Loss of central vision caused by macular degeneration affects millions of individuals, and its prevalence is expected to increase as the population ages. Many of the affected individuals develop central scotomas and can no longer use their fovea for fixation. In these cases, a fixation locus in the peripheral retina is commonly used as a pseudofovea (preferred retinal locus [PRL]).1–3 There are considerable deficits when using the peripheral retina for vision, as recently reviewed by Strasburger et al.4 In addition, other visual processes are limited by the cortical representation of the peripheral visual field, and these also contribute to poorer visual performance at eccentric retinal locations. These perceptual abnormalities include position uncertainty,5 confusion of item sequence,6 crowding interference and deficits in spatial phase resolution, and abnormal symmetry detection.7
As a result of the decreasing visual representations with increasing eccentricity, altered pattern recognition and deficits in obtaining contextual information are commonly reported in patients with central retinal disease. In individuals with age-related macular degeneration (AMD), one consequence of using a PRL is a deficit in face perception.8–10 Although sensory losses may contribute to the deficits in processing faces,11–14 previous work has reported that some aspects of face perception have a stronger relationship with reading acuity and continuous text reading9,10,13 than with static letter acuity or contrast sensitivity alone. These assertions led us to consider other factors that could play a role in decreased face perception in patients with AMD. It has been extensively documented that abnormal eye movement control is associated with eccentric viewing.15–17 Unstable fixation has been associated with poorer performance on complex tasks, such as reading. For example, Crossland et al.18 reported a linear relationship between fixation stability and reading speed in subjects with newly developed eccentric viewing. We have also reported a significant relationship between decreased reading speeds and decreased fixation stability in a group of subjects with nonexudative AMD.19 It may be that the difficulties in face perception that individuals with AMD report are related to altered scanning behavior. In support of this, difficulties with face perception are common in other groups of subjects, including patients with autism, social phobias, and schizophrenia, and it has been documented that many of these individuals have abnormal scanning patterns when viewing faces.20–24 In the present study, we examined the fixation patterns of patients with AMD when viewing an image of a face.
Nine normally sighted control subjects were recruited from the staff of Lighthouse International and New York Eye and Ear Infirmary. Nine individuals with AMD and eccentric fixation were recruited from the clinic at Lighthouse International and from the practice of one of the authors (R.R.). Patients with other major ophthalmologic problems were excluded. All subjects were screened to exclude those with neurologic disease, moderate to severe media opacities, and/or cognitive impairments (e.g., those with a Mini Mental Status Examination score <25). The study complied fully with the Declaration of Helsinki, and written informed consent was obtained from all subjects involved in the study. This study was approved by Lighthouse International’s and New York Eye and Ear Infirmary’s institutional review boards.
Clinical Vision Tests
Visual acuity was measured using an ETDRS (Early Treatment Diabetic Retinopathy Study) acuity chart. Spectral optical coherence tomography (OCT) images were obtained using an OTI-OPKO Health, Inc. SLO/OCT. The OPKO instrument provides images of the fundus using a scanning laser ophthalmoscope. The SLO and spectral OCT images are obtained through the confocal optics. The confocal optics allows pixel-to-pixel correspondence of the images, ensuring precise registration and orientation. We also used the microperimetry feature of the device to collect spot increment thresholds at local retinal locations. An OLED screen was used to present perimetry targets while collecting SLO images of the subjects’ fundus. Real-time fundus tracking was used to adjust the stimulus locations to ensure presentation of the perimetry stimuli at the intended retinal locations. The fundus tracking algorithm was also used to locate the position and stability of fixation.
In the present work, we used the various capabilities of the OPKO machine to collect data on the structural and functional status of each subject’s retina. Initially, SLO images and OCT topographies were obtained. Fig. 1A shows the images for one of the control subjects, and Fig. 2A shows the images for two of the subjects with AMD. Three-dimensional topographies were obtained, and the locus of the anatomic fovea was determined in relation to visible lesions (Fig. 3A). Next, we used the fixation tracking capability of the OPKO instrument to quantify the position and stability of each patient’s fixation. A “cross” was presented at the center of the OLED screen, and the subject was asked to fixate this target for a period of 20 seconds. The positions of fixations were calculated using the fundus tracking algorithm.
