Influence of target type and RAF rule on the measurement of near point of convergence

Transcription

1 Ophthal. Physiol. Opt : Influence of target type and RAF rule on the measurement of near point of convergence Paul M. Adler*, Mary Cregg, Ann-Julie Viollier à and J. Margaret Woodhouse *50 High Street, Stotfold, Hertfordshire SG5 4LL, UK, Faculty of Medicine and Health Science, UAE University, PO Box 17666, Al-Ain, United Arab Emirates, à Universite de Paris-Sud, Enseignement d Optometrie, Tour alpha, Rue des Millepertuis, Les Ulis, France, and School of Optometry & Vision Sciences, Cardiff University, Cardiff, UK Abstract Purpose: This study was designed to investigate if the use of different target types or the RAF rule affected the measurement of near point of convergence (NPC). Methods: The subjects comprised three groups: (1) 20 children aged 6 9 years (2) 17 children aged years (3) 14 adults aged years. Five targets were used to measure the NPC: (1) pencil tip, (2) fingertip, (3) penlight, (4) N5 letter and (5) vertical line target on the RAF rule. Results: There was no significant difference in NPC measurements between the pencil tip, fingertip and N5 target in free space. The penlight resulted in significantly more remote NPC break and recovery points compared with the fingertip and pencil tip (p < 0.05). The RAF rule influences the NPC obtained (p < 0.001). The greatest difference in NPC measurements was observed when comparing the RAF line target and the finger in free space; the former resulted in NPC values of 1.9 times (95% CI times) as much as those obtained with the finger. Conclusions: Use of the penlight and RAF rule resulted in a more distant NPC break point compared with other targets. The effect of the RAF rule was more apparent for more receded NPC points. Keywords: accommodation, age, near point of convergence, RAF rule Received: 28 April 2005 Revised form: 4 December 2005, 24 February 2006 Accepted: 26 February 2006 Correspondence and reprint requests to: Paul M. Adler. address: paul.adler@eyezone.co.uk Introduction The prevalence of convergence insufficiency has been estimated to be between 0.8% and 17.8% depending on the number of clinical criteria used to diagnose the condition (Porcar and Martinez-Palomera, 1997; Rouse et al., 1998; Lara et al., 2001; Junghans et al., 2002). The clinical significance of the near point of convergence (NPC) measurement, which is routinely included in optometric eye examinations, is that it forms one of the primary diagnostic tools in the assessment of convergence insufficiency and in determining treatment outcomes (Dwyer, 1991; Adler, 2002; Scheiman et al., 2002, 2005). Rouse et al. (1997) found that 93% of optometrist respondents to a survey indicated that they consider NPC as a factor in making a diagnosis of convergence insufficiency. Normal values for NPC break point were traditionally considered to be between 8 and 10 cm, according to conventional optometric textbooks (von Noorden, 2002; Evans, 2002; Griffin and Grisham, 2002) and this view (albeit incorrect) is still widely held by many optometrists in clinical practice. This is despite the fact that recent studies of NPC in children have shown that NPC measurements are in general <6 and 10 cm for break and recovery respectively (Hayes et al., 1998; Chen et al., 2000; Jime nez et al., 2004). Scheiman et al. (2003) concluded that the clinical cut-off for NPC in adults, viewing an accommodative target or penlight with red green glasses, was 5 cm for NPC break and 7 cm for NPC recovery. It is expected that, for a given doi: /j x

