Dynamic Interactions of Eye and Head Movements When Reading with Single-Vision and Progressive Lenses in a Simulated Computer-Based Environment

Abstract

purpose. To assess dynamic interactions of eye and head movements during return-sweep saccades (RSS) when reading with single-vision (SVL) versus progressive-addition (PAL) lenses in a simulated computer-based business environment.

methods. Horizontal eye and head movements were recorded objectively and simultaneously at a rate of 60 Hz during reading of single-page (SP; 14° horizontal [H]) and double-page (DP; 37° H) formats at 60 cm with binocular viewing. Subjects included 11 individuals with normal presbyopic vision aged 45 to 71 years selected by convenience sampling from a clinic population. Reading was performed with three types of spectacle lenses with a different clear near field of view (FOV): a SVL (60° H clear FOV), a PAL-I with a relatively wide intermediate zone (7.85 mm; 18° H clear FOV), and a PAL-II with a relatively narrow intermediate zone (5.60 mm; 13° H clear FOV).

results. Eye movements were initiated before head movements in the SP condition, and the reverse was found in the DP condition, with all three lens types. Duration of eye movements increased as the zone of clear vision decreased in the SP condition, and they were longer with the PALs than with the SVL in the DP condition. Gaze stabilization occurred later with the PALs than with the SVL in both the SP and DP conditions. The duration of head movements was longer with the PAL-II than with the SVL in both the SP and DP conditions. Eye movement peak velocity was greater with the SVL than the PALs in the DP condition.

conclusions. Eye movement and head movement strategies and timing were contingent on viewing conditions. The longer eye movement duration and gaze-stabilization times suggested that additional eye movements were needed to locate the clear-vision zone and commence reading after the RSS. Head movements with PALs for the SP condition were similarly optically induced. These eye movement and head movement results may contribute to the reduced reading rate and related symptoms reported by some PAL wearers. The dynamic interactions of eye movements and head movements during reading with the PALs appear to be a sensitive indicator of the effect of lens optical design parameters on overall reading performance, because the movements can discriminate between SVL and PAL designs and at times even between PALs.

Adaptation to and satisfaction with progressive-addition lenses (PALs) have been hampered in part by the restricted zones that may make reading difficult. ^{1 2} This includes the intermediate working distance of the typical computer user. The narrow intermediate zone of clear vision may cause an increase in the frequency of compensatory head movements, which is thought to be a major contributor to nonacceptance of PALs. ^{1 2}  

Few studies have been conducted in which both eye and head movements during reading have been objectively recorded simultaneously, ³ especially in multiple spatial dimensions. In one study ⁴ in which no special spectacle lenses were worn or compared, horizontal head and eye movements were recorded while reading Korean text placed along the midline at a distance of 35 cm. Smooth head movement occurred almost continuously during reading. However, the horizontal text extent was 90°, and hence represented a very unnatural condition that demanded initiation of extremely large eye and head movements. In a preliminary study of eye and head movements in three experienced subjects reading a text that extended 20° horizontally and using no spectacle lenses, ⁵ smooth and slow head movements were found during and immediately after the return-sweep saccade (RSS), lasting approximately 500 msec. Thus, during these periods, the eye moved to the right to read as the subject’s head was still rotating to the left. In a recent study of ours, ⁶ two-dimensional eye movements and three-dimensional head movements were recorded objectively during reading with the conventional single-vision lens (SVL) and two types of PALs with different intermediate zone widths. Single-page (SP) and double-page (DP) text formats of adult-level text were used, similar to that which a reader confronts in the typical contemporary computer-based environment, especially in the business world. In addition to quantification of the standard reading eye movement parameters (e.g., fixation duration, number of regressions), ⁷ frequency and amplitude of head movements were assessed. Several important differences in these static oculomotor and reading test parameters were found between lens conditions and text formats that provided important insights into the reading process and strategies used with PALs. 

We now report on the dynamic eye and head interactions that occurred during the RSS eye movements in the same experiment. 

Materials and Methods

Eye and Head Movement Recording

Eye and head movements during reading were recorded objectively with an integrated eye and head movement computer-based system (RK-726 PCT; ISCAN, Burlington, MA) that has been used by us in other static oculomotor and reading analyses. ⁶ Files with both sets of dynamic data (i.e., instantaneous eye and head position) were numerically identified according to the time recorded, so that temporal aspects across and between the eye and head movements could be compared quantitatively. 

