- Email: [email protected]
Differential vertical visual latency as determined with a simultaneity paradigm
Vision Research 50 (2010) 534–540
Contents lists available at ScienceDirect
Vision Research journal homepage: www.elsevier.com/locate/visres
Differential vertical visual latency as determined with a simultaneity paradigm Shephali Patel a, Steven H. Schwartz a,*, William H. Swanson b a b
State College of Optometry, State University of New York, 33 West 42nd Street, New York, NY 10036, USA School of Optometry, Indiana University, 800 East Atwater Avenue, Bloomington, IN 47405, USA
a r t i c l e
i n f o
Article history: Received 25 August 2009 Received in revised form 14 December 2009
Keywords: Visual latency Reaction time Glaucoma Visual fields Temporal-order judgment
a b s t r a c t Behavioral studies, as well as anatomical and physiological data, suggest differences in functionality for inferior and superior visual fields. Previous investigations comparing latencies of the two fields have employed motor reaction times. This approach is of limited usefulness in elderly clinical populations where various degrees of motor impairment may be present. In this report, we describe a simultaneity paradigm that allows the determination of relative latencies without dependence on motor reaction times. A slightly, but statistically significant, shorter latency (3.9 ± 5.9 ms) was found for the superior visual field. The results are not affected by age, and both within- and between-session variability are low. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction Data supporting vertical asymmetries in visual function have been obtained using a variety of paradigms and methodologies (Carlsen, Maslovat, Chua, & Franks, 2007; Chen, Wyatt, & Swanson, 2005; Danckert & Goodale, 2001; Horn, Mardin, Korth, & Martus, 1996; Khan & Lawrence, 2005; Liu, Heeger, & Carrasco, 2006; Portin, Vanni, Virsu, & Hari, 1999; Previc, 1990; Yoshii & Paarmann, 1989; Zhe, Van, & Yulong, 2006). In general, these studies show greater functionality for the inferior visual field (Levine & McAnany, 2005; Sample, Irak, Martinez, & Yamagishi, 1997; Skrandies, 1987). Both anatomical and physiological data are generally consistent with this vertical asymmetry. Photoreceptors and ganglion cells are more densely packed superiorly in the human retina than inferiorly (Curcio & Allen, 1990; Curcio, Sloan, Kalina, & Hendrickson, 1990). Asymmetries are also present in the nonhuman dorsal lateral geniculate nucleus, striate cortex, MT and V6A, with greater representation for the inferior field (Connolly & Van Essen, 1984; Galletti, Fattori, Kutz, & Gamberini, 1999; Maunsell & Van Essen, 1987; Tootell, Switkes, Silverman, & Hamilton, 1988; Van Essen, Newsome, & Maunsell, 1984). The few behavioral studies that have directly examined latency differences between superior and inferior visual fields, the topic of the current study, suggest the inferior visual field has a shorter latency. Motor reaction times for superiorly and inferiorly positioned suprathreshold stimuli show small, but in some cases, statistically
* Corresponding author. E-mail address: [email protected] (S.H. Schwartz). 0042-6989/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2009.12.010
significant differences (Maehara, Okubo, & Michimata, 2004; Payne, 1967). Electrophysiological studies using pattern electroretinograms (PERGs) and visually evoked potentials (VEPs) find that stimulation of the inferior visual field results in shorter latencies than stimulation of the superior field (Skrandies, 1987; Zhe et al., 2006). The differences are again small, but statistically significant. Basic knowledge of relative processing speeds of inferior and superior visual fields may have clinical application in the assessment of visual field defects that respect the horizontal meridian. Such defects are found in glaucoma, most often taking the form of a nasal step that shows a relative decrement in sensitivity in either the superior or inferior nasal field (Aulhorn & Karmeyer, 1977; Hart & Becker, 1982; Heijl & Lundqvist, 1984; Katz, Quigley, & Sommer, 1995). Given the small differences in reaction times found in the previously cited experiments (about 14–23 ms), variability inherent in motor responses could be an important limiting factor when considering potential clinical applications (Schwartz, 1992; Schwartz & Loop, 1982). Unstable maladies of motor function are not uncommon in the elderly (Crawford, Goodrich, Henderson, & Kennard, 1989; Fozard, Vercruyssen, Reynolds, Hancock, & Quilter, 1994; Volkow et al., 1998). The high test–retest variability expected if manual reaction times were employed to follow progressive optic neuropathies in this population seriously limits their potential clinical utility. In this paper, we describe a temporal-order judgment (TOJ) methodology to assess symmetry of visual latency that does not depend on motor reaction times. Subjects observed a pair of vertically aligned flashed lights presented at various temporal asynchronies and were asked to indicate which appeared first. The midpoint of the matching range for simultaneity was taken as a
S. Patel et al. / Vision Research 50 (2010) 534–540
measure of the latency difference between the superior and inferior visual fields. Data were obtained from a sample of healthy subjects. Within- and between-session variability was determined. 2. Methods 2.1. Subjects Twenty-two subjects were recruited from the students, faculty, and staff of the SUNY State College of Optometry. All were naïve regarding the aims and design of the experiment. To be included in the study, subjects were required have vision correctable to 20/20 in the eye tested (right eye), normal color vision, and no personal or family history of glaucoma. The subjects’ ages ranged from 21 to 79, with a mean of 40 years. 2.2. Apparatus A graphics controller (Visual Stimulus Generator [VSG 2/5] Cambridge Research Systems, Cambridge, UK) was used to present stimuli on a CRT monitor (ViewSonic Professional Series P220fb). The display resolution was 800 600 pixels with a refresh rate of 100 Hz. Head position was stabilized with a chin and headrest. Viewing distance was 40 cm. A centered cross was used to maintain fixation. In those cases where a lens was required to correct the subject for the viewing distance, it was positioned in a lens holder immediately in front of the subject’s eye. The subject responded to each trial by pressing one of four buttons on a small control box. 2.3. Stimulus Since we desired to maximize the possibility of finding a vertical asymmetry in clinical populations, all stimuli were presented in the nasal visual field. This aspect of the visual field shows differential rates of damage in glaucoma, presumably reflecting anatomical and/or physiological differences for retinal loci situated above and below the horizontal raphe (Vrabec, 1966). The stimulus consisted of a pair of vertically aligned, achromatic squares that each subtended 0.43°, the diameter of the Goldmann III stimulus. Stimulus and background luminances were 66 cd/m2 and 6 cd/m2, respectively (Minolta LS-110 luminance meter). The stimulus pair was centered on the horizontal meridian and separated by 10° when presented 15° nasally, or 15° when presented 5° nasally. These parameters were selected to sample two different loci that fall within a region of the visual field susceptible to glaucomatous visual field loss (Aulhorn & Karmeyer, 1977). The squares that constituted the stimulus pair were flashed for 100 ms (square-wave onset and offset) with a temporal asynchrony that ranged from 250 ms (top square on first) to 250 ms (bottom square on first). Stimulus duration was chosen to minimize the potential effect of eye movements toward one of the flashes. Both flash duration and asynchrony were verified using phototransistors that fed into a PC. 2.4. Procedure Prior to testing, visual acuity, color vision status, and glaucoma history were determined. Based on the subject’s age, lenses were utilized to correct for the 40 cm testing distance. The experimental procedure was approved by the SUNY State College of Optometry
535
Institutional Review Board, and informed consent was obtained prior to testing each subject. For each trial, the subject could give one of the following three responses: (1) the top appeared first, (2) the bottom appeared first, or (3) the flashes appeared simultaneously/could not tell which came on first (Jas´kowski, 1993a). The first trial was initiated when the subject pressed any of the keys on the control box. Subsequent trials were initiated when the subject pressed one of the buttons to indicate his/her response to the prior trial. Subjects were given unlimited time to respond, but encouraged to move along. They were periodically reminded to maintain central fixation, and asked to rate their accuracy of fixation. Subjects reported no difficulty maintaining fixation. Pseudo-randomly interleaved ascending and descending staircases were employed to determine the lower and upper boundaries of the simultaneity range. The ascending staircase commenced with a temporal asynchrony of 250 ms and the descending staircase with an asynchrony of 250 ms. Since the procedure may be utilized on clinical populations, the step size should be as large as possible to expedite testing without significantly affecting the measured value. Based on preliminary observations, initial inter-trial step sizes of 60 and 40 ms were selected. After two reversals, the step size was halved. For the ascending staircase, a response of ‘‘bottom first” was followed by a decrease in the asynchrony, while a response of ‘‘top first” or ‘‘cannot tell” was followed by an increase in the asynchrony. The same logic was followed for the descending staircase. A staircase continued until there were ten reversals. To assess subject attention to the task, the asynchrony for every sixth trial was increased by a value that was a fixed multiple of the final step size (5 for final steps of 30 ms and 10 for final steps of 20 ms). False negative rates, which were monitored based on responses to these ‘‘free” trials, were minimal. In an experimental session, a subject was tested twice for each of the following four different conditions: (1) 15° nasally with a vertical separation of 10°, initial step size of 60 ms; (2) 15° nasally with a vertical separation of 10°, initial step size of 40 ms; (3) 5° nasally with a vertical separation of 15°, initial step size of 60 ms; and (4) 5° nasally with a vertical separation of 15°, initial step size of 40 ms. A session lasted about 1 h. Each subject participated in two sessions, with the sessions separated by at least 1 week. 2.5. Data analysis The final eight reversals for a staircase were averaged (mean reversal) to arrive at either the lower or upper boundary of the matching range. The midpoint between an upper and lower boundary (mathematical average) was taken as the latency difference between superior and inferior retina (differential latency or DL). A DL of zero would indicate vertical symmetry, while a positive value would indicate a shorter latency for the inferior field and a negative value would show that the superior field is faster. The data were analyzed in more detail by constructing two yes– no psychometric functions, which were labeled the top and bottom functions, using data combined from both staircases (Swanson, 1993). For top function, the question was ‘‘Did the top come on first?” and for the bottom function, the question was ‘‘Did the bottom come on first?” A response of ‘‘top first” was taken as ‘‘yes” for the top function and ‘‘no” for the bottom function. ‘‘Bottom first” was coded as ‘‘yes” for the bottom function and coded as ‘‘no” for the top function. Responses of ‘‘simultaneous/not sure” were recorded as ‘‘no” for both functions. Data were fit with Weibull functions, with the upper and lower boundaries of the simultaneity range determined based on 50% seen.
