Poster #1 - A Meta-Analysis of Prospective Memory Development in Preschool Children and its Associated Moderators

• In Event: 3-106 - Poster Session 11
  In Poster Session: PS 11 Section - Attention, Learning, Memory

Integrative Statement

Children and adults alike engage in actions in their daily lives that require prospective memory (PM), the ability to remember to execute a planned action in the future. While several meta-analyses have examined age-related declines in PM across adulthood (Uttl, 2008), less is known about the development of children’s PM. The current pre-registered (http://osf.io/nzaq8) meta-analysis examines preschoolers’ PM development in relation to age and moderators influencing performance on lab-based event-based PM tasks.

In a typical lab-based event-based PM task, children are engaged in an ongoing task (OT) and are told to carry out a specific action (PM intention) when a target cue appears. Between the introduction of the PM intention and the OT, there is typically a retention interval to allow some time for forgetting. For successful PM performance, children must remember to carry out the PM intention when they encounter the target cue. As children get older, PM performance tends to increase.

A random-effects meta-regression was conducted using 18 studies with 120 conditions that encompassed data from 2,233 typically developing 2- to 7-year-olds. As suggested by Mahy et al. (2014), the following moderators were examined: (1) Age, (2) Nature of the PM intention, (3) Length and nature of the delay task, (4) Nature of the OT (5) Nature of the PM cue, and (6) frequency of PM targets (Table 1). The dependent variable was the proportion of correct PM trials in each condition (raw mean/total number of PM trials).

Initial meta-regression with no moderators examined the effect size and presence of heterogeneity. Overall summary effect was .57 (p <.001) with a significant test of heterogeneity, Q(119) = 5816, p <.001, indicating additional between group variance that may be explained by moderators (Figure 1). Mean age in years was added as a moderator and as hypothesized, age was significant, Q(1) = 19.52, p <.001, for every 1 year increase in age, there was a .10 increase in PM performance (Table 1).

All moderators were then entered into a full meta-regression model (Table 1). The overall test was significant, Q(8) = 215, p <.001. As hypothesized, PM performance improved with age, with an easy OT, and when PM targets were more frequent. PM intention complexity, delay length and external reminders were not significant. Contrary to our hypothesis, PM performance was better with an arbitrary PM intention and with non-focal PM cues.

Meta-regression result supports the age-related increase in PM performance during the preschool years and finds presence of moderators suggested by Mahy et al. (2014) that foster PM performance: easy OT and greater frequency of PM targets. Opposite to our hypothesis, PM performance was greater when the PM intention was arbitrary and with non-focal PM cues. A motivating intention may distract children from the PM instructions or may overwhelm children’s working memory capacity. Of note, the number of studies with non-focal cue was smaller (n= 23) compared to focal cues (n= 102) such that unexamined moderators in our meta-analysis might drive the effect in studies with non-focal cues.

Authors

• Stephanie Chen-Wu Gluck, University of Oregon
  Presenting Author

• Caitlin E. V. Mahy, Brock University
  Non-Presenting Author

• Tessa Mazachowsky, Brock University
  Non-Presenting Author

• Louis Moses, University of Oregon
  Non-Presenting Author