2 Influence of measurement method on NPC: P. M. Alder et al. 23 target, the NPC break point differs from the NPC recovery point (Siderov et al., 2001; Scheiman et al., 2003; Jime nez et al., 2004). Despite the general consensus on the importance of measuring NPC, there is disagreement as to the type of test target that should be used. Opinions differ as to whether the target should have an accommodative component or whether the NPC should represent a measurement of solely fusional and proximal convergence (Grosvenor, 1996; von Noorden, 1996; Hayes et al., 1998; Pensyl and Benjamin, 2006; Evans, 2002). Ciuffreda (1974) has shown that targets with a high accommodative demand result in less remote measurements of NPC. Adler (2001) suggested that the NPC is heavily influenced by the attention the subject is able to bring to the task rather than whether the target itself is accommodative or non-accommodative. The RAF rule (Neely, 1956) has traditionally formed a part of routine optometric assessments in the UK and Ireland, but there is little or no research data to support or validate its use (Haag-Streit UK, Edinburgh Way, Harlow, Essex, UK). The Bernell Accommodative Rule fulfils a similar function and is more commonly used in the USA (Bernell VTP, Mishawaka, IN, USA). Two recent studies have investigated comparative data for different target types in adults. In the first study, Siderov et al. (2001) compared three target types (a fingertip, pencil tip and RAF line target), all placed next to or mounted on the RAF rule, and found that in non-presbyopes a fingertip or pencil gave statistically significantly less remote NPC break points than the RAF line target; nevertheless the authors did not consider the difference (2.5 cm) to be clinically significant. There was no difference in NPC recovery points in non-presbyopes. The second study (Scheiman et al., 2003) compared three targets (one accommodative: 20/ 30 letter; one non-accommodative: penlight; one dissociative and non-accommodative: penlight viewed through red green glasses), all placed next to or mounted on a Bernell Accommodative Rule. In subjects with no signs or symptoms of convergence insufficiency, this study found a statistically significant difference between all three targets for NPC break points and between the penlight and the other two targets for NPC recovery points. As in the aforementioned study by Siderov et al. (2001), the authors considered the actual difference (<0.7 cm) to be clinically insignificant. Jime nez et al. (2004) set out to determine normative values for NPC break and recovery in children aged 6 12 years using both a penlight, and a penlight viewed with a red filter in front of one eye. Although they reported different values for mean NPC break and recovery for each target, this was not analysed statistically and the authors did not comment on the difference. The purpose of the present study was to evaluate the effect of different targets commonly used in optometric practice and the presence of the RAF rule on NPC measurement in asymptomatic, binocularly normal, emmetropic, pre-presbyopic subjects in different age groups. Methods Fifty-one subjects aged between 6 and 30 years were recruited for the study. All subjects were patients of one of the authors (PA), who attended for a routine optometric eye examination. From the commencement of the study, all patients presenting to the practice for an eye examination, who did not wear glasses and who met the inclusion criteria, were invited to participate in the study on a consecutive basis. The first group comprised 20 children aged 6 9 years (mean age 7.71 years, S.D. 0.85). The second group comprised 17 children aged years (mean age years, S.D. 0.98) and the third group consisted of 14 adults aged between 20 and 30 years (mean age years S.D. 3.60). This final group was originally intended to include young adults aged between 20 and 23 years but the age range was increased to years for this group because of the difficulty in recruiting suitable subjects who met the criteria for inclusion in the study. All subjects were required to meet the following criteria for inclusion in the study: unaided vision in both eyes of 6/6 or better for distance and N5 at 40 cm for near; stereo-acuity of 70 of arc or better on the Randot Stereo