The eye movement system (ETL-400; RK-726 PCT; ISCAN) uses a video-based, dark pupil-to-cornea reflection method. It consists of a head-mounted recording system attached to a stable cap that is fixed with Velcro straps onto the head, an eye imaging monitor, and a personal computer with specialized analysis and graph-producing software. An infrared light-emitting unit and video camera are mounted on the lightweight cap worn by the subject. The eye is illuminated by the infrared light, and the camera records the image of the eye. The video eye image is transmitted simultaneously to the monitor and computer. Differences between the corneal Purkinje image 1 and the pupil center are calculated as changes in eye position. The system can record horizontal and vertical eye movements simultaneously and independently, with a sampling frequency of 60 Hz. Although this moderate level of sampling-frequency may reduce or distort the true time and absolute peak velocity, we were interested in the relative temporal and peak velocity changes across lens type, and hence this lower sampling would not present a problem. Only horizontal eye movements were considered in the present study. The system has a linear range of ±20° with a resolution of 0.3° in both directions. 

The head movement system (Insidetrak; Polhemus, Colchester, VT) involves an electromagnetic field recording method, consisting of an electromagnetic emitting unit and a sensor. The small sensor is attached to the cap, and the emitting unit is placed on a table 40 cm away. Three-dimensional (horizontal, vertical, and torsional) head movements can be recorded simultaneously and independently with a sampling frequency of 60 Hz; however, only horizontal head movements were considered in the present study. The system has a linear range of 360° with a resolution of 0.1° in all three directions. 

Subjects

Subjects included 11 individuals with normal presbyopic vision (age range, 45–71 years) selected by convenience sampling from the patient population of the clinic practice. All had received a complete vision examination within 6 months of testing. None was taking any drugs or medications that would adversely affect either motor control or attentional state. Six of the subjects were habitual PAL wearers, and the other five were novice wearers. Best spectacle-corrected distance and near binocular visual acuity in each subject was 20/20 or better. Distance spectacle correction ranged from −5 to +2 D, with only two subjects having astigmatism greater than 2 D (−2.75 and −2.25 D). Near additions ranged from +1.25 to +2.25 D. The tenets of the Declaration of Helsinki were followed, and written informed consent was obtained. 

Reading Materials

Reading was performed using two hard-copy text formats: standardized single-page (SP) text (horizontal [H]14°/vertical [V]17°) positioned along the subject’s midline at a distance of 60 cm to simulate conventional reading of a page of text, such as a document page, and standardized double-page (DP) text (H 37°/V 17°) arranged so that the end of line 1 of one page of text continued with line 1 of the other text page, and so forth, for successive lines, also at a distance of 60 cm, to simulate conventional secretarial conditions in the business and legal world for typing of documents and contracts, word processing, and other computer-based reading. One page of text was positioned and centered along the subject’s midline, whereas the center of the other text page was positioned 30° to the right. The first condition simulated standardized normal reading at a computer screen, whereas the second simulated the computer work situation in which an additional page of text was situated beside the computer screen. Both texts consisted of standardized adult-level paragraphs (Visagraph 2; Taylor Associates, Huntington, NY) specially printed at 40% contrast in 9-point letter size. At this level of contrast and higher, there is no effect on reading rate in observers with normal emmetropic vision. ⁸ Such a contrast level is typical of, for example, most newspapers, inexpensive paperback books, labels, and hand-printed words on a yellow notepad. Furthermore, we thought that we might uncover some subtle dynamic changes when this contrast was effectively lowered by the defocus effects of the lens outside the near-vision channel. Reading was conducted under normal room illumination conditions (47 cd/m²) using diffuse fluorescent overhead lighting. 

Lenses

Reading was performed with three types of moderately large spectacle lenses (A, 52; B, 46; and ED, 59 mm box size). ¹ These commercially available lenses included a SVL (60° clear FOV horizontally), a PAL-I with a relatively wide intermediate zone of 7.85 mm (18° clear FOV horizontally), and a PAL-II with a relatively narrow intermediate zone of 5.60 mm (13° clear FOV horizontally). 

