• children;
  • frequency-doubling perimetry;
  • rarebit perimetry


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Purpose: To describe the outcome of visual field examinations performed with rarebit (RB) and frequency-doubling technology perimetry (FDT) in children and young adults.

Methods: Twenty-one children (aged 6.5–12 years) and 30 teenagers and young adults (aged 14–20 years), participated in the study.

Results: Reliable RB examinations were carried out in 76% of the younger group and 90% of the older group. Corresponding values for FDT were 57% and 90–95%, respectively. The RB results were very similar to those previously obtained in adult subjects, while some subjects showed borderline values in FDT, depending on the criteria used. The RB perimetry was preferred by 88% of the examined subjects.

Conclusions: Rarebit perimetry seems useful for visual field examination in children aged 7 years and over, if age-corrected normative data are established; this test was also preferred by the tested subjects. No adaptation or special instructions were needed and the children found it rather amusing.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

In adults, computerized static perimetry has become the gold standard in most diagnoses. However, when examining children, manual testing with Goldmann perimetry is still the most commonly used method (De Souza et al. 2000). The advantages of Goldmann perimetry are that the examination can be adapted to the maturity and age of the child and that there are well defined normal values for kinetic perimetry for children (Quinn et al. 1991; Wilson et al. 1991). However, the shortcomings are well known: the results are dependent on the knowledge and skills of the examiner, and the sensitivity of white-on-white perimetry to a low degree of damage in, for example, glaucoma is low (Kerrigan-Baumrind et al. 2000). A few studies on computerized perimetry in children have been published. Some have shown that computerized perimetry can be used for examinations of children from 5 years of age (Marraffa et al. 1995; Tschopp et al. 1998a, 1998b), while other authors put the limit at 7−8 years (Donahue & Porter 2001; Morales & Brown 2001; Brown et al. 2005). In many of these studies specially designed equipment was built, suprathreshold techniques were used, and good performance was rewarded (Safran et al. 1996; Tschopp et al. 1999). In studies without adaptation of the technique, reliable results have been reported in children of 10–12 years of age (Becker & Semes 2003).

Previous studies have shown that adult performance in visual sensitivity testing is reached at about 11 years of age (Tschopp et al. 1999), and that visual field (VF) extent at 12 years of age is approximate to that of an adult (Wilson et al. 1991). However, few studies report reference values from children for the different computerized techniques, and there are no age-adjusted normative databases for the paediatric population (Blumenthal et al. 2004).

During the last few decades new perimetric techniques have been developed that do not rely on differential light threshold measurements. The most recent methods for quantitative measurement of visual function within the 20- to 30-degree VF are frequency-doubling technology perimetry (FDT) and rarebit perimetry (RB). These techniques are patient-friendly and have short examination times, which presumably make them suitable for use in children.

The FDT N-30 threshold program takes less than 5 mins. In addition, the apparatus offers a very rapid screening program (C-20-5), which in normal adults takes less than 40 seconds. Several studies have indicated that FDT is sensitive to early glaucomatous damage in adults (Burnstein et al. 2000; Cello et al. 2000; Bayer et al. 2002; Martin et al. 2003; Martin & Wanger 2004; Medeiros et al. 2004).

The RB test, which is run on an ordinary computer, relies on mouse-clicks and uses a moving fixation object, thus mimicking a simple computer game. The examination time can be varied from less than 1 min to about 5 mins for a more thorough examination. In the first reports RB showed promising results regarding the early detection of neurological (Frisén 2002, 2003; Martin et al. 2004) and glaucomatous damage (Martin & Wanger 2004; Brusini et al. 2005).

The aim of the present study was to describe the outcome of VF examinations performed with these two non-conventional techniques in children and young adults.

Materials and Methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References


Two groups of subjects were recruited to the study. The younger group consisted of 12 boys and 9 girls, aged 6.5–12 years. The older group consisted of 14 males and 16 females, aged 14–20 years (Table 1).

Table 1.  Demographic data, median (95% confidence limits) and range.
 Younger group (n = 21)Older group (n = 30)
  1. F = female; M = male; sph eq = spherical equivalent.

