The age deficit on photopic counterphase flicker: Contrast, spatial frequency, and luminance effects

Abstract This study evaluated the contribution of reduced contrast sensitivity and retinal illuminance to the age-related deficit on the temporal resolution of suprathreshold spatial stimuli. The discrimination of counterphase flicker was measured in optimally refracted young and elderly observers for sinusoidal gratings of three spatial frequencies (1, 4, and 8 cycles per degree) at three contrast levels (0.11, 0.33, and 0.66). Age deficits in flicker discrimination at the two higher contrast levels and at the two lower spatial frequencies were unrelated to observer contrast sensitivity. Flicker discrimination of young observers who carried out the task through .5 ND filters to simulate a two-thirds reduction of retinal illuminance in the older eye, was similar to that of the elderly observers. An age-related reduction in retinal luminance appears to be a major determinant of the age-related spatiotemporal deficit at suprathreshold contrast levels, although neural factors may also be involved.

Considerable evidence indicates that the senescent visual system is compromised in its ability to track temporal change and to resolve spatial detail in temporally modulated target stimuli (e.g., Kline, 1991; Kline & Scialfa, 1996; Owsley & Sloane, 1990; Spear, 1993). Although optical and sensorineural factors both appear to contribute to this loss, there is little consensus regarding their relative importance. Nor is it clear if the factors that limit spatiotemporal resolution at threshold contrast levels are the same as those that do so for suprathreshold stimuli. The goals of this study were to 1. determine the effects of target contrast and spatial frequency as a function of observer age and contrast sensitivity on the discrimination of counterphase flicker in suprathreshold sinusoidal gratings, and 2. estimate the contribution of reduced retinal illuminance to agerelated spatiotemporal deficits by comparing the thresholds of old observers at high luminance with those of young observers at high and low target luminance.

Studies of the spatial contrast sensitivity function (CSF) have typically reported a relative sparing of sensitivity (the reciprocal of contrast threshold) at low spatial frequencies (i.e., 1 c/deg and below) and an increasing age deficit at intermediate and high spatial frequencies (i.e., 2 c/deg and higher) (e.g., Burton, Owsley, & Sloane, 1993; Derefeldt, Lennerstrand, & Lundh, 1979; Elliott, 1987; Elliott & Whitaker, 1992; Elliott, Whitaker, & MacVeigh, 1990; Higgins, Jaffe, Caruso, & de Monasterio, 1988; Kline, Schieber, Abusamra, & Coyne, 1983; Owsley, Sekuler, & Siemsen, 1983; Scialfa, Tyrrell, Garvey, Deering, Leibowitz, & Goebel, 1988). Several studies have also observed a deficit at low spatial frequencies (e.g., Korth, Horn, Stork, & Jonas, 1989; Nameda, Kawara, & Ohzu, 1989; Ross, Clarke, & Bron, 1985; Sloane, Owsley, & Alvarez, 1988). Crassini, Brown, and Bowman (1988) have shown that the pattern of increasing age-related loss at higher spatial frequencies in the central retina also holds for the peripheral retina (10 degrees temporally). Although increased light scatter and attenuated retinal illuminance contribute to this loss (Guriao, Gonzalez, Redondo, Geraghty, Norrby, & Artal, 1999; Scialfa, Kline, & Wood, 2002), studies that have systematically varied optical factors (e.g., Elliott et al., 1990; Sloane, Owsley, & Alvarez, 1988), manipulated visual noise (Bennett, Sekuler, & Ozin, 1999; Pardhan, Gilchrist, & Elliott, 1996), or bypassed age-related optical effects using laser interferometry (Burton et al., 1993; Elliott, 1987), indicate that sensorineural factors also play a role. A recent study by Schefrin, Tregear, Harvey, and Werner (1999) reached a similar conclusion regarding aging effects on the scotopic CSF for the nasal retinal field. Significant age deficits at and below 1.2 c/deg were attributed by Schefrin et al. to a decline in the mechanisms composing the magnocellular pathway.

Several studies have examined the temporal resolving characteristics of the aging visual system using the temporal contrast sensitivity function (TCSF), the temporal analog to the spatial CSF. The TCSF is a measurement of the contrast level needed to detect flicker as a function of its temporal rate. At its upper limit - when the contrast modulation threshold reaches 100 per cent - the TCSF is equivalent to the critical flicker frequency (CFF) task. Most studies of aging effects on the TCSF report pronounced deficits at intermediate and higher temporal flicker rates, but as yet, there is little consensus regarding the underlying causal mechanisms.

