Although the human anatomy is arranged symmetrically about a central vertical axis, the majority of persons will use one side of their body with greater frequency, facility or skill than the other. From this universal observation has emerged the functional concept of laterality dominance,1 demonstrated in physical terms as an habitual writing hand or perhaps a preferred ball-kicking foot. Notwithstanding the uniquely bilateral cortical representation of either eye, a similar distinction has arisen with respect to the ocular system: the majority of individuals will use a preferred eye for particular uni-ocular tasks or under specific viewing conditions.

Contrary to popular belief the laterality of a preferred eye is not linked or matched to limb preference, and is also not reliably indicated by habitually superior visual acuity (VA) of one or other eye.2 Furthermore, it has come to be realised that laterality of eye dominance is not as functionally rigid as the term might imply, instead varying in degree or presence depending upon test circumstances.3–5

Nevertheless, there arise clinical situations where the demonstration of a preferred eye is a necessary precursor to any proposed ophthalmic therapy or treatment. These occasions include the prescribing of symptom-relieving optical prism as a supplement in corrective spectacle lenses,6 where all (or the greater part) of the prescribed prism is located before the non-dominant eye. The knowledge of an individual's OD is useful when considering the expediency of a ‘monovision’ approach to temporary or longterm unilateral refractive correction with contact lenses or surgery.7 The monovision technique was initially devised for the convenience of presbyopic contact lens wearers,8 but ocular surgery has embraced this approach in conjunction with intra-ocular lens implantation and (laser) refractive procedures with a high rate of success and degree of patient satisfaction across the several studies now reported.9,10 In monovision the dominant eye is usually corrected for distance and its companion for near: it is considered that the sight of the dominant eye will be less easily suppressed by the relatively blurred image in the non-dominant eye.7 It is also worth remarking that the ophthalmic clinician with a preferred viewing eye might (in the early years of a career) experience difficulty when using monocular equipment (e.g., an ophthalmoscope) to examine a patient's eye or ocular adnexa or when setting up and aligning instrumentation for ophthalmic measurement (e.g., a slit lamp-mounted Goldmann applanation tonometer). It is usually the case that experience will moderate or overcome initial difficulties.

Laterality of ocular dominance might be determined in several ways.11–13 The question arises as to which approach could be considered the most appropriate. Should the test reflect visually guided behaviour using, for example, a sighting or visual alignment task? Alternatively, should reliance be placed on an assessment of the sensory modality using tests of spatial vision or perhaps an unilateral optical blurring or stereo suppression technique? And if both types of test were undertaken on an individual, would the ocular laterality results be in agreement? On this latter point consensus is lacking: while several authors have reported disagreement between the outcome of the alternative OD test approaches,14–18 others have disputed this conclusion.19,20

Given the advisabilty of establishing the existence and laterality of any eye preference before attempting an unilateral optometric or ophthalmological intervention, the results of a concise comparative investigation are reported here using two specific OD tests, one sighting-based and the other sensory.

The subject group (n=72: mean age 29.0±7.4 years) comprised 51% males (n=37: mean age 28.9±7.4 years) and 49% females (n=35: mean age 29.2±7.5 years). Age distributions of the genders were not statistically significantly different (p=0.8). Mean optimal monocular VA for the sensory dominant eye across the group was −0.089±0.052logMAR (6/4.9 Snellen), with no statistically significant difference between the mean acuity of right or left dominant eyes (p=0.6) and no statistically significant gender-based acuity differences (p>0.4).

Table 1 summarises the OD outcomes recorded with the two alternative dominance test formats across the whole study group. Because distributions for male or female subjects were not statistically significantly different within and between tests (p>0.6) combined-gender numerical data are presented and analysed here.

Assessment of the raw data indicated that there was inter-test agreement of OD laterality in only 50% (36 of 72) of instances. Seventy-one percent (51 of 72) of individuals were identified as right eye dominant using the sighting test, while only 54% (39 of 72) of persons were right eye dominant with the sensory (monocular blur) test. This represents a statistically significantly different outcome (p=0.04) between tests.

As an aside, it was found that the introduction of +1.50D of spherical monocular spectacle blur under these test conditions reduced the group mean logMAR acuity for the sensory dominant eye of these young adult subjects to +0.705±0.096 (6/30.4 Snellen) from an optimal level of −0.089±0.052 (6/4.9). As an illustration (and ignoring any effect of variables such as pupil size)23 this approximately 0.8 log unit loss corresponds to a theoretical deterioration in clinical Snellen VA of 6/4 to 6/25, or 6/5 to 6/32, or 6/6 to 6/38. A pragmatic implication of this quantitative observation will be considered later.

We specifically investigated OD under conditions of distance (6m) fixation: we cannot speculate here whether the same outcome would have been obtained for near viewing.24 What can be stated is that the value of 71% right OD found for these subjects using the sighting test is in accord with the proportion recorded in larger population-based studies reported in the literature.1 The much lower value of 54% right eyedness obtained when applying the monocular blur test to these same individuals is striking but not unrecognised,15 as is the magnitude of the level of agreement (only 50% here) between the two alternative techniques across this subject group.

These incongruent OD test outcomes are in accord with similar but larger and more extensive comparative studies (most recently Seijas et al.15) in suggesting that alternative (e.g., sighting versus sensory) OD techniques show poor intra-subject agreement. Recent reviews of the phenomenon of OD, including its identification and a consideration of the significance of eye dominance within human visual perception, have suggested a more circumscribed role for a preferred eye than the burgeoning research literature on the topic might imply.4,25 Put succinctly, perhaps a so-called ‘dominant eye’ is simply that eye which is used by habit, preference or circumstance when viewing conditions permit only monocular or unilateral fixation. Furthermore it has been suggested on more than one occasion that perhaps a strong individual indication of OD might even prove to be a barrier to that person's subsequent successful tolerance of monovision.15,26

Under clinical circumstances, when an unilateral ophthalmic/optical treatment is to be applied, does an absolute sighting task or a relative sensory technique provide the most appropriate indication of OD? Recognising that the human visual system is habitually binocular, might tests that permit the maintenance of (a degree of) binocularity during ocular preference testing be desirable?27 However, a lack of significant sensory dominance in the normal visual system has been claimed,14 in which case indications of OD derived from sensory techniques may not prove a useful basis for ophthalmic clinical decision-making. Then again, given the evolving evidence-base that suggests OD might be considered a situation-driven preference,4,25,27 might a sighting technique better reflect the reality of OD? Finally, it has to be acknowledged that although the monocular refractive blur procedure (temporarily) produces a clinical situation akin to monovision correction the simplicity of the ‘hole-in-card’ test, possibly combined with peer-imitation based upon the published literature, probably accounts for this technique's popularity and frequent selection to indicate a preferred eye in the clinic or consulting room.

As a final point, a practical proposal arises from this present work. A recent study15 that investigated a battery of sighting and sensory OD tests reported a degree of OD uncertainty in >11.5% of cases using the optical blur test as compared to no equivocation using the ‘hole-in-card’ approach. Those authors used +1.00D of optical defocus which, in young adult subjects, has been shown to produce an approximately 0.4 log unit acuity loss28; whence 6/4 Snellen reduces to 6/10, 6/5 to 6/13, and 6/6 to 6/15. The +1.50D blur utilised in the present study produced a larger (0.8 log unit) acuity loss in our subjects (see Section “Results”), and in doing so appeared to remove subjective uncertainty as to the preferred laterality: unequivocal responses were recorded across subjects. It is thus suggested that a blurring lens of this greater refractive magnitude should preferably be used if this particular sensory OD test is applied to clinical subjects.