Rapid number naming in chronic concussion: eye movements in the King-Devick test

Abstract

Objective: The King-Devick (KD) test, which is based on rapid number naming speed, is a performance measure that adds vision and eye movement assessments to sideline concussion testing. We performed a laboratory-based study to characterize ocular motor behavior during the KD test in a patient cohort with chronic concussion to identify features associated with prolonged KD reading times.

Methods: Twenty-five patients with a concussion history (mean age: 31) were compared to control participants with no concussion history (n = 42, mean age: 32). Participants performed a computerized KD test under infrared-based video-oculography.

Results: Average intersaccadic intervals for task-specific saccades were significantly longer among concussed patients compared to controls (324.4 ± 85.6 msec vs. 286.1 ± 49.7 msec, P = 0.027). Digitized KD reading times were prolonged in concussed participants versus controls (53.43 ± 14.04 sec vs. 43.80 ± 8.55 sec, P = 0.004) and were highly correlated with intersaccadic intervals. Concussion was also associated with a greater number of saccades during number reading and larger average deviations of saccade endpoint distances from the centers of the to-be-read numbers (1.22 ± 0.29° vs. 0.98 ± 0.27°, P = 0.002). There were no differences in saccade peak velocity, duration, or amplitude.

Interpretation: Prolonged intersaccadic intervals, greater numbers of saccades, and larger deviations of saccade endpoints underlie prolonged KD reading times in chronic concussion. The KD test relies upon a diffuse neurocognitive network that mediates the fine control of efferent visual function. One sequela of chronic concussion may be disruption of this system, which may produce deficits in spatial target selection and planning of eye movements.

Figures

Figure 1

LEFT panels. Test cards 1 (top), 2 (middle), and 3 (bottom) of the King–Devick test (as implemented in digital paradigm and on sideline testing). RIGHT panels. Demonstration of eye tracking tracings of saccades and fixations overlaid on KD test cards 1 (top), 2 (middle), and 3 (bottom) for a representative control subject. The blue lines with dots (data samples) represent “task‐specific” horizontal and oblique saccades and the red circles represent fixations (duration, which is displayed in milliseconds, correlates with the diameter of the circle). KD, King–Devick.

Figure 2

Main sequence plots of concussed and control subjects for all saccades during the KD test. Dotted line demarcates task‐specific saccades (2° or greater). (A). Plot of peak velocity versus amplitude showing that as saccade amplitude increases, the peak velocity increases in an asymptotic distribution. Parameters: Control Vm = 545.41, c = 6.21; Concussion Vm = 565.43, c = 6.17. (B) Plot of duration versus amplitude. No differences are seen in these relationships between concussed and control subjects in A or B. Parameters: Control D = 9.13, n = 0.62; Concussion D = 9.00, n = 0.62. KD, King–Devick.

Figure 3

Distributions of saccade amplitudes and intersaccadic intervals. (A) Histograms and subject‐by‐subject raster plots of intersaccadic interval values. (B) Histograms and subject‐by‐subject raster plots of saccade amplitudes. In both panels, histograms for control subjects are plotted in black, and in blue for concussion subjects. Histograms highlight the overall distribution of values, while rasters highlight any individual differences in the pattern of ISI values relative to the group.

Figure 4

Distributions of spatial errors. For each saccade endpoint, spatial error was defined as the difference between the saccade position and the nearest position on the screen corresponding to a KD number. (A) Horizontal errors. (B) Median endpoint errors. KD, King–Devick.

Figure 5

Correlation plots. (A) Correlations between median intersaccadic intervals (ISI) values and King–Devick (KD) test times. (B) Correlations between the median number of saccades below 20° produced by each subject and that subject's KD test time. In both panels, dashed lines are the best fitted line through the origin that minimizes total squared error of the fit to the control data. This fit was chosen on the assumption that the limiting case of a KD test time of zero would also be associated with zero ISI and saccade amplitudes. Fitted lines are displayed to highlight the linear pattern and consistency of results only. Product‐moment correlations are also computed, and are independent of the displayed lines.