Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Apr 1.
Published in final edited form as: Vision Res. 2014 Mar 7;97:100–107. doi: 10.1016/j.visres.2014.02.010

Adaptive optics without altering visual perception

Koenig DE 1, Hart NW 1,1, Hofer HJ 1
PMCID: PMC4013550  NIHMSID: NIHMS573733  PMID: 24607992

Abstract

Adaptive optics combined with visual psychophysics creates the potential to study the relationship between visual function and the retina at the cellular scale. This potential is hampered, however, by visual interference from the wavefront-sensing beacon used during correction. For example, we have previously shown that even a dim, visible beacon can alter stimulus perception (Hofer, H. J., Blaschke, J., Patolia, J., & Koenig, D. E. (2012). Fixation light hue bias revisited: Implications for using adaptive optics to study color vision. Vision Research, 56, 49-56). Here we describe a simple strategy employing a longer wavelength (980nm) beacon that, in conjunction with appropriate restriction on timing and placement, allowed us to perform psychophysics when dark adapted without altering visual perception. The method was verified by comparing detection and color appearance of foveally presented small spot stimuli with and without the wavefront beacon present in 5 subjects. As an important caution, we found that significant perceptual interference can occur even with a subliminal beacon when additional measures are not taken to limit exposure. Consequently, the lack of perceptual interference should be verified for a given system, and not assumed based on invisibility of the beacon.

Keywords: adaptive optics, psychophysics, color appearance, perceptual interference, wavefront beacon

1. Introduction

Adaptive optics correction of the eye's aberrations allows imaging and presentation of visual stimuli with spatial detail as fine as single retinal receptors (as described first by Liang, Williams, & Miller, 1997; and as reviewed recently in Rossi et al, 2011), creating the potential to probe the neural limits on vision and the relationship between visual function and the retina at this same scale (e.g. Hofer, Singer, & Williams, 2005; Makous et al, 2006; Sincich et al, 2009; Rossi, & Roorda, 2010a). However, this potential is hampered by visual interference from the wavefront-sensing beacon used during correction of aberrations. To reduce this interference, most current vision science adaptive optics systems use near infrared wavefront-sensing beacons, ranging from ∼780-850 nm (e.g. Chen et al, 2007; Artal, et al, 2010; Rossi & Roorda, 2010a; Guo et al, 2012; Sawides et al, 2011; Guo, Atchison, & Birt, 2008; Li et al, 2009; Murray et al, 2010). However, even at these wavelength the required powers (∼5-65 μW at the cornea) are high enough that the beacon is visible and disruptive in most psychophysical tasks.

Two strategies are commonly used to mitigate the impact of the beacon. One is to turn it off after the initial aberration correction, leaving the mirror static during stimulus trial blocks (e.g. Liang, Williams, & Miller, 1997; Yoon & Williams, 2002; Marcos et al, 2008; Dalimier, Dainty, & Barbur, 2008). While this strategy completely avoids interference from the beacon, such static, or ‘open-loop’, aberration correction is suboptimal (Hofer, et al, 2001a; Hofer et al, 2001b; Diaz-Santana et al, 2003) and not sufficient for evaluating the finest retinal and neural limits on visual function.

Another strategy is to correct aberrations dynamically with the beacon displaced from the location of the visual stimulus (e.g. Hofer, Singer, & Williams, 2005; Chen, et al, 2007, Guo, Atchison, & Birt, 2008, Dalimier and Dainty, 2010). While this strategy allows excellent optical correction, so long as the distance between the beacon and stimulus is on the order of 1° or less (Bedggood, Daaboul, Ashman, Smith, & Metha, 2008), even a dim, displaced beacon can significantly impact perception. For example, Hofer et al. (2012) found that a 1 μW, 840 nm beacon caused significant shifts in red-green appearance for small point stimuli, similar to those previously described for large stimuli when using colored fixation targets (Jameson and Hurvich, 1967).

Here we describe a simple strategy for eliminating the impact of the wavefront-sensing beacon on both detection and perception of visual stimuli that requires only minimal changes to the standard system configuration, namely replacing the wavefront-sensing beacon with a longer wavelength source. Interestingly, when testing this modified system configuration, we discovered that the beacon can still interfere with the perceived appearance of visual stimuli, even when dim enough that subjects say they are unable to see it. While investigators should be aware of this potential interference, we've found it can be eliminated with careful restriction on beacon exposure and placement.

2. Methods

2.1. Wavefront Sensor and Adaptive Optics System

We modified an existing adaptive optics system (Hofer, et al. 2012) to accommodate a long wavelength 980 nm beacon (super luminescent diode, SLD, QPhotonics LLC) instead of the original, more typical, 840 nm beacon (SLD, Volga Technology Ltd.). Figure 1 describes the current system.

Figure 1.

Figure 1

Open in a new tab

Adaptive optics system for psychophysics and imaging. Aberrations are measured and corrected in closed-loop with a Shack-Hartmann wavefront sensor consisting of a thermoelectrically-cooled, electron-multiplying, charge-coupled device (EMCCD) camera (PhotonMax 512, Princeton Instruments) and a 24 mm focal length micro-lenslet array (Adaptive Optics Associates), coupled with a 97-channel deformable mirror (Xinetics). Lenslet spacing is 0.4 mm and the clear aperture of the deformable mirror is 7.26 mm in the pupil plane. The beam splitter (BS, top right) transmits infrared light for wavefront sensing (>900 nm) and reflects visible light for imaging or stimulus display. Computerized shutters (S) in pupil conjugate planes (P) control timing and exposure duration of all light sources. During adaptive optics psychophysics and retinal imaging pupil (P*) is set at 6 mm. To reduce defocus measured at the wavefront sensing camera subjects use a stabilizing bitebar mounted on a translating Badal optometer (eye, bottom right). Longitudinal chromatic aberration between wavefront-sensing and stimulus/imaging wavelengths is corrected by adjustment of a focus correction slider (upper left). Fixation target (far right) and stimuli (OLED or point stimuli) are seen through Maxwellian view with unit and 3.33 magnification, respectively.

