Theta oscillations in top-down and bottom-up processing during early childhood

• In Event: The theta rhythm as a neural signature of early learning processes

Abstract

Diverse cognitive processes like attention, learning, and memory have been associated with neural activity in the theta frequency range (3-6 Hz in infants; 4-7 Hz in adults). Still, the precise functional role of theta, particularly in early childhood is matter of ongoing debate (Begus, & Bonawitz, 2020). Cavanagh and Frank (2014) propose that it reflects a signal used to realize the need for cognitive control, which may arise top-down through regulation due to task demands or bottom-up through stimuli containing surprising information. However, little is known about whether top-down and bottom-up processing in young children is similarly reflected in their theta oscillations despite immaturity of theta’s neural generators. Here, we will present two electroencephalography (EEG) studies in which we explored the involvement of frontal theta power in top-down and bottom-up processing of others’ actions during early childhood.
Study 1 investigated whether frontal theta power is modulated by top-down intentions of 4-year-old children. We tested whether children’s theta power is sensitive to task engagement and to different cognitive task demands. In a within-subjects design, children received three different instructions before watching videos; either no demands (No Task), language-related demands (Color-naming Task), or action-related demands (Imitation Task). We analyzed children’s theta power with cluster-based permutation tests in two contrasts: (1) Task vs. No Task and (2) Color-naming vs. Imitation Task. Children’s frontal theta power was higher when they were engaged in a task than when they had no task. Also, their theta power increased during cognitive engagement and left fronto-temporal theta power was higher for the language- compared to the action-related demands. These findings support the claim that frontal theta reflects a top-down regulation when there is need for cognitive control, extending this effect to preschoolers.
Study 2, with infants, investigated bottom-up factors such as variability and surprise in the environment that may modulate frontal theta power. We measured 14-month-olds’ EEG while they observed demonstrations of three actions. Within subjects, actions were presented either with constant (normal or high) or variable (normal and high) amplitude movements. Frontal theta power was significantly higher in the variable compared to the other conditions (see Figure 1). In a post-hoc analysis, we tested whether trial-by-trial fluctuations in theta power were related to the bottom-up surprise of each stimulus. We quantified the level of surprise elicited by each action demonstration in an information-theoretic framework, and show that frontal theta power correlates with the level of surprise of the actions. A model comparison procedure allowed us to prefer this model over other competing models that did not include surprise as a predictor. Thus, surprise induced by variability in action demonstrations was associated with infants’ frontal theta power. Implications of these findings on top-down and bottom-up processing in infants and preschoolers will be discussed with respect to the theoretical framework of frontal theta signaling the need for cognitive control.

Authors

• Marlene Meyer, Radboud Universiteit Nijmegen
  Presenting Author

• Johanna Van Schaik, Radboud University
  Non-Presenting Author

• F. Poli, Radboud University
  Non-Presenting Author

• Sabine Hunnius, Radboud University
  Non-Presenting Author