Temporal Response Properties of the Macular Cone System: Effect of Normal Aging and Age-Related Maculopathy

Abstract

purpose. To evaluate the influence of aging and age-related maculopathy (ARM) on the temporal frequency response function (TFR) of macular focal electroretinography.

methods. Macular (18°) focal electroretinograms (FERGs) in response to sinusoidal flicker, modulated at TFs between 3.7 and 52 Hz, were recorded from 13 young (age range, 14–29 years) and 9 old (age range, 55–80 years) healthy subjects and from 18 patients with ARM (stage 2 disease; age range, 55–80 years; visual acuity ≥0.4). Amplitude and phase of the Fourier-analyzed response fundamental (1F) and seconnd harmonic (2F) were measured.

results. In young healthy subjects, mean 1F TFR showed a maximum amplitude at 41 Hz, a secondary peak at 3.7 Hz, a minimum at 8 Hz, and a high TF (32–52 Hz) roll-off. Mean 1F TFR of old, compared with young, healthy subjects showed amplitude enhancement at 10 to 14 Hz and a small loss at high TF. Mean 2F TFR of young and old healthy subjects had a maximum at 5.7 to 8 Hz and an attenuation beyond 10 Hz. Mean 1F and 2F TFRs of ARM patients were similar to those of old healthy subjects but were depressed in mean amplitude. FERG TFR changes of old healthy subjects and ARM patients were not mimicked by reducing stimulus retinal illuminance or modulation depth in young healthy subjects.

conclusions. FERG temporal properties are affected by normal aging and ARM. Because FERG TFR is shaped mainly by postreceptoral activity, the findings suggest that photoreceptor and postsynaptic dysfunction underlie aging- and ARM-related FERG changes.

Age-related macular degeneration is a degenerative disease of the macula characterized in the early stage by large soft drusen, hyperpigmentation or hypopigmentation of the retinal pigment epithelium (RPE), and moderate loss of central vision (age-related maculopathy [ARM], based on the international classification ¹ ). Late stages of the disease—geographic atrophy of the RPE or the subretinal neovascular membranes—are associated with more severe central visual impairment and can be considered a leading cause of irreversible blindness in elderly persons in the developed world. ²  

Histopathologic studies ³ have shown that drusen, the basic ARM lesions resulting from extracellular deposits between the RPE and Bruch membrane, are associated with photoreceptor degeneration (deflected or truncated outer segments and broadened inner segments) and altered synaptic terminals of photoreceptors to second-order neurons. Degenerative changes of photoreceptors have been also described in normal aging retinas, ⁴ suggesting that the retinal abnormalities found in ARM eyes represent a continuum of lesion severity from minimal/mild in normal aging eyes to moderate/severe in ARM eyes. ⁵ These anatomic changes may have a significant impact on visual function. There is indeed evidence that subtle visual losses, involving a variety of functions mediated by subpopulations of photoreceptors or postreceptoral neurons, can be detected in normal aging and ARM. ^{6 7 8 9 10 11 12 13} Temporal cone flicker sensitivity (CSF), among these functions, has been proposed as a potentially useful clinical test. ^{14 15} Temporal CFS losses of 0.2 to 0.4 log units at several temporal frequencies of the flicker stimulus were indeed found to be associated with early fundus alterations and to discriminate between ARM-affected eyes that convert to the exudative form and nonconverting eyes. ^{14 16}  

Temporal responsiveness of the macular cone system can be directly evaluated by using the focal electroretinogram (FERG), ^{17 18 19} a signal generated from the outer retina and middle retina in the macular region, in response to sinusoidal flicker stimulation. The FERG signal consists mainly of two harmonic components, a fundamental harmonic ²⁰ at the same temporal frequency as the stimulus (1F) and a second harmonic at twice the stimulation frequency (2F). It has been proposed that FERG 1F and 2F components reflect the activity of, at least in part, different generators located in distal and proximal retina, respectively. ^{20 21 22} Results in monkeys provide evidence that different mechanisms, probably in parallel, may contribute to 1F and 2F components, depending on temporal frequency. ^{23 24 25}  

The aim of the present study was to evaluate whether aging and ARM affect the temporal frequency response (TFR) functions of the 1F and 2F components of macular FERG. Although it is already documented that the FERG can be altered in early ARM eyes ^{26 27} and that signal abnormalities are associated with the severity of ARM lesions, ^{18 19 28} none of the previous studies analyzed the FERG TFR. Results showed significant TFR changes associated with normal aging and ARM and supported the presence of a postsynaptic dysfunction underlying these changes. 