Test (Stereo Optical Company, Chicago, IL, USA); anisometropia 1.00 D in any meridian; spherical component of between )0.25 DS and DS with no more than )0.50 DC of astigmatism; asymptomatic; normal near point of accommodation in both eyes; exophoria <6 D or esophoria<2 D (all compensated as demonstrated on Mallett Unit) and no strabismus or oculomotor defects such as incomitant eye movements. The research followed the tenets of the Declaration of Helsinki and approval for the research protocol was obtained from the School of Optometry and Vision Science Research Ethics Committee at Cardiff University, Number Informed consent was obtained from all subjects after the nature of the experiment and possible consequences of the study had been fully explained and an information sheet provided. In the case of children, parental consent was also obtained. If the subject was 12 years old but less than 16 years, a signature of the child was also obtained and witnessed by a third party so as to comply with UK Data Protection rules. The RAF rule consists of a 50-cm-long square section rule with a plastic slider holding a rotating

3 24 Ophthal. Physiol. Opt : No. 1 Figure 1. The RAF rule. four-sided drum, on the outside of which a choice of four different visual targets is mounted, one of which is a vertical line with central dot for convergence fixation. A plastic cheek rest is attached to one end of the rule to allow the device to be comfortably placed on the patient s cheeks. The targets can be moved closer towards the patient along the rule until the NPC is recorded (Figure 1). Measurements of NPC were obtained using six different targets. Four targets were presented in free space: (1) the sharpened tip of a pencil, (2) the examiner s finger, (3) a penlight, (4) a single N5 letter. Two targets were mounted on the RAF rule (5) a single N5 letter (6) the RAF target of vertical line with a central dot. For each subject the targets were presented in a random order using tables generated by Microsoft Excel 2000Ò. The subjects were seated comfortably in a well-lit room and the task was explained to them. The targets were presented at a slight downwards angle as described by Neely (1956) and the target moved towards the subject at a speed of approximately 1 2 cm per second as described by Siderov et al. (2001). The subject was instructed to report when the target became double (subjective break point). This point was measured to the nearest 0.5 cm using a tape measure and was recorded as the NPC break point. The examiner also noted the position when the eyes ceased to fixate (objective break point). The target was then moved away from the subject until single vision was recorded (subjective recovery point) or the examiner observed that both eyes had re-fixated (objective recovery point). All measurements were taken by the same examiner (A-JV). For each target type and condition, the NPC break and recovery were each taken to be the average of three successive measurements. The data were entered into SPSS (Version 11 for Windows, SPSS Inc., Chicago, IL, USA) for statistical analysis. Table 1. Near point of convergence break and recovery (Rec) points (mean ± S.D.) in cm for different target types in comparison to other studies Pencil tip Finger Penlight N5 N5 on RAF Line on RAF Break Rec Break Rec Break Rec Break Rec Break Rec Break Rec Age 6 9 years (n ¼ 20) 5.5 ± ± ± ± ± ± ± ± ± 7.5 a 11.9 ± 7.8 a 9.9 ± 7.0 a 12.7 ± 7.4 a Age years (n ¼ 17) 5.5 ± ± ± ± ± ± ± ± ± 6.2 a 10.9 ± 6.9 a 8.6 ± 6.5 a 10.7 ± 7.4 a Age years (n ¼ 14) 4.5 ± ± ± ± ± ± ± ± ± 8.6 a 10.9 ± 9.3 a 8.8 ± 8.1 a 10.3 ± 9.0 a 7.9 ± 0.6 a 8.6 ± 1.5 a 7.7 ± 0.5 a 7.9 ± 1.2 a 5.3 ± 1.9 a 7.6 ± 1.9 a Siderov et al. (2001), mean age 25.1 ± 6.3 years (n ¼ 14) 2.1 ± ± ± 1.7 b 4.3 ± 2.7 b Scheiman et al. (2003), age years (n ¼ 175) a Indicates the target was mounted on the RAF rule. b This target was a single 6/9 letter mounted on a Bernell Accommodative Rule.

