Optometry & Visual Performance 266 Volume 13 | Issue 4 | December 2025 The autorefractor data was limited for two subjects due to data loss during each trial. One subject had only five available trials, and the second had only four available trials, which would otherwise have resulted in missing cells for ANOVA. These two subjects were excluded from the Posture and Neck data analysis but were included when comparing the craniovertebral angle data since their photographs were good. Posture and eye analysis Two primary analyses were performed on the accommodative data. The first was on the mean accommodation response in each trial, and the second was on the fluctuations of accommodation in each trial. Mean accommodation response Figure 2 depicts the mean accommodative error for each neck posture (Neutral and Forward head posture) and for each eye (right and left eye) under binocular viewing. The accommodative error data are based on a 3 m viewing distance, for which the perfect autorefraction would be -0.33 D. An autorefraction value of -0.33 D corresponds to an accommodative error of zero diopters. The positive values in Figure 2 indicate a lead in accommodation, whereas the negative values display a lag in accommodation. Both eyes and neck postures displayed an accommodative lag, except for the left eye during the Forward head posture, which displayed a lead in accommodation. Differences between conditions appear to be very small. The assumptions of ANOVA were tested prior to analysis. Four data sets were tested: Neutral posture and right eye, Neutral posture and left eye, Forward posture and right eye, and Forward posture and left eye. The Kolmogorov–Smirnov test28 showed that all four distributions did not depart significantly from the normal distribution (p > 0.2). The Brown–Forsythe test for homogeneity of variances29 showed no significant difference between same-eye conditions (p > 0.58). However, all left-eye data sets had a greater variance than the right-eye data sets (p ranging between 0.0006 and 0.0032). The variance ratios in these cases ranged between 3.21 and 3.72. Because these variance ratios were elevated, Box’s correction factor for departure from homogeneity of variances was applied to the ANOVA. Using a two-factor within-subjects ANOVA, neither posture (Factor A; F1,21 = 2.45, p = 0.13) nor eye (Factor B; F1,21 = 0.41, p = 0.53) were statistically significant factors, and there was no significant interaction between neck posture and eye (A × B; F1,9 = 0.84, p = 0.38). The confidence intervals for these nonsignificant differences were exceptionally narrow. The 95% confidence interval for Factor A (neck posture) had a low of -0.068 D and a high of +0.010 D. The 95% confidence interval for Factor B (eye) had a low of -0.185 D and a high of +0.098 D. Therefore, we can be sure that any dioptric differences due to these two factors are minimal in the population. Fluctuations of accommodation response The standard deviation of the accommodation response in each trial was used to measure overall accommodation microfluctuations. This measure includes low-frequency and highfrequency components of the accommodation microfluctuations.27 All data sets were normal by the Kolmogorov–Smirnov test (p > 0.2).28 All pairs of data sets had no significant difference in variance by the Brown–Forsythe test for homogeneity of variances (p > 0.22).29 Thus, the ANOVA assumptions were met. The ANOVA showed no significant effect of Posture (A; F1,21 = 2.57, p = 0.13), Eye (B; F1,21 = 0.24, p = 0.63), or Posture by Eye interaction (F1,21 = 1.67, p = 0.21). Craniovertebral angle analysis Figure 3 displays the cranio-vertebral angle for each neck posture (Neutral and Forward head posture) and before and after each trial. The data for Trials 1 and 2 were averaged for this analysis. The posture changed very little over the 2-minute trial in both the Neutral and Forward postures. The assumptions of ANOVA were tested prior to analysis. The Kolmogorov–Smirnov test28 showed Figure 2. Accommodative error for neutral and forward neck posture in right and left eyes. A positive value of accommodative error indicates a lead, and a negative value indicates a lag. Error bars indicate ±1 s.e.m. in the group.
RkJQdWJsaXNoZXIy MjEyMDQ0Nw==