When Wright and Drasdo (1985) determined TCSFs for a yellow (550 nm) LED light source, they observed modest age-related deficits at lower temporal frequencies (3.3 and 10 Hz) and a pronounced deficit at 30 Hz. Noting that reduced luminance exerted a greater effect at high than low spatial and temporal frequencies, they concluded that the older observers' sensitivity loss was due to the attenuation of retinal illuminance that results from the older eye's smaller pupil (i.e., senile miosis) rather than to neural factors. Using a long-wavelength red (660 nm) LED light source to reduce absorption by the ocular media and a high luminance level (120 cd/ml) to minimize pupil effects, Mayer, Kim, Svingos, and Glucs (1988) determined TCSFs for flicker rates from 1.8 to 50 Hz for young (18 to 42 years) and old (65 to 86 years) observers. They found that older observers' sensitivity to flicker was most reduced at high flicker rates (10 to 45 Hz). Even after the stimuli were equated for age differences in retinal luminance, an age-related deficit remained. The deficit was manifest primarily as a downward rather than horizontal displacement of the TCSF, leading the authors to conclude that it reflected an age-related loss of contrast sensitivity for temporally modulated targets and not visual slowing. When Kim and Mayer (1994) subsequently compared the foveal flicker sensitivities of observers age 18 to 77, they observed a decline past age 44 that was again more consistent with a decline in sensitivity than a loss of temporal resolution. Zhang and Sturr (1995) also concluded that the age differences that they observed on temporal summation as a function of luminance reflected a decline in sensitivity (i.e., amplitude) rather than to slowing of the senescent visual system. Tyler (1989), however, arrived at just the opposite conclusion in large-scale study of the TCSFs of 1,000 observers aged 5 to 75 years. Using a high-luminance (400 cd/ml) 660-nm light source, he found a leftward shift of the TCSF with age and concluded that, beyond 16 years of age, the visual system slowed by about 20% over a 60-year span. Kuyk and Wesson (1991) also found evidence for a neurally mediated age decline on the TCSF that increased with temporal frequency. Their data, however, did not reveal a leftward age shift of the TCSF toward lower frequencies as suggested by the general slowing hypothesis of Tyler (1989). They also argued that a diffuse loss of receptors or neurons as suggested by Mayer, Kim, and their co-workers (Kim & Mayer, 1994; Mayer, Kim, Svingos, & Glucs, 1988) would produce a uniform loss of sensitivity across the temporal frequency range rather than the frequency-- specific deficit that has been observed in most studies.

Age-related deficits on the TCSF appear to vary with retinal location. When Casson, Johnson, and Nelson-- Quigg (1993) compared the flicker sensitivity of young, middle-aged and old observers at three temporal frequencies (2, 8, and 16 Hz) across the central 54 degrees of the visual field, they found that age-related losses were greater in the peripheral retina than on the fovea, particularly so at 16 Hz. Older observers also showed greater asymmetry between the superior and inferior visual field at 16 Hz than did young observers. The authors noted that neither the rate-specific decrements in flicker sensitivity, nor the interaction between age and eccentricity that were observed could be attributed to changes in retinal illuminance.

Age-related temporal losses at contrast threshold have been shown to vary with the spatial characteristics of the target stimuli, with age differences in contrast sensitivity thresholds for flickered or moving gratings being most pronounced at higher spatial frequencies (e.g., Elliott et al., 1990; Nameda et al., 1989; Royer & Gilmore, 1985; Scialfa, Garvey, Tyrrell, & Leibowitz, 1992; Tulunay-Keesey, Ver Hoeve, & Terkla-McGrane, 1988; Wright & Drasdo, 1985). Royer and Gilmore (1985) measured contrast sensitivity thresholds for sinusoidal gratings (0.5 to 12.0 c/deg) that were counterphase flickered at 0.3 and 6.3 Hz. Finding that their older observers showed greater losses at higher spatial frequencies, especially so at the highest flicker rate, they postulated an increase in the time constant of the different visual channels that is proportional to their initial sensitivity. When Scialfa, Garvey, Tyrrell, and Leibowitz, (1992) compared the sensitivity of young and old adults for gratings that traveled along a circular path at 5, 10, or 15 deg/s they found a similar result. Older observers exhibited diminished sensitivity at lower target velocities than did the young, and age differences in velocity effects were greater at intermediate and higher spatial frequencies. Nameda, Kawara, and Ohzu (1989) measured contrast sensitivity to static and drifting gratings directly on the retina of young and old observers using interference fringes. Motion enhancement was found in older observers at all velocities, but in young observers, only at the fastest velocity tested (3.5 deg/s). The authors suggested that changes before age 40 reflect a sensitivity loss in the high spatial frequency channels and a shift in the low spatial frequency channels to lower temporal frequencies and lower sensitivities. They hypothesized that both systems experience a temporal decline after age 40. Elliott, Whitaker, and MacVeigh (1990) measured the contrast thresholds of young and old subjects for counterphase-- modulated gratings at combinations of three spatial frequencies (1, 4, and 8 cycles per degree, c/deg) and three temporal frequencies (0, 4, and 16 Hz). They found an age deficit in spatiotemporal contrast sensitivity at all but the lowest combinations of spatial and temporal frequency. Simulations of senile miosis using a pupil-constricting drug and the age-related increase in lenticular light absorption using a neutral density filter in young observers could not account for the results. This led the authors to propose a sensitivity loss in the sustained (i.e., parvo) visual pathway.

If the age-related deficit is due in part or whole to diminished sensitivity for temporally modulated contrast and not to visual slowing, it might be expected to vary inversely with both target contrast and the observer's contrast sensitivity. Previous research findings are inconsistent on this matter, however. In support of the diminished sensitivity hypothesis, Tulunay-Keesey et al., (1988) reported contrast sensitivity losses at high spatial and temporal frequencies among older observers, but no deficit on a suprathreshold contrast matching task on which the stimuli were either static or counter modulated at 5 Hz. They suggested that older adults may experience less impairment in suprathreshold tasks than threshold tasks because of functional compensation, based on either a different gain mechanism or an independent high contrast neural system that is relatively unaffected by aging. Little support for this suggestion, however, has been seen in the several studies that have shown oscillatory displacement thresholds (ODTs) for high contrast stimuli to be highly sensitive to aging effects (e.g., Barrett, Davison, & Eustace, 1994; Buckingham, Whitaker, & Banford, 1987; Elliott, Whitaker, & Thompson, 1989; Hiller & Kline, 2001; Kline, Culham, Bartel, & Lynk, 2001).