4 Influence of measurement method on NPC: P. M. Alder et al. 25 Results Effect of age The results of NPC for different age groups are given in Table 1. In all cases, the objective break point was the same as the subjective break points. For comparative purposes, the results obtained from other studies comparing target types are also listed. To analyse whether there was a difference between age groups, a Repeated Measures ANOVA was performed (NPC break and recovery points in cm as dependent variable, age group as factor variable). Results were adjusted using Greenhouse Geisser correction to account for violations of assumptions of independence of errors. There was no overall difference between age groups (F ¼ 0.41, df ¼ 4, p > 0.05) but there was a significant difference between target types for NPC break and recovery points obtained (F ¼ 22.66, df ¼ 10, p < 0.001). The interaction of target type and age group was not significant (F ¼ 1.03, df ¼ 20, p > 0.05). Effect of target type For the entire group of subjects as a whole (Table 2), the NPC break point was significantly different for different target types (F ¼ 20.34, df ¼ 2.58, p < 0.01). Subsequent pairwise analysis showed that the NPC measured with targets mounted on the RAF rule was significantly more remote than those presented in free space (p < 0.001), and that, to a lesser extent, the NPC measured with penlight target differed to the finger and pencil tip (p < 0.05). It is important to point out here that the data were not strictly normally distributed, there was a high degree of variability as measured by the standard deviation and there were some outliers. To account for skewness, log transformation of the data was carried out and the analysis repeated. The results were similar in so far as the targets mounted on the RAF rule were significantly different to the other targets (p < 0.001) and the penlight differed from both the finger and the pencil tips (p < 0.05). Results for those targets which showed a significant difference are given in Table 3. The greatest difference was between the NPC break points obtained with the RAF line target mounted on the RAF rule, and the finger in free space. The geometric mean of the NPC obtained with the RAF line target is 1.9 times as much as that obtained with a finger target in free space (95% confidence interval times as much). The type of target had an overall effect on NPC recovery point obtained (F ¼ 3.48, df ¼ 3.11, p < 0.05), (see Table 2). Subsequent pairwise analysis showed that the penlight differed from the pencil tip (p < 0.01) and the finger (p < 0.05); similarly the Table 2. Near point of convergence break and recovery (Rec) points (cm) for different target types for entire group (n ¼ 51) Target in free space Mounted on RAF rule Pencil tip Finger Penlight N5 N5 on RAF Line on RAF Break Rec Break Rec Break Rec Break Rec Break Rec Break Rec n ¼ 51 Mean ± S.D. a 5.2 ± ± ± ± ± ± ± ± ± ± ± ± 8.0 Median a Geometric mean a Lower reference range a Upper reference range a a As the data were not strictly normally distributed, a log transformation was carried out. For the purposes of simplicity and ease of comparison with other studies, the data is reported with mean and standard deviation taken from the untransformed data with the median also given as an estimation of central tendency; the 95% reference ranges for the data are taken from the transformed data. (The geometric mean is the value obtained when the mean of the transformed data is back-transformed, i.e. the anti-log of the mean of the logarithmic form of the data.

5 26 Ophthal. Physiol. Opt : No. 1 Target NPC more receded than Mean difference a (ratio) 95% Confidence interval Significance level NPC break points Penlight Pencil Finger Line on RAF rule Pencil <0.001 Finger <0.001 Penlight <0.001 N <0.001 N5 on RAF rule Pencil <0.001 Finger <0.001 Penlight <0.001 N <0.001 NPC recovery points Penlight Pencil Finger RAF (N5) Pencil Table 3. Targets which resulted in a significant difference in near point of convergence (NPC) break points and recovery points (n ¼ 51) a Log-transformed values were used; therefore, the differences between targets in the logtransformed scale represent a ratio in the original scale. The geometric mean of NPC break point with the penlight is 1.2 times more (or 20% more receded) than the NPC break point obtained with the pencil, etc. targets mounted on the RAF rule differed from the pencil tip and the fingertip (p < 0.05). To eliminate skewness, we considered it prudent to carry out log transformation of the data. Subsequent analysis confirms the differences between the penlight and both pencil tip and fingertip (p < 0.05); the N5 target mounted on the RAF rule differed from the pencil tip (p < 0.05). The geometric mean of the NPC recovery points obtained with the penlight is 1.18 times as much (95% confidence interval ) as that obtained with a pencil tip target in free space (i.e. 18% increase). Difference between NPC break and recovery points To assess whether there was a significant difference in break point and recovery point for different targets, a paired t-test was carried out for each target type. As expected, the recovery point was significantly more remote from the break point at a significance level of in all cases. The mean difference between break and recovery points was less for the targets mounted on the RAF rule compared with targets in free space. Confidence intervals for mean differences for those targets mounted on the RAF rule do not overlap with other targets (Table 4). Discussion Effect of age using different targets The difference between target type was consistent across all age groups studied, i.e. there was no age effect. We Table 4. Mean difference in NPC break and recovery points in cm for different target types (n ¼ 51) Target Mean difference 95% Confidence interval Significance Pencil tip < Finger < Penlight < N < Line on RAF rule < N5 on RAF rule < originally selected age groups based on primary and secondary school age groups as we had postulated that attention may be a factor in NPC measurement and that the younger children might show greater interest in some targets relative to others. However, it appears that for children of school age, any differences in NPC measurements obtained with various targets are not influenced by age. Comparison to previous studies comparing target types Although two previous studies (Siderov et al., 2001; Scheiman et al., 2003) have examined the effect of different target types on NPC measurement using adult observers, to our knowledge this is the first study to examine the effect of different target types on NPC measurement in children. Although Jime nez et al. (2004) reported NPC measurements in children using both a penlight and a penlight viewed through a red filter, they did not analyse or comment on the differences found.

