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 in variance between any pair of conditions (p > 0.11). The ANOVA showed a highly significant effect of Neck Posture (Factor A; F1,23 = 113, p < 0.0001), but no effect of before or after the trial (Factor B; F1,23 = 0.72, p = 0.40) and no significant Posture by Eye interaction (F1,23 = 0.17, p = 0.69). The confidence intervals for Factor A show that the difference between Forward versus Neutral Posture was very different from zero (95% CI: -16.1° to -10.8°). Confidence intervals for before versus after the trial show tight bounds, indicating that the subjects did not move their heads very much during their 2-minute session (95% CI: -0.48° to +0.20°). Figure 4 displays the change in craniovertebral angle from Neutral to Forward Posture. The abscissa shows a large range of changing craniovertebral angles in the group. The mean change was -13.5°, with a range of -30.3° to -3.2°. Therefore, a principal axis regression30 was performed to determine whether subjects who adopted greater changes in posture from Neutral to Forward positions also had greater changes in their accommodation responses. The results showed a minimal slope (+5.7×10-4 D/ degree) that was not significantly different from zero (p = 0.85). The 95% confidence interval for the slope was -5.9×10-3 to +7.1×10-3 D/degree. The correlation coefficient (r) of +0.04 was also minimal. Therefore, the accommodation change from Neutral to Forward head posture was not greater for those subjects who could bend their necks more than the other subjects. Discussion Relationship between neck posture and accommodation According to our data, there was no significant effect of a forward neck posture on eye accommodation. The confidence intervals on these non-significant differences were extremely narrow (less than one-tenth of a diopter), meaning we can be sure that dioptric differences are minimal. This finding does not support the PRI hypothesis presented in the Introduction that neck posture could affect subjective refraction. However, there are three limitations to this conclusion in the current study. First, our study examined posture using a 2-minute duration, which is about half the time of a total subjective refraction.25 Hence, future studies may consider longer posture durations to find if these affect accommodation. For example, sympathetic innervation to accommodation is slow and takes one minute or more to act compared to a fraction of a second for parasympathetic innervation.31 Thus, if neck posture affects sympathetic tone, a longer duration might lead to a greater effect. Second, instead of using a distance target as we did, future studies might use a near target, as this may invoke some visual stress. For example, Zetterberg et al.12 measured accommodation and found that visually demanding experimental near work significantly elevated eye, neck, and shoulder discomfort. Domkin et al.13 found a strong and significant correlation between the level of ciliary muscle contraction force and that of trapezius muscle activity during near tasks. Optometry & Visual Performance 267 Volume 13 | Issue 4 | December 2025 Figure 3. Craniovertebral angle for neutral and forward neck postures before and after each trial. Error bars indicate ±1 s.e.m. in the group. Figure 4. Change in autorefraction (D) as a function of change in craniovertebral angle (degrees) from Neutral to Forward Posture. A negative value on the abscissa indicates that the neck bent forward, lowering the craniovertebral angle.
RkJQdWJsaXNoZXIy MjEyMDQ0Nw==