Mimicking Speech and Gesture During a Lesson may not be Beneficial for Early Learners

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Cognitive Processes

Abstract

Introduction
Early patterning skills often predict formal mathematics achievement (Burgoyne et al., 2017). However, little is known about mechanisms that support pattern learning. Abstract speech (e.g., this pattern goes ABB) can be beneficial for pattern learning, especially when children spontaneously adopt (i.e., mimic) this speech (Fyfe et al., 2015). Further, in other areas of mathematics, instructional speech is often beneficial in combination with gestures (Cook et al., 2008)—and children’s spontaneous mimicry of an instructor’s gestures is related to enhanced problem solving (Vest et al., in press). Here, using a randomized experiment, we asked whether instructing children to mimic an instructor’s speech and gesture benefited their pattern performance.

Method
128 children (M age = 5.87 years, range = 4.0-7.92, SD = 1.03) met individually with an experimenter and completed a lesson on repeating patterns including six example items and six solve items with feedback. Children were randomly assigned to one of four lesson conditions based on a crossing of two factors: gesture (yes vs. no) and mimicry (yes vs. no). In all conditions, the experimenter used abstract speech labels to explain the patterns (e.g., “Look! This pattern goes ABBABB.”). In gesture conditions, the experimenter also used gestures (e.g., point to A object, then vee-point to adjacent B objects in ABB pattern). In mimicry conditions, children were required to repeat the experimenter’s speech (and gesture, if relevant). After the lesson, participants completed a six-item posttest with novel items.

Results
Performance on the solve items during the lesson was highest in the gesture without mimicry condition (M = 5.74 out of 6.00). In a regression with gesture, mimicry, age, and all possible interactions entered as predictors, there was no significant effect of gesture, a significant negative effect of mimicry, b = -3.30, p = 0.005 and a significant mimicry-by-age interaction, b = 0.04, p = 0.012 (see Figure 1); the negative effect of mimicry was only present for younger children. Similar to lesson performance, performance on the posttest was highest in the gesture without mimicry condition (M = 5.11 out of 6.00). In a similar regression model, there was a significant negative effect of mimicry, b = -3.39, p = 0.031 and a significant mimicry-by-age interaction, b = 0.04, p = 0.047 (see Figure 2); the negative effect of mimicry was only present for younger children.

Conclusion
Children performed well on repeating pattern tasks during a lesson and on a posttest. However, younger children performed less well when instructed to mimic the experimenter’s speech and gestures. Although prior work suggests that spontaneous mimicry may be beneficial, perhaps forcing children to mimic an instructor produces rote behaviors that are ineffective for learning, especially for younger children (i.e., early learners). Unexpectedly, we did not find any effects of gesture on pattern performance, suggesting a need for future research to examine the use of speech and gesture across a variety of learning tasks and ages.

Authors

• Nicholas Allan Vest, University of Wisconsin - Madison
  Presenting Author

• Giulia Alexandra Borriello, Indiana University - Bloomington
  Non-Presenting Author

• Emily Fyfe, Indiana University - Bloomington
  Non-Presenting Author