Procedures

Subjects sat in front of the reading material at a distance of 60 cm without a chin rest, with the restriction of maintaining a constant reading distance. During testing, the experimenter carefully observed the subjects to prevent any change in reading distance. They were allowed to move their heads, but they were not encouraged to do so. A practice trial preceded the test trials in all subjects. During the experiment, subjects silently read each of the test texts with each pair of spectacles. They were instructed to read for comprehension and to keep each fixated word clear. Type of spectacle lens and text format were counterbalanced across subjects to prevent order effects. 

Dynamic Parameter Analysis

The parameters investigated in this study were:

1.  
   Peak saccadic eye movement velocity.
2.  
   Peak step head movement velocity.
3.  
   Duration of the RSS eye movement.
4.  
   Duration of eye movement from initiation of the RSS to the beginning of the first reading fixation.
5.  
   Duration of head movement.
6.  
   Time difference between initiation of eye and head movement.
7.  
   Time difference between completion of eye and head movement.
8.  
   Time difference between eye and head movement peak velocity.
9.  
   Time difference between initiation of RSS and head movement peak velocity.
10.  
    Gaze stabilization time from completion of RSS to the first reading fixation.
11.  
    Percentage of responses showing nonzero gaze-stabilization times.

Figure 1 shows some of these dynamic parameters. Determination of eye and head position changes was based on the associated velocity. The velocity criteria for each parameter were: eye movement (EM) 2.8 deg/s, head movement (HM) 1.7 deg/s, and gaze 3.3 deg/s. These values are somewhat lower than typically found in the literature; however, they were noise-based values for detection of any type of movement (i.e., slow or fast) and not just the most rapid, such as occurs with a saccade or a steplike head movement. These thresholds were based on the noise levels of the respective motor responses when the subjects were instructed to keep their eyes and heads as steady as possible during midline fixation in primary position. Hence, there was no or minimal movement or change in eye or head position during these periods. 

For all but one of the statistical analyses, a one-way ANOVA was performed for each text condition (i.e., SP or DP), followed by the post hoc least-significant-difference (LSD) test, because we were interested in lens-related performance changes within each text condition. The only exception was for the parameters of gaze-stabilization time and percentage of the responses showing nonzero gaze-stabilization times for the SP text condition in which a one-tailed, paired means t-test was used. In all cases, α was set at 0.05. 

Results

Figure 2 presents representative combined horizontal eye (EM) and head (HM) movements that were found during SP text reading with the SVL and PALs. With the SVL and its very wide field of clear vision, reading was accomplished almost solely by eye movements; head movements were negligible. After the RSS, the reading-related fixation commenced immediately. In contrast, with the PALs and their very narrow field of clear vision, reading was now preceded by combined slow and coordinated eye and head movements, with the latter demanded by the optical design of the lens and being necessary to project the optical field of clear vision onto the text region of interest. We thus refer to this post-RSS slow eye movement as an optically induced compensated slow eye movement. It appeared with the PALs, but not with the SVL. 

This optically induced compensated slow eye movement is shown in Figure 3 (EM) for the SP text reading with the PAL-II, but now with greater spatial and temporal resolution. It highlights the detailed dynamic oculomotor activity that occurs in and around the RSS. Subsequent to completion of the RSS, there was a period of approximately 250 ms in which both the eye and head were relatively stationary, followed by a 700-ms period of combined eye and head movements, with this slow movement being initiated by the head (Fig. 3 , HM) to place the narrow optical field of clear vision of the PAL-II onto the text region of interest, before the actual reading commenced. 

Figure 4 presents the eye movement main sequence plot for the RSSs (Fig. 4A) and the head movement main sequence for the associated steplike head movements (Fig. 4B) combined in both the SP and DP text formats across the three lens types. The values fell within normal limits and were consistent with those reported by Zee and Robinson ⁹ for eye movements, and Zangemeister et al. ¹⁰ for head movements. The tabulated data for all six test conditions and associated parameters are presented in Table 1 , with detailed statistics in the ^Appendix in Table A1. Eye movement peak velocity with the DP text format was slower with both PALs than with the SVL. Head movement peak velocity was greater with the PAL-II than with the SVL with the SP text format. Head movement amplitude was greater with both PALs than with the SVL in the SP text condition. In the DP text condition, the head movement amplitude was greater with the PAL-II than the SVL. 