The beacon wavelength was intended to be long enough to not disturb or impact vision, yet short enough for accurate wavefront sensing and correction with our existing wavefront camera. (Wavefront sensing accuracy decreases with wavelength despite the relative constancy of the eye's higher order aberrations (Fernandez & Artal, 2008), due to the effects of increased scatter and diffraction on the localizability of the Shack-Hartmann spots.) Therefore, we considered the following factors: the quantum efficiency of the existing wavefront sensing camera (PhotonMAX 512, Princeton Instruments, Trenton, NJ - Fig 2), the eye's spectral sensitivity (Baylor, Nunn, & Schnapf, 1987; Fig 2), and the increase in Shack-Hartmann spot position error as spot size increases with diffraction at longer wavelengths, Hardy, 1998).

Figure 2.

Figure 2

Open in a new tab

Estimated sensitivity of the human eye and adaptive optics (AO) system at near infrared wavelengths. The point where the WFS camera sensitivity function (solid curve) crosses the human eye sensitivity function (dashed curve) is the shortest wavelength at which WFS is predicted to become possible for beacon powers below visual threshold. The inset compares photopic quantum efficiency of the human eye (dashed curve - normalized to unit peak efficiency) to that of our current wavefront sensing camera (solid curve - normalized to unit peak efficiency).

We estimated visual and wavefront sensor sensitivity at longer wavelengths from measurements of the dark-adapted visual threshold for a continuously viewed 840 nm beacon (two subjects, method of adjustment) and the minimum power required for satisfactory adaptive correction, given the following assumptions:

  1. Retinal reflectance (Berendschot, Kraats, Kanis, & van Norren, 2010) and ocular transmittance (Boettner, & Wolter, 1962) are relatively constant with wavelength in this regime.

  2. Visual threshold decreases 2.3 log units per 100 nm (extrapolated from Baylor, Nunn, & Schnapf, 1987).

  3. Higher order ocular aberrations are relatively constant with wavelength (Fernandez and Artal, 2008).

  4. A proportional increase in beacon power with wavelength offsets the impact of increased Shack-Hartmann spot position error maintaining a constant wavefront sensor signal to noise ratio.

While some of these assumptions are simplistic, we considered them a reasonable starting point given the level of uncertainty associated with several of the relevant factors. For example, the relative balance of visual and wavefront sensor sensitivity depends on both ocular transmission and retinal reflectance. While ocular transmission is known to vary with wavelength, with a relatively narrow dip in transmission near 980 nm and then decreasing more sharply after ∼1300 nm (Boettner and Wolter, 1962), the behavior of retinal reflectance is less clear, with previous data suggesting both increases and decreases with longer wavelengths, perhaps depending on the level of pigmentation (e.g. van de Kraats et al, 1996; Zagers et al, 2002; Elsner et al, 1996). The role of retinal reflectance is further complicated as the penetration depth increases with wavelength, resulting in reflection from multiple layers, which may impact both wavefront sensitivity and visual sensitivity.

Figure 2 shows the estimated thresholds of the human eye and our adaptive optics system incorporating these assumptions for near infrared wavelengths. The point where the adaptive optics system sensitivity function crosses the human eye sensitivity function, ∼980nm, is the shortest beacon wavelength predicted to allow accurate wavefront sensing without being visible to a dark-adapted subject.

After modification to incorporate the 980 nm beacon we found it necessary to increase the wavefront-sensing camera exposure time from 30 to 100 ms (system rate of ∼ 9.1Hz) and decrease the loop gain, or the fraction of the residual aberration compensated for on each iteration of adaptive correction, from 0.3-0.4 to 0.15-0.25 to compensate for the reduced camera efficiency and increased spot position noise. Five to eight iterations were typically required for a total correction time of ∼600-900 ms.

With these parameters effective correction (rms < 0.1 over a 6 mm pupil) still required 11-14 μW at the eye's pupil, 2-3 times more power than predicted (6-9 times more than predicted when also considering the increased exposure time). The larger required power may be due in part to our failure to accurately account for the wavelength dependence of ocular transmittance. However, the measured and predicted visual thresholds at 980 nm were extremely close (4.3 vs 4.8 μW at the cornea, Subject 4, method of adjustment), suggesting that any decrease in ocular transmittance is offset by increased retinal reflectance, at least some of which is visually effective.

At these powers, the 980 nm beacon was invisible in the light-adapted retina but was still somewhat visible to most subjects when dark-adapted and displayed in a dark visual field. For example, the required powers were 0.4-0.5 log units above Subject 4's dark-adapted visual threshold for continuous viewing. The beacon generally appeared as a small red spot when fixated, but sometimes appeared as a more diffuse gray spot when off the visual axis. This diffuse gray appearance may possibly reflect the activity of rods, the two-photon/second harmonic generation effect (where light of one half the beacon wavelength is created and potentially seen, Zaidi & Pokorny, 1988), or both. We discuss potential interference from half wavelength generated light when using longer wavelength infrared beacons further in the Discussion.