The frame rate of the SLO was 8 frames per second (fps) (125 milliseconds scan time). During the 20-second fixation epoch, a total of 160 data points could potentially be collected. In all but two subjects, more than 95% of the points were recorded (median, 97.5%). Although 8 fps is a low sampling rate compared with that used by pupil image eye trackers, one must consider the timing of the events that we are measuring. Average fixation duration is approximately 200 to 300 milliseconds, and there is an average of 2 to 3 saccades per second, depending on the stimulus and the task.25–27 Longer fixation durations are observed when viewing faces.28 A sampling rate of 8 fps is well above the Nyquist frequency for sampling long-duration fixations. Although pupil image eye trackers can have much higher sampling rates, measuring multiple data points at essentially the same fixation locus would not alter the calculated bivariate contour ellipse area (BCEA). We also examined data to determine the effect of the number of data points on the calculated BCEA. Using an AMD patient’s data sampled at 8 fps, the BCEA for the first 25% of the points was 3.8 log minutes of arc squared; for the second 25% of the points, it was 3.6; for the third, it is was 3.6; and for the last 25% of the points, it was 3.7. The area calculated for all of the points was 3.7 log minutes of arc squared. In further support, we did not find a significant correlation in our data between the number of points and BCEA across all individuals (r = 0.31, p = 0.41).
The positions of fixations were mapped onto the image of the retina (Figs. 1B, 2B, and 3B). This allowed us to locate the area of the retina used for fixation by each subject (fovea in control subjects or PRL in AMD patients). Using the digitized X and Y positions of each fixation, we fitted a bivariate contour ellipse to each patient’s fixation data to quantify fixation stability 18,29,30 and calculated the 2-D means to define the retinal position of the PRL. Using this information, we calculated the eccentricity of the PRL relative to the anatomic fovea (Fig. 3C). In our sample of patients with relatively good acuity (≤0.9 logMAR), we could identify the location of the anatomic fovea with reasonable certainty. Determination of location was aided by identifying the probable location of the fovea based on identifying a region of interest based on the relative location of the optic disc (as previously done in Sunness et al.,31 Timberlake et al.,32 and Seiple et al.33). To do this, we measured the lateral position of the anatomic fovea relative to the bottom center of the optic disk in 10 control subjects. The average position of the foveal pit in this group of normal subjects was 15.1 (±0.83) degrees temporal to the disk. When attempting to locate the fovea in our patients, we used this measure as the starting point for the search. An OCT b-scan through the PRL (Fig. 3D) shows that fixation for this individual with AMD is located at a point where an intact inner segment/outer segment junction is closest to the fovea.
After structural and functional data collection, we mapped fundus movements when the subject viewed an image of a face. The subject was instructed to fixate an X drawn at the center of the OLED screen. The subject was then told that an image would appear on the screen and was instructed to “look at the image.” A color image of the face of the painting “Mona Lisa” by Leonardo da Vinci was then displayed on the OLED. The image of the face subtended 8 degrees, which is approximately the size of a face at a distance of 1 m (conversation distance). During the next 16 seconds, a SLO movie of fundus movements was recorded (at 8 fps). Offline, the image of the face and the SLO fundus movie were overlaid in Adobe Premiere Elements 8.0 (Figs. 1C and 2C). On each frame of the movie, the location of fixation (fovea or PRL) was recorded. The locations of all fixations for a control subject and for two AMD subjects are plotted in Figs. 1D and 2D, respectively. Although using a SLO movie as an eye tracker does not allow for the temporal precision of standard eye-tracking devices, it does eliminate the possible confounds caused by assuming, first, that one knows the retinal locus of fixation and second, that this same retinal locus is used for fixation during calibration. These assumptions are inherent in eye trackers that use corneal reflections. This is of special concern when attempting to follow the eye movements of individuals with central scotomas and unsteady fixation. The SLO allows direct view of fundus movements, and therefore, the location of diseased retina can be precisely mapped on each frame.
Table 1 lists the clinical data for the nine individuals with AMD who participated in this study (four females and five males). Seven of the subjects had exudative AMD, and two had dry AMD. The average age of these subjects was 75 years (range, 61 to 87 years), and the average acuity of this group was 0.48 logMAR. All subjects had eccentric PRLs, as confirmed with the OPKO instrument. The PRL eccentricities ranged from 0.5 to 4.1 degrees from the patients’ anatomic fovea.