536
S. Patel et al. / Vision Research 50 (2010) 534–540
3. Results
3.4. Within- and between-session variability
3.1. Data analysis Based on 344 datasets across all subjects and stimulus conditions, the average mean reversal for a descending staircase was 63.0 with a SD of 35.9 ms, while the mean frequency of seeing (FOS) estimate for the upper boundary was 56.8 with a SD of 35.4 ms. (This does not include 8 data sets for subject 22, who fell asleep during the experiment.) Equivalent values for the ascending staircase were 70.8 ± 36.0 ms and 64.6 ± 33.9 ms, respectively. The absolute difference between the mean reversal and FOS estimate was calculated for each dataset; the mean difference was 9.4 ms with a standard deviation of 11.5 ms. Further analyses, described in the following sections of this paper, were performed using only those datasets for which the mean reversal values for both ascending and descending staircases were within 40 ms of the FOS estimates. Employing this criteria resulted in inclusion of 94.2% of the datasets (N = 324). FOS curves for these data had an average slope of 30.0 ± 23.3 compared to an average slope of 11.0 ± 14.8 for data in which the mean reversal values differed from the FOS estimates by 40 or more msec. The corresponding width estimates based on the 10% and 90% points of the psychometric functions are 96.1 ± 86.5 and 260.6 ± 151.1, respectively (Kuss, Jäkel, & Wichmann, 2005).
During a session, two measures of DL were each obtained for all the four experimental conditions. Differences between the first and second DLs for a given condition are plotted as a histogram in Fig. 2A (N = 154) (K–S: d = 0.054, p > 0.20; SW: W = 0.981, p = 0.029). Plotting within-session variability as a function of the averaged values for the two sessions suggests that extreme DLs do not affect within-session variability (r = 0.061, p = 0.452) (Fig. 2B) (Bland & Altman, 2003). Neither retinal location (F = 0.089, p = 0.767) nor inter-trial step size (F = 0.60, p = 0.440) significantly affected within-session DL variability (within-subjects ANOVA). To ascertain a subject’s between-session variability, the average DL obtained in the first session for a given condition was subtracted from the average value obtained in the second session (Fig. 3A; N = 70) (K–S: d = 0.072, p > 0.20, but SW: W = 0.979, p = 0.311). Plotting between-session variability as a function of the averaged values for the two sessions suggests that betweensession variability is not affected by extreme DL values (r = 0.019, p = 0.879) (Fig. 3B) (Bland & Altman, 2003). Neither retinal location (F = 0.958, p = 0.331) nor inter-trial step size (F = 1.25, p = 0.267) had a significant effect on between-session variability (within-subjects ANOVA).
3.2. Differential latencies
4. Discussion
Fig. 1A shows the distribution of all differential latencies across all sessions and conditions for all subjects (N = 324) (Kolmogorov– Smirnov (K–S): d = 0.057, p > 0.20; Shapiro–Wilk (SW): W = 0.987, p = 0.005). The scatterplot in Fig. 1B shows DL values grouped by the mean subject DL. Fourteen of the twenty-two subjects (63.6%) manifested shorter latencies for the superior visual field. The mean subject DL (based on 16 measurements for most subjects) ranges from 4.9 to 14.8 ms, with an average (per subject) value of 3.9 ± 5.9 ms. This is significantly different than the expected value of zero that would be found in the case of perfect vertical symmetry (t = 3.09, p = 0.0056). Neither retinal location (F = 0.012, p = 0.912) nor inter-trial step size (F = 0.60, p = 0.442) significantly affected DL (within-subjects ANOVA). The age distribution of the subjects was bimodal, with a group of 12 subjects in their 20 s and the remaining 10 with ages between 48 and 79. DL did not vary significantly between these two age groups (t = 0.798; p = 0.435).
4.1. Data analysis
3.3. Comparison to results obtained with FOS analysis DL was also calculated using the simultaneity ranges as determined with the FOS curves. Across subjects, the average value was 4.1 ± 6.2 ms, which is not statistically different than that obtained using the mean reversal data (t-test, dependent groups: t = 0.418, p = 0.680). As is the case for the DL values that are based on mean reversals, the average DL is significantly different than the expected value of zero that would be found in the case of perfect vertical symmetry (t = 3.14, p = 0.0049). The upper boundary of the matching range given by mean reversals was 64.3 ± 29.3 ms, while the FOS estimate for this boundary was 58.9 ± 28.6 ms. Comparable data for the lower boundary are 72.1 ± 31.0 and 67.1 ± 28.1 ms, respectively. In both cases, the absolute mean reversal values are slightly, but significantly, larger than the FOS estimates (t-test, dependent samples: t = 6.07, p < 0.05 and t = 5.23, p < 0.05). The average matching range based on mean reversals (136.4 ± 59.0) is significantly wider than that based on FOS estimates (126.1 ± 55.3; t-test, dependent samples: t = 5.20, p < 0.005).
For the most part, the results from the mean reversal and FOS analysis were very similar, but the range was slightly, but significantly, wider when determined with the mean reversal data. Compared to the data obtained from the FOS curves, the range is increased approximately symmetrically at both upper and lower boundaries such that the DL obtained with mean reversals and FOS curves are statistically the same. A comment is in order regarding the use of yes–no psychometric functions (with threshold taken as 50% correct) rather than a two-alternative forced choice (2AFC). We initially considered requiring subjects to indicate which came on first, without providing the ‘‘simultaneous/not sure” option. The decision to provide a ‘‘simultaneous/not sure” option is supported by the TOJ work of Jas´kowski (1993a) who reported that forcing a choice between ‘‘top first” and ‘‘bottom first” introduces an artifactual bias toward one of these choices when the subject is not sure which came on first. He found that giving the subject a third option of ‘‘simultaneously” reduced the bias. This is consistent with our experiences in preliminary 2AFC observations: when an option of ‘‘simultaneous/not sure” was not available, there was a temporal offset that was not apparent when subjects were allowed to select ‘‘simultaneous/not sure.” 4.2. Comparison to previous studies Previous studies employing visually evoked potentials and motor reaction times have tended to find slightly, but significantly, shorter latencies for the inferior visual field. The most comprehensive electrophysiological data have been reported by Skrandies (1987), who found evoked potential latencies for inferior stimulation to be 8.6–16.5 ms shorter, depending on the mode of stimulus presentation (onset, offset, or pattern reversal). Examining motor reactions times to small (15 min arc) suprathreshold flashes presented in a circle centered on the point of fixation (15° radius), Payne (1967) found latencies to be shortest for the inferior nasal field and longest in the superior temporal field,
537
S. Patel et al. / Vision Research 50 (2010) 534–540
A
80 70
No. of obs.
