Infant EEG Contributes to Toddler Executive Function through Infant Attention but Not Infant Temperament

• In Event: 3-013 - Advances in the Developmental Cognitive Neuroscience of Executive Function

Integrative Statement

Researchers typically examine early direct predictors of later executive functions (EF). For example, infant frontal lobe development measured using EEG (Kraybill & Bell, 2013) and infant attention measured using looking behaviors (Cuevas & Bell, 2014) are linked with toddler and early childhood EF. There are similar findings for temperament. Specifically, negative temperament has the potential to undermine early developing EF (Ferrier et al., 2014; Rochette & Bernier, 2016). Development is complex, however, and examining direct predictors gives a simplified view of EF. Considering multiple processes can greatly inform our knowledge of early developing EF. As an example, Whedon and colleagues (2016) reported that infant frontal lobe development (assessed via EEG) predicted early age 3 EF through age 2 attention. Because temperament is also linked with frontal processes (Fox, 1989), we expanded on Whedon’s model by including infant negative temperament. We predicted negative links between 5-mo frontal EEG and 10-mo infant attention (i.e., shorter looking time is linked with greater processing efficiency; Colombo et al, 2010) and between 5-month frontal EEG and 10-month negative temperament (Fox, 1994). We predicted negative links between infant attention and toddler EF (Colombo & Cheatham, 2006; Cuevas & Bell, 2014). Finally, we predicted that infant frontal EEG would predict toddler EF through infant attention and negative temperament.
Participants were part of an ongoing longitudinal study examining the early integration between cognition and emotion. Data for this analysis came from lab visits at 5, 10, and 24 months and included 303 infants (half females) and their mothers (wide ranging education levels). From the 5-month visit, a composite score was created for frontal EEG power values during a baseline recording. At 10 months, infant attention to a novel toy was coded and maternal-report of infant negative temperament was collected using IBQ-r. The EF composite score at 24 months consisted of pre-switch DCCS, Tongue Task, and pig-bull test and was created after standardizing each variable.
We tested our hypothesis with Mplus Version 8 (Muthén & Muthén, 2015) using Maximum Likelihood (ML) estimation to handle missing data. Our analyses indicated that the model was a good fit, χ2 = .687, df = 1, p = .407, CFI = 1, RMSEA = .00. We found both direct and indirect paths from infancy through toddlerhood. Specifically, 5-month EEG (p<.001) and 10-month attention (p<.001) directly predicted toddler EF. Frontal EEG power at 5 months (p<.000) directly predicted 10-month attention and indirectly predicted toddler EF through 10-month attention (b = 0.09, SE = 0.02, β = 3.68, 95% CI [.041, .134]).
Infant negative affect did not explain toddler EF nor was it linked with infant frontal EEG as hypothesized. This may be because we used a frontal composite rather than frontal EEG asymmetry as our frontal indicator (Smith et al., 2016).
Thus, infant frontal lobe development (i.e., EEG) explained the development of toddler EF through infant attention. Studying early EF within the context of multiple indicators is necessary to gather a more coherent story of development.

Authors

• Tatiana Garcia-Meza, Virginia Tech
  Presenting Author

• Martha Ann Bell, Virginia Tech
  Non-Presenting Author