The mean fixation stability (log minutes of arc squared) of the patients was significantly poorer than that of the control subjects (patients = 3.09 log minutes of arc squared (range, 2.3 to 3.7); control = 2.13 log minutes of arc squared (range, 1.77 to 2.45); t = 3.27, p = 0.007). The two patients who had “normal” visual acuity had the smallest BCEA of this sample, 2.6 and 2.3 log minutes of arc squared. Within the patient group, stability was significantly related to acuity (r = 0.73, p = 0.025); increased fixation stability was associated with better acuity. Stability was also related to PRL eccentricity, with decreasing stability as a function of increasing eccentricity (r = 0.79, p = 0.01). We also found a significant relationship between acuity and PRL eccentricity (r = 0.85, p = 0.003), with decreasing acuity related to increasing eccentricity. However, fixation stability was not significantly related to the percent of fixations on external features of the face (r = 0.13, p = 0.73)
To analyze fixations, we binned fixation location by areas of the face: internal features (eyes, nose, and mouth) and external features (not eyes, nose, or mouth) (Fig. 4A). The AMD subjects had fewer fixations on the internal features and more fixations on the external features than those in the control subjects. The percent of fixations on each facial area is summarized for each group in Fig. 4B. A two-way repeated-measures analysis of variance found significant differences between the control and patient groups (F = 9.00, p = 0.04) and among locations (F = 6.01, p = 0.01). A significant interaction between group and location was also found (F = 5.25, p = 0.01). Post hoc comparison found a significantly greater proportion of fixations on external features for the AMD group than for the control group (p < 0.001). However, age, acuity, PRL eccentricity, and fixation stability were not related to the percentage of fixation on external features.
We next compared the order of fixations for each subject. Examples of these data are shown for two control subjects and two subjects with AMD in Fig. 5. These examples were chosen to illustrate the range of eye movements in each group. In general, there were fewer and shorter saccades for the control subjects than for the AMD patients. To quantify these movements, we plotted the direction and distance of each saccade from the point of last fixation. Examples of this process are shown in Fig. 6 for the same two control subjects and the same two patients. We hypothesized that control subjects would have a greater number of vertical and horizontal saccades along the axes of the salient features of the face and that AMD subjects would have a uniform distribution of saccade angles (especially given the results of Yu and Chung34). The angle of each saccade was measured and binned into horizontal, oblique, or vertical movements. Both groups had approximately one-half oblique saccades and one-half horizontal and vertical saccades (Fig. 7). There were no significant differences in the distributions of the angles of saccades between the two groups. The control group averaged significantly fewer saccades (control = 22.2 ± 5.7, AMD = 37.1 ± 8.7; p = 0.008) and significantly shorter saccade lengths than those in the patient group (p < 0.05). Average fixation durations were 26% longer for the control group (t =1.99, p = 0.06). For both groups, average fixation durations on the nose were 19.6% and on the mouth 25.7% longer than those on the eyes.
We tested for the effects of reduced acuity of the fixation patterns using four control subjects. We were not attempting to simulate disease effects with this manipulation but merely to assess the effects of optical blur on the fixation patterns of control subjects. This was done to eliminate optic blur as a cause of the differences in fixation patterns between the two groups of subjects. The image of the face was Gaussian blurred to an equivalent acuity level of 20/60 and 20/100. For both blurred images, fixations remained centered on the internal features of the face. The locations of fixation for one control subject when viewing this degraded image are shown in Fig. 8. The results were similar for the three additional control subjects tested with these blurred images.
One explanation for the differing distribution of fixation locations is that the AMD subjects were using multiple PRLs while scanning the image of the face. It has been documented that the position of a subject’s PRL can change depending on the requirement of the task.35,36 Generally, fixation behavior has been examined using pupil reflection eye trackers, and calibrations have been made assuming a single fixation locus (either the fovea or the PRL).35,37–39 The subject is given a single target to view, and the position of the calibrated fixation locus is quantified during the course of viewing. This is a valid method when there is only a single target in the field and when one assumes that this target is being fixated at all times by the calibrated locus. In the current work, we used a single stimulus paradigm to measure the locus of fixation (Figs. 1B and 2B). In this paradigm, we observed the use of a single PRL for both control subjects and patients because the calibrated fixation locus fell on or near the cross fixation target on every fixation. If a subject was to use multiple fixation loci, a bimodal (or multimodal) distribution of fixations would be observed. We have demonstrated this by instructing a control subject to alternately view the fixation cross with his fovea and with a retinal locus inferior to the fovea. Under these viewing conditions, a bimodal distribution of fixations was recorded (e.g., multiple PRLs; Fig. 9A).
We also wished to examine our data for the possible use of multiple loci when viewing the image of a face. In control subjects, we located the position of the fovea on each frame of the SLO movie and then marked the region of the face that was imaged by the fovea (i.e., eyes, red dots; nose, green dot; and mouth, black dot). For the control subject shown in Fig. 9B, all of the features (eyes, nose, and mouth) were imaged by the fovea (central clump of points at the fundus location of the fovea). In other words, as expected, the subject used a single PRL (the fovea) to view the image.