Age (years)9 (8.6–10.1)17 (16.5–17.8)
Range 6.5–12Range 14–20
Gender (F/M)9/1216/14
Visual acuity1.0 (1.0–1.1)1.2 (1–1.2)
Range 1.0–1.3Range 1.0–1.6
Refraction0 (− 0.22–0.08)0 (− 0.33–0.1.7)
(sph eq)Range − 0.5 to + 0.5Range − 3.6 to + 1.5)

All subjects were recruited from among the families and friends of staff at St. Erik's Eye Hospital, Stockholm. The inclusion criteria were good health, best corrected visual acuity (VA) of at least 1.0 (20/20), no history of any eye disease or reading difficulties (Pammer & Wheatley 2001), and no use of medications that might influence performance or VF results. All subjects were examined with RB perimetry. All children in the younger group and 20 subjects in the older group were also examined with FDT.


No special efforts were made to adapt the examination methods and no special rewards were offered to the examined subjects. The VF examinations were performed by the author, strictly according to the manufacturer's recommendations (Johnson et al. 1998; Rarebit Visual Field Test, Handbook 2000, included in the RB program). The subjects were informed in practically the same way as adult patients and no special training session was used. To avoid the effect of learning and fatigue, only the right eye was examined in all subjects. The tests were performed in random order. In FDT, both the screening program (C-20-5) and the threshold program (N-30, Version 3.00/2.00) were performed. After completion of the RB and FDT examinations, all subjects were asked which method they preferred.

Rarebit perimetry

The rarebit technique has been extensively described elsewhere (Frisén 2002; the software is available free of charge from In the current study, Version 3.0 was used. The test depends on presentations of one or two high-contrast, minuscule (one-half of normal minimum angle of resolution [MAR]) light spots (microdots), briefly (200 ms) presented against a dark background on an ordinary 15-inch liquid crystal display (LCD) screen. The subjects respond by single or double mouse-clicks, depending upon the number of perceived dots. A total of 10% of the presentations were used for control purposes and contained only one dot or none at all. The number of false positives were calculated and expressed as ‘errors’ and used as a measure of the reliability of the test results. Stimuli were presented in 30 separate test areas within the 30-degree VF. Every area consisted of a 5-degree circle and was tested repeatedly, with a total of 10 presentations per area with new dot localizations every time. Testing distance was 0.5 m, except for the four most central test locations. These locations were probed at the end of the test and required a 1.0-m test distance in order to adjust for normal MAR. The RB test is thought to target the receptive fields of the parvo cellular pathway, which is considered to mediate finely detailed vision, such as resolution (Merigan & Katz 1990; Merigan et al. 1991; Wassle & Boycott 1991; Dacey 1993). The results were expressed as percentages of the overall mean hit rate, that is, the number of targets seen relative to all target presentations. The number of locations with a hit rate below 90% is shown in Fig. 1. Abnormal values are indicated with an asterisk.


Figure 1. Printout from a normal RB perimetry (97%) from a 12-year-old boy. Empty circles indicate that all dots were perceived. The more filled a circle is, the larger proportion of non-perceived dots in this location.

Download figure to PowerPoint

A normal person has a complete receptor matrix without any gaps between the receptive fields, and will have a hit rate close to 100%. Loss of receptive fields gives a lower hit rate (Frisén 2002). Fixation was encouraged by dynamically changing the location of the fixation mark. For this study, the examiner observed the tested subjects' eyes during the entire procedure. The examinations were performed with an undilated pupil and with appropriate correction for the examination distance if the refraction errors exceeded + 1.0 D or − 2.00 D. Three ranges for normal results in adults have been published: 88–100% (Frisén 2002); 78–100% (Martin & Wanger 2004), and 78–98% (Brusini et al. 2005). In the current study, abnormality for an RB examination was determined according to both the manufacturer's indicators (Rarebit Visual Field Test Handbook 2000, included in the program) and the classification suggested by Brusini et al. (2005). An abnormal RB test was defined as having at least one of the following: mean hit rate of less than 80%; more than 15 areas with a hit rate of less than 90%; at least two areas with a hit rate of less than 50%; or at least one area with a hit rate of 30% or less (Brusini et al. 2005).

