Performance consistency of normal observers in forced-choice tachistoscopic visual line bisection

Pseudoneglect (PN) refers to the leftward error exhibited by normal observers on line bisection tasks (Bowers and Heilman, Neuropsychologia, 18, (1980) 491-8). Although a thorough review of the literature has shown PN to be relatively robust (Jewell and McCourt, Neuropsychologia, 38, (2000) 93-110), controversy remains concerning the reliability of the phenomenon, with some studies reporting a relatively high incidence of normal subjects with rightward bisection errors. The present experiment assesses the consistency of bisection performance in normal young observers. Right-handed subjects (N=22) participated in a tachistoscopic forced-choice line bisection task. Each subject participated in 7-16 experimental sessions separated by at least 24 h (total bisection measurements=317). Individual bisection performance could thus be evaluated with respect to within-subject variability measures. An eyetracker recorded gaze position during the task in one session. A highly significant mean group bisection error of -0.26 degrees (P<0.001) was obtained (left negative), and individual subject means ranged from -0.55 degrees to +0.03 degrees. Of the 317 total bisection measurements, 9% (28) deviated rightward. Significant (P<0.05) mean leftward errors occurred in 91% (20/22) of subjects. Mean bisection error in two subjects was not significantly different from zero. No subject possessed a significant rightward error. Mean gaze deviation from screen (and line) center ranged from +/-0.9 degrees, and was positively correlated (P<0.05) with bisection error. It is concluded that forced-choice tachistoscopic line bisection measures are highly reliable; a mean correlation of +0.87 exists between mean error based on 15 trials and means estimated from a random sample of only two trials. The incidence of true rightward bisection error in the population of normal right-handed subjects is thus estimated to be less than 5%.