60 50 40 30 20 10 0 -50
-40
-30
-20
-10
0
10
20
30
Differential Latency (msec)
B
40
Differential Latency (msec)
30 20 10 0 -10 -20 -30 -40 -50 -60 -16
-14
-12
-10
-8
-6
-4
-2
0
2
4
6
Subject Mean (msec) Fig. 1. A. Histogram showing the distribution of differential latency (DL) for 22 subjects (N = 324). B. DLs grouped by mean subject DL. The solid line is the mean of all data points, and the dashed lines represent the 95% CI. Values greater than zero indicate a shorter latency for the inferior field, while those less than zero show a shorter latency for the superior field.
with a difference of about 23 ms. Maehara et al. (2004) found similar results using stimuli ranging from 16–64 min arc (133 ms duration) presented 7° superior or inferior to fixation on various chromatic backgrounds. The average motor reaction time for the inferior visual field was 359 ms, compared to 373 ms for the superior visual field, a significant difference. The data reported herein, obtained with a simultaneity paradigm, are consistent with previously published reaction time data in showing approximate vertical symmetry for superior and inferior field latencies. Motor reaction time studies show slightly shorter latencies for the inferior field, while simultaneity data reveal slightly shorter latencies for the superior field. The different stimulus sizes, luminances, chromaticities, and locations that were used for the motor reaction time and simultaneity experiments
may contribute to the discrepant results. Moreover, TOJ and reaction times may not equivalently measure visual latency, possibly due to mediation by different processing streams (Cardoso-Leite, Gorea, & Mamassian, 2007; Jas´kowski, 1993b, 1996; Tappe, Niepel, & Neumann, 1994). Retinal location and inter-trial step sizes were not significant factors in the current study. The simultaneity data were robust in that there was little variability between subjects and little withinand between-session variability. These issues are not addressed in previous motor reaction time studies. Based on the results of our study and those employing motor reactions times, it appears that any latency difference between the superior and inferior fields in healthy visual systems is rather small and, for clinical purposes, essentially nil.
538
S. Patel et al. / Vision Research 50 (2010) 534–540
A
40 35
No. of obs.
30 25 20 15 10 5 0 -40
-30
-20
-10
0
10
20
30
40
50
DL1 - DL2 (msec)
B
60 50 40
DL1-DL2 (msec)
30 20 10 0 -10 -20 -30 -40 -40
-30
-20
-10
0
10
20
30
Mean: DL1, DL2 (msec) Fig. 2. A. Histogram showing differences between the first and second DL measured in a given session (DL1 – DL2) for the same set of experimental conditions (N = 154). B. Within-session differences in DL plotted as a function of the average of the two values that were subtracted to obtain this difference. The regression line (solid) and 95% CI (dashed) are given.
4.3. Clinical implications While a motor reaction time methodology may provide useful information in a younger, healthy population, its usefulness in an elderly clinical population, where varying degrees of motor impairment are not uncommon, is limited (Crawford et al., 1989; Fozard et al., 1994; Volkow et al., 1998). Motor impairment is not expected to be a factor when a simultaneity paradigm is employed to assess differential latency, and no age-related differences were found in the current study. Visual latency increases with decreasing visibility, so it would be expected that the increased visual thresholds found in visual field defects would result in increased latencies for these loci (Becker, Vonthein, Volpe, & Schiefer, 2005; Flammer, Drance, & Schulzer, 1984; Mihaylova, Stomonyakov, & Vassilev, 1999; Roufs,
1974; Ueno, 1979). Static perimetric studies of glaucoma patients find that while threshold reaction times are not significantly different in damaged and normal regions of the visual field, they are increased for high luminance stimuli (0 dB, which is about 3183 cd/ m2) presented with the Humphrey perimeter (Wall, Kutzko, & Chauhan, 2002; Wall, Maw, Stanek, & Chauhan, 1996). The increase was attributed to the decreased visibility/detectability of stimuli that fall within glaucomatous defects. Neurological damage to the optic nerve fibers that occurs in glaucoma frequently manifests as an altitudinal nasal visual field defect, referred to as a nasal step (Aulhorn & Karmeyer, 1977; Hart & Becker, 1982; Heijl & Lundqvist, 1984; Katz et al., 1995). Decreased sensitivity in a damaged region is expected to result in an increased latency relative to normal regions (Artes, McLeod, & Henson 2002; Flammer et al., 1984; Nowomiejska et al., 2008; Wall
539
S. Patel et al. / Vision Research 50 (2010) 534–540
A
20 18 16
No. of obs.
14 12 10 8 6 4 2 0 -35
-25
-15
-5
5
15
25
DLS1 - DLS2 (msec)
B
40 30
DL S1 - DL S2 (msec)
20 10 0 -10 -20 -30 -40 -30
-25
-20
-15
-10
-5
0
5
10
15
Mean: DL S1, DL S2 (msec) Fig. 3. A. Histogram showing differences between the average DLs obtained in sessions 1 and 2 (DLS1 – DLS2) for the same set of experimental conditions (N = 70). B. Between-session differences in DL plotted as a function of the average of the two values that were subtracted to obtain this difference. The regression line (solid) and 95% CI (dashed) are given.
et al., 1996). For the simultaneity paradigm introduced in this paper, DLs are predicted to be asymmetric when testing nasal steps if one of the paired stimuli falls in a region of the visual field less damaged than the fellow stimulus. Sensitivities to inferior and superior achromatic stimuli, presented with a Humphrey Visual Field Analyzer in the approximate loci tested in the current experiment, are essentially the same (about 31 dB, or 2.5 cd/m2) (Heijl, Lindgren, & Olsson, 1987). Based on considerations of sensitivity alone, it is not unexpected that the suprathreshold stimuli employed in the current study (66 cd/m2) elicit approximately vertically symmetrical latencies. Additionally, our simultaneity data suggest that anatomical and physiological asymmetries found in the primate visual system would not result in vertical latency asymmetry that could be a potentially confounding factor in suprathreshold clinical testing.
The data reported herein show that DL has relatively little variability across subjects and strong within- and between-session reliability. These results point to its possible clinical utility to compare functionality of superior and inferior visual fields. Because testing utilizes a paired stimulus, with each of the stimulus components presented at a different retinal locus, the technique has the potential advantage of comparing function of disparate retinal loci with nearly simultaneous stimulation. In visual fields and other traditional procedures, thresholds determined for individual points are compared with those determined previously or thereafter during the session. While these measurements may be close in time, they are not simultaneous, introducing the potential of sensitivity drifts. Use of paired stimuli may provide a more direct comparison of retinal function.