Frequency-doubling technology perimetry

Frequency-doubling technology perimetry was performed using the FDT visual field instrument provided by Humphrey®Welch-Allyn (Zeiss Humphrey Systems, Dublin, California, USA). An extensive description of the technique can be found elsewhere (Johnson et al. 1998). The FDT stimulus consists of light and dark stripes with low spatial frequency (0.25 cycle/degree), flickering with high frequency (25 Hz), giving an illusion of doubled spatial frequency. The contrast in the stimuli was gradually increased, and the subject responded whenever a movement was perceived somewhere in the VF. The screening program (C-20-5) tests 17 locations within the 20-degree field, with a stimulus intensity corresponding to a 95% chance of being seen by a normal subject. This examination takes about 35 seconds in healthy adult subjects. If the stimulus is not seen in one or more locations the result is considered to be abnormal.

The N-30 threshold program tests two additional locations on the nasal side. Examination time is normally less than 4.5 mins. The results are numerically expressed in two indices, as mean deviation (MD) and pattern standard deviation (PSD), with asterisks indicating the age-corrected probability for pathology. The total deviation and the pattern deviation maps are either clear white (normal) or show one of four possible levels of shading corresponding to age-normative significance levels for each location tested (Fig. 2). Frequency-doubling technology perimetry was originally designed for the detection of early glaucomatous damage (Johnson & Samuels 1997), and several authors have suggested different definitions of limits for abnormality, especially for glaucoma diagnosis and follow-up. These definitions use the number and placement of locations with subnormal thresholds, together with the probability for abnormality in the MD and PSD indices (Burnstein et al. 2000; Sample et al. 2000; Milano et al. 2001; Medeiros et al. 2004). In the current study both the manufacturer's probability limits (Johnson et al. 1998) and the definition of glaucomatous damage suggested by Medeiros et al. (2004) were used; these were the presence of at least two test areas with p < 0.05 or worse on the pattern deviation plot, or the presence of a PSD with p < 0.05. The examinations were performed with an undilated pupil and, in line with the manufacturer's recommendations, no correction for refractive errors was used (Johnson et al. 1998).


Figure 2. Printout from an abnormal FDT perimetry with clustered shaded locations and an abnormal PSD from the same 12-year-old boy as in Fig. 1.

Download figure to PowerPoint

Reliability of the tests

Both methods provide indices of test result reliability. In RB perimetry, the moving fixation mark enhances fixation and the examiner supervises the co-operation of the tested subject. Automatic tests of fixation accuracy are not made in RB perimetry, but one of the test areas is placed so as to at least partially overlap with the blind spot, in the expectation that good fixation should cause a substantial fraction of probes to be missed in this location (Frisén 2002). In addition, the number of errors is automatically counted to give an index of reliability. For adult subjects, three or less errors are considered normal (Frisén 2002). The number of errors and the percentage of misses in the blind spot area were used as criteria for reliability, in addition to the examiner's judgement.

The FDT C-20-5 examinations' fixation errors and false positives are given. They are also given for the N-30 program, together with false negatives. Johnson (1998) reported that more than 30% erroneous responses were flagged as being outside normal limits for reliability at the printout. Becker & Semes (2003) independently used a stricter criterion for unreliability in the C-20-1 program when examining healthy children, where one or more false positive or one or more fixation error, or two or more locations were flagged as being outside the normal limits. In the current study, both criteria were used.


For statistical analysis the non-parametric Mann–Whitney U-test, Fisher's exact test and the linear regression anova test were used. A p-value of less than 0.05 was regarded as significant.

The study was approved by the local ethical committee and conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all subjects and their parents prior to enrolment.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Data regarding age, gender, VA and refraction (spherical equivalent) are given in Table 1.

Examination time

The examination times for the FDT C-20-5 test, N-30 test, and the RB, together with the reaction times for RB in the different age groups, are given in Table 2. There was a significant difference between the age groups in examination times for RB, but not for FDT.