2.2 Psychophysical Assessment of Visual Interference

2.2.1 Subjects

A total of five subjects participated, subjects 1 and 2 were chosen because the original 840 nm wavefront-sensing beacon was known to impact their color reports in a prior study (Hofer et al. 2012), while subjects 3-5 were new recruits. Subjects 1 and 4 were the authors and subjects 3 and 5 were practiced psychophysical observers naïve to the purposes of this study. Subjects 1 and 3 were female and subjects 2, 4, and 5 were male. All subjects had normal vision correctable to at least 20/20 and normal color vision as assessed by the Hardy-Rand-Rittler 4th edition pseudoisochromatic plates (Richmond Products, 2002; Hardy et al., 1954, Croner, 1961). The research followed the tenets of the Declaration of Helsinki and all subjects gave informed consent after an explanation of the study procedure and any possible risks. All study procedures were approved by the Institutional Review Board of the University of Houston.

2.2.2 Experimental Stimuli and Procedures

We assessed whether the 11-14 μW 980 nm beacon interfered with visual perception by comparing color appearance for threshold foveal point stimuli with and without the beacon present in three different experimental scenarios, described below.

For all scenarios, stimuli were created by either illuminating a 25 μm pinhole (subtending ∼0.2’) with a white light emitting diode (LED) or with a white light organic LED microdisplay (EMA-100503 SXGA Monochrome White XL Microdisplay, eMagin Corporation, Bellevue, WA). Stimulus wavelength of 580 nm was chosen based on previous research (Cicerone and Nerger, 1989; Krauskopf and Srebro, 1965) to maximize variability in color appearance and was controlled with a narrow band (10 nm bandwidth) interference filter. We sought to determine the impact of the wavefront beacon on color appearance by comparing color-reports with the beacon present to when the beacon was absent. Since adaptive correction of the eye's aberrations cannot be achieved without the wavefront-sensing beacon, stimuli in both conditions were displayed without adaptive optics with the adaptive mirror in a flattened state (i.e. conventional refraction) through a 2 mm artificial pupil. A longer duration was required with the microdisplay (30 ms) than the LED (6 ms). The LED was used by subjects 1, 2, and 3 in the third experimental scenario. The microdisplay was used by all subjects in the first two experimental scenarios. Color ratings with the LED and microdisplay stimuli, for subjects who used both (subjects 1, 2, and 3), differed only in the mean yellow-blue direction in a manner consistent with typical intra-subject variation across days.

Stimuli were monocularly presented to the central fovea after at least 10 minutes of dark adaptation in an otherwise completely dark visual field, save for 2 dim (≲ 0.01 cd/m2), broadband white fixation dots, vertically separated by 2.25° (Figure 1) and presented in Maxwellian view. Subjects fixated with the test eye between the dim white dots on all trials and the non-test eye remained dark-adapted. No dilating agents were used and subjects wore their habitual spectacle correction, or were corrected with either trial lenses and/or by translating a movable stage in a Badal optometer.

We performed three variations of the same experiment to assess the impact of the longer wavelength (980 nm) wavefront-sensing beacon on reported color appearance of dim, monochromatic, point stimuli. In the first scenario the beacon was present in the visual field, displaced horizontally 1° from the center of fixation and stimulus presentation. In the second scenario the 980 nm beacon was continuously displayed and co-localized with, or masked by, the uppermost fixation dot (i.e. ∼1° above the center of fixation and stimulus presentation). In this second scenario the wavefront beacon was generally not visible (even after dark adaptation) when fixating in between the two white fixation dots. In the third scenario the 980 beacon was co-localized with the uppermost fixation dot, but exposure was limited by an electronic shutter to the wavefront-sensing interval (i.e. duration < 1 second immediately preceding stimulus presentation on each trial), which further reduced visibility. The shutter was not used with the first scenario as shuttering the visible beacon was found to be highly distracting.

For each scenario subjects performed randomly intermixed blocks of trials with the beacon either present or absent over the course of 1 or 2 days. Blocks consisted of up to ∼80 trials and included 5 intensity levels (including blanks) spanning the psychometric function, with typically 400-500 trials completed per condition. The number of trials reflects the typical number in an adaptive optics color experiment. Stimulus timing and intensity were controlled by a custom Matlab program (The Mathworks, Natick, MA) incorporating Psychophysics Toolbox routines (Brainard, 1997; Pelli, 1997).

Subjects rated color appearance with the 5 color name method described by Hofer et al. (2012) and Koenig and Hofer (2012). Briefly, subjects were instructed to rate seen stimuli according to hue and saturation, and not brightness, by distributing 10 key presses among five categories: white, green, blue, yellow, or red; in any manner they felt best reflected the appearance of the stimulus on each trial (Koenig & Hofer, 2012). For example, a stimulus appearing saturated green would be rated by placing all 10 key presses in the green category, whereas a moderately desaturated orange might be rated as 3 red, 2 yellow, and 5 white. Subjects used a separate ‘I don’t know key' to report any ‘colorless’ or ‘indescribable’ stimuli (Bouman & Walraven, 1957; Hofer, Singer, and Williams, 2005; Krauskopf, 1978), and we excluded these responses from color analysis. Subjects were excused from rating color appearance for stimuli that were not seen. Trials were self-paced and subjects entered ratings with a small handheld numeric keypad. The response process included numerous safeguards to prevent or identify response errors as previously described (Hofer et al, 2012; Koenig & Hofer, 2012).