A similar exercise could be done for the subjects with AMD using the fixation locus measured in the cross fixation task. However, we cannot assume the locus of fixation or the locus of attention in the same way we can with control subjects. The image of the face has multiple areas of interest, and if the predetermined PRL does not fall at a particular internal feature of the face, we cannot know if the subject is viewing a different feature with a different retinal locus. To test this, we plotted the retinal locations of all internal features of the face on each frame. If multiple PRLs were used, we expected to find two or more grouping of fixations. For example, if the predetermined PRL was used to view the eyes and mouth and a second locus was used for viewing the nose, fixation data similar to that shown on Fig. 9C would be recorded. For the AMD patients, we failed to find a single locus on the retina at which all relevant facial features were imaged (Fig. 9D). These findings do not support the use of multiple PRLs when viewing this image of a face.
This work has demonstrated that subjects with AMD show a pattern of scanning an image of a face that is different from that of control subjects. For the control subjects, an average of 87% of the fixations were on internal features (eyes, nose, and mouth) and only 13% on external features. The AMD subjects had significantly fewer fixations on the internal features (average of 62%) and significantly more fixations on the external features (38%) than the control subjects. There is an enormous amount of literature devoted to exploring the particular effects of parametric manipulations of face stimuli and instructions to the subjects on eye movement behavior. The literature is filled with studies using faces with different emotions, images with hair or with hair masked, faces that have been spatial frequency filtered, and so on. In these past studies, the subject’s tasks have ranged from recognition of familiar faces to face-matching tasks, and various instructions have been given to subjects. In the current study, we have shown that, when control subjects and patients were given the same instructions, had no practice trials, and were shown an image of a well-known face, individuals with AMD have a different scanning behavior from control subjects.
Difficulties with face perception are commonly reported by individuals with AMD. Among the possible explanations for diminished face perception in individuals with central retinal disease is a loss of sensory capabilities; that is, individuals with AMD who use a PRL have visual acuity and contrast sensitivity losses that may contribute to altered face perception. In support of the sensory locus of reduced face perception, Tejeria et al.10 reported that performance on a familiar face-recognition task was correlated with distance acuity. Bullimore et al.9 reported that normally sighted subjects could recognize faces at a distance that was 10 times farther than individuals with AMD could. McCulloch et al.11 reported that, when detecting a simplified face stimulus, performance decreased as a function of reduced visual acuity and elevated contrast thresholds. Barnes et al.12 used a matching task to quantify face perception in individuals with AMD. They reported that individuals with AMD were significantly less accurate on a face-matching task than age-similar control subjects and that accuracy decreased with reductions in visual acuity.
However, other data suggest a weaker relationship between sensory capabilities and face perception. Alexander et al.8 measured face expression in a group of individuals with AMD and concluded that visual acuity had only a weak correlation with the ability to see faces. Tejeria et al.10 reported a stronger relationship between performance on a facial expression–discrimination task and continuous text reading acuity than with static acuity, and Bullimore et al.9 reported that word reading acuity was the best predictor of face-recognition performance. Thompson et al.40 found that face-recognition accuracy was maintained with crude pixelized vision. Similarly, the role of contrast sensitivity in face perception is not always clear. Owsley et al.14 reported that, although clinically measured contrast threshold elevations were related to perception, contrast sensitivity losses alone were not always significantly related to face perception. Lott et al.13 reported that high contrast acuity and age best predicted face recognition; however, these authors also reported that adding contrast to the multiple regression analysis did not increase the R2 value. Bullimore et al.9 found that face recognition thresholds had a low correlation with contrast sensitivity, and Rubin and Schuchard41 reported no significant correlation between contrast sensitivity and face recognition. Furthermore, Rizzo et al.42 found that impairments in contrast sensitivity were not related to the development of prosopagnosia.
Based on the lack of strong singular evidence for the role of sensory losses in face perception difficulties, we hypothesized that deficits in eye movement control observed in individuals with AMD might play a role in face perception. Eye movement control is very poor when using an eccentric fixation locus. Fixation stability decreases and saccade rate and drift velocity increase with eccentricity.1,15,16,43 The eye movement system develops as a mechanism for placing an object on the fovea for inspection of fine details (foveation), and appropriate eye movements are necessary for the analysis of simple patterns.44,45 An important aspect of vision perception is moving fovea to image stimuli that are relevant to ongoing cognitive and behavioral activities.46,47 When the fovea is diseased and a PRL is used for inspection, a new oculomotor strategy must be adopted. This is not easily accomplished. White and Bedell48 reported that only one of 10 individuals with AMD shifted the reference point of eye movements to their PRL and that oculomotor control improved with duration of disease onset. Lingnau et al.49 demonstrated that reading performance was better when eccentric gaze and attention were at the same retinal location, suggesting that adaptation to the use of eccentric viewing locus requires attention and eye movement control remapping.