Table 2.  Examination times for the FDT C-20–5, N-30 tests, and the RB, together with reaction time for RB in the different age groups, median (95% confidence limits) and range. Times are in seconds.
 Younger group (n = 21)Older group (n = 30)Significance
RB exam time306 (294–330)270 (270–288)p = 0.0014
Range 245–384Range 252–324) 
RB reaction time0.76 (0.73–0.92)0.69 (0.68–0.77)NS
Range 0.59–1.39Range 0.57–1.18 
FDT C-20–5 exam time49.8 (43.8–55.9)46.1 (43–49.2)NS
Range 34–72Range 40–64(n = 20)
FDT N-30 exam time264 (258–276)282 (270–293)NS
Range 240–306Range 210–324(n = 20)

Reliability of the RB VF examination

According to the examiner's judgement, all subjects except one maintained fixation during the entire RB examination. The exception was the youngest subject, aged 6.5 years, who was unable to co-operate throughout a full examination due to boredom and fatigue. In two children in the younger group and two in the older, no blind spot was detected, indicating dubious fixation. The mean of the observed percentage missed in the blind spot area was 25% in the younger group and 31% in the older group. This indicates detection of the blind spot, although less clearly than in adults (Frisén 2002). The number of errors ranged from 0 to 9 with a modal number of 2 in the younger group, and from 0 to 7 with a modal number of 1 in the older group. Five children in the younger group (24%) and three in the older group (10%) had more than three errors. There was a significant correlation (r = 0.32, p = 0.02) between age and RB errors, but not between RB hit rate and RB errors (r = 0.14, p = 0.33). Two of the youngest children had some difficulty with the double mouse-click, which often resulted in a ‘four-fold’ click. However, this did not influence the examination results. According to the criteria used for reliability, 16 out of 21 (76%) subjects in the younger group and 27 out of 30 (90%) in the older group showed reliable RB results.

Reliability of the FDT VF examination

Twelve of the 21 children (57%) between 6.5 and 12 years of age had reliable C-20-5 screening FDT VFs, according to the criteria defined by the manufacturer (Johnson et al. 1998). All unreliable VFs were due to fixation errors. When using the criteria defined by Becker & Semes (2003), only three of the 21 children (14%) produced reliable results. Seven children had more than two flagged locations (range 3–6). One 10-year-old girl did not perceive any stimulus in the C-20-5 test and was unable to perform this test. Nineteen of the 20 subjects (95%) in the older group had reliable C-20-5 tests using FDT criteria. However, when using the Becker criteria, four of these 20 subjects showed unreliable results due to two or more flagged locations (range 2–6).

Twelve of the 21 children (57%) in the younger group showed reliable N-30 visual fields. The youngest child did not want to perform this longer test. Of the 12 younger children with reliable C-20 fields, six also had reliable N-30 fields. Three of the eight children who did not fulfil the criteria for reliability in the C-20-5 showed reliable N-30 fields. Eighteen of 20 (90%) subjects in the older group showed reliable N-30 threshold VFs. Sixteen of these also performed reliable C-20-5 VFs. There was no significant correlation between age and errors or between FDT MD and errors.

A total of 36 of the 41 subjects examined with both FDT and RB stated that they preferred the RB perimetry.

Rarebit VF results

Table 3 shows the median mean hit rate and the number of locations, with less than a 90% hit rate in the two different age groups for all 51 subjects. There was no significant difference in results between the VFs classified as reliable and those classified as unreliable, according to the criteria used. The median mean RB hit rates differed significantly between the younger and older groups (p = 0.0085). The median hit rate was the same in the older group as the rates previously published for adult subjects (Martin & Wanger 2004). The range was almost the same in both age groups as it is in adults (Frisén 2002; Martin & Wanger 2004; Brusini et al. 2005). The numbers of depressed locations also differed significantly between the age groups (p = 0.0134). Table 4 shows the number of subjects with significantly abnormal RB hit rates and the number of locations with a hit rate of less than 90% of VF test in the different age groups. The correlation between RB hit rate and age in the younger group was r = 0.64 (p = 0.002), but no correlation was found in the older group (r = 0.03, p = 0.8).

