of Chicago
  Presenting Author

• Jessica Chandler, University of Chicago
  Non-Presenting Author

• Amanda Woodward, University of Chicago
  Non-Presenting Author