When Buckingham, Whitaker, and Banford (1987) measured oscillatory displacement thresholds (ODTs) for 2 c/deg sinusoidal gratings of 65% contrast, age-- related losses were observed at temporal frequencies ranging from 1 to 20 Hz. Elliott, Whitaker, and Thompson (1989) determined ODTs as a function of age for high-contrast targets oscillating at 2 Hz in the presence of a stable reference line. The age-related increase in ODTs could not be accounted for by attenuated retinal illuminance, a finding that led the investigators to conclude that neural factors were responsible. When Whitaker, Elliott, and MacVeigh (1992) observed age-related threshold increases on oscillatory displacement at 2 Hz, but no aging effect on the static vernier thresholds of the same observers, they too implicated neural factors in the loss. Barrett, Davison, and Eustace (1994) found that ODTs for high-contrast targets oscillating at 2 Hz were relatively unaffected by optical degradation but rose steadily with increasing age. More recently, Hiller and Kline (2001) reported that the elevation in older observers' ODTs, which varied directly with oscillation rate, was also highly correlated with contrast sensitivity. Interestingly, however, the age deficit did not vary with target contrast. Kline, Culham, Bartel, and Lynk (2001) found that thresholds for discrimination of static vernier displacement were almost identical for young and old observers, but a marked age-related deficit unrelated to target contrast (8 or 64%) emerged on the same task when target oscillation was increased. No relationships were seen, however, between static or dynamic vernier thresholds and contrast sensitivity for either age group.

There is recent neurophysiological evidence of deterioration in the visual processing speed of suprathreshold stimuli at the level of the visual cortex. Mendelson and Wells (2002) recorded the responses of cells in areas 17 and 18 of young and aged rats to both moving light bars and flashing lights. They found that cortical cells of the young rats had a higher preferred speed for the moving bars, and that they were able to entrain to higher frequencies of flashing lights than were old rats. They also reported that the simple cells of young rats showed lower light flicker thresholds as well as a preference for slower movement speeds compared to complex and hypercomplex cells. In the aged animals, however, no differences between the three cell types were seen for preferred speed or flicker thresholds, leading the authors to suggest that aging effects on temporal processing speed might be greatest for complex and hypercomplex cells.

Prior studies have implicated a variety of contributors to the age-related loss of visual temporal resolution, including reduced retinal illuminance, diminished sensitivity to contrast, and a fundamental slowing of the visual system. The relative importance of these factors in aging effects on the resolution of temporally modulated stimuli at suprathreshold contrast levels is unclear due to the absence of prior research to examine systematically the effects of target contrast, spatial frequency, and luminance among the same young and old observers. Thus, the present study compared the counterphase grating flicker thresholds of old observers at high luminance with those of young observers at high and low luminance for low, intermediate, and high contrast targets of varied spatial frequency. Flicker thresholds were expected to vary inversely with spatial frequency and directly with contrast for both young and old observers. Consistent with a hypothesized decline in the temporal resolving properties of the visual system, it was expected that the flicker thresholds of the young observers at both low and high luminance would exceed those of the older observers at all contrast levels. Greater age differences, however, were anticipated for gratings of low contrast and high spatial frequency (i.e., an age by contrast by spatial frequency interaction).

Method

PARTICIPANTS

Three groups, each composed of 12 community-resident volunteers (6 men and 6 women), participated in the study: an elderly group (mean age 66.2 years; range 55-77) and two young groups (overall mean age 20.9 years; range 18-28). One of the two young groups as well as the elderly group viewed the display at full luminance (FL); the second young group carried out the task through 0.5 neutral density (ND) filters to simulate the two-thirds reduction in retinal illuminance estimated to occur from age 20 to 60 years (Weale, 1961). The age of the young FL (M = 21.4, SD = 1.83) and ND groups (M = 20.3, SD = 2.81) did not differ, t(22) = 1.120, p = 0.275.

All participants had at least high-school education. The mean level of formal education was higher for the old (M = 16.2 years., SD = 4.11) than the young participants (M = 14.2 years, SD = 1.56), t(34) = 2.123, p < .05. Scores on the vocabulary subtest of the Wechsler Adult Intelligence Scale - Revised (WAIS-R; Wechsler, 1981), used as a control measure for cohort comparability, were very similar for the old participants (mean 58.3, SD = 10.14) and combined young groups (mean 56.0, SD = 7.39) and t(34) = .760, p = .453. A comparison of the two young groups indicated that they were comparable in regard to both education, FL M = 14.5 (SD = 1.44) and ND M = 13.7 (SD = 1.64), t(22) = 1.190, p = .247, and mean WAIS-R vocabulary scores, FL M = 58.6, (SD = 6.78), and ND M = 53.4 (SD = 7.33), /(22) = 1.793, p = .087. The participants in both age groups were in good self-reported general and visual health.

After correction for the 83-cm test distance, the right (tested) eye acuity of all participants was 1.20 minarc (equivalent to Snellen 20/24) or better. The best-corrected acuity of the old observers (Snellen equivalent = 20/18; M = 0.93 minarc, SD = 0.151), however, was somewhat worse than that of their younger counterparts (Snellen equivalent = 20/15; M = 0.77 minarc, SD = 0.081), t(34) = 4.012, p <.001. The acuities of the young FL and ND groups were very similar, -0.77 and 0.78 minarc, respectively, t(22) = 0.123, p = .903. Consistent with the age-related recession of the near point (i.e., presbyopia), the young and old observers differed in the correction required for best acuity. Twenty of the young observers were emmetropic for the test distance, three myopic (mean add -0.75 D; SD = 0.43) and one hyperopic (add = +0.25 D); 11 of the older observers were presbyopic (mean add = +1.14 D, SD = 0.56) and one was myopic (mean add = 0.50 D).