6 Influence of measurement method on NPC: P. M. Alder et al. 27 There are some differences in the results obtained in the present study for the 20- to 30-year-old group compared with the study by Siderov et al. (2001). It is interesting to note that their group obtained more remote mean values for NPC break point using the pencil tip and the fingertip given that the targets were placed alongside the RAF rule. This phenomenon does not extend across the rest of the results obtained when the RAF line target was used (in both studies the RAF line target was mounted on the RAF rule), the mean values obtained by Siderov s group were less remote than the values obtained in our study. The values obtained in the present study fall outside the norms given by Scheiman et al. (2003); NPC values in our study were more remote. Assessment of the standard deviation suggests that the NPC break points obtained show greater variability in range compared with the study by Siderov et al. (2001) and Scheiman et al. (2003). Both previous studies comparing target types (Scheiman et al., 2003) used adult subjects, all of whom were optometry students. As acknowledged in the latter study, this may limit the applicability of the results to the normal population. In our study, objective NPC values were equal to subjective NPC values indicating that the inattention of the observers was not a factor. Variability may be accounted for by the fact that the subjects in this study were inexperienced observers with no knowledge about the aims of the test; experienced observers may be inclined to give more consistent results. Anticipation may influence the NPC obtained which might occur if an observer is familiar with the aims of the exercise. Kumar et al. (2002) showed that anticipatory vergence movements often occur when a stimulus moves in a predictable fashion. The indications are that these movements rely on working memory being influenced by prior vergence responses. Other factors may also have had an effect. All subjects in our study were emmetropic (within defined limits) with excellent unaided vision and were not wearing spectacles. Myopic or hyperopic subjects in the studies by Siderov et al. (2001) and Scheiman et al. (2003) may have been helped or hindered by base-in or base-out prismatic effects of spectacle lenses. They were allowed to use their habitual spectacle prescription for the experiment; this could have introduced bias because of induced prismatic effect depending on the power and centration of the spectacles worn. The inter-pupillary distances and centration distance of the lenses were neither measured nor taken into account. Conversely, although all subjects in our study had minor (if any) refractive errors, these were uncorrected and may have affected the accommodative response and thereby the NPC measurement. Nevertheless, as our subjects were asymptomatic and had not been prescribed spectacles for visual tasks, our assessments reflect the type of examination that would occur in optometric practice, i.e. if a subject does not wear spectacles, the optometrist examines the patient s ability to converge appropriately under everyday conditions (without refractive correction in place). Moreover, we chose target scenarios that are commonly used in practice. The RAF rule is usually only used with a mounted target; if a fingertip or pencil tip are used then these are usually presented in free space. In the previous studies, all measurements were taken with either the RAF rule or the Bernell Accommodative Rule in apposition to the face. Proximal accommodation and convergence might, therefore, have influenced the results. Influence of accommodation on the NPC response Ciuffreda (1974) has shown that targets with high accommodative demand result in better or closer NPC measurements, presumably because accommodative convergence is involved. In theory, using an accommodative target maximises all three components of the convergence response: fusional, proximal and accommodative. In this study, the Ônon-accommodativeÕ target (penlight) gave statistically more remote NPC break and recovery points relative to the pencil tip and fingertip. We had expected that the N5 target would maximise the accommodative response; instead it gave slightly more remote values and so could not be differentiated from the penlight. It appears that the detail on the target is irrelevant but that a fingertip or pencil tip is an effective target in differentiating an ÔaccommodativeÕ NPC response from a Ônon-accommodativeÕ NPC response as obtained using a penlight. It could also be the case that the fingertip and the pencil tip being three-dimensional targets are more effective in eliciting a convergence response. Rouse et al. (2002) examined the repeatability and reliability of NPC measurements and reported that a typical between session measurement error in NPC assessments is 1 cm but differences can be as high as 5 cm. If we assume that a difference of >5 cm is clinically significant, our finding that the NPC value with the RAF rule is 1.9 times as much as the NPC obtained with the finger target (i.e. increase of 90%) implies that the greater the NPC obtained with a target in free space the greater the likelihood of obtaining a clinically significant difference in NPC with the RAF rule. Similar conclusions can be drawn when comparing other targets in free space compared with those on the RAF rule (see Table 3). Effect of the RAF rule One unexpected finding of the study was that the presence of the RAF rule adversely affected the NPC

