Development In Action: Using Hand Tracking to Capture Age-Related Improvements in Flexibility and Cognitive Control

• In Event: A Neuroscientific Perspective on Key Transitions in the Development of Cognitive Control

Abstract

Developmental change in cognitive control (CC) is commonly assessed with response time (RT) and accuracy in congruency tasks such as the Eriksen flanker task (Eriksen & Eriksen, 1974). This approach has greatly informed our current understanding of CC development across the lifespan, however, it offers limited insight into how the processes underlying CC unfold over time (Erb et al., 2016). To address this limitation, researchers are increasingly using hand-tracking techniques that allow RTs to be separated into initiation times (ITs; time elapsed between stimulus presentation and response initiation) and movement times (MTs; time elapsed between response initiation and response completion).
Previous research indicates that MTs in the flanker task reflect a controlled selection process that is particularly important on trials that require participants to flexibly unbind the stimulus and response features from the previous trial in order to bind the same response to a new stimulus in the current trial. Further, developmental work indicates that this process undergoes a relatively protracted development, reaching adult-like levels in late childhood (Erb et al., 2018).
Recently, researchers have started to link the dynamics observed in hand tracking to dynamics observed in continuous neural measures (e.g., Tafuro et al., 2020). However, these approaches have not been used to investigate this link across development. To address this gap, the current study targets the neural correlates of the age-related improvements observed in MT between middle childhood and adulthood.
Participants in the current study completed a computerized version of the flanker task using three-button response box. Participants initiated each trial by pressing and holding a central button until the stimulus array appeared, and then completed their response by reaching to press one of two response buttons. This method allows us RTs to be separated into ITs (when the central button is released) and MTs (when the response button is pushed).
Data collection for this study is ongoing, with a current sample size of 36 adults and 26 8- and 9-year-olds (target N = 72). Preliminary analyses revealed an interaction between current congruency, response type (response repetition or response switch) and age group in MTs [F(1,58) = 13.40, p < .001, η2 = 0.26], with age-related gains in performance being specific to trials featuring the same response as the previous trial [72ms; F(1,58) = 93.96, p < .001, η2 = 0.61]. Thus, our behavioural results replicate previous work indicating that age-related gains in flanker performance between middle childhood and adulthood are driven by trials that place greater demands on flexible binding processes.
Initial EEG analyses revealed an interaction between current congruency, response type, and age group in the mean amplitude of the P3 at FCz between 300 and 500 ms after stimulus presentation [F(1,58) = 7.28, p = .009, η2 = 0.12]. Comparably to MTs, age-related change in activity during this time period is driven primarily by response repetition trials. These effects present candidate neural correlates for the controlled selection process and bring us one step closer to linking the behavioural and neural dynamics of control across development.

Authors

• Katie Ann Smith, University of Auckland
  Presenting Author

• Inanna K Haddon

• Paul M Corballis

• Christopher D Erb, University of Auckland