Table 3.  Rarebit VF results in the different age groups, median (95% confidence limits) and range.
 Younger group (n = 21)Older group (n = 30)Significance
  1. (Hit rate = percentage of microdots correctly identified.

  2. No. of locations < 90% = number of locations with a hit rate of less than 90%.

Hit rate93% (90–95)97% (96–98)p = 0.0085
Range 78–100Range 89–100 
No. of locations5 (4–9)2 (2–4)p = 0.0118
< 90%Range 0–21Range 0–10 
Table 4.  Number of subjects in the different age groups with significantly abnormal RB hit rates and abnormal number of locations with a hit rate < 90% according to the manufacturer's indicators, and the number of abnormal fields according to the criteria suggested by Brusini et al. (2005).
 Younger group (n = 21)Older group (n = 30)Significance
Hit rate50p = 0.05
No of locations < 90%50p = 0.05
Brusini et al. 200520NS

Frequency-doubling technology C-20-5 VF results

One of the children in the younger group was unable to undertake the FDT C-20-5 screening program. Of the remaining 20 young children, eight had normal results (i.e. no location with abnormal thresholds). Twelve young children had abnormal thresholds in one to six locations. One subject in the older group had three locations flagged as abnormal; the remainder had normal FDT C-20-5 screening results.

Frequency-doubling technology N-30 VF results

Table 5 shows the FDT MD and PSD values and the number of shaded test areas in the pattern deviation map from all examined subjects. There were significant differences regarding both FDT MD (p = 0.0001) and FDT PSD (p = 0.006) between the age groups. There was no significant difference in results between the VFs, classified as reliable or unreliable according to the manufacturer's indicators. The numbers of significantly abnormal results, according to the indices, are shown in Table 6. Two of the younger children had significant deviations in both MD and PSD, seven in MD only, and four in PSD only. Fifteen of the subjects, 10 in the younger and five in the older group, had a cluster of at least two adjacent test areas reaching a probability of p < 0.05 in the pattern deviation map and a PSD with p < 0.05 (see Fig. 2 for an example). This indicated abnormal results using the definition of glaucomatous damage according to Medeiros et al. (2004). There was no significant correlation between FDT MD and age in either the younger or the older group.

Table 5.  FDT N-30 VF results in the different age groups, median (95% confidence limits) and range.
 Younger group (n = 21)Older group (n = 20)Significance
  1. MD = mean deviation.

  2. PSD = pattern standard deviation.

  3. PDM p < 0.05 = number of test areas with p < 0.05 in the pattern deviation map.

FDT MD−2.5 (− 4.6 to − 1.2)0.8 (− 0.22 to + 1.33)p = 0.0007
Range − 13.68 to + 6.25Range − 2 to + 3.33 
FDT PSD4.7 (4.18–7.89)3.35 (2.08–3.92)p = 0.005
Range 3.05–19.88Range − 5.06 to + 4.74 
PDM < 0.054 (3.36–7.03)1 (0.78–2.3)p = 0.0007
Range 1–14Range 0–5 
Table 6.  Number of subjects with significantly abnormal FDT MD and FDT PSD, according to the manufacturer's indicators and number of subjects classified as abnormal according to the criteria used by Medeiros et al. (2004), in the different age groups.
 Younger group (n = 21)Older group (n = 20)Significance
  1. MD = mean deviation.

  2. PSD = pattern standard deviation.

  3. Numbers outside the bracket = the number of abnormal VFs in pattern deviation map only, within brackets = the number of abnormal VFs both according to the pattern deviation map and the PSD index.

Medeiros et al. 200410 (6)5 (1)NS


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The results in the current study indicate that children from 7 years of age can perform RB perimetry with reliable and good results. Frequency-doubling technology perimetry, however, seems to be more difficult for younger children. The proportion of subjects with significantly abnormal FDT screening results in the current study was almost the same as that in the study by Nesher et al. (2004), as were the threshold data.