Subjects and Methods

Subjects

Thirteen young healthy subjects (six males, seven females; age range, 14–29 years; mean age, 22 years) and nine old healthy subjects (four men, five women; age range, 55–80 years; mean age, 66 years) provided normative FERG data. None of the older control subjects showed ARM lesions during fundus examination performed by direct ophthalmoscopy and retinal biomicroscopy. 

Eighteen patients were evaluated (eight men, 10 women; age range, 55–80 years; mean age, 65 years). Patients had a Snellen visual acuity ≥0.4 and a clinical diagnosis of ARM, established by direct and indirect ophthalmoscopy and retinal biomicroscopy in accordance with the classification of Jackson et al. ¹³ Each subject or patient underwent complete general and ophthalmologic examination. All showed clear optical media, normal color vision (Farnsworth Panel D-15), and no concomitant ophthalmic or systemic disease (e.g., ocular hypertension or glaucoma, diabetes, systemic hypertension). None of the subjects or patients was taking medication known to affect macular function Clinical diagnosis of ARM was established when any of the following primary lesions in the macular area were identified: soft, distinct or indistinct drusen; areas of hyperpigmentation associated with drusen; or areas of hypopigmentation of the RPE, without any visibility of choroidal vessels, associated with drusen. Funduscopic evidence of more than one large (size 63 μm) drusen in the foveal region or focal hyperpigmentation of RPE, assigned to ARM stage 2 of Jackson et al., ¹³ was found in all ARM study eyes. All patients and healthy control subjects had central and steady fixation. 

Informed consent to participate in the study was obtained from all subjects and patients; the research followed the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board. 

Apparatus and Procedure

FERGs were recorded from the macular region at different temporal frequencies (TFs) superimposed on an equiluminous steady adapting background, used to minimize stray-light modulation, according to a published protocol. ²⁹ Briefly, the stimulus consisted of a circular flickering uniform field generated by an array of eight red LEDs (λ maximum, 660 nm; mean luminance, 93 cd/m²) sinusoidally driven by a custom-made digital frequency generator ³⁰ and presented on the rear of a Ganzfeld bowl (white-adapting background) illuminated at the same mean luminance as the stimulus. A diffusing filter placed in front of the LED array made it appear as a circle of uniform red light. The stimulus, presented such that the white light was not returned from the diffuser in front of the LEDs, had a diameter subtending the central 18° at a viewing distance of 33 cm. FERGs were recorded in response to the sinusoidal luminance modulation (95% modulation depth) of the uniform field presented in the macular region. Stimulus TF was varied between 3.7 and 52 Hz (3.7, 5.7, 8, 9, 10.3, 14, 21, 32, 41, and 52 Hz) for the study protocol. A small central fixation mark allowed the stimulus field to be centered on the fovea. Patients and healthy subjects fixated monocularly at a small fixation mark (0.25°) placed in the center of the stimulating field. In each patient, eyelid opening and blink frequency during the recording session were judged to be normal by an observer who monitored patient fixation. In all subjects and patients, pupils were pharmacologically (1% tropicamide and 2.5% phenylephrine hydrochloride) dilated to a diameter of 8 mm or greater, and all subjects started the FERG testing protocol after a preadaptation period of 20 minutes to the stimulus mean illuminance, to avoid gradual changes in light adaptation during the recording session. No significant differences across the different study groups were found in the mean pupillary diameter during FERG recordings (mean ± SD; healthy subject young, 9 ± 0.2; healthy subject old, 9 ± 0.25; ARM patient, 9 ± 0.2). 

Three healthy young observers participated in additional experimental sessions in which FERG TF functions were evaluated under different mean illuminance levels (stimulus and background) or modulation depth. These special protocols were applied to evaluate the effects of the optical factors related to eye senescence, which may reduce stimulus retinal illuminance or the effective retinal contrast. For the retinal illuminance protocol, the FERG TF response function was recorded at a mean retinal illuminance of 3.6 and 3.2 log photopic trolands, with stimulus and background viewed through a maximally dilated pupil. For the modulation depth protocol, the FERG TF response function was recorded at the modulation depth of 93.8%, 63.6%, and 33.1%, whereas mean stimulus and background illuminance were kept at 3.6 log photopic trolands. 