APPARATUS AND MATERIALS

Acuity was measured to the nearest two Snellen equivalent lines using a custom Landolt C chart presented at a luminance of 200 cd/mr. A Canon R-22 Autorefractor and American Optical Master Phoroptor were used to carry out the retractions and an R. H. Burton Model TLS lens set was used to optimize right-eye acuity for the task distance of 83 cm. So that contrast sensitivity could be determined for the same spatial frequencies as were used in the counterphase test gratings (i.e., 1, 4, and 8 c/deg), a Vistech contrast sensitivity chart (Model VCTS 6500 Form A) was used at a nonstandard viewing distance (203 cm); its luminance was 130 cd/mI. The left eye was patched during testing.

To achieve the millisecond resolution required for the flicker task, stimuli were presented using a Gerbrands Model T-3B-1 Harvard three-field tachistoscope and 61159 Logic Interface. The stimuli were pairs of vertical sinusoidal gray-scale gratings, each pair member 180 deg out of phase with respect to the other. Nine grating-pair combinations were tested based on three spatial frequencies (1, 4, and 8 c/deg) at three Michelson contrast levels (0.11, 0.33, and 0.66). The phase difference between the gratings was established individually for each observer using a gimbaled half-silvered mirror to align a set of red vertical vernier vertical target lines (0.7 deg x 0.1 deg) at the top centre of each tachistoscopic field. Stimuli were viewed monocularly with the right eye through a circular mask 3.5 deg in diameter. A .5 ND filter was placed in the mask when testing the young ND group. Photocells mounted in each of the tachistoscopic fields, interfaced with digital multimeters and calibrated against a Minolta Model LS-110 spot photometer, were used to maintain the space-averaged luminance of the gratings at a constant 30.0 cd/m^sup 2^. To minimize transient light/dark adaptation effects, the ambient illumination level was maintained at a similar level.

PROCEDURE

After the completion of contrast sensitivity and acuity testing, participants were adapted to the ambient test luminance conditions for eight minutes as they were familiarized with the flicker discrimination task. After the appearance of flicker produced by counterphase alternation was described, the participants were shown samples for three different spatial-frequency/contrast combinations: 1 c/deg at .11 contrast, 4 c/deg at .33 contrast, and 8 c/deg at .66 contrast. For each combination, participants viewed samples of both the "motion" (i.e., low flicker rate) and "steady" states (i.e., high flicker rate). Observers then received a series of practice trials on which flicker rates were progressively increased and then decreased. They were asked to give a "motion" response whenever motion or flicker was perceived and a "steady" response otherwise. The practice session was continued until observers' "steady" and "motion" responses appeared to be stable and they indicated that they were comfortable with the task.

The ability to discriminate counterphase flicker was established for each of the nine spatial-frequency/contrast conditions using a single staircase adaptive procedure. Flicker rates of the continuously present display were varied by increasing or decreasing the exposure duration of each grating in the pair in 1-ms steps. Observers provided a "motion" or "steady" response with each experimenter-initiated rate change. Testing in each new stimulus condition began with an "ascending" trial wherein flicker rate was increased until "motion" responses gave way to two consecutive "steady" responses. The flicker rate was then decreased from this level (a descending trial) until "steady" responses were supplanted by two consecutive "motion" responses. The trial threshold was the flicker rate at which the first of two consecutive "steady/motion" response changes occurred. The flicker threshold for each spatial-frequency/contrast condition was taken as the average of two ascending and two descending trials.

Flicker thresholds were established in spatial frequency blocks, an observer receiving the same order of contrast within each block. With the limitation that test trials would not begin with the most visually demanding condition (i.e., 8 c/deg and .11 contrast), test order was counterbalanced by testing one member of each age by sex group using one of six randomized orders that resulted from the pairing of the six possible spatial frequency orders with the six possible contrast orders. The vocabulary test was given at the end of the test session.

Results

CONTRAST SENSITIVITY

In response to possible violations of the assumption of compound symmetry in univariate repeated measures analysis of variance (ANOVA), a multivariate analysis of variance (MANOVA) was performed on the log contrast sensitivity data. It revealed significant main effects for group, F(2,33) = 8.210, p < .001, and spatial frequency, mult,F(2,32) = 55.423, p < .001. The age group by spatial frequency interaction was not significant, p =0.461. T-- tests on log mean contrast sensitivity collapsed across spatial frequency indicated that the young FL and ND groups did not differ significantly, p = .096. The log mean sensitivity of the two young groups combined (see Table 1), however, significantly exceeded that of the old, t(34) = 3.57, p < .001. Paired sample t-tests comparing thresholds across spatial frequency showed that log mean sensitivity at 4 c/deg exceeded that at both 1 c/deg, t(35) = 9.18, p < .001 and at 8 c/deg, t(35) = 7.35, p < .001; log sensitivity at 8 c/deg was higher than at 1 c/deg, t(35) = 2.53, p < .017.

FLICKER THRESHOLDS

A Group (3) x Spatial Frequency (3) x Contrast (3) MANOVA was performed on the flicker thresholds. Significant main effects were obtained for Groups, F(2,33) = 9.382, p = .001, Spatial Frequency, multF(2,32) = 147.794, p < .001, and Contrast, multF(2,32) = 209.412, p < .001. The Group by Spatial Frequency, multF(4, 64) = 5.521, p = .001, Group by Contrast, mult,F(4, 64) = 4.041, p < .01, and the Spatial Frequency by Contrast, multF(4,30) = 7.381, p < .001 interactions were also significant. As shown in Figure 1, the thresholds for young FL observers were considerably higher than those of both the old FL and young ND groups, while those of the latter two groups were quite similar. One-way ANOVAs between the three groups at each spatial frequency indicated significant group differences at 1 c/deg, F(2,33) = 19.168, p < .001, and 4 c/deg, F(2,33) = 8.795, p = .001, but not at 8 c/deg, p = .167. Bonferroni-corrected t-tests indicated that thresholds for the FL young were significantly higher than those of the FL old at 1 c/deg, t(22) = 5.23, p < .001, and 4 c/deg, t(22) = 4.23, p < .001, as well as the ND young at 1 c/deg, t(22) = 6.01, p < .001, and 4 c/deg, t(22) = 3.71, p = .001. Thresholds for FL old and ND young did not differ at either 1 c/deg, p = .653, or 4 c/deg, p = .898.