A higher percentage (76%) of the younger group was able to produce reliable RB results compared to the C-20-5 and N-30 FDT tests (57%). Reliability between the tests was similar for the older group (90–95%). The main reasons for unreliability were false positive responses in RB and fixation losses in FDT. However, it was difficult to compare the reliability data and it is possible that if fixation checks had been performed in RB in the same way as in FDT, the results might have been different. In FDT, where the fixation target was static, perception was sometimes reported by the subjects to be disturbed by the Troxler effect. In RB, the fixation target was dynamic and presented at different locations on the screen, thus enhancing fixation. This may be the reason why RB seemed easier for the younger group of tested subjects. In previous studies of FDT in children, reliable results from the C-20-1 screening program and the C-20 threshold test were obtained at 5–7 years of age (Becker & Semes 2003; Nesher et al. 2004). These results were not confirmed in the current study, which might be partly due to the stricter inclusion criteria regarding attention span used by Nesher et al. (2004). Pammer & Wheatley (2001) reported that children with reading disabilities produced subnormal FDT results compared to normal readers. None of the children in the younger group in this study had any reading or writing difficulties.

Regarding reaction time in RB, the observed values conform to values at threshold (≈ 0.7 seconds) in conventional perimetry reported by Wall et al. (2002), but are longer than supra-threshold stimuli (≈ 0.4 seconds) in the same study. Thus, the time for decision making at threshold seems to be similar to the time for discrimination between one and two dots. In a previous study of 18-year-old subjects, the median reaction time was also ≈ 0.7 seconds (Martin et al. 2004), and the same reaction time was noted in a study conducted in adults aged 28–88 years (unpublished data).

According to the FDT criteria for glaucomatous damage suggested by different authors (Burnstein et al. 2000; Sample et al. 2000; Milano et al. 2001; Medeiros et al. 2005), approximately 50% of the younger and at least four of the older subjects could be classified as having glaucomatous damage when examined with the FDT N-30 program. Regarding the RB examination, only two subjects in the younger group could be regarded as outside normal limits according to the criteria suggested by Brusini et al. (2005), and four more subjects according to the manufacturer's criteria. Frequency-doubling technology perimetry was developed in the assumption that it would measure the integrity of a particular subgroup of retinal ganglion cells (Anderson & O'Brian 1997). However, this assumption has been questioned (Anderson & Johnson 2002; White et al. 2002), and the method may instead be measuring grating detection or contrast sensitivity rather than spatial frequency doubling (McKendrick et al. 2003). Children are supposed to reach adult-like levels of contrast sensitivity (measured with contrast sensitivity distance charts) at 7 years of age, but the variability is rather high (Scharre et al. 1990). Immaturity in this system may partly explain the borderline results observed in some children in this study and in the study by Nesher et al. (2004), who examined children from the age of 5 years. Another explanation may be that the FDT results are not expressed as the actually measured thresholds, but as deviations from age-corrected, average normal thresholds in the built-in database, which does not contain any subjects younger than 18 years of age. When comparing threshold data, the sensitivities observed in the current study correspond very well with those previously obtained by Nesher et al. (2004), but they differ from those reported by Pammer & Wheatley (2001). A possible explanation for this may be that the current study was performed under standard conditions, as in a clinical setting, but without any special adaptations for children and without extra training.

Tschopp et al. (1999) proposed that the difference in visual sensitivity, measured using computerized perimetry, between young children and adults might be explained by difference in attention. In the current study, no significant correlation was found between either RB hit rate and RB errors, or between FDT MD and number of errors.

Rarebit perimetry was preferred by 88% of the examined subjects, which is in accordance with previous findings when examining adults (Martin & Wanger 2004). Most of the children were very familiar with computer games, and only the youngest (age 6.5 years) had some difficulties with double-clicking, probably due to immature motor function (De Souza et al. 2000). Several of the younger and the majority of the older subjects found the examination rather amusing. Regarding FDT, several children were uncertain about whether they really saw anything or not. In the younger group, almost every child had difficulty in fitting into the headrest. Several children were disturbed by the Troxler effect, induced by a stable fixation mark, especially those in whom the non-dominant eye was examined.