7 28 Ophthal. Physiol. Opt : No. 1 measurement. Measurements of NPC break point were invariably more remote when the RAF rule was used. As the data are skewed, the more remote the NPC the greater the effect of the RAF rule. The use of the line target on the RAF rule increases the NPC break point by 90% compared with the NPC obtained with the finger. Even the apparently Ônon-accommodativeÕ target (penlight) gave less remote values of NPC break compared with the accommodative stimuli (N5 target and RAF line) which were mounted on the RAF rule. The maximum NPC extended up to 36 cm using the RAF rule. Without the RAF rule, the maximum NPC recorded was 26 cm. Although it was expected that the break point would differ from the recovery point (Siderov et al., 2001; Scheiman et al., 2003; Jime nez et al., 2004), it was interesting to note from the results in this study that the mean difference between NPC break and recovery points appears lower for targets mounted on the RAF rule relative to those obtained with targets in free space. This may be accounted for by the fact that the RAF rule-mounted targets are different from all others in terms of NPC break point but different only from the pencil tip on NPC recovery. Thus the mean difference between break and recovery for each target mounted on the RAF rule is decreased. As this is one of the factors that should be taken into account in making a diagnosis of convergence insufficiency (Scheiman et al., 2003), the use of the RAF rule may mask borderline cases in this regard. The more remote NPC obtained with the RAF rule could be due to a number of effects. Because of the position of the cheek rest it was not possible to obtain an NPC measurement <2 cm on the RAF rule although, as this study was completed, a modification to the RAF rule has been suggested to overcome this problem (Adler, 2004). This does not, however, account for the difference in NPC break points observed. Another possible factor is the physical distraction of the RAF rule in that attention may be drawn towards the rule and thus away from the target itself. The NPC target is moving while the RAF rule is stationary. Thus the ruler may provide a reference point in space against which the moving target is judged. In a sense, the presence of the RAF rule may provide a Ôconflicting cueõ to the vergence response. Erkelens (2001) has speculated that depth and vergence may result from independent streams of disparity processing. He has suggested that magnocellular layers process disparities that drive vergence, while a parvocellular stream of disparity processing is involved in depth perception. Sheliga and Miles (2003) estimated that perceived depth can elicit vergence movements under certain conditions. Results from the work of McGraw et al. (2003) showed that the visual system often combines inconsistent sources of visual information including object luminance, motion, depth and textural cues to form an overall approximation of spatial position; however, this merging of visual characteristics results in a decrease in localisation accuracy. As the RAF rule itself is seen in physiological diplopia, this may affect the NPC obtained. Tanimoto et al. (2004) have shown that the dynamics of the initial vergence response can be influenced by peripheral disparity and that this is probably unrelated to the perception of motion in depth. Another possibility is the contribution of proprioceptive and proximal convergence feedback. The subject is aware of the cheek rest of the RAF rule, thus the proprioceptive feedback might indicate that the target is closer than it actually is. This might provide a stimulus to cease convergence and accommodation at an earlier point, resulting in a more remote NPC. Although the effects of this Ôcheek proprioceptionõ have not been investigated, Han and Lennerstrand (1998) showed that the dynamics of monocularly-driven accommodative vergence movements were influenced by extra-retinal signals from neck proprioception. In relation to the proximity of the RAF rule, there is evidence that this is unlikely to have an effect. Hung et al. (1996) found that the relative contributions of proximal accommodation and proximal convergence were normally insignificant under simulated naturalistic viewing situations. Do we really need to use the RAF rule? One might assume that using the RAF rule would lead to more accurate measurements but the standard deviation of the measurements was greater when using the RAF rule. One does not read with a ruler projecting from the book or page of print. It seems more logical to measure the NPC using an object in free space that correlates with the real world scenario. A piece of string or tape attached to a target, which can then be measured against a ruler, might suffice. From this study, we suggest that care should be taken comparing NPC break points obtained with the RAF rule-mounted target with norms obtained from studies where the target was presented in free space and vice versa. Although Scheiman et al. (2003) found no clinically important difference in target type for asymptomatic subjects without convergence insufficiency, they did find a significant difference in the NPC response with different targets in symptomatic patients who had convergence insufficiency. It would be interesting to assess the effect of targets mounted on the RAF rule relative to targets in free space in subjects with convergence insufficiency.

