Using Reach Tracking to Document the Role of Inhibition in Selective Learning

• In Event: Mechanisms underlying imitation, social learning, and social preferences

Abstract

Preschoolers are not credulous of others’ information, but rather are selective (e.g., Koenig & Harris, 2005). What are the underlying cognitive mechanisms behind selective learning? Some suggest young children have an initial “default bias to trust” in which they believe others because they are unable to inhibit false information (Jaswal et al., 2010, 2014). These data suggest that inhibitory control is necessary and sufficent for children to express selective learning capacities. Others suggest inhibitory control is one of several cognitive capacities that is necessary, but not sufficient for selective learning (Sobel & Kushnir, 2013).
We used Reach Trajectory Tracking, which measures finger position as children make manual responses in 3D space over time, to track the time between the onset of a stimulus signally children to make a choice and the flight of their finger (Initiation Time), but also the deviance in curvature of the trajectory of their finger to the intended response (Curvature). Previous findings (e.g., Erb, Moher, Song & Sobel, 2018) have documented that initiation time corresponds to a global inhibitory process and dACC functioning, monitoring for conflict between stimuli and response, and prevents a manual response from occurring. Curvature corresponds to competitive inhibitory process and prefrontal functioning, which allows a conflict resolution system to unfold and control the manual response that is made (see Munakata et al., 2011).
Using a design similar to Jaswal et al. (2010), we tracked responses of 3- to 6-year-olds (N=18, anticipating N=40, testing paused due to COVID) as they chose where to look for an object hidden in one of two locations over 20 trials. The informant was either a human or an arrow, which was either always reliable or misleading (See Figure 1).
Our analysis strategy involved constructing GEE models on whether children found the hidden object, and initiation time and curvature for accurate responses, looking at the effects of informant (human vs. arrow), accuracy (reliable vs. misleading), trial, and age. All models showed a main effect of trial (the first three trials were different from the final 17, as children adjusted to the procedure). For accuracy and initation time, there was a main effect of informant and age, with participants more accurate and faster for information from arrows than human beings (p-values < .05) and more accurate and faster as they got older, but no significant effect of reliability. Only the curvature model showed a significant effect of reliability, B = 0.06, SE = 0.02, 95% CI [-0.01, 0.07], Wald 2 (1) = 3.80, p = .05, but no other significant effects.
These results, while preliminary, suggest that inhibitory control does not globally relate to selective learning and resolving unreliable information. While global inhibition is involved in resolving the nature of the information, conflict resolution processes in frontal cortex functioning seem to be more involved at tracking and resolving the conflict between reliable and unreliable stimuli. These results suggest that inhibitory control might interact with other cognitive processes to form the basis of a mechanism for selective learning.

Authors

• Yuyi Taylor, Brown University
  Presenting Author

• Joo-Hyun Song, Brown University
  Non-Presenting Author

• David Marc Sobel, Brown University
  Non-Presenting Author