In conclusion, RB perimetry seems to be useful for VF examination of children from 7 years of age, if age-corrected normative data are established. This test was also preferred by the subjects. No adaptation and no special instructions were needed, and the children found it rather entertaining. Studies for the evaluation of its diagnostic efficiency in different conditions, such as paediatric glaucoma, are underway.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References
  • Anderson AJ & Johnson CA (2002): Mechanisms isolated by frequency-doubling technology perimetry. Invest Ophthalmol Vis Sci 43: 398401.
  • Anderson RS & O'Brian C (1997): Psychophysical evidence for selective loss of M ganglion cells in glaucoma. Vision Res 37: 10791083.
  • Bayer AU, Maag KP & Erb C (2002): Detection of optic neuropathy in glaucomatous eyes with normal standard visual fields using a test battery of short-wavelength automated perimetry and pattern electroretinography. Ophthalmology 109: 13501361.
  • Becker K & Semes L (2003): The reliability of frequency-doubling technology (FDT) perimetry in a paediatric population. Optometry 74: 173199.
  • Blumenthal EZ, Haddad A, Horani A & Anteby I (2004): The reliability of frequency-doubling perimetry in young children. Ophthalmology 111: 435439.
  • Brown SM, Bradley JC, Monhart MJ & Baker DK (2005): Normal values for Octopus tendency-oriented perimetry in children 7 through 13 years old. Graefes Arch Clin Exp Ophthalmol 2005, April 15 (Epub ahead of print).
  • Brusini P, Salvetat ML, Parisi L & Zeppieri M (2005): Probing glaucoma visual damage by rarebit perimetry. Br J Ophthalmol 89: 180118.
  • Burnstein Y, Ellish NJ, Magbalon M & Higginbotham EJ (2000): Comparison of frequency-doubling perimetry with Humphrey visual field analysis in a glaucoma practice. Am J Ophthalmol 129: 328333.
  • Cello KE, Nelson-Quigg JM & Johnson CA (2000): Frequency-doubling technology perimetry for detection of glaucomatous visual field loss. Am J Ophthalmol 129: 314322.
  • Dacey DM (1993): The mosaic of midget ganglion cells in the human retina. J Neurosci Dec 13: 53345355.
  • De Souza EC, Berezovsky A, Morales PH, De Arruda Mello PA, De Oliveira Bonomo PP & Salamao SR (2000): Visual field defects in children with congenital glaucoma. J Pediatr Ophthalmol Strabismus 37: 266272.
  • Donahue SP & Porter A (2001): SITA visual field testing in children. J AAPOS 5: 114117.
  • Frisén L (2002): New sensitive window on abnormal spatial vision: rarebit probing. Vision Res 42: 19311939.
  • Frisén L (2003): Spatial vision in visually asymptomatic subjects at high risk for multiple sclerosis. J Neurol Neurosurg Psychiat 74: 11451147.
  • Johnson CA & Samuels SJ (1997): Screening for glaucomatous visual field loss with frequency-doubling perimetry. Invest Ophthalmol Vis Sci 38: 413425.
  • Johnson CA, Wall M & Murray F et al. (1998): A Primer for Frequency Doubling Technology. Dublin, California: Humphrey Systems.
  • Kerrigan-Baumrind LA, Quigley HA, Pease ME, Kerrigan DF & Mitchell RS (2000): Number of ganglion cells in glaucoma eyes compared with threshold visual field tests in the same persons. Invest Ophthalmol Vis Sci 41: 741748.
  • Marraffa M, Pucci V, Marchini G, Morselli S, Belluci R & Bonomi L (1995): HPR perimetry and Humphrey perimetry in glaucomatous children. Doc Ophthalmol 189: 383386.
  • Martin L, Ley D, Marsal K & Hellström A (2004): Visual function in young adults following intrauterine growth restriction. J Pediatr Ophthalmol Strabismus 41: 212218.
  • Martin L & Wanger P (2004): New perimetric techniques: a comparison between rarebit and frequency-doubling technology perimetry in normal subjects and glaucoma patients. J Glaucoma 13: 268272.