Since detection sensitivity can be highly variable over time, and color naming responses to tiny flashes have been shown to vary with both stimulus intensity and detection criterion (especially in the Blue-Yellow color direction; Koenig and Hofer, 2012), we collected data for both conditions of each experimental comparison on the same day with beacon absent and beacon present trial blocks alternated within a single test session. Tests of significance (mean difference) were performed only on test and control data acquired within such a session.

2.2.3. Data analysis

The impact of the wavefront-sensing beacon on visual appearance is expected to be greatest for threshold stimuli. For this reason we restricted data analysis to stimuli with frequencies of seeing between 20% and 85%. Mean ratings were computed for each subject and condition after performing an arcsine transform to uniformly distribute variance (Abramov, Gordon, & Chan, 2009). Two-tailed z-tests were used to assess the significance of differences in mean hue ratings across conditions, with results verified using non-parametric permutation tests (i.e. tests not dependent on normal distribution of the raw color ratings data; Good, 2006; Hofer et al, 2012). We also performed chi-squared analysis across hue categories and conditions to test for potential differences in the distribution of color ratings that may not be reflected in the mean ratings.

Data were visualized with Uniform Appearance Diagrams (UAD; Abramov, Gordon, & Chan, 2009). Figure 3 shows an example UAD illustrating the variation in color appearance for repeated 580 nm foveal small spot stimuli for one subject (subject 1). The variation in color appearance shown in this diagram is typical of small, threshold, foveally presented monochromatic stimuli and similar to that previously reported for the same stimuli, which encompass roughly 10 – 20 cones (Hofer et al, 2012, Koenig & Hofer, 2012; Krauskopf & Srebro, 1965). Hofer et al. (2005) have shown even greater variability with smaller stimuli (encompassing 1 – 3 cones) presented with adaptive optics correction 1° from the fovea.

Figure 3.

Figure 3

Open in a new tab

Uniform Appearance Diagram (UAD) illustrating typical variation in appearance of monochromatic (580 nm), threshold (20–85% seen), foveally viewed point stimuli for one subject (subject 1). The green minus the red rating (after the arcsine transform) is plotted versus the yellow minus the blue rating for each of 464 seen stimuli. Stimuli rated as purely white fall at the origin, while stimuli rated as purely colored (saturated) lie along the edges (diagonal lines) of the diagram. The weight of each point indicates the number of stimuli with that rating, with the darkest points representing the most stimuli. The variation in color appearance reported by this subject is typical, with stimuli generally varying in appearance along an orange-bluegreen axis and infrequently, or rarely, appearing yellowgreen or violet.

3. Results

Two subjects (subjects 1 and 2) participated in the first experimental scenario, in which the beacon was displaced horizontally ∼1° from fixation. Although the 11-14 uW 980 nm beacon was dim and below threshold in photopic conditions, it was still generally visible when dark adapted. While the frequency of seeing was not impacted (two-tailed z-test p-value across subjects = 0.88), both subjects reported the 580 nm stimulus significantly greener with the beacon present than absent (two-tailed z-test p-values for mean green-red differences were 0.045 for both subjects, with significance verified by permutation testing), as shown in Figure 4. Despite the marked reduction in the salience and subjective brightness of the beacon, the red-green hue shift was of the same magnitude as found previously for these subjects when the 840 nm wavefront beacon was present, but not fixated (∼1, and 1.5 UAD units for the 980 nm beacon and ∼1 and 2 units for the 840 nm beacon for subjects 1 and 2, respectively; Hofer et al, 2012).

Figure 4.

Figure 4

Open in a new tab

Mean hue is significantly greener and/or less red for two subjects when viewing foveally presented, threshold, 580 nm stimuli with the wavefront-sensing beacon present (Pr), than with the beacon absent (Ab). (a) Distribution of color ratings among the five categories: red, yellow, green, blue, and white; shows increased green and/or less red ratings. (b) Mean color ratings in the green-red and yellow-blue directions for beacon present (red) and beacon absent (black) conditions. Data points are labeled with subject number and represent the mean of all seen spots plus or minus 1 standard error. Note that only the central portion of the UAD is shown (5, -5).

Three subjects (subjects 1, 2, and 3) participated in the second experimental scenario. The 980 nm beacon co-localized under the top fixation spot (masked) was generally not visible or very subtle when fixating between the two white spots. As with the first scenario, presence of the beacon did not impact the detectability of the stimulus (two-tailed z-test p-value across subjects = 0.82). However, as shown in Figure 5, the 980 nm beacon caused the 580 nm stimuli to appear redder for all subjects, with significant impact for subjects 2 and 3. Two-tailed z-test p-values for subject 2's mean red ratings and subject 3's mean green, red, and green-red ratings were 0.0003, and 0.003, 0.03, and 0.003 (significance verified by permutation testing). Pearson's chi-square test across the five color categories, with the beacon present and absent, and with 4 degrees of freedom was significant for subject 2 with p = 0.0006, and borderline for subject 3 with p = 0.07. These results are also consistent with our prior report (Hofer et al, 2012), where we found that the direction of the beacon-related hue shift depended on whether it was located in an area of the visual field used for fixation, or present in the field but not fixated. Furthermore, the magnitude of the color shifts were similar to those we reported when fixating an 840 nm beacon or using other red fixation targets (Hofer et al, 2012), despite the fact that 2 subjects (1 and 3) reported that they could not see the 980 nm beacon in this configuration.