Separate ANOVAs performed at each contrast level (see Figure 2) indicated significant differences between the three groups at contrasts of .33, F(2,33) = 11.914, p < .001, and .66, F2,33) = 11.864, p < .001; the difference at .11 was not significant after Bonferroni correction, p = .028. Pairwise t-tests across groups demonstrated that flicker thresholds in the young FL group were significantly higher than those of the old FL observers at both .33 contrast, t(22) = 4.25, p < .001, and .66 contrast, t(22) = 3.87, p = .001. The young ND group also had significantly lower thresholds than the young FL group at .33, t(22) = 4.33, p < .001, and .66 contrast, t(22) = 4.58, p < .001. The thresholds of the young ND group, however, were not different from those of the old FL group at either .33 (P = .939) or .66 (p = .497) contrast.

To determine the overall effects of contrast as a function of spatial frequency, the mean flicker threshold increment for all observers produced by each contrast increment was calculated and compared across the different spatial frequencies using paired sample t-tests with Bonferroni correction. They showed that the mean threshold increment produced by the contrast increment between .11 and .33 was significantly greater at 1 c/deg than at 8 c/deg, that between .33 and .66, it was greater for 1 c/deg than 4 c/deg, and that between .11 and .66, it was greater for 1 c/deg than for either 4 c/deg or 8 c/deg (all p < .005). No other differences were significant, indicating that increasing contrast elevated temporal thresholds more at low than at high spatial frequencies.

RELATIONSHIPS BETWEEN CONTRAST SENSITIVITY AND FLICKER THRESHOLDS

Pearson correlation coefficients were calculated to estimate the relationships at each spatial frequency and contrast level between individual log contrast sensitivity levels and flicker thresholds. None of them, however, reached significance after Bonferroni correction (p > .05).

Discussion

Old observers were markedly less able than their young counterparts to discriminate counterphase flicker in high contrast and low spatial frequency suprathreshold gratings. This loss was unrelated to the age-related decline on contrast sensitivity. Flicker thresholds for old observers tested at high luminance were very similar to those for the ND young low-luminance group, suggesting that the age-related reduction in retinal illumination may explain much of the loss with age in the temporal resolution of suprathreshold spatial stimuli.

The present findings offer little support for either the suggestion that age-related temporal losses are attributable to sensitivity declines (e.g., Kim & Mayer, 1994; Mayer et al., 1988; Zhang & Sturr, 1995), or that they are compensated at suprathreshold levels (Tulunay-- Keesey et al., 1988). They are, however, consistent with prior studies reporting an age-related decline in the temporal resolution of high-contrast targets (e.g., Barrett et al., 1994; Buckingham et al., 1987; Hiller & Kline, 2001; Kline et al., 1990; Kline et al., 2001; Whitaker et al., 1992). When contrast was increased, counterphase flicker thresholds for both young and old observers at full luminance were elevated, but age differences were not reduced. In fact, there was no support for the hypothesis that the deficit on flicker thresholds would vary inversely with grating contrast and directly with spatial frequency. The age deficit at full-- luminance was actually greater at intermediate and high contrast levels than it was at low contrast. The age deficit was also prominent for gratings of 1 c/deg and 4 c/deg, whereas at 8 c/deg, the age difference was not significant. Finally, the age differences on flicker thresholds did not "track" those seen on the contrast sensitivity measure nor was there any relationship between observers' contrast sensitivity and their flicker thresholds at the same spatial frequency.

When ND filters were used to attenuate luminance by two-thirds to simulate reduced retinal illuminance in the older eye, the flicker thresholds of young observers were similar to those of the elderly at full luminance. This suggests that age-related optical factors play a large role in limiting flicker discrimination at suprathreshold levels, similar to their hypothesized adverse effect on temporal contrast sensitivity (Wright & Drasdo, 1985). That is not to conclude that neural factors do not contribute at all to aging effects on the discrimination of suprathreshold flicker. First, the .5 log unit ND reduction of luminance may have been excessive for a photopic task (Kuyk & Wesson, 1991). Second, the attenuation of retinal illuminance by ND filters does not take into account the compensatory benefits of a smaller pupil on image quality (e.g., Winn, Whitaker, Elliott, & Phillips, 1994). When Calver, Cox, and Elliott (1999) measured the modulation transfer function (MTF) and the extent of monochromatic wavefront aberrations with the pupil's diameter fixed at the same size, they found that older eyes were worse than young eyes. When natural pupil diameters were used, however, the MTFs of the two age groups were very similar and wave-front aberrations were actually less in the older eye. Third, prior studies have provided psychophysical (e.g., Kline & Orme-Rogers, 1978; Lachenmayr, Kojetinsky, Ostermaier, Angstwurm, Vivell, & Schaumberger, 1994) as well as neurophysiological evidence (e.g., Mendelson & Wells, 2002) that aging effects on neural processes degrade temporal resolution at suprathreshold contrast levels. The evidence for neural factors, however, is not obvious in the present study.