The time difference parameters are presented in Table 2 , with detailed statistics in the ^Appendix in Table A2. None of the comparisons between lens types for each text format was statistically significant. Head movement peak velocity (Fig. 5 , HM Peak vel) occurred after that of the eye movement (Fig. 5 , EM Peak vel) and RSS completion. In fact, in all six test conditions, eye movements were completed before the correlated head movement. With the SP text format, on average, the eye movements were initiated before the head movements, whereas with the DP text format, the opposite occurred (Fig. 6) . 

The duration of the RSS, duration of total eye movement including the RSS and optically induced compensated slow eye movement, and duration of the associated head movement showed several statistically significant differences (Table 3 , detail in Table A3 in the ^Appendix). The total eye movement durations were different for the various PAL and SVL comparisons, as presented in Table 3 . In the SP text condition, duration increased as the clear-vision zone extent decreased, and it was longer with the PALs versus the SVL in the DP text condition. This is presented in Figure 7 for the SP text condition and in Figure 8 for both text conditions. Head movement duration increased with the PAL-II in both text formats (Fig. 9) . 

Finally, the two gaze-stabilization parameters exhibited many differences between lens types and text conditions (Table 4 , with detailed statistics in the ^Appendix in Table A4; Figs. 10 11 ). With the SVL, gaze stabilization always occurred immediately after completion of the RSS for the SP text condition. The percentage of nonzero times in the SP text condition was greater for the PAL-II than the PAL-I. In the DP text condition, there was a difference between the SVL and the PALs in both parameters. The time to gaze stabilization was longer with the PALs than with the SVL (Fig. 10A) . The percentage of responses exhibiting nonzero gaze-stabilization times was higher with the PALs than with the SVL (Fig. 10B) . Representative recordings of gaze-stabilization changes with the three lenses for the SP text condition are presented in Figure 11 . 

Discussion

Despite significant lens design differences between the SVL and PALs, many of the parameters did not discriminate between them. First, saccadic eye movements and the steplike head movements maintained their normal main sequence neurologic control relationship. Thus, although a variety of timed and complex, combined eye and head movement interactions occurred, normal peak velocities in both persisted. This could be due to nonlinear velocity-based eye and head interactions, ¹¹ or perhaps more simply to linear interactions with peak velocity timing differences sufficiently large that the peak velocities of the saccadic eye movements were always substantially larger than the associated nonpeak velocities of the head movements occurring at that same instant. These interactions became more prevalent and prominent as eye and head movements progressively increased beyond 15° in amplitude. Furthermore, the main sequence approach involves first-order dynamics, and hence may not be sensitive to more subtle higher-order dynamic effects. The present head movement peak velocities were in the low normal range of velocities in Zangemeister et al., ¹⁰ and this may be explained by their faster sampling rate (150 Hz), as well as instructions to the subjects to “move the head as fast as possible,” which suggests development of time-optimal control strategies. Second, timing differences for the various parameters between lenses within each text format were not significantly different. This could be attributable to sufficient similarity of the various eye and head movement interactions or perhaps to the relatively large variability that could obscure any small and subtle temporal differences. 

However, there were also several important differences in many of the parameters that discriminated between the SVL and PALs and even between the PALs themselves. First, differences in eye movement peak velocities between the SVL and the PALs in the DP text format can be attributed to the associated smaller eye movement amplitudes made with the PALs, as dictated by the main sequence neurologic relationship. ⁹ In addition, there may be an interaction between the head and eye movements, as their completion time differences were considerably closer in the DP than in the SP text format. Second, there was a consistent difference in eye and head initiation times between the SP and DP text formats independent of lens type. It appears that the horizontal angular extent of the text dictates whether the eye or head moves first. For the SP text condition, eye movements preceded head movements, whereas the reverse was true for the DP text condition. This difference appears to be related to the much wider total angular extent of the reading material (37°) for the DP versus for the narrower field SP (14°) text condition. For the combined SP-SVL condition, the results are consistent with those of Bahill et al., ¹² who found that most naturally occurring saccades in the absence of any head movement were 15° or smaller in amplitude. However, for the SP-PAL condition, head movement was present, despite the small angular extent of the text, presumably because of the optical limitations imposed by the much smaller clear vision zone of the PALs. With the DP text format, head movement always preceded eye movement, presumably because of the larger angular extent of the DP text. However, this may also reflect more general motor control principles involving the desire for time optimality and preprogramming related to the predictable head and eye movements afforded by the fixed text dimensions and not to clarity of vision, per se. 

