Assessing ocular activity during performance of motor skills using electrooculography

Abstract

Eye-tracking research has revealed that, compared to novices, experts make longer ocular fixations on the target of an action when performing motor skills; that is, they have a longer quiet eye. Remarkably, the reason why a longer quiet eye aids movement has yet to be established. There is a need for interdisciplinary research and new measures to accelerate progress on the mechanistic understanding of the phenomenon. With the aim to provide researchers with new tools, we assessed the utility of electrooculography (EOG) to examine ocular activity while 10 experts and 10 novices putted golf balls. We measured quiet eye durations, distinguishing its pre- and postmovement initiation components, and developed a novel time-varying index of ocular activity, eye quietness, computed as the variability of the EOG in short time intervals: lower values correspond with greater quietness. Finally, we measured movement durations using a combination of infrared and sound sensors. Experts had longer postmovement initiation quiet eye compared to novices; however, total and premovement quiet eye durations did not differ between groups. Eye quietness was inversely correlated with quiet eye duration, and was greatest immediately after movement initiation. Importantly, movement duration correlated positively with postmovement initiation quiet eye and negatively with eye quietness shortly after movement initiation. This study demonstrates the utility of assessing ocular activity during performance of motor skills using EOG. Additionally, these findings provide evidence that expert-novice differences in ocular activity may reflect differences in the kinematics (e.g., movement duration) of how experts and novices execute motor skills.

Keywords: EOG; eye quietness; golf putting; ocular behavior; quiet eye.

Figure 1

(a) Horizontal and vertical (left and right eye) EOG signals. Voltages (µV) are represented as function of time (s). Voltage increases indicate eye movements to the left or upward; voltage decreases indicate movements to the right or downward. Eyeblinks are evident in the vertical EOG signals. (b) Output of the QE algorithm with 60 and 20 µV thresholds. Thick colored lines indicate the quiet eye period in its premovement initiation (QE_pre) and postmovement initiation (QE_post) components

Figure 2

(a,b,c) Eye quietness (i.e., HEOG‐SD, µV) and (d,e,f) its variability across putts (i.e., SD HEOG‐SD, µV) as a function of time (s) from −4 to 2 s and either group (expert, novice) or outcome (holed, missed). HEOG‐SD is inversely related to eye quietness: lower values indicate greater quietness. (a,d) Group × Time effects. Error bars indicate between‐subjects SE. (b,e) Outcome × Time effects for the experts. (c,f) Outcome × Time effects for the novices. Error bars indicate within‐subject SE computed through normalization of the outcome factor (Cousineau, 2005)

Figure 3

Durations (s) of (a) total (QE_total), (b) premovement initiation (QE_pre), and (c) postmovement initiation (QE_post) quiet eye, as a function of threshold (µV and corresponding degrees of visual angle). The solid and dashed lines represent mean durations, respectively, for experts and novices. The two colored bars above the x axis indicate r² and p values associated with the independent samples t tests conducted on group differences (df = 18). Shaded areas represent the SE of each group's means and were computed using pooled estimates, hence corresponding with the independent samples t tests (Pfister & Janczyk, 2013)