Eye movements are an integral part of pattern perception. Pattern recognition, in general, is related to an organized sequence of fixations. Face perception, in particular, has been related to a stereotypic pattern of fixations on the internal features of the face.47,50,51 The observed patterns of fixation in our control subjects were typical of those reported in the literature for normally sighted individuals.52–54 The observed patterns of fixations of our subjects with AMD were similar to those observed in other types of patients who have a significantly greater proportion of fixations on external features of faces associated with difficulties in face perception, including individuals with social phobias, Williams syndrome, autism, schizophrenia, or prosopagnosia.20–24
A question remains about the locus of abnormal oculomotor performance in individuals who use eccentric retina for fixation. A complex distributed cortical network is involved in processing faces; this network includes the inferior occipital gyrus, the fusiform gyrus, the superior temporal sulcus, the hippocampus, the amygdala, the inferior frontal gyrus, and the orbitofrontal cortex.55,56 Because of the involvement of multiple cortical areas, inappropriate patterns of fixations could result from a lack of sensory information on which to base the next saccade (“bottom up,” salience16,43) and/or from abnormal cognitive control of attention (“top down,” attention).57
In support of a “bottom up” influence on perception and eye movements, Pitcher et al.58 reported that areas of the visual cortex, specifically the occipital face area, preferentially code the eyes, nose, and mouth of face stimuli and that an intact occipital face area is necessary for accurate face perception. Published work, as a whole, shows a decrease in face perception with eccentricity in line with losses of sensory processing in the peripheral visual system.59–63 However, the rate of decline depends on the spatial frequency and contrast demands of the task. For example, Bayle et al.59 found that the reaction times to discriminate sex were longer than those to detect expression, and that the difference increased with increasing eccentricity. The differences were only statistically significant for eccentricity greater than 30 degrees. Although they did not examine this, the different losses may be caused by the spatial frequency demands underlying the task.
Another aspect to consider is the interplay between sensory input and perceptual processing. We and others have explored this in the past, with the hypothesis that even minor alterations in sensory input could alter perception. We found that individuals with retinitis pigmentosa (RP) who had visual acuities of 20/30 or better had abnormal symmetry discrimination.7 We ruled out sensory explanations for this deficit and suggested that even minor alterations of sensory input may affect perceptual encoding. In further support, Alexander et al.8 reported that individuals with RP who had only minor reductions in their visual acuity could have severely compromised motion perception, and Turano and Schuchard64 reported that some individuals with RP had poor spatial/position precision on a bisection task that could not be predicted by their visual acuities. Similarly, Temme et al.65 reported an abnormal perceptual magnification of the central visual field in some individuals with retinal disease.
The data of Dulin et al.66 support an exaggerated effect of nonspecific sensory losses on perception. They investigated how specific input alterations affected complex cognitive performance. Performance on recognition of famous faces was compared between a group of subjects with RP who had severe peripheral field losses and good acuities and a group of Stargardt dystrophy subjects who had central scotomas and poor acuities. Although both groups of patients had significantly poorer face recognition than that of the control group, performance did not differ significantly between patient groups. Despite vastly different sensory losses, the resulting perceptual performance was similar.
Abnormal perception may result in abnormal eye movements. The role of perceptual/cognitive control of attention (“top down,” attention) is emphasized by the findings that individuals with affective disorders may have face perception abnormalities and abnormal scanning behavior despite intact visual sensory input. Numerous studies of “top down” influences can be found in the literature (see recent reviews by Tatler et al.67 and Schütz et al.68 In addition, the pattern of scanning has been shown to be altered by the nature of the task to be performed, cognitive demand, and mood, as well as the particulars of the visual stimulus.54,69–74 It also has been suggested that the generation of saccades is preceded by a shift in covert attention.75 Manipulation of spatial attention has been shown to influence the cortical responses to faces with neutral expressions.76 Because the allocation of covert attention decreases with increasing eccentricity,77,78 generation of saccades based on allocation of attention in AMD patients with eccentric PRLs may be abnormal. The role of an attention component in scanning has been demonstrated by Schmalzl et al.79 These authors reported that training by attracting attention to specific characteristics of the internal features of a face improved performance on face recognition and increased the percentage of fixations directed at the internal features. That report gives hope that eye movement control training and training of allocation of attention could improve face perception and eye scanning behavior in individuals with AMD.
111 East 59th St
New York, NY 10022
This work was supported in part by a grant from the US Department of Veterans Affairs and by the Bendheim Family Retina Center of the New York Eye and Ear Infirmary.
Received June 6, 2012; accepted October 1, 2012.
1. Whittaker SG, Budd J, Cummings RW. Eccentric fixation with macular scotoma. Invest Ophthalmol Vis Sci 1988; 29: 268–78.