Figure 5.

Figure 5

Open in a new tab

Mean hue is significantly redder for two of three subjects (2-circles, and 3-diamonds) when viewing foveally presented, threshold, 580 nm stimuli with the wavefront-sensing beacon present (Pr) but co-localized with the top fixation spot (masked), than with the beacon absent (Ab). (a) Distribution of color ratings among the five categories: red, yellow, green, blue, and white; shows increased red and decreased green ratings. (b) Mean color ratings in the green-red and yellow-blue directions for beacon present (red) and beacon absent (black) conditions show a redward shift for all subjects that is significant for subjects 2 and 3. Data points are labeled with subject numbers and represent the mean of all seen spots plus or minus 1 standard error. Note that only the central portion of the UAD is shown (5, -5).

To give a sense of the magnitude of this shift, subject 3's mean color ratings correspond to a moderately (∼40%) saturated orange with the beacon, but nearly white without [an estimated mean change in CIExy coordinates from (0.45, 0.36) to (0.34, 0.34)].

Five subjects participated in the 3rd experimental scenario, in which the beacon was co-localized with the uppermost fixation spot and its duration was restricted to the wavefront sensing interval (500 – 900 ms just prior to stimulus presentation). In this configuration the beacon was not visible when fixating and there was no significant impact on detectability (paired t-test p-value = 0.61) or color reports for any of the five subjects. Figure 6 shows the distribution of color ratings and mean hue ratings for all subjects and Table 1 shows two-tailed z-test p-values for the mean ratings differences in the green-red directions. Pearson's chi-square tests across the five color categories and two conditions were also not significant for any of the five subjects, indicating that other aspects of the distribution of color ratings were not significantly impacted by the beacon for any subject.

Figure 6.

Figure 6

Open in a new tab

Distribution of color ratings and mean color ratings for stimuli presented in the presence (Pr) or absence (Ab) of the masked and duration-limited wavefront-sensing beacon for five subjects. (a) The distributions of color ratings for a 580 nm point stimulus were not significantly different whether a 980 nm wavefront-sensing beacon was present (masked) or absent (shutter closed during stimulus presentation). (b) The mean color ratings for a 580 nm point stimulus were also not significantly different whether a 980 nm wavefront-sensing beacon was present or absent. Data points (subjects 1-circles, 2-squares, 3-diamonds, 4-x's, 5-triangles) are labeled with subject numbers and represent the mean of all seen spots plus or minus 1 standard error. Note that only the central portion of the UAD is shown (3, -3).

Table 1.

Magnitude and two-tailed z-test p-values for the differences in mean color ratings in the green-red direction with the beacon present vs. absent, when the beacon was co-localized with the top fixation spot and presented only briefly by shuttering. There was no significant impact on color reports for any of the five subjects.

Green-Red
difference p-value
Subject 1 0.053 0.823
Subject 2 0.490 0.067
Subject 3 0.366 0.623
Subject 4 0.076 0.810
Subject 5 0.545 0.277
Open in a new tab

Although the beacon did not significantly impact reported color for any individual subject when both masked and presented briefly, inspection of Figure 6 and Table 1 show a small red-ward shift in hue for all subjects. Paired t-tests (p-value 0.04) indicate this shift in appearance is statistically significant in aggregate, although the magnitude is well within the typical intra- and inter- subject variation.

Inspection of figures 4 and 5 shows that color naming varied across days, for the same condition (e.g. color naming with the beacon absent for those subjects participating in multiple experiment scenarios), on a similar scale (±1 linear UAD unit) as the differences in mean color naming response between beacon absent and beacon present conditions (which were always measured within the same session) in experimental scenarios 1 and 2. This variability in color naming may be due in part to variability in detection sensitivity and criterion (Koenig and Hofer, 2012). The size of this impact is also comparable to the differences in color naming for tiny adaptive optics correction stimuli across individuals with different cone ratios reported in Hofer, Singer, & Williams, 2005, where color naming responses appeared to vary systematically with long- and medium-wavelength sensitive cone ratios and were therefore unlikely to be attributable solely to criterion differences or context effects. The results of Hofer et al, 2005 suggest assessing and controlling for all of these factors will be important when the intention is to evaluate the relationship between color responses and cone ratios across different subjects or different retinal areas within a subject. Here, to mitigate the confounding effect of longer term variations in color appearance judgments on the evaluation of the impact of the beacon, control (beacon absent) and test (beacon present) conditions were alternated within a single session.

4. Discussion and Conclusions

Our goal was to enable adaptive optics psychophysics without altering either detection or perception of psychophysical stimuli, a requirement for realizing the full potential to uncover the most sensitive retinal and neural limits on vision. We achieved this goal by modifying an existing adaptive optics system to incorporate a long wavelength, 980 nm beacon that was masked by dim, neutral fixation dots and shuttered to limit exposure to the 500-900 ms wavefront-sensing interval immediately preceding stimulus presentation.