  • Martin L, Wanger P, Vancea L & Gothlin B (2003): Concordance of high-pass resolution perimetry and frequency-doubling technology perimetry results in glaucoma: no support for selective ganglion cell damage. J Glaucoma 12: 4044.
  • McKendrick AM, Anderson AJ, Johnson CA & Fortune B (2003): Appearance of the frequency-doubling stimulus in normal subjects and patients with glaucoma. Invest Ophthalmol Vis Sci 44: 11111116.
  • Medeiros FA, Sample PA & Weinreb RN (2004): Frequency-doubling technology perimetry abnormalities as predictors of glaucomatous visual field loss. Am J Ophthalmol May 137: 863871.
  • Merigan WH & Katz LM (1990): Spatial resolution across the macaque retina. Vision Res 30: 985991.
  • Merigan WH, Katz LM & Maunsell JH (1991): The effects of parvocellular lateral geniculate lesions on the acuity and contrast sensitivity of macaque monkeys. J Neurosci 11: 9941001.
  • Milano G, Rossi GCM, Djeugoue A & Clemente A (2001): Comparison of achromatic automated perimetry, short-wavelength automated perimetry and frequency-doubling technique perimetry in the diagnosis of early glaucoma. In: WallM (ed). Perimetry Update 2000/2001. The Hague, the Netherlands: Kugler Publications 217224.
  • Morales J & Brown SM (2001): The feasibility of short automated static perimetry in children. Ophthalmology 108: 157162.
  • Nesher R, Norman G, Stern Y, Gorck L, Epstein E, Raz Y & Assia E (2004): Frequency-doubling technology threshold testing in the paediatric group. J Glaucoma 13: 278282.
  • Pammer K & Wheatley C (2001): Isolating the M(y)-cell response in dyslexia using the spatial frequency-doubling illusion. Vision Res 41: 21392147.
  • Rarebit Visual Field Test Handbook 2000 (included in the RB program).
  • Quinn GE, Fea AM & Minguini N (1991): Visual fields in 4- to 10-year-old children using Goldmann and double-arc perimeters. J Pediatr Ophthalmol Strabismus 28: 314319.
  • Safran AB, Laffi GL, Bullinger A, Viviani P, De Weisse C, Desangles D, Tschopp C & Mermoud C (1996): Feasibility of automated visual field examination in children between 5 and 8 years of age. Br J Ophthalmol 80: 515518.
  • Sample PA, Boswoth CF, Blumenthal EZ, Girkin C & Weinreb RN (2000): Visual function-specific perimetry for indirect comparison of different ganglion cell populations in glaucoma. Invest Ophthalmol Vis Sci 41: 17831790.
  • Scharre JE, Cotter SA, Block SS & Kelly SA (1990): Normative contrast sensitivity data for young children. Optom Vis Sci 67: 826832.
  • Tschopp C, Safran AB, Viviani P, Bullinger A, Reicherts M & Mermoud C (1998a): Automated visual field examination in children aged 5–8 years. Part I. Experimental validation of a testing procedure. Vision Res 38: 22032210.
  • Tschopp C, Safran AB, Viviani P, Bullinger A, Reicherts M & Mermoud C (1998b): Automated visual field examination in children aged 5–8 years. Part II. Normative values. Vision Res 38: 22112218.
  • Tschopp C, Viviani P, Reicherts M, Bullinger A, Rudaz N, Mermoud C & Safran AB (1999): Does visual sensitivity improve between 5 and 8 years? A study of automated visual field examination. Vision Res 39: 11071119.
  • Wall M, Kutzko KE & Chauhan BC (2002): The relationship of visual threshold and reaction time to visual field eccentricity with conventional automated perimetry. Vision Res 42: 781787.
  • Wassle H & Boycott BB (1991): Functional architecture of the mammalian retina. [Review.] Physiol Rev 71 (2): 447480.
  • White AJR, Sun H, Swanson WH & Lee BB (2002): An examination of physiological mechanisms underlying the frequency-doubling illusion. Invest Ophthalmol Vis Sci 43: 35903599.
  • Wilson M, Quinn G, Dobson V & Breton M (1991): Normative values for visual fields in 4- to 12-year-old children using kinetic perimetry. J Pediatr Ophthalmol Strabismus 28: 151153.