2. Timberlake GT, Mainster MA, Peli E, Augliere RA, Essock EA, Arend LE. Reading with a macular scotoma. I. Retinal location of scotoma and fixation area. Invest Ophthalmol Vis Sci 1986; 27: 1137–47.
3. Crossland MD, Culham LE, Kabanarou SA, Rubin GS. Preferred retinal locus development in patients with macular disease. Ophthalmology 2005; 112: 1579–85.
4. Strasburger H, Rentschler I, Juttner M. Peripheral vision and pattern recognition: a review. J Vis 2011; 11: 1–82.
5. Levi DM, Klein SA, Yap YL. Positional uncertainty in peripheral and amblyopic vision. Vision Res 1987; 27: 581–97.
6. Hess RF, McCarthy J. Topological disorder in peripheral vision. Vis Neurosci 1994; 11: 1033–6.
7. Szlyk JP, Seiple W, Xie W. Symmetry discrimination in patients with retinitis pigmentosa. Vision Res 1995; 35: 1633–40.
8. Alexander MF, Maguire MG, Lietman TM, Snyder JR, Elman MJ, Fine SL. Assessment of visual function in patients with age-related macular degeneration and low visual acuity. Arch Ophthalmol 1988; 106: 1543–7.
9. Bullimore MA, Bailey IL, Wacker RT. Face recognition in age-related maculopathy. Invest Ophthalmol Vis Sci 1991; 32: 2020–9.
10. Tejeria L, Harper RA, Artes PH, Dickinson CM. Face recognition in age-related macular degeneration: perceived disability, measured disability, and performance with a bioptic device. Br J Ophthalmol 2002; 86: 1019–26.
11. McCulloch DL, Loffler G, Colquhoun K, Bruce N, Dutton GN, Bach M. The effects of visual degradation on face discrimination. Ophthalmic Physiol Opt 2011; 31: 240–8.
12. Barnes CS, De L’Aune W, Schuchard RA. A test of face discrimination ability in aging and vision loss. Optom Vis Sci 2011; 88: 188–99.
13. Lott LA, Haegerstrom-Portnoy G, Schneck ME, Brabyn JA. Face recognition in the elderly. Optom Vis Sci 2005; 82: 874–81.
14. Owsley C, Sekuler R, Boldt C. Aging and low-contrast vision: face perception. Invest Ophthalmol Vis Sci 1981; 21: 362–5.
15. McMahon TT, Hansen M, Stelmack J, Oliver P, Viana MA. Saccadic eye movements as a measure of the effect of low vision rehabilitation on reading rate. Optom Vis Sci 1993; 70: 506–10.
16. McMahon TT, Hansen M, Viana M. Fixation characteristics in macular disease. Relationship between saccadic frequency, sequencing, and reading rate. Invest Ophthalmol Vis Sci 1991; 32: 567–74.
17. Timberlake GT, Peli E, Essock EA, Augliere RA. Reading with a macular scotoma. II. Retinal locus for scanning text. Invest Ophthalmol Vis Sci 1987; 28: 1268–74.
18. Crossland MD, Culham LE, Rubin GS. Fixation stability and reading speed in patients with newly developed macular disease. Ophthalmic Physiol Opt 2004; 24: 327–33.
19. Seiple W, Grant P, Szlyk JP. Reading rehabilitation of individuals with AMD: relative effectiveness of training approaches. Invest Ophthalmol Vis Sci 2011; 52: 2938–44.
20. Stephan BC, Caine D. Aberrant pattern of scanning in prosopagnosia reflects impaired face processing. Brain Cogn 2009; 69: 262–8.
21. Riby DM, Doherty-Sneddon G, Bruce V. The eyes or the mouth? Feature salience and unfamiliar face processing in Williams syndrome and autism. Q J Exp Psychol (Hove) 2009; 62: 189–203.
22. Manor BR, Gordon E, Williams LM, Rennie CJ, Bahramali H, Latimer CR, Barry RJ, Meares RA. Eye movements reflect impaired face processing in patients with schizophrenia. Biol Psychiatry 1999; 46: 963–9.
23. Loughland CM, Williams LM, Gordon E. Schizophrenia and affective disorder show different visual scanning behavior for faces: a trait versus state-based distinction? Biol Psychiatry 2002; 52: 338–48.
24. Horley K, Williams LM, Gonsalvez C, Gordon E. Face to face: visual scanpath evidence for abnormal processing of facial expressions in social phobia. Psychiatry Res 2004; 127: 43–53.