As a cautionary note, our data also demonstrate that ‘invisibility’ of the wavefront-sensing beacon is not a sufficient condition for eliminating perceptual interference - as color reports were significantly altered by the continuous presence of the beacon, even when subjects said they were unable to see it (as in the second experimental scenario). Although masking and shuttering the beacon eliminated its impact on color reports for each individual subject, there was still a small, yet statistically significant, redward shift on aggregate. This suggests that, even with this strategy, the potential impact of the beacon should not be ignored when considering population data or a large quantity of data (≳2500 trials) for an individual subject. It was also apparent from our data that variability in color responses was an important, potentially confounding factor. That is, evaluation and measurement of the impact of the beacon on color reports was complicated by longer term variability in color naming arising in part from variability in both sensitivity and criterion, especially in less practiced subjects. For future comparisons between color responses in multiple conditions a more elaborate response paradigm, and more careful control of both sensitivity and criterion, are expected to reduce the amount of response variability. In their early investigations into color appearance of adaptive optics corrected stimuli, Hofer et al. (2005) were careful not to draw conclusions from their data that depended on the absolute color responses, fearing a potential (yet untested) impact of the wavefront beacon on hue. Consequently, our current findings do not alter the conclusions of their prior study. Rather, we reaffirm the importance of understanding and controlling the visual impact of the light sources used to image or achieve optical correction when performing adaptive optics psychophysics.

Additional solutions for reducing visual interference exist. For example, an even longer wavelength beacon could be used. Artal et al. (2012) recently reported successful aberration measurement and spherical aberration correction with a 1050 nm beacon and infrared sensitive wavefront camera, and it is possible that beacon timing and placement would be less critical with this strategy. However, we were wary of pursuing even longer wavelengths due to concerns about increased wavefront sensor noise (given that we did not wish to replace our wavefront camera with an infrared sensitive model) and the potential for even greater visual interference from two-photon/second harmonic half-wavelength light. The strength of the two photon/second harmonic effect depends on the specific spatial and temporal parameters of the incident light, but generally increases with wavelength (Vasilenko, Chebotayev, & Troitskiy, 1965; Sliney et al, 1976; Zaidi & Pokorny, 1988).

Another potential option for performing adaptive optics psychophysics without perceptual interference would be to use a ‘sensorless’ correction method (Booth, 2007, Biss et al., 2007), where optical quality is directly optimized based on some image quality metric. While we found this strategy to provide excellent optical correction in a reflective confocal system (Hofer et al, 2011), when we implemented it in the current system, which contains numerous refractive surfaces, irreducible back reflections resulted in highly variable correction. Successful implementation of this method in a refractive system may be possible with customized anti-reflection coatings.

In summary, we have described a simple strategy for eliminating the impact of the wavefront-sensing beacon on both detection and perception of visual stimuli that requires only minimal changes to the standard adaptive optics system configuration. Importantly, our data caution that even an ‘invisible’ beacon can significantly interfere with stimulus perception. However, this interference can be eliminated with careful restrictions on beacon timing and placement.

Article highlights are listed below.

  • Adaptive optics wavefront beacons interfere with stimulus detection and perception

  • Even a subliminal infrared (980 nm) beacon can significantly impact reported color

  • This impact can be eliminated with careful restrictions on placement and timing

Acknowledgments

Research supported by: NIH RO1 EY019069, T35 EY07088, and P30 EY07551. We thank Lukasz Sterkowicz, Hope Queener, Chris Kuether, and Chueh Ting for assistance with software and the stimulus apparatus. The results described in this paper were reported in part at the Association for Research in Vision and Ophthalmology (Hart, Koenig, & Hofer, 2012), and as partial requirement for the degree of Doctor of Philosophy in Physiological Optics, University of Houston College of Optometry (Koenig, 2012).

Footnotes

Conflicts of interest: DK none, NH none, HH none

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Koenig DE, Email: dkoenig@optometry.uh.edu.

Hart NW, Email: nate@camerafinder.org.

Hofer HJ, Email: hhofer@optometry.uh.edu.