FERGs were recorded by an Ag-AgCl electrode taped on the skin over the lower eyelid. A similar electrode, placed over the eyelid of the contralateral patched eye, was used as reference (interocular recording). ³¹ FERG signals were amplified (100,000-fold), bandpass filtered between 1 and 100 Hz (6 dB/oct), and averaged (12-bit resolution, 2-kHz sampling rate, 200–600 repetitions in 2–6 blocks). Averaging time (i.e., recording epoch duration) was varied according to the stimulus period. Single sweeps exceeding a threshold voltage (5 μV) were rejected to minimize noise coming from blinks or eye movements. Off-line discrete Fourier analysis was performed to isolate response fundamental harmonic (1F) at the same TF as the stimulus and second harmonic (2F) at twice the stimulation frequency, whose peak-to-peak amplitudes (in μV) and phases (in degrees) were measured. ³² Averaging and Fourier analysis were also performed on signals sampled asynchronously at 1.1 times the TF of the stimulus, to give an estimate of the background noise at 1F and 2F. In all records, the noise amplitudes ranged from 0.05 to 0.12 μV for either 1F or 2F. Under the present experimental conditions, both FERG components recorded from control subjects and ARM patients were above the noise level (1F signal-to-noise ratio [SNR] range, 5–27; 2F SNR range, 6–32) and were sufficiently reliable (variation coefficient for amplitude was typically <25%, and phase SD was <30°). 

Data Analysis

Data from the right eyes of healthy subjects and patients were included in the analysis. FERG amplitudes were compared across the two groups of control subjects and of ARM patients by two-way analysis of variance (ANOVA) in which the group (healthy young, healthy old, ARM) was the between-subjects factor and TF was the within-subjects factor. The ratio of FERG 1F amplitude to 41 and 10 Hz stimuli (41/10 Hz amplitude ratio) was compared across the study groups by one-way ANOVA. Log transformation was applied to better approximate normal distribution. The post hoc Tukey honest significance difference [HSD] test was used for multiple between-group comparisons. FERG phase data were analyzed by circular statistics, ³³ with several preliminary assumptions (see Iarossi et al. ³⁴ ) about phase lag as a function of TF, and were compared across groups by two-way ANOVA similar to that applied for amplitude data. FERG results from ARM patients were correlated with the corresponding ages and visual acuities of patients by Pearson correlation analysis. For this correlation analysis, only the responses at 41 Hz and 8 Hz (for the 1F and the 2F component, respectively) were considered because previous studies ^{18 21} have shown that FERG components at these frequencies are the most sensitive to clinical parameter changes in macular diseases. In all the analyses, P < 0.05 was considered statistically significant. 

Results

Representative examples of normal FERG waveforms have been published. ^{18 19 28} Figure 1Ashows the group averaged (± SE) 1F amplitudes and phases as a function of TF for young (open circles) and old (solid circles) healthy subjects. It can be noted that in young healthy subjects, the average TFR of 1F amplitude has a maximum at 41 Hz, a secondary peak at 3.7 Hz, a minimum at 10 Hz, and a high frequency cutoff at approximately 50 Hz. The TFR of the old differed significantly from that of the young controls (two-way ANOVA, interaction effect of group by TF, F = 3.5; df = 9,180; P < 0.01), showing a broad maximum at 14 Hz. Amplitude losses (1F) in old compared with young controls were found only at high TF (32–52 Hz). There was no significant effect of group (F = 2.5; df = 1,20; P = NS), whereas the effect of TF was highly significant (F = 11.8; df = 9,180; P < 0.001). The difference in the shape of the 1F TFR between young and old healthy subjects can be further appreciated in the Figure 1Ainset, where the mean 1F amplitude change recorded in old healthy compared with young healthy subjects is expressed as percentage change as a function of TF. Error bars in the plot indicate 95% confidence intervals of the means. The mean FERG 1F phase lagged progressively with TF in both groups of the healthy control population. The shape of the average FERG phase function did not differ significantly between young and old healthy controls. In Figure 1B , the mean (±SE) 1F amplitudes of ARM patients (cross center circles) are compared at the different TFs with those of the old, age-matched controls (solid circles). It can be seen that both TFR functions are similar in shape. ARM patient function shows only a uniform scaling in mean amplitude (on average, –0.2 log units) compared with normal function. Two-way ANOVA indicated a significant effect of group (F = 5.8; df = 1,25; P < 0.05) and TF (F = 9.8; df = 9,225; P < 0.001) but no significant interaction of group by TF (F = 0.9; df = 9,225; P = NS). The mean percentage of 1F amplitude change in ARM patients, compared with the age-matched controls (Fig. 2B , inset), revealed that mean changes tended to be uniform at all but one TF (9 Hz). 