The age deficits in flicker discrimination at 1 and 4 c/deg but not 8 c/deg are generally consistent with the hypothesized decline of diminished effectiveness of the transient magnocellular pathway (e.g., Kline & Schieber, 1981; Schefrin, Tregear, Harvey, Sr Werner, 1999; Wood & Bullimore, 1995). Although not age-related, the interaction between contrast level and spatial frequency may also be explicable in terms of the spatiotemporal response functions of the magno and parvo channels. When contrast was increased from 0.11 to 0.66, there was a significantly greater elevation in flicker thresholds for the 1 c/deg grating than for the 4 c/deg grating. This is consistent with the finding of greater contrast gains in the magnocellular than parvocellular pathway (Kaplan & Shapley, 1986; Shapley, Kaplan, & Soodak, 1981). The contrast response characteristics of the magnocellular and parvocellular pathways, however, would also suggest that a decline in the magno pathway would produce a three-way interaction between contrast, spatial frequency, and age, such that older observers would have showed the greatest threshold decrements at combinations of low contrast and low spatial frequency. The absence of such an interaction questions the utility of any simple hypotheses of disjunctive age-related decline.

Most research on aging effects on contrast sensitivity (e.g., Burton et al., 1993; Elliott, 1987; Kline et al., 1983; Owsley et al., 1983) has shown a pattern of increasing age-related loss at higher spatial frequencies. Although the age deficit observed here at all three spatial frequencies, including the lowest, is not inconsistent with some prior studies (e.g., Korth et al., 1989; Sloane et al., 1988), it may also reflect limitations of the measure used. The contrast step sizes between sequential test gratings on the Vistech test (Vistech Consultants, 1988) are fairly coarse and the low spatial frequency rows contain relatively few cycles per grating. Age differences in willingness to guess grating orientation from the three choices available, despite being encouraged to do so, may have favoured the young observers, more so perhaps at lower spatial frequencies.

In conclusion, the present results indicate a loss in the temporal resolving properties of the senescent visual system for suprathreshold targets of low and intermediate spatial frequency. This deficit is not a function of low target contrast nor does it appear to be related to observer contrast sensitivity. Although age-related optical factors that limit retinal luminance appear to explain most this deficit, neural factors may also be involved. The relative importance of each as a function of task type will be addressed in future research.

This research was supported by a grant (No. OGP0046593) from the Natural Sciences and Engineering Research Council of Canada (NSERC).

Jody C. Culham, now at the Department of Psychology, University of Western Ontario.

Our appreciation is expressed to the participants in this research; without their invaluable contributions it would not have been possible.

[Reference]

References

[Reference]

Barrett, B. T., Davison, P. A., & Eustace, P. (1994). Assessing retinal/neural function in patients with cataract using oscillatory displacement thresholds. Optometry & Vision Science, 71, 801-808.

[Reference]

Bennett, P. J., Sekuler, A. B., & Ozin, L. (1999). Effects of aging on calculation efficiency and equivalent noise. Journal of the Optical Society of America A, 16, 654-668.

Buckingham, T., Whitaker, D., & Banford, D. (1987). Movement in decline? Oscillatory movement displacement thresholds increase with aging. Ophthalmic and Physiological Optics, 7, 411-413.

Burton, K. B., Owsley, C., & Sloane, M. E. (1993). Aging and neural spatial contrast sensitivity: Photopic vision. Vision Research, 33, 939-946.

Calver, R. J., Cox, M. J., & Elliott, D. B. (1999). Effects of aging on the monochromatic aberrations of the human eye. Journal of the Optical Society of America A, 16, 2069-2078.

[Reference]

Casson, E. J., Johnson, C. A., Sr Nelson-Quigg, J. M. (1993). Temporal modulation perimetry: The effects of aging and eccentricity on sensitivity in normals. Investigative Ophthalmology & Vision Science, 34, 3096-3102.

Crassini, B., Brown, B., & Bowman, K. (1988). Age-related changes in contrast sensitivity in central and peripheral retina. Perception, 17, 315-332.

Derefeldt, G., Lennerstrand, G., & Lundh, B. (1979). Age variations in normal human contrast sensitivity. Acta Ophthalmologica, 57, 679-690.

Elliott, D. B. (1987). Contrast sensitivity decline with ageing: A neural or optical phenomenon? Ophthalmic and Physiological Optics, 7, 415-419.

Elliott, D. B., & Whitaker, D. (1992). Clinical contrast sensitivity chart evaluation. Ophthalmic & Physiological Optics, 12, 275-280.

[Reference]

Elliott, D., Whitaker, D., & MacVeigh, D. (1990). Neural contribution to spatiotemporal contrast sensitivity decline in healthy ageing eyes. Vision Research, 30, 541-547.

Guirao, A., Gonzalez, C., Redondo, M., Geraghty, E., Norrby, S., & Artal, P. (1999). Average optical performance of the human eye as a function of age in a normal population. Investigative Ophthalmology and Visual Science, 40, 203-213.

[Reference]

Higgins, K. E., Jaffe, M. J., Caruso, R. C., & de Monasterio, F. M. (1988). Spatial contrast sensitivity. Effects of age, test-retest, and psychophysical method. Journal of the Optical Society ofAmerica A, 5, 2173-2180.

Hiller, N. J., & Kline, D. W. (2001). Diminished spatial summation contributes to the age deficit in the discrimination of low-contrast vernier oscillation. Optometry and Vision Science, 78, 616-622.

Kaplan, E., & Shapley, R. M. (1982). The primate retina contains two types of ganglion cells, with high and low contrast sensitivity. Proceedings of the National Academy of Science USA, 83, 2755-2757.

Kim, C. B., & Mayer, M. J. (1994). Foveal flicker sensitivity in healthy aging eyes: II. Cross-sectional aging trends from 18 through 77 years of age. Journal of the Optical Society ofAmerica, 11, 1958-1969.

[Reference]

Kline, D. W. (1991). Light, ageing and visual performance. In J. Marshall Sr J. R. Cronly-Dillon (Eds.), Vision and visual dysfunction: Vol. 16. The susceptible visual apparatus (pp. 150-161). London: MacMillan.