Furthermore, these results demonstrate the adaptive aspect of the dual-mode model proposed by Barnes, ¹³ wherein the eye-head versus head-eye latency criterion cutoff limit was angularly reduced because of the imposed lens optics of the PALs. Third, we have identified two new and objective vestibulo-ocular reflex (VOR)-related eye and head movement parameters—optically induced compensated slow eye movement duration and gaze-stabilization time—which reveal the more subtle manner in which eye and head movements can adversely affect reading. The time after the RSS itself was completed was followed by a period of several hundred milliseconds in which both eye and gaze had not stabilized, and thus actual reading could not commence until stability was attained. This was typically found with each of the PALs in both the SP and DP text conditions and also found at times in the DP condition for the SVL when large head movements were of necessity executed. We attribute this period of relative oculomotor instability to the lens design and, more specifically, to the nature of the intermediate clear zone. Once the RSS was completed, the head slowly rotated to acquire a clear and focused image of the targeted word through the narrow optical lens channel. 

The same is true of gaze-stabilization times. Similar results have been reported in nonreading conditions with helmet-mounted optical sighting systems that have narrow optical channels. ¹⁴ This slow, VOR-mediated movement is different from that described by Kowler et al. ⁵ In their study, movement occurred as the subject began reading the next line of text, whereas ours was completed before the reading commenced. Furthermore, movement in their study was probably initiated due to the greater horizontal extent of the text, which was greater than 15°, thus necessitating a combined eye and head movement, ¹² whereas in our study it was optically based, as described earlier. Our slow VOR-mediated movement followed by a gaze shift is consistent with the notion of a conditioned linkage between these two oculomotor systems. ¹⁵  

There were no relatively large and consistent differences in dynamic eye and head movement parameters in their mean values and associated response variability between habitual versus new wearers, as well as across the individual subjects, when analyzed with a nonparametric response variability approach. There was sufficient overlap of data across subgroups and parameters to warrant the conclusion of no difference, which could be attributed to a variety of factors. First, oculomotor neural adaptation is known to be very rapid. ⁷ Second, variability across subjects within each subgroup was relatively large. Third, and perhaps most critical, the number of subjects within each subgroup was relatively small (six versus five). Hence, no obvious trends were evident. Therefore, this important question must await further investigation in a study with a much larger subgroup sample size. 

By its very physical nature, the optical design of a PAL is a complex affair. ^{1 2} Several optical factors have been implicated in the symptoms reported by PAL wearers. These include nonuniform optical magnification effects throughout the near and far periphery of the lens producing the phenomenon of “swim” (i.e., illusory and variable seesawlike movement of the visual field associated with lateral head movement); uncorrected astigmatism producing slight blur in the same lens regions as where “swim” is produced; and, the narrow channel width requiring increased head movement to superimpose this clear-vision region optically onto the text portion of interest. Therefore, a variety of optical and perceptual factors may be involved in a patient’s ability to learn how to use and adapt to such a lens in an efficient and symptom-free manner under a variety of viewing conditions. 

In addition to these factors, there are a variety of additional dynamic eye and head movement motor factors that must be considered. In an attempt to provide clear retinal imagery of the text through the narrow optical zone, the observer has not only to maintain the viewing distance relatively constant (especially in an absolute presbyope), but must also learn to execute appropriate and purposeful lateral eye and head movements concurrently. Thus, in addition to the increased frequency and amplitude of head movements found in our earlier investigation, ⁶ as well as by Jones et al., ¹⁶ there are now the added increased eye movement duration and gaze-stabilization times with the PALs requiring additional slow ramplike head movements of a more subtle, dynamic nature before oculomotor control and gaze become stable, and reading can commence. Hence, a variety of oculomotor, sensory, and perceptual problems associated with PAL lenses should be considered in a comprehensive and integrated manner to assist in development of more optimal lens designs in the future. 

Appendix

Table A1.