Figure 2ABcompares the group averaged (± SE) FERG 2F amplitudes and phases as a function of TF for the healthy young (open squares) and the healthy old (solid squares) subjects (Fig. 2A)and for the healthy old subjects and ARM patients (cross squares; Fig. 2B ). In all the three examined groups, the 2F TFR shows a broad maximum at medium TFs (8–10 Hz). No significant differences were found between young and old healthy eyes (two-way ANOVA; interaction effect of group by TF, F = 0.92; df = 9,180; P = NS), though a small increase in mean amplitude was observed for old subjects at the two lowest TFs. The difference in shape of the 2F TFR between young and old healthy subjects can be better appreciated in Figure 2Ainset, where the mean 2F amplitude changes of old relative to young observers are reported as percentages as a function of TF. Error bars in the plot indicate 95% confidence intervals of the means. Compared with the age-matched controls, ARM eyes showed 2F losses in mean amplitude at low-medium and high TFs. ANOVA indicated a significant effect of group (F = 11.91; df = 1,25; P < 0.01). The effect of TF was also highly significant (F = 28.6; df = 9,225; P < 0.001), whereas the interaction of group by TF did not reach statistical significance (F = 1.93; df = 9,225; P = NS). The Figure 2Binset shows the mean percentage of 2F amplitude in ARM patients, compared with age-matched controls, as a function of TF. Losses tended to be nonselective and to involve low and high TF regions. The FERG 2F phase did not differ between subjects and patients and showed two different slopes for separate frequency ranges (3.7–10 Hz and 14–52 Hz). 

In Figure 3 , the log ratio between the 1F amplitudes recorded at 10 Hz and 41 Hz is plotted as a function of age (years) for each examined group. In old healthy subjects and ARM patients, the ratio between the 1F amplitudes recorded at high (41 Hz) and middle (10 Hz) TFs tended to decrease compared with healthy young subjects. Mean log ratio changed significantly across the groups (ANOVA, effect of group; F = 5.7; df = 2,37; P < 0.01). Multiple comparisons indicated that the mean ratio was significantly reduced in ARM patients compared with young controls (Tukey HSD, P < 0.05). The other comparisons (i.e., healthy young subjects vs. old subjects, healthy old subjects vs. ARM patients) approached but did not reach statistical significance. 

In Figure 4 , the FERG 2F amplitudes recorded at 8 Hz in individual ARM eyes are plotted as a function of corresponding visual acuity (Snellen equivalent) values. There was a significant positive correlation between 8 Hz 2F amplitude and visual acuity of individual patients, even though most of our patients had only moderate visual acuity losses, and several patients with acuities of 1.0 had low amplitude equivalent to that in patients with poorer acuity. FERG 1F and 2F amplitudes recorded at the other TF (41 Hz) were not correlated with visual acuity, indicating that these responses might have been less sensitive to visual acuity changes in ARM. 

Figure 5Ashows the effect of reducing the retinal illuminance of the stimulus (by 0.4 log units), mimicking a reduction in photoreceptor quantum catching ability, on the FERG 1F TF spectrum in three healthy young subjects. The shape of the FERG 1F function was substantially unaffected by reducing mean illuminance, indicating that the differences in the FERG 1F TFR function found between old and young healthy subjects cannot be accounted for by a reduction in photoreceptor quantum catching ability. Figure 5Bshows the effect of reducing the modulation depth of the stimulus (93.8%, 63.6%, and 33.1%), mimicking contrast attenuation resulting from increased optical lens density, on the FERG 1F TF spectrum in three healthy young subjects. The shape of the mean FERG 1F TFR was similar at the three contrast levels, indicating that the differences in TFR between young and old healthy subjects were not likely to have resulted from age-related lens density changes. 