25. Hooge IT, Erkelens CJ. Control of fixation duration in a simple search task. Percept Psychophys 1996; 58: 969–76.
26. Salthouse TA, Ellis CL. Determinants of eye-fixation duration. Am J Psychol 1980; 93: 207–34.
27. Sereno SC, Rayner K. The when and where of reading in the brain. Brain Cogn 2000; 42: 78–81.
28. Guo K, Mahmoodi S, Robertson RG, Young MP. Longer fixation duration while viewing face images. Exp Brain Res 2006; 171: 91–8.
29. Timberlake GT, Sharma MK, Grose SA, Gobert DV, Gauch JM, Maino JH. Retinal location of the preferred retinal locus relative to the fovea in scanning laser ophthalmoscope images. Optom Vis Sci 2005; 82: 177–85.
30. Steinman RM. Effect of target size, luminance, and color on monocular fixation. J Opt Soc Am. 1965; 55: 1158–65.
31. Sunness JS, Applegate CA, Haselwood D, Rubin GS. Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt disease. Ophthalmology 1996; 103: 1458–66.
32. Timberlake GT, Sharma MK, Grose SA, Maino JH. Retinal locus for scanning text. J Rehabil Res Dev 2006; 43: 749–60.
33. Seiple W, Szlyk JP, McMahon T, Pulido J, Fishman GA. Eye-movement training for reading in patients with age-related macular degeneration. Invest Ophthalmol Vis Sci 2005; 46: 2886–96.
34. Yu D, Chung ST. Critical orientation for face identification in central vision loss. Optom Vis Sci 2011; 88: 724–32.
35. Duret F, Issenhuth M, Safran AB. Combined use of several preferred retinal loci in patients with macular disorders when reading single words. Vision Res 1999; 39: 873–9.
36. Lei H, Schuchard RA. Using two preferred retinal loci for different lighting conditions in patients with central scotomas. Invest Ophthalmol Vis Sci 1997; 38: 1812–8.
37. Crossland MD, Crabb DP, Rubin GS. Task-specific fixation behavior in macular disease. Invest Ophthalmol Vis Sci 2011; 52: 411–6.
38. Tarita-Nistor L, Gonzalez EG, Markowitz SN, Steinbach MJ. Fixation characteristics of patients with macular degeneration recorded with the mp-1 microperimeter. Retina 2008; 28: 125–33.
39. Crossland MD, Sims M, Galbraith RF, Rubin GS. Evaluation of a new quantitative technique to assess the number and extent of preferred retinal loci in macular disease. Vision Res 2004; 44: 1537–46.
40. Thompson RW Jr., Barnett GD, Humayun MS, Dagnelie G. Facial recognition using simulated prosthetic pixelized vision. Invest Ophthalmol Vis Sci 2003; 44: 5035–42.
41. Rubin GS, Schuchard RA. Does Contrast Sensitivity Predict Face Recognition Performance In Low-Vision Observers? Noninvasive Assessment of the Visual System, OSA Technical Digest Series, vol. 3. Washington, DC: Optical Society of America; 1990: 130–7.
42. Rizzo M, Corbett JJ, Thompson HS, Damasio AR. Spatial contrast sensitivity in facial recognition. Neurology 1986; 36: 1254–6.
43. Whittaker SG, Cummings RW, Swieson LR. Saccade control without a fovea. Vision Res 1991; 31: 2209–18.
44. Schlingensiepen KH, Campbell FW, Legge GE, Walker TD. The importance of eye movements in the analysis of simple patterns. Vision Res 1986; 26: 1111–7.
45. Kowler E, Steinman RM. The role of small saccades in counting. Vision Res 1977; 17: 141–6.
46. Findlay JM, Brown V, Gilchrist ID. Saccade target selection in visual search: the effect of information from the previous fixation. Vision Res 2001; 41: 87–95.
47. Henderson JM, Williams CC, Falk RJ. Eye movements are functional during face learning. Mem Cognit 2005; 33: 98–106.
48. White JM, Bedell HE. The oculomotor reference in humans with bilateral macular disease. Invest Ophthalmol Vis Sci 1990; 31: 1149–61.
49. Lingnau A, Schwarzbach J, Vorberg D. Adaptive strategies for reading with a forced retinal location. J Vis 2008; 8: 6.1–18.
50. Mertens I, Siegmund H, Grusser OJ. Gaze motor asymmetries in the perception of faces during a memory task. Neuropsychologia 1993; 31: 989–98.
51. Rizzo M, Hurtig R, Damasio AR. The role of scanpaths in facial recognition and learning. Ann Neurol 1987; 22: 41–5.
52. Althoff RR, Cohen NJ. Eye movement–based memory effect: a reprocessing effect in face perception. J Exp Psychol Learn Mem Cogn 1999; 25: 997–1010.