References

  1. Abramov I, Gordon J, Chan H. Color appearance: Properties of the uniform appearance diagram derived from hue and saturation scaling. Attention, Perception, and Psychophysics. 2009;71:632–643. doi: 10.3758/APP.71.3.632. [DOI] [PubMed] [Google Scholar]
  2. Artal P, Manzanera S, Piers P, Weeber H. Visual effect of the combined correction of spherical and longitudinal chromatic aberrations. Optics Express. 2010;18(2):1637–1648. doi: 10.1364/OE.18.001637. [DOI] [PubMed] [Google Scholar]
  3. Artal P, Schwarz C, Canovas C, Mira-Agudelo A. Night myopia studied with an adaptive optics analyzer. PlosOne. 2012;7(7):e40239, 1–6. doi: 10.1371/journal.pone.0040239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baylor DA, Nunn BJ, Schnapf JL. Spectral sensitivity of cones of the monkey Macaca fascicularis. Journal of Physiology. 1984;390:145–160. doi: 10.1113/jphysiol.1987.sp016691. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bedggood P, Daaboul M, Ashman R, Smith G, Metha A. Characteristics of the human isoplanatic patch and implications for adaptive optics retinal imaging. Journal of Biomedical Optics. 2008;13(2):024008, 1–7. doi: 10.1117/1.2907211. [DOI] [PubMed] [Google Scholar]
  6. Berendschot TTJM, van de Kraats J, Kanis MJ, van Norren D. Directional model analysis of the spectral reflection from the fovea and para-fovea. Journal of Biomedical Optics. 2010;15(6):065005-1–065005-9. doi: 10.1117/1.3523370. [DOI] [PubMed] [Google Scholar]
  7. Biss DP, Webb RH, Zhou Y, Bifano TG, Zamira P, Lin CP. An adaptive optics biomicroscope for mouse retinal imaging. Proc SPIE. 2007;6467:646703–646708. [Google Scholar]
  8. Boettner EA, Wolter JR. Transmission of the ocular media. Investigative Ophthalmology and Visual Science. 1962;1(6):776–783. [Google Scholar]
  9. Booth MJ. Adaptive optics in microscopy. Philosophical Transactions of the Royal Society A. 2007;365:2829–2843. doi: 10.1098/rsta.2007.0013. [DOI] [PubMed] [Google Scholar]
  10. Bouman MA, Walraven PA. Some color naming experiments for red and green lights. Journal of the Optical Society of America. 1957;47:834–839. doi: 10.1364/josa.47.000834. [DOI] [PubMed] [Google Scholar]
  11. Brainard DH. The psychophysics toolbox. Spatial Vision. 1997;10:433–436. [PubMed] [Google Scholar]
  12. Chen L, Artal P, Gutierrez D, Williams DR. Neural compensation for the best aberration correction. Journal of Vision. 2007;7(10):9, 1–9. doi: 10.1167/7.10.9. [DOI] [PubMed] [Google Scholar]
  13. Cicerone CM, Nerger JL. The relative numbers of long-wavelength sensitive to middle-wavelength-sensitive cones in the human fovea centralis. Vision Research. 1989;29:115–128. doi: 10.1016/0042-6989(89)90178-8. [DOI] [PubMed] [Google Scholar]
  14. Crone RA. Quantitative diagnosis of defective color vision: A comparative evaluation of the Ishihara test, Farnsworth dichotomous test, and the Hardy-Rand-Rittler polychromatic plates. American Journal of Ophthalmology. 1961;51:298–305. [PubMed] [Google Scholar]
  15. Dalimier E, Dainty C, Barbur JL. Effects of higher-order aberrations on contrast acuity as a function of light level. Journal of Modern Optics. 2008;55(4-5):791–803. [Google Scholar]
  16. Dalimier E, Dainty C. Role of ocular aberrations in photopic spatial summation in the fovea. Optics Letters. 2010;35(4):589–591. doi: 10.1364/OL.35.000589. [DOI] [PubMed] [Google Scholar]
  17. Diaz-Santana L, Torti C, Munro I, Gasson P, Dainty C. Benefit of higher closed-loop bandwidths in ocular adaptive optics. Optics Express. 2003;11(20):2597–2605. doi: 10.1364/oe.11.002597. [DOI] [PubMed] [Google Scholar]
  18. Elsner AE, Burns SA, Weiter JJ, Delori FC. Infrared imaging of subretinal structures in the human ocular fundus. Vision Research. 1996;36(1):191–205. doi: 10.1016/0042-6989(95)00100-e. [DOI] [PubMed] [Google Scholar]
  19. Fernandez EJ, Artal P. Ocular aberrations up to the infrared range: from 632.8 to 1070 nm. Optics Express. 2008;16(26):21199–21208. doi: 10.1364/oe.16.021199. [DOI] [PubMed] [Google Scholar]
  20. Good PI. Resampling methods: A practical guide to data analysis. 3rd. Boston: Birkhauser; 2006. [Google Scholar]
  21. Guo H, Atchison DA, Birt BJ. Changes in through-focus spatial visual performance with adaptive optics correction of monochromatic aberrations. Vision Research. 2008;48:1804–1811. doi: 10.1016/j.visres.2008.04.033. [DOI] [PubMed] [Google Scholar]
  22. Guo H, Fallah HR, Dainty C, Goncharov AV. Subjective evaluation of intraocular lenses by visual acuity measurement using adaptive optics. Optics Letters. 2012;37(12):2226–2228. doi: 10.1364/OL.37.002226. [DOI] [PubMed] [Google Scholar]
  23. Hardy JW. Adaptive optics for astronomical telescopes. New York: Oxford University Press; 1998. [Google Scholar]
  24. Hardy LH, Rand G, Rittler C. H-R-R polychromatic plates. Journal of the Optical Society of America. 1954;44(7):509–523. [Google Scholar]
  25. Hofer HJ, Artal P, Singer B, Aragon JL, Williams DR. Dynamics of the eye's wave aberration. Journal of the Optical Society of America A. 2001a;18(3):497–506. doi: 10.1364/josaa.18.000497. [DOI] [PubMed] [Google Scholar]
  26. Hofer HJ, Chen L, Yoon GY, Singer B, Yamauchi Y, Williams DR. Improvement in retinal image quality with dynamic correction of the eye's aberrations. Optics Express. 2001b;8(11):631–643. doi: 10.1364/oe.8.000631. [DOI] [PubMed] [Google Scholar]
  27. Hofer H, Singer B, Williams DR. Different sensations from cones with the same photopigment. Journal of Vision. 2005;5(5):444–454. doi: 10.1167/5.5.5. [DOI] [PubMed] [Google Scholar]