Discussion

Temporal Dynamics of the Macular Cone System in Normal Aging and ARM Retinas

The major finding of this study was that the TFR of the macular FERG 1F component was significantly altered in healthy aging eyes and in ARM eyes. In healthy aging eyes but not young eyes, 1F amplitude losses were found only at high TFs (32–52 Hz) whereas at medium frequencies (10–20 Hz) there was even an amplification of responses, in contrast to the results in healthy young eyes. This resulted in a significant average change in the shape of the 1F TFR, displaying in old eyes a broad maximum in the TF region at which a dip was present in young eyes. These changes were not mimicked by artifactually reducing stimulus retinal illuminance or modulation depth, suggesting that neural rather than optical factors should contribute to the observed variations in the 1F TFR. Although the shape of the 2F TFR did not differ significantly between young and old observers, a small amplitude enhancement was observed at the lowest TFs, consistent with 1F amplitude changes. The FERG TFR functions of ARM eyes were comparable to those of old eyes, but attenuated 1F and 2F amplitude values, with a rather uniform scaling, were observed. These results indicated a change in the 1F temporal response dynamics in normal aging eyes compared with young eyes and an amplitude loss that was frequency independent compared with age-matched controls in ARM eyes. 

Interpreting the Effects of Aging and Disease: Postsynaptic Dysfunction of the Retinal Cone System

In the past, classic electrophysiology studies indicated that the FERG 1F shows functional properties of photoreceptors, ^{35 36 37} has a peak current source density in the distal retina, ³⁸ is comparable to the local receptor potential, ³⁹ and may contain a contribution from bipolar cells and glial cells. ³⁶ The FERG 2F has several retinal sources, ³⁸ with at least one in the inner retina ²³ and another in the outer retina at about the same level as the 1F source. More recently, experimental studies in monkeys ²⁵ have begun to elucidate in greater detail the retinal generators of cone-mediated ERGs evoked by sinusoidal stimuli. Pharmacologic blockade of synaptic transmission from photoreceptors to ON and OFF cone bipolar cells demonstrated that the frequency response function of the sinusoidal flicker ERG is shaped mainly by the activity of ON and OFF bipolars, with a minor contribution of cone photoreceptors only at low TFs (<5 Hz). Postsynaptic blockade involving the ON channel or the OFF channel, or both, resulted in a dramatic change in the shape of the TFR, with a selective or dominant amplitude loss at around the maximum of the normal frequency function (30–40 Hz) and at higher frequencies. Vector modeling analysis of the 1F flicker responses also predicted that postsynaptic dysfunction should result in a reduction in the high to mid TF amplitude ratio (namely, the amplitude ratio between 32- and 8-Hz amplitudes in the monkey experiments, or the 41/10 Hz ratio in the present experiment) and a selective phase delay for the high-frequency (30–50 Hz) responses. The ratio can, therefore, be taken as an index of postsynaptic dysfunction of the retinal cone system, and these changes in the photopic flicker ERG could reflect specifically a postsynaptic abnormality. 

On the basis of these experimental findings, the major implication of the present results is that normal aging may affect temporal response properties at the level of the photoreceptor/bipolar cell complex (as reflected in the 1F changes), with a minor involvement of more proximal retinal layers (as signaled by the 2F changes). ARM-related changes consist only in a response amplitude loss, with a TFR similar to that of healthy old controls for 1F or 2F. Although a direct comparison between ARM eyes and healthy young eyes was not planned in this study, it is clearly apparent, from inspection of Figure 1 , that the TFR changes found in normal aging, compared with young, eyes are also maintained in ARM eyes, suggesting that altered temporal dynamics are indeed present in this group. The present findings, showing a decrease of the 41/10 Hz amplitude ratio in ARM eyes, further support the hypothesis that postsynaptic dysfunction, in addition to photoreceptoral losses, underlies ARM-related FERG changes. 

We did not find significant changes in response phase for the 1F and 2F components associated with aging or ARM. One possible reason is that the phase changes were too small to be detected by the present technique. Another possibility is that significant phase delays may be detectable only by reducing the stimulus modulation depth in ARM eyes. ¹⁸ Given that the TFR function of the FERG was recorded at the maximum possible modulation depth (93%), the present technique might have missed phase changes associated with amplitude TFR changes. Jackson et al., ⁵ studying age-related changes in the multifocal electroretinogram (MERG), have shown that the amplitudes of the first negative and positive peaks of the first-order MERG waveform were reduced with age. In addition, they observed that older adults had reduced amplitudes in the second-order components of the MERG and shifts in the latencies of the first-order component, suggesting temporal disturbances in the MERG of older adults. Jackson et al. ⁵ proposed that an anatomic dropout of photoreceptors and postreceptoral cells in the aged eye or a slowed temporal adaptation in the aged retina may contribute to aging changes in the MERG. 