8 Influence of measurement method on NPC: P. M. Alder et al. 29 Conclusions We agree with previous studies that the use of an accommodative target produces a more accurate assessment of convergence ability than a non-accommodative target such as a penlight. The detail of this accommodative target is irrelevant as we found no significant difference between a fingertip, pencil tip or N5 letter presented in free space. The value of the RAF rule lies in the convenience of measuring the actual NPC position but the use of the rule itself results in more remote NPC break points and thus a decrease in the relative difference between NPC break and recovery. If the RAF rule is used, it is important that the norms used for comparison in interpreting results are taken from NPC norms established using the RAF rule; similarly for targets presented in free space. If instituting therapy for improving the NPC and using NPC as one of the outcome measures, one must use either the RAF rule or free-space methods on successive occasions. References Adler, P. (2001) Optometric evaluation of binocular vision anomalies. In: Binocular Vision and Orthoptics (eds S. Doshi and B. J. W. Evans), Butterworth-Heinemann, Oxford, pp Adler, P. M. (2002) Efficacy of treatment for convergence insufficiency using vision therapy. Ophthalmic Physiol. Opt. 22, Adler, P. M. (2004) RAF near point rule modification. Ophthalmic Physiol. Opt. 24, Chen, A. H., O Leary, D. J. and Howell, E. R. (2000) Near visual function in children. Ophthalmic Physiol. Opt. 20, Ciuffreda, J. K. (1974) Near point of convergence as a function of target accommodative demand. Opt. J. Rev. Optom. 111, Dwyer, P. S. (1991) Clinical criteria for vergence accommodation dysfunction. Clin. Exp. Optom. 74, Erkelens, C. J. (2001) Organisation of signals involved in binocular perception and vergence control. Vis. Res. 41, Evans, B. J. W. (2002) Pickwell s Binocular Vision Anomalies: Investigation and Treatment. Butterworth-Heinemann, Oxford. Griffin, J. R. and Grisham, J. D. (2002) Binocular Anomalies: Diagnosis and Vision Therapy, 4th edn. Butterworth-Heinemann, Boston. Grosvenor, T. P. (1996) Primary Care Optometry: Anomalies of Refraction and Binocular Vision, 3rd edn. Butterworth- Heinemann, Boston. Han, Y. and Lennerstrand, G. (1998) Effects of neck muscle proprioceptive activation on the dynamics of monocularly driven horizontal vergence movements. Acta Ophthalmol. Scand. 76, Hayes, G. J., Cohen, B. E., Rouse, M. W. and De Land, P. N. (1998) Normative values for the near point of convergence of elementary schoolchildren. Optom. Vis. Sci. 75, Hung, G. K., Ciuffreda, K. J. and Rosenfield, M. (1996) Proximal contribution to a linear static model of accommodation and vergence. Ophthal. Physiol. Opt. 16, Jime nez, R., Pérez, M. A., Garcı a, J. A. and Gonza lez, M. D. (2004) Statistical normal values of visual parameters that characterize binocular function in children. Ophthal. Physiol. Opt. 24, Junghans, B., Kiely, P. M., Crewther, D. P. and Crewther, S. G. (2002) Referral rates for a functional vision screening among a large cosmopolitan sample of Australian children. Ophthal. Physiol. Opt. 