540
S. Patel et al. / Vision Research 50 (2010) 534–540
Acknowledgments This research was supported by a Grant from Vision Service Plan (Schwartz), Beta Sigma Kappa (Patel), and the SUNY College of Optometry Office of Graduate Studies. Dr. Swanson was supported by NIH Grant EY007716. A portion of this research was presented at the 2008 meeting of the Association for Research in Vision and Ophthalmology in Ft. Lauderdale, FL. We appreciate the thoughtful comments of the referees. References Artes, P. H., McLeod, D., & Henson, D. B. (2002). Response time as a discriminator between true- and false-positive responses in suprathreshold perimetry. Investigative Ophthalmology and Visual Science, 43, 129–132. Aulhorn, E., & Karmeyer, H. (1977). Frequency distribution in early glaucomatous defects. In E. L. Greve (Ed.). Documenta ophthalmologica proceeding series (Vol. 14, pp. 75–83). The Hague: Dr. W. Junk bv Publishers. Becker, S. T., Vonthein, R., Volpe, N. J., & Schiefer, U. (2005). Factors influencing reaction time during automated kinetic perimetry on the Tubingen Computer Campimeter. Investigative Ophthalmology and Visual Science, 46, 2633–2638. Bland, J. M., & Altman, D. G. (2003). Applying the right statistics: Analysis of measurement studies. Ultrasound in Obstetrics and Gynecology, 22, 85–93. Cardoso-Leite, P., Gorea, A., & Mamassian, P. (2007). Temporal order judgment and simple reaction times: Evidence for a common procession system. Journal of Vision, 7(6), 11. 1–14. Carlsen, A. N., Maslovat, D., Chua, R., & Franks, I. M. (2007). Perceptual processing differences owing to visual field asymmetries. Neuroreport, 18(10), 1067–1070. Chen, Y., Wyatt, H. J., & Swanson, W. H. (2005). Pupillary evaluation of retinal asymmetry: Development and initial testing of a technique. Vision Research, 45, 2549–2563. Connolly, M., & Van Essen, D. C. (1984). The representation of the visual field in parvocellular and magnocellular layers of the lateral geniculate nucleus in the macaque monkey. The Journal of Comparative Neurology, 226(4), 544–564. Crawford, T., Goodrich, S., Henderson, L., & Kennard, C. (1989). Predictive responses in Parkinson’s disease: Manual keypresses and saccadic eye movements to stimulus events. Journal of Neurology, Neurosurgery, & Psychiatry, 52, 1033–1042. Curcio, C. A., & Allen, K. A. (1990). Topography of ganglion cells in human retina. The Journal of Comparative Neurology, 300(1), 5–25. Curcio, C. A., Sloan, K. R., Kalina, R. E., & Hendrickson, A. E. (1990). Human photoreceptor topography. The Journal of Comparative Neurology, 292(4), 497–523. Danckert, J., & Goodale, M. A. (2001). Superior performance for visually guided pointing in the lower visual field. Experimental Brain Research, 137, 303–308. Flammer, J., Drance, S. M., & Schulzer, M. (1984). Covariates of the long-term fluctuations of differential light threshold. Archives of Ophthalmology, 102, 880–882. Fozard, J. L., Vercruyssen, M., Reynolds, S. L., Hancock, P. A., & Quilter, R. E. (1994). Age differences and changes in reaction time: The Baltimore longitudinal study of aging. Journal of Gerontology, 49(4), 179–189. Galletti, C., Fattori, P., Kutz, D. F., & Gamberini, M. (1999). Brain location and visual topography of cortical area V6A in the macaque monkey. European Journal of Neuroscience, 11, 575–582. Hart, W. M., Jr., & Becker, B. (1982). The onset and evolution of glaucomatous visual field defects. Ophthalmology, 89(3), 268–279. Heijl, A., Lindgren, G., & Olsson, J. (1987). Normal variability of static perimetric threshold values across the central visual field. Archives of Ophthalmology, 105, 1544–1549. Heijl, A., & Lundqvist, L. (1984). The frequency distribution of earliest glaucomatous visual field defects documented by automatic perimetry. Acta Ophthalmologica, 62(4), 658–664. Horn, F., Mardin, C., Korth, M., & Martus, P. (1996). Quadrant pattern ERG and SLO stimulation in normals and glaucoma patients. Graefe’s Archive for Clinical and Experimental Ophthalmology, 234(Suppl. 1), S174–S179. Jas´kowski, P. (1993a). Selective attention and temporal-order judgment. Perception, 22, 681–689. Jas´kowski, P. (1993b). Temporal-order judgment and reaction time to stimuli of different rise times. Perception, 22, 963–970. Jas´kowski, P. (1996). Simple reaction time and perception of temporal order: Dissociations and hypothesis. Motor and Perceptual Skills, 82(3), 707–730.
Katz, J., Quigley, H. A., & Sommer, A. (1995). Repeatability of the Glaucoma Hemifield Test in automated perimetry. Investigative Ophthalmology and Visual Science, 36(8), 1658–1664. Khan, M. A., & Lawrence, G. P. (2005). Differences in visuomotor control between the upper and lower visual fields. Experimental Brain Research, 164, 395–398. Kuss, M., Jäkel, F., & Wichmann F. A. (2005). Approximate Bayesian inference for psychometric functions using MCMC sampling. Technical Report No. 135. Max Planck Institute for Biological Cybernetics. Levine, M. W., & McAnany, J. J. (2005). The relative capabilities of the upper and lower hemifields. Vision Research, 45(21), 2820–2830. Liu, T., Heeger, D. J., & Carrasco, M. (2006). Neural correlates of the visual vertical meridian asymmetry. Journal of Vision, 6, 1294–1306. Maehara, G., Okubo, M., & Michimata, C. (2004). Effects of background color on detecting spot stimuli in the upper and lower visual fields. Brain and Cognition, 55, 558–563. Maunsell, J. H. R., & Van Essen, D. C. (1987). Topographic representation of the middle temporal visual area in the macaque monkey: Representational biases and the relationship to callosal connections and myeloarchitectocnic boundaries. The Journal of Comparative Neurology, 266(4), 535–555. Mihaylova, M., Stomonyakov, V., & Vassilev, A. (1999). Peripheral and central delay in processing high spatial frequencies: Reaction time and VEP latency studies. Vision Research, 39(4), 699–705. Nowomiejska, K., Vonthein, R., Paetzold, J., Zagorski, Z., Kardon, R., & Schiefer, U. (2008). Reaction time during semi-automated kinetic perimetry (SKP) in patients with advanced visual field loss. Acta Ophthalmologica, 1, 5. Payne, W. H. (1967). Visual reaction times on a circle about the fovea. Science, 155, 481–482. Portin, K., Vanni, S., Virsu, V., & Hari, R. (1999). Stronger occipital cortical activation to lower than upper visual field stimuli: Neuromagnetic recordings. Experimental Brain Research, 124, 287–294. Previc, F. H. (1990). Functional specialisation in the lower and upper visual fields in humans: Its ecological implications. Behavior and Brain Sciences, 13, 519–542. Roufs, J. A. (1974). Dynamic properties of vision. V. Perception lag and reaction time in relation to flicker and flash thresholds. Vision Research, 14(9), 853–869. Sample, P. A., Irak, I., Martinez, G. A., & Yamagishi, N. (1997). Asymmetries in the normal short-wavelength visual field: Implications for short-wavelength perimetry. American Journal of Ophthalmology, 124, 46–52. Schwartz, S. H. (1992). Reaction time distributions and their relationship to the transient/sustained nature of the neural discharge. Vision Research, 32(11), 2087–2092. Schwartz, S. H., & Loop, M. S. (1982). Evidence for transient luminance and quasisustained color mechanisms. Vision Research, 22(4), 445–447. Skrandies, W. (1987). The upper and lower visual field of man: Electrophysiological and functional differences. In D. Ottoson (Ed.). Progress in sensory physiology (Vol. 8, pp. 1–93). Berlin: Springer-Verlag. Swanson, W. H. (1993). Analysis of Rayleigh match data with psychometric functions. Journal of the Optical Society of America A, 10(8), 1807–1817. Tappe, T., Niepel, M., & Neumann, O. (1994). A dissociation between reaction time to sinusoidal gratings and temporal-order judgment. Perception, 23, 335–347. Tootell, R. B. H., Switkes, E., Silverman, M. S., & Hamilton, S. L. (1988). Functional anatomy of macaque striate cortex. II. Retinotopic organization. The European Journal of Neuroscience, 8(5), 1531–1568. Ueno, T. (1979). Spatial summation in human vision: Simple reaction time measurements. Journal of the Optical Society of America, 69(7), 1023–1028. van Essen, D. C., Newsome, W. T., & Maunsell, J. H. R. (1984). The visual field representation in striate cortex of the macaque monkey: Asymmetries, anisotropies, and individual variability. Vision Research, 24, 429–448. Volkow, N. D., Gur, R. C., Wang, G.-J., Fowler, J. S., Moberg, P. J., Ding, Y.-S., et al. (1998). Association between decline in brain dopamine activity with age and cognitive and motor impairment in healthy individuals. American Journal of Psychiatry, 155, 344–349. Vrabec, F. (1966). The temporal raphe of the human retina. Journal of Ophthalmology, 62(5), 926–938. Wall, M., Kutzko, K. E., & Chauhan, B. C. (2002). The relationship of visual threshold and reaction time to visual field eccentricity with conventional automated perimetry. Vision Research, 42, 781–787. Wall, M., Maw, R. J., Stanek, K. E., & Chauhan, B. C. (1996). The psychometric function and reaction times of automated perimetry in normal and abnormal areas of the visual field in patients with glaucoma. Investigative Ophthalmology and Visual Science, 37, 878–885. Yoshii, M., & Paarmann, A. (1989). Hemiretinal stimuli elicit different amplitudes in the pattern electroretinogram. Documenta Ophthalmologica, 72(1), 21–30. Zhe, Q., Van, S., & Yulong, D. (2006). Asymmetry between the upper and lower visual fields: An event-related potential study. Chinese Science Bulletin, 51(5), 536–541.