 

Detailed Statistics for Table 1

	Text Format	df Effect	MS Effect	df Error	MS Error	F	P	Post Hoc LSD Test (P)
																SVL/PAL-I	SVL/PAL-II	PAL-I/PAL-II
EM-Peak velocity	SP	2	1554.3	20	906.2	1.715	0.205	0.297	0.080	0.448
	DP	2	8418.5	18	1691.9	4.975	0.019	0.013	0.014	0.972
HM-Peak velocity	SP	2	100.9	8	28.4	3.549	0.079	0.489	0.032	0.100
	DP	2	509.6	16	396.9	1.284	0.304	0.182	0.186	0.991
RSS amplitude	SP	2	3.480	20	1.626	2.141	0.144	0.197	0.055	0.491
	DP	2	32.083	18	16.0	2.008	0.163	0.228	0.063	0.472
HM amplitude	SP	2	33.8	8	2.6	13.059	0.003	0.105	0.001	0.012
	DP	2	120.6	16	13.7	8.791	0.003	0.029	0.001	0.094

Table A2.

 

Detailed Statistics for Table 2

	Text Format	df Effect	MS Effect	df Error	MS Error	F	P	Post Hoc LSD Test (P)
																SVL/PAL-I	SVL/PAL-II	PAL-I/PAL-II
Initiation	SP	2	0.006	6	0.013	0.474	0.644	0.880	0.480	0.398
	DP	2	0.001	16	0.005	0.155	0.857	0.707	0.595	0.875
Completion	SP	2	0.037	6	0.021	1.715	0.258	0.150	0.171	0.926
	DP	2	0.001	16	0.015	0.055	0.947	0.835	0.910	0.748
Peak velocity	SP	2	0.009	4	0.060	0.155	0.861	0.846	0.748	0.611
	DP	2	0.015	16	0.010	1.509	0.251	0.136	0.888	0.173
Saccade start to HM peak velocity	SP	2	0.003	4	0.034	0.088	0.917	0.795	0.702	0.900
		DP	2	0.013	16	0.010	1.395	0.276	0.190	0.887	0.150

Table A3.

 

Detailed Statistics for Table 3

	Text Format	df Effect	MS Effect	df Error	MS Error	F	P	Post Hoc LSD Test (P)
																SVL/PAL-I	SVL/PAL-II	PAL-I/PAL-II
RSS duration	SP	2	0.000	20	0.001	0.027	0.974	0.881	0.823	0.940
	DP	2	0.001	20	0.005	0.243	0.787	0.699	0.766	0.495
EM duration	SP	2	0.978	20	0.220	44.214	0.000	0.000	0.000	0.001
	DP	2	0.220	20	0.037	5.974	0.009	0.049	0.003	0.194
HM duration	SP	2	0.346	6	0.067	5.149	0.049	0.247	0.019	0.105
	DP	2	0.077	16	0.027	2.883	0.085	0.446	0.032	0.135

Table A4.

 

Detailed Statistics for Table 4

 

A. Student’s t-Test for the SP Format

	df	Pearson Correlation	t Statistic	P	t Critical One-Tail	Variance
												PAL-I	PAL-II
Time	10	−0.023	−1.050	0.159	1.812	77729.29	57296.88
Percentage	10	−0.104	−2.534	0.015	1.812	0.063	0.058

B. One-way ANOVA for the DP Format
	df Effect	MS Effect	df Error	MS Error	F	P	Post Hoc LSD Test (P)
														SVL/PAL-I	SVL/PAL-II	PAL-I/PAL-II
Percentage	2	1.070	20	0.057	18.836	0.000	0.000	0.000	0.209
Time	2	0.479	20	0.053	9.024	0.002	0.006	0.001	0.346

 Supported by The Spectacle Lens Group of Johnson & Johnson Vision Care, Inc.

 Submitted for publication May 28, 2002; revised October 18, 2002; accepted November 4, 2002.

 Commercial relationships policy: F.

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

 

Figure 1.

 

Eye, head, and gaze position as a function of time depicting several of the dynamic parameters tested during the RSS period.

Figure 2.

 

Eye and head position as a function of time during SP reading with the three lenses.

Figure 3.

 

Details of eye and head position as a function of time during the RSS in the SP/PAL-II condition.