22, Kumar, A. N., Han, Y., Garbutt, S. and Leigh, R. J. (2002) Properties of anticipatory vergence responses. Invest. Ophthalmol. Vis. Sci. 43, Lara, F., Cacho, P., Garcia, A. and Megias, R. (2001) General binocular disorders: prevalence in a clinic population. Ophthal. Physiol. Opt. 21, McGraw, P. V., Whitaker, D., Badcock, D. R. and Skillen, J. (2003) Neither here nor there: localizing conflicting visual attributes. J. Vis. 3, Neely, J. C. (1956) The RAF near point rule. Br. J. Ophthal. 40, von Noorden, G. K. (2002) Examination of the Patient-II: Motor signs in heterophoria and heterotropia. In: Binocular Vision Motility and Ocular Motility. (ed G. K. von Noorden), Mosby, St Louis, pp Pensyl, C. D. and Benjamin, W. J. (2006) Ocular motility. In: Borish s Clinical Refraction (ed. W. J. Benjamin), W.B. Saunders Company, Philadelphia, PA, pp Porcar, E. and Martinez-Palomera, A. (1997) Prevalence of general binocular dysfunctions in a population of university students. Optom. Vis. Sci. 74, Rouse, M. W., Hyman, L., Hussein, M. and the Convergence Insufficiency and Reading Study (CIRS) Group. (1997) How do you make the diagnosis of convergence insufficiency? Survey results. Quoted in: Scheiman M, Gallaway M, Frantz KA, Peters RJ, Hatch S, Cuff M. and Mitchell GL. (2003). Nearpoint of convergence: test procedure, target selection, and normative data. Optom. Vis. Sci. 80, Rouse, M. W., Hyman, L., Hussein, M. and Solan, H. and the Convergence Insufficiency and Reading Study (CIRS) Group. (1998) Frequency of convergence insufficiency in optometry clinic settings. Optom. Vis. Sci. 75, Rouse, M. W., Borsting, E., Deland, P. N. and the Convergence Insufficiency and Reading Study (CIRS) Group. (2002) Reliability of binocular vision measurements used in the classification of convergence insufficiency. Optom. Vis. Sci. 79, Scheiman, M., Wick, B. and Steinman, B. (2002) Clinical Management of Binocular Vision: Heterophoric, Accommodative, and Eye Movement Disorders, 2nd edn. Lippincott Williams & Wilkins, Philadelphia, PA.

9 30 Ophthal. Physiol. Opt : No. 1 Scheiman, M., Gallaway, M., Frantz, K. A., Peters, R. J., Hatch, S., Cuff, M. and Mitchell, G. L. (2003) Nearpoint of convergence: test procedure, target selection, and normative data. Optom. Vis. Sci. 80, Scheiman, M., Mitchell, G. L., Cotter, S., Taylor, Kulp, M., Cooper, J., Rouse, M., Borsting, E., London, R., Wensveen, J. and the Convergence Insufficiency Treatment Trial Group. (2005) A randomized clinical trial of vision therapy/orthoptics vs pencil pushups for the treatment of convergence insufficiency in young adults. Optom. Vis. Sci. 82, Sheliga, B. M. and Miles, F. A. (2003) Perception can influence the vergence responses associated with open-loop gaze shifts in 3D. J. Vis. 3, Siderov, J., Chiu, S. C. and Waugh, S. J. (2001) Differences in near point of convergence with target type. Ophthal. Physiol. Opt. 21, Tanimoto, N., Takagi, M., Bando, T., Abe, H., Hasegawa, S., Usui, T., Miki, A. and Zee, D. S. (2004) Central and peripheral visual interactions in disparity-induced vergence eye movements. I. Spatial interaction. Invest. Ophthalmol. Vis. Sci. 45,