  28. Hofer H, Sredar N, Li C, Queener H, Porter J. Wavefront sensorless adaptive optics ophthalmoscopy in the human eye. Optics Express. 2011;19:14160–14171. doi: 10.1364/OE.19.014160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hofer HJ, Blaschke J, Patolia J, Koenig DE. Fixation light hue bias revisited: Implicatinos for using adaptive optics to study color vision. Vision Research. 2012;56(1):49–56. doi: 10.1016/j.visres.2012.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jameson D, Hurvich LM. Fixation-light bias: An unwanted by-product of fixation control. Vision Research. 1967;7:805–809. doi: 10.1016/0042-6989(67)90043-0. [DOI] [PubMed] [Google Scholar]
  31. Koenig DK, Hofer HJ. Do color appearance judgments interfere with detection of small threshold stimuli? Journal of the Optical Society of America A – Optics Image Science and Vision. 2012;29:A258–A267. doi: 10.1364/JOSAA.29.00A258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. van de Kraats J, Berendschot TTJM, van Norren D. The pathways of light measured in fundus reflectometry. Vision Research. 1996;36(15):2229–2247. doi: 10.1016/0042-6989(96)00001-6. [DOI] [PubMed] [Google Scholar]
  33. Krauskopf J. On identifying detectors. In: Armington JC, Krasukopf J, editors. Visual psychophysics and physiology. New York: Academic Press; 1978. [Google Scholar]
  34. Krauskopf J, Srebro R. Spectral sensitivity of color mechanisms: Derivation from fluctuations of color appearance near threshold. Science. 1965;150:1477–1479. doi: 10.1126/science.150.3702.1477. [DOI] [PubMed] [Google Scholar]
  35. Li S, Xiong Y, Li J, Wang N, Dai Y, Xue L, Zhao H, Jiang W, Zhang Y, He JC. Effects of monochromatic aberration on visual acuity using adaptive optics. Optometry and Vision Science. 2009;86(7):868–874. doi: 10.1097/OPX.0b013e3181adfdff. [DOI] [PubMed] [Google Scholar]
  36. Liang J, Williams DR, Miller DT. Supernormal vision and highresolution retinal imaging through adaptive optics. Journal of the Optical Society of America A – Optics and Image Science. 1997;14:2884–2892. doi: 10.1364/josaa.14.002884. [DOI] [PubMed] [Google Scholar]
  37. Makous W, Carroll J, Wolfing JI, Lin J, Christie N, Williams DR. Retinal microscotomas revealed with adaptive-optics microflashes. Investigative Ophthalmology and Visual Science. 2006;47(9):4160–4167. doi: 10.1167/iovs.05-1195. [DOI] [PubMed] [Google Scholar]
  38. Marcos S, Sawides L, Gambra E, Dorronsoro C. Influence of adaptive-optics ocular aberration correction on visual acuity at different luminances and contrast polarities. Journal of Vision. 2008;8(13):1, 1–12. doi: 10.1167/8.13.1. [DOI] [PubMed] [Google Scholar]
  39. Murray IJ, Elliott SL, Pallikaris A, Werner JS, Choi S, Tahir HJ. The oblique effect has an optical component: Orientation-specific contrast thresholds after correction of high-order aberrations. Journal of Vision. 2010;10(11):10, 1–12. doi: 10.1167/10.11.10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Pelli DG. The VideoToolbox software for visual psychophysics: Transforming numbers into movies. Spatial Vision. 1997;10:437–442. [PubMed] [Google Scholar]
  41. Rossi EA, Roorda A. The relationship between visual resolution and cone spacing in the human fovea. Nature Neuroscience. 2010a;13(2):156–157. doi: 10.1038/nn.2465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rossi EA, Roorda A. Is visual resolution after adaptive optics correction susceptible to perceptual learning? Journal of Vision. 2010b;10(12):11, 1–14. doi: 10.1167/10.12.11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rossi EA, Chung M, Dubra A, Hunter JJ, Merigan WH, Williams DR. Imaging retinal mosaics in the living eye. Eye. 2011;25:301–308. doi: 10.1038/eye.2010.221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sawides L, de Gracia P, Dorronsoro C, Webster M, Marcos S. Adapting to blur produced by ocular high-order aberraations. Journal of Vision. 2011;11(7):21, 1–11. doi: 10.1167/11.7.21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sincich LC, Zhang Y, Tiruveedhula P, Horton JC, Roorda A. Resolving single cone inputs to visual receptive fields. Nature Neuroscience. 2009;12(8):967–969. doi: 10.1038/nn.2352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sliney DH, Wangemann RT, Franks JK, Wolbarsht ML. Visual sensitivity of the eye to infrared laser radiation. Journal of the Optical Society of America. 1988;66(4):339–342. doi: 10.1364/josa.66.000339. [DOI] [PubMed] [Google Scholar]
  47. Vasilenko LS, Chebotayev VP, troitskiy YV. O visual'nom nablyudenii infrakrasnogo izlucheniya opticheskogo kvantovogo generator. Translation by: NASA Center for Aerospace Information 1966 Zhurnal Eksperimental'noy i Teoreticheskoy Fiziki. 1965;48:777–778. [Google Scholar]
  48. Yoon GY, Williams DR. Visual performance after correcting the monochromatic and chromatic aberrations of the eye. Journal of the Optical Society of America A. 2002;19(2):266–275. doi: 10.1364/josaa.19.000266. [DOI] [PubMed] [Google Scholar]
  49. Zagers NPA, van de Kraats J, Berendschot TTJM, van Norren D. Simultaneous measurement of foveal spectral reflectance and cone-photoreceptor directionality. Applied Optics. 2002;41(22):4686–4696. doi: 10.1364/ao.41.004686. [DOI] [PubMed] [Google Scholar]
  50. Zaidi Q, Pokorny J. Appearance of pulsed infrared light: second harmonic generation in the eye. Applied Optics. 1988;27(6):1064–1068. doi: 10.1364/AO.27.001064. [DOI] [PubMed] [Google Scholar]

RESOURCES