Figure 4.

 

Main sequence results. (A) RSS eye movements. (B) Associated head movements.

Table 1.

 

Mean Peak Velocity and Amplitude of the RSS Eye Movement and Associated Head Movement

	EM Peak Velocity			EM Amplitude			HM Peak Velocity			HM Amplitude
		Deg/s	n	Deg	n	Deg/s	n	Deg	n
SP-SVL	293 ± 20	11	12 ± 0.35	11	12 ± 1.04	5	1.7 ± 0.46, ‡	11
SP-PAL-I	279 ± 20	11	11 ± 0.37	11	16 ± 1.95	9	4.3 ± 1.06, ‡	11
SP-PAL-II	269 ± 22	11	11 ± 0.48	11	20 ± 1.44, †	11	6.0 ± 0.45	11
DP-SVL	358 ± 29*	11	25 ± 1.38	10	64 ± 8.77	9	20.9 ± 1.14*	9
DP-PAL-I	296 ± 25	10	23 ± 1.06	10	75 ± 6.24	11	24.7 ± 1.44	11
DP-PAL-II	289 ± 32	11	21 ± 1.41	10	75 ± 5.49	11	28.2 ± 1.29	11

Table 2.

 

Time Difference between Eye and Head Movements

	Initiation			Completion			Peak Velocity			Saccade Start to HM Peak Velocity
		Difference	n	Difference	n	Difference	n	Difference	n
SP-SVL	82 ± 44	5	333 ± 76	4	237 ± 103	4	261 ± 92	4
SP-PAL-I	89 ± 39	8	178 ± 75	8	279 ± 84	8	273 ± 67	8
SP-PAL-II	38 ± 40	11	159 ± 53	11	240 ± 48	11	263 ± 49	11
DP-SVL	−94 ± 30	9	74 ± 33	9	230 ± 25	9	249 ± 29	9
DP-PAL-I	−82 ± 26	11	78 ± 38	11	178 ± 31	10	229 ± 40	11
DP-PAL-II	−108 ± 44	11	89 ± 46	11	217 ± 40	11	251 ± 33	11

Figure 5.

 

Detailed examples of eye and head position and velocity as a function of time during the RSS in the (A) SP and (B) DP text conditions.

Figure 6.

 

Initiation time differences between eye and head movements for the six reading conditions (mean + 1 SEM).

Table 3.

 

Durations of RSS, Eye Movement, and Head Movements

	RSS			EM			HM
		Duration	n	Duration	n	Duration	n
SP-SVL	93 ± 10	11	93 ± 10*	11	291 ± 53	4
SP-PAL-I	95 ± 8	11	431 ± 72*	11	540 ± 75	9
SP-PAL-II	95 ± 11	11	687 ± 48*	11	765 ± 69, †	11
DP-SVL	141 ± 33	11	589 ± 57	11	779 ± 52	9
DP-PAL-I	129 ± 18	11	760 ± 67, †	11	934 ± 107	11
DP-PAL-II	150 ± 19	11	870 ± 72, †	11	1024 ± 73, †	11

Figure 7.

 

Eye and head position as a function of time showing the different eye movement completion times with the three lenses for the SP text format.

Figure 8.

 

Duration of the RSS and eye movement in the six reading conditions (mean + 1 SEM).

Figure 9.

 

Duration of head movement in the six reading conditions (mean + 1 SEM).

Table 4.

 

Gaze-Stabilization Time and Percentage of Responses Showing Nonzero Gaze Stabilization Time

	Percentage	Time (ms)
SP-SVL	0	0
SP-PAL-I	45 ± 7.6	465 ± 84
SP-PAL-II	73 ± 7.3*	583 ± 72
DP-SVL	33 ± 11, †	242 ± 83, †
DP-PAL-I	79 ± 9.5	547 ± 87
DP-PAL-II	92 ± 4	642 ± 54

Figure 10.

 

(A) Gaze stabilization time. (B) Percentage of responses showing nonzero gaze-stabilization times, in the six reading conditions (mean + 1 SEM).

Figure 11.

 

Gaze as a function of time with the three lenses and the SP text format. Arrow: end of gaze-stabilization times for each lens.

The authors thank Elizabeth Bauer for statistical assistance and Wayne Spencer for technical assistance.