Kline, D.W., Culham, J.C., Bartel, P., & Lynk, L. (2001). Aging effects on vernier hyperacuity: A function of oscillation rate but not target contrast. Optometry and Vision Science, 78, 676-682.

Kline, D. W., & Orme-Rogers, C. (1978). Examination of stimulus persistence as a basis for superior visual identification performance among older adults. Journal of Gerontology, 33, 76-81.

Kline, D. W., & Schieber, F. (1981). Visual aging: A transient/sustained shift? Perception & Psychophysics, 29, 181-182.

[Reference]

Kline, D. W., Schieber, F., Abusamra, L. C., & Coyne, A. C. (1983). Age and the visual channels: Contrast sensitivity and response speed. Journal of Gerontology, 38, 211216.

[Reference]

Kline, D. W., & Scialfa, C. T. (1996). Visual and auditory aging. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging Oth ed., pp. 181-203). San Diego, cA: Academic Press.

Kline, D. W., Scialfa, C. T., Lyman, B. J., & Schieber, F. (1990). Age differences in the temporal continuity of gratings as a function of their spatial frequency. Experimental Aging Research, 16, 61-65.

Korth, M., Horn, F., Storck, B., Jonas, J. B. (1989). Spatial and spatiotemporal contrast sensitivity of normal and glaucoma eyes. Graefe's Archive for Clinical and Experimental Ophthalmology, 227,428-435.

Lachenmayr, B. J., Kojetinsky, S., Ostermaier, N., Angstwurm, K., Vivell, P. M. O., & Schaumberger, M. (1994). The different effects of aging on normal sensitivity in flicker and light sense perimetry. Investigative Ophthalmology and Visual Science, 35, 2741-2748.

Mayer, M. J., Kim, C. B. Y., Svingos, A., & Glucs, A. (1988). Foveal flicker sensitivity in healthy aging eyes: I. Compensating for pupil variation. Journal of the Optical Society of America A, 5, 2201-2209.

Mendelson, J. R., & Wells, E. F. (2002). Age-related changes in the visual cortex. Vision Research, 42, 695-703. Nameda, N., Kawara, T., & Ohzu, H. (1989). Human visual

spatio-temporal frequency performance as a function of age. Optometry and Vision Science, 66, 760-765.

Owsley, C., Sekuler, R., & Siemsen, D. (1983). Contrast sensitivity throughout adulthood. Vision Research, 23, 689699.

[Reference]

Owsley, C., & Sloane, M. E. (1990). Vision and aging. In F. Boiler & J. Grafman (Eds.), Handbook of neuropsychology (Vol. 4, pp. 229-249). Amsterdam: Elsevier.

Pardhan, S., Gilchrist, J., & Elliott, D.B. (1996). A comparison of sampling efficiency and internal noise level in young and old subjects. Vision Research, 36, 1641-1648.

[Reference]

Ross, J. E., Clarke, D. D., & Bron, A. J. (1985). Effect of age on contrast sensitivity function: Uniocular and binocular findings. British Journal of Ophthalmology, 69, 51-56.

Royer, F. L., & Gilmore, G. C. (1985). Spatiotemporal factors and developmental changes in visual processes. Bulletin of the Psychonomic Society, 23, 404-406.

Schefrin, B. E., Tregear S. J., Harvey, L. O., Jr., & Werner, J. W. (1999). Senescent changes in scotopic contrast sensitivity. Vision Research, 39, 3728-3736.

Scialfa, C. T., Garvey, P. M., Tyrrell, R. A., & Leibowitz, H. W. (1992). Age differences in dynamic contrast thresholds. Journal of Gerontology, 47, P172-P175.

Scialfa, C. T., Kline, D. W., & Wood, P. K. (2002). Structural modeling of contrast sensitivity in adulthood. Journal of the Optical Society of America A, 19,158-165.

Scialfa, C. T., Tyrrell, R. A., Garvey, P. M., Deering, L. M., Leibowitz, H. W., & Goebel, C. C. (1988). Age differences in Vistech near contrast sensitivity. American Journal of Optometry and Physiological Optics, 65, 951956.

[Reference]

Shapley, R., Kaplan, E., & Soodak, R. (1981). Spatial summation and contrast sensitivity of X and Y cells in the lateral geniculate nucleus of the macaque. Nature, 292, 543-545.

[Reference]

Sloane, M. E., Owsley, C., & Alvarez, S. L. (1988). Aging, senile miosis and spatial contrast sensitivity at low luminance. Vision Research, 28, 1235-1246.

Spear, P. D. (1993). Neural bases of visual deficits during aging. Vision Research, 33, 2589-2609.

Tulunay-Keesey, U., Ver Hoeve, J. N., & Terkla-McGrane, C. (1988). Threshold and suprathreshold spatiotemporal

[Reference]

response throughout adulthood. Journal of the Optical Society ofAmerica A, 5, 2191-2200.

Tyler, C. W. (1989). Two processes control variations in flicker sensitivity over the life span. Journal of the Optical Society of America A, 6, 481-490.

Vistech Consultants, Inc., (1988). Vistech Contrast Sensitivity Test, Ohio, IL.

Weale, R. A. (1961). Retinal illumination and age. Transactions of the Illuminating Engineering Society, 26, 95-100.

[Reference]

Wechsler, D. (1981). WAIS-R Manual: Wechsler Adult Intelligence Scale - Revised. New York: Psychological Review.

[Reference]

Whitaker, D., Elliott, D., & MacVeigh, D. (1992). Variations in hyperacuity performance with age. Ophthalmic and Physiological Optics, 12, 29-32.

