Sustained Visual Attention Predicts Young Children’s Recognition Memory During Naturalistic Activities

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Attention, Learning, Memory

Abstract

Researchers use visual attention to index a wide range of processes; yet, many studies fail to find a consistent relation between looking and learning (Aslin, 2007; Kirkorian, Choi, & Pempek, 2016). A clear understanding of attention-memory associations during naturalistic activities would have implications for many areas of study and would inform the development of educational materials (e.g., TV; Anderson, 2004).

The inconsistent relation between attention and memory may be due partly to how visual attention is operationalized: Many studies measure cumulative looking time across an entire session. However, a viewer’s engagement likely changes as attention is sustained over time (Richards & Anderson, 2004), marked by heart-rate deceleration (Pempek et al., 2010; Richards & Casey, 1991; Richards & Cronise, 2000) and resistance to distraction (Anderson, Choi, & Lorch, 1987; Richards & Turner, 2001). Consistent with this hypothesis, Burns & Anderson (1993) found that adults’ recognition memory for TV scenes increased as attention was sustained. Put differently, adults were more likely to remember scenes that appeared after they had been looking at the TV screen for an extended period of time than those that occurred soon after they reengaged with the screen. We sought to replicate this finding in young children.

Children (N = 40, 3-5 years, 48% girls) played in a comfortably furnished laboratory room (Figure 1) in the presence of child- and adult-directed TV (8 minutes each, order counterbalanced). An experimenter then tested recognition memory in two ways: (1) Following Burns & Anderson (1993), children viewed short (3-5 second) video clips and indicated whether they had seen each clip during the session. Half of the clips (n = 16) were shown during the session; the rest came from other episodes in the series. (2) Children were shown 13 sets of four screenshots from the child-directed series and asked which of the screenshots was seen during the session.

Given that a child was looking at the TV when a test item first appeared, the probability of the child recognizing the test item later increased as a function of time the look to TV had been in progress (i.e., sustained attention toward the TV), B = 0.35, SD = 0.17, p = .036. The reverse was also true: Given that a child was not looking at the TV when a test item first appeared, the probability of the child recognizing the test item later decreased as a function of time since the last look at the TV (i.e., sustained attention away from the TV), B = -0.63, SD = 0.28, p = 0.028.

Our findings replicate Burns and Anderson (1993) in a sample of young children and extend research on children’s in-the-moment engagement (heart-rate deceleration, resistance to distraction) to its effects on subsequent memory. These findings underscore the need to consider attention measures other than cumulative looking time (e.g., median look length; Richards & Anderson, 2004), and they have implications for the design of educational programs for young children (e.g., building sustained engagement before presenting new information; Anderson, 2004).

Authors

• Heather Kirkorian, University of Wisconsin - Madison
  Presenting Author

• Seung Heon Yoo, University of Rochester
  Non-Presenting Author