The aging- and disease-related changes observed in the temporal dynamics of the FERG 1F may have an histopathologic counterpart in morphologic studies, ³ showing that in ARM eyes the synaptic architecture of the outer retina is altered and the synaptic transmission in photoreceptors overlying drusen might be compromised. Drusen-associated changes in photoreceptors were demonstrated by immunostaining with antibodies against synaptic vesicle proteins. Immunolabeling with antibodies against synaptophysin was normally restricted to rod and cone axon terminals within the outer plexiform layer. When drusen were present, however, synaptophysin immunolabeling was often absent, and the number of synaptophysin-immunopositive rod and cone axon terminals was significantly reduced. Drusen-associated abnormalities appeared to be limited to photoreceptor cells because antibodies that label horizontal, bipolar, amacrine, and ganglion cells did not reveal significant alterations in the morphology of these second- and third-order neurons. 

Clinical Implications

Our results indicate that the analysis of the FERG TFR in normal aging and ARM may be more informative than protocols exploring only one or two TFs ^{18 27 28} about cone system temporal responsiveness. Electrophysiological differences between normal aging and ARM eyes found in the present study could be of clinical interest for predicting the development of visual loss in an individual patient, similar to that observed by Mayer et al., ⁴⁰ for flicker sensitivity. However, a substantial overlap between normal aging and ARM eyes was found in this study for the 1F 41/10 Hz amplitude ratio, suggesting that this ratio is not really likely to be a good prognostic indicator. Similarly, although the FERG 2F recorded at 8 Hz in ARM patients showed a significant positive correlation with visual acuity, many ARM eyes with good acuity had low 2F amplitudes, such as those with poorer acuity. Longitudinal studies are needed to support or reject the prognostic value of these measurements, alone or in combination with psychophysical ¹⁶ flicker sensitivity, for predicting visual loss in ARM patients. 

In conclusion, the present results indicate that the FERG TFR undergoes specific age-related changes that indicate altered response dynamics in the outer and inner retina in the macular cone system. In addition, the study showed that the FERG TFR may detect early abnormalities consisting of response amplitude loss, whose predictive significance can be clarified only by longitudinal studies of patients with ARM. 

 Supported by MIUR–Fondi di Ateneo ex 60% 2005–2006.

 Submitted for publication March 13, 2007; revised April 26, 2007; accepted June 15, 2007.

 Disclosure: B. Falsini, None; L. Ziccardi, None; G. Stifano, None; G. Iarossi, None; E. Merendino, None; A.M. Minnella, None; A. Fadda, None; E. Balestrazzi, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

(A) Group-averaged 1F amplitudes and phases (±SEM) as a function of TF for control (healthy young) subjects and healthy old subjects. Inset: percentage of mean 1F amplitude loss for old versus young healthy subjects as a function of TF. Error bars indicate 95% confidence intervals of the means. (B) Group-averaged 1F amplitudes and phases (±SEM) as a function of TF for control subjects (healthy old) and ARM patients. Inset: the percentage of mean 1F amplitude loss for ARM patients versus old healthy subjects as a function of TF. Error bars indicate 95% confidence intervals of the means.

Figure 2.

 

(A) Group-averaged 2F amplitudes and phases (±SEM) as a function of TF for control subjects (healthy young) and healthy old subjects. Inset: percentage of mean 2F amplitude loss for old versus young healthy subjects as a function of TF. Error bars indicate 95% confidence intervals of the means. (B) Group-averaged 2F amplitudes and phases (±SEM) as a function of TF for control subjects (healthy old) and ARM patients. Inset: percentage of mean 2F amplitude loss for ARM patients versus old healthy subjects as a function of TF. Error bars indicate 95% confidence intervals of the means.

Figure 3.

 

The logarithm of the ratio between the 1F amplitudes recorded at 41 Hz and 10 Hz plotted as a function of age (years) for each examined group.

Figure 4.

 

FERG 2F amplitudes of individual ARM eyes plotted as a function of corresponding visual acuity (Snellen equivalent) values.

Figure 5.

 

Effect of reducing (A) mean retinal illuminance of the stimulus (by 0.4 log units) and (B) contrast level (three different modulation depths) on the FERG 1F TF spectrum.