Contents lists available at ScienceDirect
Vision Research journal homepage: www.elsevier.com/locate/visres
Differential vertical visual latency as determined with a simultaneity paradigm Shephali Patel a, Steven H. Schwartz a,*, William H. Swanson b a b
State College of Optometry, State University of New York, 33 West 42nd Street, New York, NY 10036, USA School of Optometry, Indiana University, 800 East Atwater Avenue, Bloomington, IN 47405, USA
a r t i c l e
i n f o
Article history: Received 25 August 2009 Received in revised form 14 December 2009
Keywords: Visual latency Reaction time Glaucoma Visual fields Temporal-order judgment
a b s t r a c t Behavioral studies, as well as anatomical and physiological data, suggest differences in functionality for inferior and superior visual fields. Previous investigations comparing latencies of the two fields have employed motor reaction times. This approach is of limited usefulness in elderly clinical populations where various degrees of motor impairment may be present. In this report, we describe a simultaneity paradigm that allows the determination of relative latencies without dependence on motor reaction times. A slightly, but statistically significant, shorter latency (3.9 ± 5.9 ms) was found for the superior visual field. The results are not affected by age, and both within- and between-session variability are low. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction Data supporting vertical asymmetries in visual function have been obtained using a variety of paradigms and methodologies (Carlsen, Maslovat, Chua, & Franks, 2007; Chen, Wyatt, & Swanson, 2005; Danckert & Goodale, 2001; Horn, Mardin, Korth, & Martus, 1996; Khan & Lawrence, 2005; Liu, Heeger, & Carrasco, 2006; Portin, Vanni, Virsu, & Hari, 1999; Previc, 1990; Yoshii & Paarmann, 1989; Zhe, Van, & Yulong, 2006). In general, these studies show greater functionality for the inferior visual field (Levine & McAnany, 2005; Sample, Irak, Martinez, & Yamagishi, 1997; Skrandies, 1987). Both anatomical and physiological data are generally consistent with this vertical asymmetry. Photoreceptors and ganglion cells are more densely packed superiorly in the human retina than inferiorly (Curcio & Allen, 1990; Curcio, Sloan, Kalina, & Hendrickson, 1990). Asymmetries are also present in the nonhuman dorsal lateral geniculate nucleus, striate cortex, MT and V6A, with greater representation for the inferior field (Connolly & Van Essen, 1984; Galletti, Fattori, Kutz, & Gamberini, 1999; Maunsell & Van Essen, 1987; Tootell, Switkes, Silverman, & Hamilton, 1988; Van Essen, Newsome, & Maunsell, 1984). The few behavioral studies that have directly examined latency differences between superior and inferior visual fields, the topic of the current study, suggest the inferior visual field has a shorter latency. Motor reaction times for superiorly and inferiorly positioned suprathreshold stimuli show small, but in some cases, statistically
* Corresponding author. E-mail address: [email protected] (S.H. Schwartz). 0042-6989/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2009.12.010
significant differences (Maehara, Okubo, & Michimata, 2004; Payne, 1967). Electrophysiological studies using pattern electroretinograms (PERGs) and visually evoked potentials (VEPs) find that stimulation of the inferior visual field results in shorter latencies than stimulation of the superior field (Skrandies, 1987; Zhe et al., 2006). The differences are again small, but statistically significant. Basic knowledge of relative processing speeds of inferior and superior visual fields may have clinical application in the assessment of visual field defects that respect the horizontal meridian. Such defects are found in glaucoma, most often taking the form of a nasal step that shows a relative decrement in sensitivity in either the superior or inferior nasal field (Aulhorn & Karmeyer, 1977; Hart & Becker, 1982; Heijl & Lundqvist, 1984; Katz, Quigley, & Sommer, 1995). Given the small differences in reaction times found in the previously cited experiments (about 14–23 ms), variability inherent in motor responses could be an important limiting factor when considering potential clinical applications (Schwartz, 1992; Schwartz & Loop, 1982). Unstable maladies of motor function are not uncommon in the elderly (Crawford, Goodrich, Henderson, & Kennard, 1989; Fozard, Vercruyssen, Reynolds, Hancock, & Quilter, 1994; Volkow et al., 1998). The high test–retest variability expected if manual reaction times were employed to follow progressive optic neuropathies in this population seriously limits their potential clinical utility. In this paper, we describe a temporal-order judgment (TOJ) methodology to assess symmetry of visual latency that does not depend on motor reaction times. Subjects observed a pair of vertically aligned flashed lights presented at various temporal asynchronies and were asked to indicate which appeared first. The midpoint of the matching range for simultaneity was taken as a
S. Patel et al. / Vision Research 50 (2010) 534–540
measure of the latency difference between the superior and inferior visual fields. Data were obtained from a sample of healthy subjects. Within- and between-session variability was determined. 2. Methods 2.1. Subjects Twenty-two subjects were recruited from the students, faculty, and staff of the SUNY State College of Optometry. All were naïve regarding the aims and design of the experiment. To be included in the study, subjects were required have vision correctable to 20/20 in the eye tested (right eye), normal color vision, and no personal or family history of glaucoma. The subjects’ ages ranged from 21 to 79, with a mean of 40 years. 2.2. Apparatus A graphics controller (Visual Stimulus Generator [VSG 2/5] Cambridge Research Systems, Cambridge, UK) was used to present stimuli on a CRT monitor (ViewSonic Professional Series P220fb). The display resolution was 800 600 pixels with a refresh rate of 100 Hz. Head position was stabilized with a chin and headrest. Viewing distance was 40 cm. A centered cross was used to maintain fixation. In those cases where a lens was required to correct the subject for the viewing distance, it was positioned in a lens holder immediately in front of the subject’s eye. The subject responded to each trial by pressing one of four buttons on a small control box. 2.3. Stimulus Since we desired to maximize the possibility of finding a vertical asymmetry in clinical populations, all stimuli were presented in the nasal visual field. This aspect of the visual field shows differential rates of damage in glaucoma, presumably reflecting anatomical and/or physiological differences for retinal loci situated above and below the horizontal raphe (Vrabec, 1966). The stimulus consisted of a pair of vertically aligned, achromatic squares that each subtended 0.43°, the diameter of the Goldmann III stimulus. Stimulus and background luminances were 66 cd/m2 and 6 cd/m2, respectively (Minolta LS-110 luminance meter). The stimulus pair was centered on the horizontal meridian and separated by 10° when presented 15° nasally, or 15° when presented 5° nasally. These parameters were selected to sample two different loci that fall within a region of the visual field susceptible to glaucomatous visual field loss (Aulhorn & Karmeyer, 1977). The squares that constituted the stimulus pair were flashed for 100 ms (square-wave onset and offset) with a temporal asynchrony that ranged from 250 ms (top square on first) to 250 ms (bottom square on first). Stimulus duration was chosen to minimize the potential effect of eye movements toward one of the flashes. Both flash duration and asynchrony were verified using phototransistors that fed into a PC. 2.4. Procedure Prior to testing, visual acuity, color vision status, and glaucoma history were determined. Based on the subject’s age, lenses were utilized to correct for the 40 cm testing distance. The experimental procedure was approved by the SUNY State College of Optometry
535
Institutional Review Board, and informed consent was obtained prior to testing each subject. For each trial, the subject could give one of the following three responses: (1) the top appeared first, (2) the bottom appeared first, or (3) the flashes appeared simultaneously/could not tell which came on first (Jas´kowski, 1993a). The first trial was initiated when the subject pressed any of the keys on the control box. Subsequent trials were initiated when the subject pressed one of the buttons to indicate his/her response to the prior trial. Subjects were given unlimited time to respond, but encouraged to move along. They were periodically reminded to maintain central fixation, and asked to rate their accuracy of fixation. Subjects reported no difficulty maintaining fixation. Pseudo-randomly interleaved ascending and descending staircases were employed to determine the lower and upper boundaries of the simultaneity range. The ascending staircase commenced with a temporal asynchrony of 250 ms and the descending staircase with an asynchrony of 250 ms. Since the procedure may be utilized on clinical populations, the step size should be as large as possible to expedite testing without significantly affecting the measured value. Based on preliminary observations, initial inter-trial step sizes of 60 and 40 ms were selected. After two reversals, the step size was halved. For the ascending staircase, a response of ‘‘bottom first” was followed by a decrease in the asynchrony, while a response of ‘‘top first” or ‘‘cannot tell” was followed by an increase in the asynchrony. The same logic was followed for the descending staircase. A staircase continued until there were ten reversals. To assess subject attention to the task, the asynchrony for every sixth trial was increased by a value that was a fixed multiple of the final step size (5 for final steps of 30 ms and 10 for final steps of 20 ms). False negative rates, which were monitored based on responses to these ‘‘free” trials, were minimal. In an experimental session, a subject was tested twice for each of the following four different conditions: (1) 15° nasally with a vertical separation of 10°, initial step size of 60 ms; (2) 15° nasally with a vertical separation of 10°, initial step size of 40 ms; (3) 5° nasally with a vertical separation of 15°, initial step size of 60 ms; and (4) 5° nasally with a vertical separation of 15°, initial step size of 40 ms. A session lasted about 1 h. Each subject participated in two sessions, with the sessions separated by at least 1 week. 2.5. Data analysis The final eight reversals for a staircase were averaged (mean reversal) to arrive at either the lower or upper boundary of the matching range. The midpoint between an upper and lower boundary (mathematical average) was taken as the latency difference between superior and inferior retina (differential latency or DL). A DL of zero would indicate vertical symmetry, while a positive value would indicate a shorter latency for the inferior field and a negative value would show that the superior field is faster. The data were analyzed in more detail by constructing two yes– no psychometric functions, which were labeled the top and bottom functions, using data combined from both staircases (Swanson, 1993). For top function, the question was ‘‘Did the top come on first?” and for the bottom function, the question was ‘‘Did the bottom come on first?” A response of ‘‘top first” was taken as ‘‘yes” for the top function and ‘‘no” for the bottom function. ‘‘Bottom first” was coded as ‘‘yes” for the bottom function and coded as ‘‘no” for the top function. Responses of ‘‘simultaneous/not sure” were recorded as ‘‘no” for both functions. Data were fit with Weibull functions, with the upper and lower boundaries of the simultaneity range determined based on 50% seen.