53. Keating CF, Keating EG. Visual scan patterns of rhesus monkeys viewing faces. Perception 1982; 11: 211–9.
54. Walker-Smith GJ, Gale AG, Findlay JM. Eye movement strategies involved in face perception. Perception 1977; 6: 313–26.
55. Ishai A, Schmidt CF, Boesiger P. Face perception is mediated by a distributed cortical network. Brain Res Bull 2005; 67: 87–93.
56. Haxby JV, Petit L, Ungerleider LG, Courtney SM. Distinguishing the functional roles of multiple regions in distributed neural systems for visual working memory. Neuroimage 2000; 11: 380–91.
57. Loomis JM. Lateral masking in foveal and eccentric vision. Vision Res 1978; 18: 335–8.
58. Pitcher D, Walsh V, Duchaine B. The role of the occipital face area in the cortical face perception network. Exp Brain Res 2007; 209: 481–93.
59. Bayle DJ, Schoendorff B, Henaff MA, Krolak-Salmon P. Emotional facial expression detection in the peripheral visual field. PLoS One 2011; 6: e21584.
60. Guo K, Liu CH, Roebuck H. I know you are beautiful even without looking at you: discrimination of facial beauty in peripheral vision. Perception 2011; 40: 191–5.
61. Gutierrez A, Nummenmaa L, Calvo MG. Enhanced processing of emotional gist in peripheral vision. Span J Psychol 2009; 12: 414–23.
62. Makela P, Nasanen R, Rovamo J, Melmoth D. Identification of facial images in peripheral vision. Vision Res 2001; 41: 599–610.
63. McKone E. Isolating the special component of face recognition: peripheral identification and a Mooney face. J Exp Psychol Learn Mem Cogn 2004; 30: 181–97.
64. Turano K, Schuchard RA. Space perception in observers with visual field loss. Clin Vis Sci 1991; 6: 289–99.
65. Temme LA, Maino JH, Noell WK. Eccentricity perception in the periphery of normal observers and those with retinitis pigmentosa. Am J Optom Physiol Opt 1985; 62: 736–43.
66. Dulin D, Cavezian C, Serriere C, Bachoud-Levi AC, Bartolomeo P, Chokron S. Colour, face, and visuospatial imagery abilities in low-vision individuals with visual field deficits. Q J Exp Psychol (Hove) 2011; 64: 1955–70.
67. Tatler BW, Hayhoe MM, Land MF, Ballard DH. Eye guidance in natural vision: reinterpreting salience. J Vis 2011; 11: 5.
68. Schütz AC, Braun DI, Gegenfurtner KR. Eye movements and perception: a selective review. J Vis 2011; 5: 11.1–30.
69. Yarbus AL. Eye Movements and Vision. New York, NY: Plenum Press; 1967.
70. Buchan JN, Pare M, Munhall KG. Spatial statistics of gaze fixations during dynamic face processing. Soc Neurosci 2007; 2: 1–13.
71. Castelhano MS, Mack ML, Henderson JM. Viewing task influences eye movement control during active scene perception. J Vis 2009; 9: 6.1–15.
72. Isaacowitz DM, Toner K, Neupert SD. Use of gaze for real-time mood regulation: effects of age and attentional functioning. Psychol Aging 2009; 24: 989–94.
73. Vuilleumier P, Armony JL, Driver J, Dolan RJ. Effects of attention and emotion on face processing in the human brain: an event-related fMRI study. Neuron 2001; 30: 829–41.
74. Ellis HD, Shepherd JW, Davies GM. Identification of familiar and unfamiliar faces from internal and external features: some implications for theories of face recognition. Perception 1979; 8: 431–9.
75. Deubel H, Schneider WX. Delayed saccades, but not delayed manual aiming movements, require visual attention shifts. Ann N Y Acad Sci 2003; 1004: 289–96.
76. Wojciulik E, Kanwisher N, Driver J. Covert visual attention modulates face-specific activity in the human fusiform gyrus: fMRI study. J Neurophysiol 1998; 79: 1574–8.
77. Mackeben M. Sustained focal attention and peripheral letter recognition. Spat Vis 1999; 12: 51–72.
78. Seiple W, Clemens C, Greenstein VC, Holopigian K, Zhang X. The spatial distribution of selective attention assessed using the multifocal visual evoked potential. Vision Res 2002; 42: 1513–21.
79. Schmalzl L, Palermo R, Green M, Brunsdon R, Coltheart M. Training of familiar face recognition and visual scan paths for faces in a child with congenital prosopagnosia. Cogn Neuropsychol 2008; 25: 704–29.