Poster #1 - Early Childhood Interindividual Differences and Intraindividual Change: Predictors of the Dimensional Change Card Sort Performance

• In Event: Poster Session 02

Abstract

Of the three component executive functions, shifting takes the longest to develop, with 13-year-olds not yet reaching adult-level competencies (Diamond, 2013). Shifting performance throughout early childhood is generally attributable to development of memory (Davidson et al., 2006) and effortful control (EC: Quiñones-Camacho et al., 2019). On the prototypical early childhood shifting task, Dimensional Change Card Sort (DCCS), successful rule switching occurs around age 5 (Zelazo, 2006). Meta-analysis of DCCS identified age and verbal instruction (i.e., a focus on conflict in switch condition) as predictors of performance (Doebel & Zelazo, 2015). However, longitudinal development and predictors of that developmental change on this classic shifting task have been understudied.
Participants were 119 typically developing children. We examined age 3 predictors of DCCS performance at age 4; and ages 3 and 4 predictors of DCCS performance change from ages 4 to 6. At ages 3 and 4, memory (forward digits span task) and EC (Child Behavior Questionnaire parent report) were assessed. Raw score on the Peabody Picture Vocabulary Test (PPVT) was included as a covariate. We focused on DCCS post switch at ages 4 and 6. For children who succeeded on 5 of 6 post switch trials, we also examined DCCS borders at ages 4 and 6.
Post switch proportion correct represents the correct sorting after a switch in dimensional sorting rule, e.g. color to shape. Hierarchical regression results are in Table 1. Age 4 post switch was predicted by age 3 vocabulary (beta = .419). Change in post-switch proportion correct from ages 4 to 6 was predicted by age 3 vocabulary score only at step one (beta = -.386). In step two, only the age 4 post switch proportion correct predicted the change score (beta= -.948). This association was probed using PROCESS (Hayes, 2022; Model 4), confirming mediation with a significant indirect path (95% Bootstrap CI {-.0155, -.0092}).
In the borders condition, children were instructed to sort by a given dimension only for a card with a border. Age 3 vocabulary predicted age 4 borders (beta = .319). Change in borders proportion correct from ages 4 to 6 was predicted only at step two by age 4 borders proportion correct (beta =-.624). This relationship was also probed using PROCESS confirming mediation with a significant indirect path (95% Bootstrap CI {-.0056, -.0007}).
Statistically, age 4 performance would be a strong predictor of the age 4 to 6 change score. Interestingly, however, this was a negative beta weight, suggesting a regression in performance from ages 4 to 6 for many of the children (see Figure 1). It may be that instability or unexamined person-situation factors have a critical role in DCCS performance development. Significant indirect paths indicate that age 3 vocabulary is associated with later performance growth only through age 4 performance. Perhaps initial advantages offered by a verbal understanding of the task leveled as participants developed.
Our findings demonstrate that examination of longitudinal assessment of DCCS performance yields valuable information not only of interindividual differences but also intraindividual change.

Authors

• J. Douglas Harrison, Virginia Polytechnic Institute and State University
  Presenting Author

• Martha Ann Bell, Virginia Polytechnic Institute and State University