536
S. Patel et al. / Vision Research 50 (2010) 534–540
3. Results
3.4. Within- and between-session variability
3.1. Data analysis Based on 344 datasets across all subjects and stimulus conditions, the average mean reversal for a descending staircase was 63.0 with a SD of 35.9 ms, while the mean frequency of seeing (FOS) estimate for the upper boundary was 56.8 with a SD of 35.4 ms. (This does not include 8 data sets for subject 22, who fell asleep during the experiment.) Equivalent values for the ascending staircase were 70.8 ± 36.0 ms and 64.6 ± 33.9 ms, respectively. The absolute difference between the mean reversal and FOS estimate was calculated for each dataset; the mean difference was 9.4 ms with a standard deviation of 11.5 ms. Further analyses, described in the following sections of this paper, were performed using only those datasets for which the mean reversal values for both ascending and descending staircases were within 40 ms of the FOS estimates. Employing this criteria resulted in inclusion of 94.2% of the datasets (N = 324). FOS curves for these data had an average slope of 30.0 ± 23.3 compared to an average slope of 11.0 ± 14.8 for data in which the mean reversal values differed from the FOS estimates by 40 or more msec. The corresponding width estimates based on the 10% and 90% points of the psychometric functions are 96.1 ± 86.5 and 260.6 ± 151.1, respectively (Kuss, Jäkel, & Wichmann, 2005).
During a session, two measures of DL were each obtained for all the four experimental conditions. Differences between the first and second DLs for a given condition are plotted as a histogram in Fig. 2A (N = 154) (K–S: d = 0.054, p > 0.20; SW: W = 0.981, p = 0.029). Plotting within-session variability as a function of the averaged values for the two sessions suggests that extreme DLs do not affect within-session variability (r = 0.061, p = 0.452) (Fig. 2B) (Bland & Altman, 2003). Neither retinal location (F = 0.089, p = 0.767) nor inter-trial step size (F = 0.60, p = 0.440) significantly affected within-session DL variability (within-subjects ANOVA). To ascertain a subject’s between-session variability, the average DL obtained in the first session for a given condition was subtracted from the average value obtained in the second session (Fig. 3A; N = 70) (K–S: d = 0.072, p > 0.20, but SW: W = 0.979, p = 0.311). Plotting between-session variability as a function of the averaged values for the two sessions suggests that betweensession variability is not affected by extreme DL values (r = 0.019, p = 0.879) (Fig. 3B) (Bland & Altman, 2003). Neither retinal location (F = 0.958, p = 0.331) nor inter-trial step size (F = 1.25, p = 0.267) had a significant effect on between-session variability (within-subjects ANOVA).
3.2. Differential latencies
4. Discussion
Fig. 1A shows the distribution of all differential latencies across all sessions and conditions for all subjects (N = 324) (Kolmogorov– Smirnov (K–S): d = 0.057, p > 0.20; Shapiro–Wilk (SW): W = 0.987, p = 0.005). The scatterplot in Fig. 1B shows DL values grouped by the mean subject DL. Fourteen of the twenty-two subjects (63.6%) manifested shorter latencies for the superior visual field. The mean subject DL (based on 16 measurements for most subjects) ranges from 4.9 to 14.8 ms, with an average (per subject) value of 3.9 ± 5.9 ms. This is significantly different than the expected value of zero that would be found in the case of perfect vertical symmetry (t = 3.09, p = 0.0056). Neither retinal location (F = 0.012, p = 0.912) nor inter-trial step size (F = 0.60, p = 0.442) significantly affected DL (within-subjects ANOVA). The age distribution of the subjects was bimodal, with a group of 12 subjects in their 20 s and the remaining 10 with ages between 48 and 79. DL did not vary significantly between these two age groups (t = 0.798; p = 0.435).
4.1. Data analysis
3.3. Comparison to results obtained with FOS analysis DL was also calculated using the simultaneity ranges as determined with the FOS curves. Across subjects, the average value was 4.1 ± 6.2 ms, which is not statistically different than that obtained using the mean reversal data (t-test, dependent groups: t = 0.418, p = 0.680). As is the case for the DL values that are based on mean reversals, the average DL is significantly different than the expected value of zero that would be found in the case of perfect vertical symmetry (t = 3.14, p = 0.0049). The upper boundary of the matching range given by mean reversals was 64.3 ± 29.3 ms, while the FOS estimate for this boundary was 58.9 ± 28.6 ms. Comparable data for the lower boundary are 72.1 ± 31.0 and 67.1 ± 28.1 ms, respectively. In both cases, the absolute mean reversal values are slightly, but significantly, larger than the FOS estimates (t-test, dependent samples: t = 6.07, p < 0.05 and t = 5.23, p < 0.05). The average matching range based on mean reversals (136.4 ± 59.0) is significantly wider than that based on FOS estimates (126.1 ± 55.3; t-test, dependent samples: t = 5.20, p < 0.005).
For the most part, the results from the mean reversal and FOS analysis were very similar, but the range was slightly, but significantly, wider when determined with the mean reversal data. Compared to the data obtained from the FOS curves, the range is increased approximately symmetrically at both upper and lower boundaries such that the DL obtained with mean reversals and FOS curves are statistically the same. A comment is in order regarding the use of yes–no psychometric functions (with threshold taken as 50% correct) rather than a two-alternative forced choice (2AFC). We initially considered requiring subjects to indicate which came on first, without providing the ‘‘simultaneous/not sure” option. The decision to provide a ‘‘simultaneous/not sure” option is supported by the TOJ work of Jas´kowski (1993a) who reported that forcing a choice between ‘‘top first” and ‘‘bottom first” introduces an artifactual bias toward one of these choices when the subject is not sure which came on first. He found that giving the subject a third option of ‘‘simultaneously” reduced the bias. This is consistent with our experiences in preliminary 2AFC observations: when an option of ‘‘simultaneous/not sure” was not available, there was a temporal offset that was not apparent when subjects were allowed to select ‘‘simultaneous/not sure.” 4.2. Comparison to previous studies Previous studies employing visually evoked potentials and motor reaction times have tended to find slightly, but significantly, shorter latencies for the inferior visual field. The most comprehensive electrophysiological data have been reported by Skrandies (1987), who found evoked potential latencies for inferior stimulation to be 8.6–16.5 ms shorter, depending on the mode of stimulus presentation (onset, offset, or pattern reversal). Examining motor reactions times to small (15 min arc) suprathreshold flashes presented in a circle centered on the point of fixation (15° radius), Payne (1967) found latencies to be shortest for the inferior nasal field and longest in the superior temporal field,
537
S. Patel et al. / Vision Research 50 (2010) 534–540
A
80 70
No. of obs.