Winn, B., Whitaker, D., Elliott, D. B., & Phillips, N. J. (1994). Factors affecting light-adapted pupil size in normal human subjects. Investigative Ophthalmology & Visual Science, 35, 1132-1137.

Wood, J. M., & Bullimore, M. A. (1995). Changes in the lower displacement limit or motion with age. Ophthalmic and Physiological Optics, 15, 31-36.

Wright, C. E., & Drasdo, N. (1985). The influence of age on the spatial and temporal contrast sensitivity function. Documenta Ophthalmologica, 59, 385-395.

Zhang, L., & Sturr, J. (1995). Aging, background luminance, and threshold duration functions for detection of low spatial frequency sinusoidal gratings. Optometry and Vision Science, 72, 198-204.

[Reference]

Sommaire

[Reference]

Meme si les facteurs optiques et sensorineuraux semblent tous deux jouer un role dans les pertes liees a l'age qui touchent la resolution spatio-temporelle, leur importance relative demeure imprecise. Les objectifs de cette etude etaient done de : 1) determiner les effets du contraste de la cible et de la frequence spatiale sur la resolution lors de la presentation d'une mire de Foucault en franges sinusoidales intermittentes et en opposition de phase mis en relation avec Page de l'observateur et de sa sensibilite au contraste; 2) evaluer le role de la diminution de l'eclairement retinien dans les pertes spatio-temporelles liees a l'age, en comparant les seuils en situation de luminance elevee obtenus par les observateurs ages a ceux obtenus par de jeunes observateurs devant une cible de luminance elevee et faible. Comme la diminution hypothetique des proprietes de resolution temporelle du systeme visuel permet de l'envisager, nous pouvions prevoir que les seuils intermittents observes chez les jeunes observateurs dans les conditions experimentales de luminance elevee et de

[Reference]

luminance faible depasseraient ceux des observateurs ages a tous les niveaux de contraste. De grandes differences d'age etaient prevues dans le cas des mires de Foucault presentant un contraste faible et une frequence spatiale elevee (c.-a-d., l'interaction entre l'age x contraste x la frequence spatiale).

Methode

[Reference]

Participants. Trois groupes de 12 participants benevoles, provenant de la collectivite, ayant obtenu la correction optimale au test de distance visuelle, ont pris part a l'etude : un groupe de personnel agees (Age moyen : 66,2 ans) et deux groupes de jeunes gens (Age moyen global : 20,9 ans). L'un des deux groupes de jeunes gens, tout comme le groupe de personnel ages, ont visionne une presentation visuelle caracterisee par une luminance totale (LT). Le second groupe de jeunes gens a effectue la the, a l'aide de filtres de densite neutre (DN) fixee a 0,5, in de reproduire une reduction des deux tiers de l'eclairement

[Reference]

retinien. Chez les observateurs ages, la correction maximale de l'acuite visuelle (20/18) etait plutot mediocre par rapport A leurs homologues des groupes de jeunes gens (20/15); l'acuite des jeunes gens, dans les conditions experimentales LT et DN, etait semblable.

Procedure. La capacite de distinguer le papillotement en opposition de phase d'une mire de Foucault a franges sinusoidales a ete calculee pour neuf combinaisons espace-frequence et de contraste, produites par trois frequences spatiales (1, 4 et 8 c/deg) chacune, en fonction de trois niveaux de contraste (0,11, 0,33 et 0,66).

Une procedure d'adaptation unique faisant appel A la methode de haut en bas a ete employee pour etablir la vitesse de scintillement observee au moment ou le taux de la. mire de Foucault etait constante pour la premiere fois (essais vers le haut) ou tout de suite avant le scintillement (essais vers le bas). On a obtenu le seuil de scintillement de chaque condition experimentale a partir de la moyenne des deux essais vers le haut et des deux essais vers le bas.

Resultats

[Reference]

Les seuils de scintillement des jeunes observateurs de la condition experimentale LT etaient superieurs a ceux des personnel agees et ceux des groupes de jeunes gens faisant partie de la condition experimentale DN a 1 et a 4 c/deg (figure 1) et aux contrastes de 0,33 et

[Reference]

0,66 (figure 2). Les seuils obtenus par les jeunes de la condition experimentale DN et les personnes agees de la condition experimentale LT, dans l'une ou l'autre des conditions, ne differaient pas de maniere significative. Aucune des correlations calculees dans le but d'evaluer les relations respectives entre les donnees tirees de la sensibilite au contraste et les seuils de scintillement etablis en fonction de chaque condition de frequence/contraste ayant atteint des valeurs significatives.

Discussion

[Reference]

En comparaison avec les participants plus jeunes, les participants ages faisaient preuve d'une aptitude moindre a distinguer le papillotement par opposition de phase, produit par une mire de Foucault supraliminaire presentant une faible frequence spatiale. Le deficit n'etait lie qua la perte de sensibilite au contrasts lies a Page. Les seuils de scintillement observes chez les participants ages a qui l'on a soumis le test dans la condition de luminance elevee etaient fort semblables a ceux observes chez le groups de jeunes gens dans la condition experimentale DN. Meme si les facteurs optiques lies a Page qui reduisent l'eclairement retinien semblent expliquer la plupart des deficits lies a l'age, sur le plan de la resolution temporelle de stimuli supraliminaires spatiaux, les facteurs neuraux pourraient aussi intervenir dans ce phenomene.

[Author Affiliation]

JODY C. CULHAM, University of Western Ontario DONALD W. KLINE, University of Calgary

[Author Affiliation]

Correspondence regarding this article should be addressed to Donald Kline, Vision and Aging Lab (PACE), Department of Psychology, University of Calgary, 2500 University Drive N.W., Calgary, Alberta T2N 1N4.