60 50 40 30 20 10 0 -50
-40
-30
-20
-10
0
10
20
30
Differential Latency (msec)
B
40
Differential Latency (msec)
30 20 10 0 -10 -20 -30 -40 -50 -60 -16
-14
-12
-10
-8
-6
-4
-2
0
2
4
6
Subject Mean (msec) Fig. 1. A. Histogram showing the distribution of differential latency (DL) for 22 subjects (N = 324). B. DLs grouped by mean subject DL. The solid line is the mean of all data points, and the dashed lines represent the 95% CI. Values greater than zero indicate a shorter latency for the inferior field, while those less than zero show a shorter latency for the superior field.
with a difference of about 23 ms. Maehara et al. (2004) found similar results using stimuli ranging from 16–64 min arc (133 ms duration) presented 7° superior or inferior to fixation on various chromatic backgrounds. The average motor reaction time for the inferior visual field was 359 ms, compared to 373 ms for the superior visual field, a significant difference. The data reported herein, obtained with a simultaneity paradigm, are consistent with previously published reaction time data in showing approximate vertical symmetry for superior and inferior field latencies. Motor reaction time studies show slightly shorter latencies for the inferior field, while simultaneity data reveal slightly shorter latencies for the superior field. The different stimulus sizes, luminances, chromaticities, and locations that were used for the motor reaction time and simultaneity experiments
may contribute to the discrepant results. Moreover, TOJ and reaction times may not equivalently measure visual latency, possibly due to mediation by different processing streams (Cardoso-Leite, Gorea, & Mamassian, 2007; Jas´kowski, 1993b, 1996; Tappe, Niepel, & Neumann, 1994). Retinal location and inter-trial step sizes were not significant factors in the current study. The simultaneity data were robust in that there was little variability between subjects and little withinand between-session variability. These issues are not addressed in previous motor reaction time studies. Based on the results of our study and those employing motor reactions times, it appears that any latency difference between the superior and inferior fields in healthy visual systems is rather small and, for clinical purposes, essentially nil.
538
S. Patel et al. / Vision Research 50 (2010) 534–540
A
40 35
No. of obs.
30 25 20 15 10 5 0 -40
-30
-20
-10
0
10
20
30
40
50
DL1 - DL2 (msec)
B
60 50 40
DL1-DL2 (msec)
30 20 10 0 -10 -20 -30 -40 -40
-30
-20
-10
0
10
20
30
Mean: DL1, DL2 (msec) Fig. 2. A. Histogram showing differences between the first and second DL measured in a given session (DL1 – DL2) for the same set of experimental conditions (N = 154). B. Within-session differences in DL plotted as a function of the average of the two values that were subtracted to obtain this difference. The regression line (solid) and 95% CI (dashed) are given.
4.3. Clinical implications While a motor reaction time methodology may provide useful information in a younger, healthy population, its usefulness in an elderly clinical population, where varying degrees of motor impairment are not uncommon, is limited (Crawford et al., 1989; Fozard et al., 1994; Volkow et al., 1998). Motor impairment is not expected to be a factor when a simultaneity paradigm is employed to assess differential latency, and no age-related differences were found in the current study. Visual latency increases with decreasing visibility, so it would be expected that the increased visual thresholds found in visual field defects would result in increased latencies for these loci (Becker, Vonthein, Volpe, & Schiefer, 2005; Flammer, Drance, & Schulzer, 1984; Mihaylova, Stomonyakov, & Vassilev, 1999; Roufs,
1974; Ueno, 1979). Static perimetric studies of glaucoma patients find that while threshold reaction times are not significantly different in damaged and normal regions of the visual field, they are increased for high luminance stimuli (0 dB, which is about 3183 cd/ m2) presented with the Humphrey perimeter (Wall, Kutzko, & Chauhan, 2002; Wall, Maw, Stanek, & Chauhan, 1996). The increase was attributed to the decreased visibility/detectability of stimuli that fall within glaucomatous defects. Neurological damage to the optic nerve fibers that occurs in glaucoma frequently manifests as an altitudinal nasal visual field defect, referred to as a nasal step (Aulhorn & Karmeyer, 1977; Hart & Becker, 1982; Heijl & Lundqvist, 1984; Katz et al., 1995). Decreased sensitivity in a damaged region is expected to result in an increased latency relative to normal regions (Artes, McLeod, & Henson 2002; Flammer et al., 1984; Nowomiejska et al., 2008; Wall
539
S. Patel et al. / Vision Research 50 (2010) 534–540
A
20 18 16
No. of obs.
14 12 10 8 6 4 2 0 -35
-25
-15
-5
5
15
25
DLS1 - DLS2 (msec)
B
40 30
DL S1 - DL S2 (msec)
20 10 0 -10 -20 -30 -40 -30
-25
-20
-15
-10
-5
0
5
10
15
Mean: DL S1, DL S2 (msec) Fig. 3. A. Histogram showing differences between the average DLs obtained in sessions 1 and 2 (DLS1 – DLS2) for the same set of experimental conditions (N = 70). B. Between-session differences in DL plotted as a function of the average of the two values that were subtracted to obtain this difference. The regression line (solid) and 95% CI (dashed) are given.
et al., 1996). For the simultaneity paradigm introduced in this paper, DLs are predicted to be asymmetric when testing nasal steps if one of the paired stimuli falls in a region of the visual field less damaged than the fellow stimulus. Sensitivities to inferior and superior achromatic stimuli, presented with a Humphrey Visual Field Analyzer in the approximate loci tested in the current experiment, are essentially the same (about 31 dB, or 2.5 cd/m2) (Heijl, Lindgren, & Olsson, 1987). Based on considerations of sensitivity alone, it is not unexpected that the suprathreshold stimuli employed in the current study (66 cd/m2) elicit approximately vertically symmetrical latencies. Additionally, our simultaneity data suggest that anatomical and physiological asymmetries found in the primate visual system would not result in vertical latency asymmetry that could be a potentially confounding factor in suprathreshold clinical testing.
The data reported herein show that DL has relatively little variability across subjects and strong within- and between-session reliability. These results point to its possible clinical utility to compare functionality of superior and inferior visual fields. Because testing utilizes a paired stimulus, with each of the stimulus components presented at a different retinal locus, the technique has the potential advantage of comparing function of disparate retinal loci with nearly simultaneous stimulation. In visual fields and other traditional procedures, thresholds determined for individual points are compared with those determined previously or thereafter during the session. While these measurements may be close in time, they are not simultaneous, introducing the potential of sensitivity drifts. Use of paired stimuli may provide a more direct comparison of retinal function.
540
S. Patel et al. / Vision Research 50 (2010) 534–540
Acknowledgments This research was supported by a Grant from Vision Service Plan (Schwartz), Beta Sigma Kappa (Patel), and the SUNY College of Optometry Office of Graduate Studies. Dr. Swanson was supported by NIH Grant EY007716. A portion of this research was presented at the 2008 meeting of the Association for Research in Vision and Ophthalmology in Ft. Lauderdale, FL. We appreciate the thoughtful comments of the referees. References Artes, P. H., McLeod, D., & Henson, D. B. (2002). Response time as a discriminator between true- and false-positive responses in suprathreshold perimetry. Investigative Ophthalmology and Visual Science, 43, 129–132. Aulhorn, E., & Karmeyer, H. (1977). Frequency distribution in early glaucomatous defects. In E. L. Greve (Ed.). Documenta ophthalmologica proceeding series (Vol. 14, pp. 75–83). The Hague: Dr. W. Junk bv Publishers. Becker, S. T., Vonthein, R., Volpe, N. J., & Schiefer, U. (2005). Factors influencing reaction time during automated kinetic perimetry on the Tubingen Computer Campimeter. Investigative Ophthalmology and Visual Science, 46, 2633–2638. Bland, J. M., & Altman, D. G. (2003). Applying the right statistics: Analysis of measurement studies. Ultrasound in Obstetrics and Gynecology, 22, 85–93. Cardoso-Leite, P., Gorea, A., & Mamassian, P. (2007). Temporal order judgment and simple reaction times: Evidence for a common procession system. Journal of Vision, 7(6), 11. 1–14. Carlsen, A. N., Maslovat, D., Chua, R., & Franks, I. M. (2007). Perceptual processing differences owing to visual field asymmetries. Neuroreport, 18(10), 1067–1070. Chen, Y., Wyatt, H. J., & Swanson, W. H. (2005). Pupillary evaluation of retinal asymmetry: Development and initial testing of a technique. Vision Research, 45, 2549–2563. Connolly, M., & Van Essen, D. C. (1984). The representation of the visual field in parvocellular and magnocellular layers of the lateral geniculate nucleus in the macaque monkey. The Journal of Comparative Neurology, 226(4), 544–564. Crawford, T., Goodrich, S., Henderson, L., & Kennard, C. (1989). Predictive responses in Parkinson’s disease: Manual keypresses and saccadic eye movements to stimulus events. Journal of Neurology, Neurosurgery, & Psychiatry, 52, 1033–1042. Curcio, C. A., & Allen, K. A. (1990). Topography of ganglion cells in human retina. The Journal of Comparative Neurology, 300(1), 5–25. Curcio, C. A., Sloan, K. R., Kalina, R. E., & Hendrickson, A. E. (1990). Human photoreceptor topography. The Journal of Comparative Neurology, 292(4), 497–523. Danckert, J., & Goodale, M. A. (2001). Superior performance for visually guided pointing in the lower visual field. Experimental Brain Research, 137, 303–308. Flammer, J., Drance, S. M., & Schulzer, M. (1984). Covariates of the long-term fluctuations of differential light threshold. Archives of Ophthalmology, 102, 880–882. Fozard, J. L., Vercruyssen, M., Reynolds, S. L., Hancock, P. A., & Quilter, R. E. (1994). Age differences and changes in reaction time: The Baltimore longitudinal study of aging. Journal of Gerontology, 49(4), 179–189. Galletti, C., Fattori, P., Kutz, D. F., & Gamberini, M. (1999). Brain location and visual topography of cortical area V6A in the macaque monkey. European Journal of Neuroscience, 11, 575–582. Hart, W. M., Jr., & Becker, B. (1982). The onset and evolution of glaucomatous visual field defects. Ophthalmology, 89(3), 268–279. Heijl, A., Lindgren, G., & Olsson, J. (1987). Normal variability of static perimetric threshold values across the central visual field. Archives of Ophthalmology, 105, 1544–1549. Heijl, A., & Lundqvist, L. (1984). The frequency distribution of earliest glaucomatous visual field defects documented by automatic perimetry. Acta Ophthalmologica, 62(4), 658–664. Horn, F., Mardin, C., Korth, M., & Martus, P. (1996). Quadrant pattern ERG and SLO stimulation in normals and glaucoma patients. Graefe’s Archive for Clinical and Experimental Ophthalmology, 234(Suppl. 1), S174–S179. Jas´kowski, P. (1993a). Selective attention and temporal-order judgment. Perception, 22, 681–689. Jas´kowski, P. (1993b). Temporal-order judgment and reaction time to stimuli of different rise times. Perception, 22, 963–970. Jas´kowski, P. (1996). Simple reaction time and perception of temporal order: Dissociations and hypothesis. Motor and Perceptual Skills, 82(3), 707–730.
Katz, J., Quigley, H. A., & Sommer, A. (1995). Repeatability of the Glaucoma Hemifield Test in automated perimetry. Investigative Ophthalmology and Visual Science, 36(8), 1658–1664. Khan, M. A., & Lawrence, G. P. (2005). Differences in visuomotor control between the upper and lower visual fields. Experimental Brain Research, 164, 395–398. Kuss, M., Jäkel, F., & Wichmann F. A. (2005). Approximate Bayesian inference for psychometric functions using MCMC sampling. Technical Report No. 135. Max Planck Institute for Biological Cybernetics. Levine, M. W., & McAnany, J. J. (2005). The relative capabilities of the upper and lower hemifields. Vision Research, 45(21), 2820–2830. Liu, T., Heeger, D. J., & Carrasco, M. (2006). Neural correlates of the visual vertical meridian asymmetry. Journal of Vision, 6, 1294–1306. Maehara, G., Okubo, M., & Michimata, C. (2004). Effects of background color on detecting spot stimuli in the upper and lower visual fields. Brain and Cognition, 55, 558–563. Maunsell, J. H. R., & Van Essen, D. C. (1987). Topographic representation of the middle temporal visual area in the macaque monkey: Representational biases and the relationship to callosal connections and myeloarchitectocnic boundaries. The Journal of Comparative Neurology, 266(4), 535–555. Mihaylova, M., Stomonyakov, V., & Vassilev, A. (1999). Peripheral and central delay in processing high spatial frequencies: Reaction time and VEP latency studies. Vision Research, 39(4), 699–705. Nowomiejska, K., Vonthein, R., Paetzold, J., Zagorski, Z., Kardon, R., & Schiefer, U. (2008). Reaction time during semi-automated kinetic perimetry (SKP) in patients with advanced visual field loss. Acta Ophthalmologica, 1, 5. Payne, W. H. (1967). Visual reaction times on a circle about the fovea. Science, 155, 481–482. Portin, K., Vanni, S., Virsu, V., & Hari, R. (1999). Stronger occipital cortical activation to lower than upper visual field stimuli: Neuromagnetic recordings. Experimental Brain Research, 124, 287–294. Previc, F. H. (1990). Functional specialisation in the lower and upper visual fields in humans: Its ecological implications. Behavior and Brain Sciences, 13, 519–542. Roufs, J. A. (1974). Dynamic properties of vision. V. Perception lag and reaction time in relation to flicker and flash thresholds. Vision Research, 14(9), 853–869. Sample, P. A., Irak, I., Martinez, G. A., & Yamagishi, N. (1997). Asymmetries in the normal short-wavelength visual field: Implications for short-wavelength perimetry. American Journal of Ophthalmology, 124, 46–52. Schwartz, S. H. (1992). Reaction time distributions and their relationship to the transient/sustained nature of the neural discharge. Vision Research, 32(11), 2087–2092. Schwartz, S. H., & Loop, M. S. (1982). Evidence for transient luminance and quasisustained color mechanisms. Vision Research, 22(4), 445–447. Skrandies, W. (1987). The upper and lower visual field of man: Electrophysiological and functional differences. In D. Ottoson (Ed.). Progress in sensory physiology (Vol. 8, pp. 1–93). Berlin: Springer-Verlag. Swanson, W. H. (1993). Analysis of Rayleigh match data with psychometric functions. Journal of the Optical Society of America A, 10(8), 1807–1817. Tappe, T., Niepel, M., & Neumann, O. (1994). A dissociation between reaction time to sinusoidal gratings and temporal-order judgment. Perception, 23, 335–347. Tootell, R. B. H., Switkes, E., Silverman, M. S., & Hamilton, S. L. (1988). Functional anatomy of macaque striate cortex. II. Retinotopic organization. The European Journal of Neuroscience, 8(5), 1531–1568. Ueno, T. (1979). Spatial summation in human vision: Simple reaction time measurements. Journal of the Optical Society of America, 69(7), 1023–1028. van Essen, D. C., Newsome, W. T., & Maunsell, J. H. R. (1984). The visual field representation in striate cortex of the macaque monkey: Asymmetries, anisotropies, and individual variability. Vision Research, 24, 429–448. Volkow, N. D., Gur, R. C., Wang, G.-J., Fowler, J. S., Moberg, P. J., Ding, Y.-S., et al. (1998). Association between decline in brain dopamine activity with age and cognitive and motor impairment in healthy individuals. American Journal of Psychiatry, 155, 344–349. Vrabec, F. (1966). The temporal raphe of the human retina. Journal of Ophthalmology, 62(5), 926–938. Wall, M., Kutzko, K. E., & Chauhan, B. C. (2002). The relationship of visual threshold and reaction time to visual field eccentricity with conventional automated perimetry. Vision Research, 42, 781–787. Wall, M., Maw, R. J., Stanek, K. E., & Chauhan, B. C. (1996). The psychometric function and reaction times of automated perimetry in normal and abnormal areas of the visual field in patients with glaucoma. Investigative Ophthalmology and Visual Science, 37, 878–885. Yoshii, M., & Paarmann, A. (1989). Hemiretinal stimuli elicit different amplitudes in the pattern electroretinogram. Documenta Ophthalmologica, 72(1), 21–30. Zhe, Q., Van, S., & Yulong, D. (2006). Asymmetry between the upper and lower visual fields: An event-related potential study. Chinese Science Bulletin, 51(5), 536–541.