ࡱ > @ c bjbjFF , , Z
4N 4N 4N 8 lN | N ) * O Z ( [ [ [ _ _ _ $ S R L Ι c ^ _ c c Ι [ [ k k k c &/ [ [ * ~ k c k k Ɠ [ O R 4N - ޓ 0 ) - " _ c` ^ k ` L
a _ _ _ Ι Ι $I 4N O 4N ELECTRONIC TEXT DISPLAYS: READING REHABILITATION OF LOW VISION PATIENTS WITH AGE-RELATED MACULAR DEGENERATION
By
Kathy Aquilante
DISSERTATION
In partial satisfaction of the requirements for the degree of
Doctor of Philosophy
in
Vision Science
Graduate program State University of New York
State College of Optometry
March 26, 2003
Acknowledgments
A special thank you to my mentor and thesis advisor Dr. Dean Yager
Thank you to my committee memebers
Dr. William H. Swanson
Dr. Philip Kruger
And for making this work possible thank you to the National Institutes of Health for grants K23-EY-00366 (KA) and R01-EY-11617 (DY) and to the Schnurmacher Institute for Vision Resarch at SUNY State College of Optometry
ELECTRONIC TEXT DISPLAYS: READING REHABILITATION OF LOW VISION PATIENTS WITH AGE-RELATED MACULAR DEGENERATION
Background and Significance
According to the World Health Organization (WHO) blindness is the most expensive cause of serious disablement worldwide ADDIN EN.CITE WHO-Study-Group19733211WHO-Study-Group1973The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518GenevaWorld Health Organization(WHO-Study-Group, 1973). Based on fragmentary data, thought to underestimate the true of number of afflicted, the WHO study group estimates that there are at least 15 million blind persons. Blindness defined by WHO is best-corrected visual acuity of 20/400 or worse in the better eye ADDIN EN.CITE WHO-Study-Group19733211WHO-Study-Group1973The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518GenevaWorld Health Organization(WHO-Study-Group, 1973). Most industrialized nations have adopted a standard that is different than the one developed by WHO for defining legal blindness, using best-corrected visual acuity that is equal to or worse than 20/200 in the better eye. The WHO Study Group did not provide world-wide estimates of the number of persons with visual impairment (defined as best-corrected visual acuity of less than 20/70 but better than 20/400) even though this level of vision loss can cause disability ADDIN EN.CITE WHO-Study-Group19733211WHO-Study-Group1973The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518GenevaWorld Health Organization(WHO-Study-Group, 1973). WHO defines disability as the lost "ability to perform a specific task" resulting from any measurable loss of functional capability (impairment) ADDIN EN.CITE Leat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211RepeatMassof19953160Massof, R.W Dagnelie, G Deremeik, J.L Suleiman, S.A Glasner, N.M1995Low vision rehabilitation in the US health care system.Journal of Vision Rehabilitation93-25WHO-Study-Group19733211WHO-Study-Group1973The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518GenevaWorld Health Organization(Leat, Legge & Bullimore, 1999; Massof, Dagnelie, Deremeik, Suleiman & Glasner, 1995; WHO-Study-Group, 1973).
The International Classification of Impairments, Disabilities and Handicaps (ICIDH) developed by WHO attempts to standardize the description of the consequences of disease (WHO Study Group, 1973; ADDIN EN.CITE Dickinson19983331Dickinson, C.1998Low Vision. Principles and PracticeOxfordButterworth HeinemannLeat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211RepeatMassof19953160Massof, R.W Dagnelie, G Deremeik, J.L Suleiman, S.A Glasner, N.M1995Low vision rehabilitation in the US health care system.Journal of Vision Rehabilitation93-25(Dickinson, 1998; Leat, et al., 1999; Massof, et al., 1995). In the WHO standard there are two categories of visual impairment: (1) visual acuity less than 20/70 but better than 20/200, and (2) visual acuity less than 20/200 but better than 20/400. Impairment may result from any physiological "disorder". For example, the disorder of age-related maculopathy involves the progressive degeneration of photoreceptors in the macula. A disorder results in an impairment when a patient does not perform as well as "normal" on a standardized test, such as visual acuity ADDIN EN.CITE Dickinson19983331Dickinson, C.1998Low Vision. Principles and PracticeOxfordButterworth HeinemannLeat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211Repeat(Dickinson, 1998; Leat, et al., 1999). If the impairment results in a lost capacity to perform a specific task that is necessary for maintaining that individuals desired lifestyle, then a disability is present. If no compensation can be made to overcome the disability, then the individual becomes handicapped ADDIN EN.CITE Leat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211RepeatMassof19953160Massof, R.W Dagnelie, G Deremeik, J.L Suleiman, S.A Glasner, N.M1995Low vision rehabilitation in the US health care system.Journal of Vision Rehabilitation93-25WHO-Study-Group19733211WHO-Study-Group1973The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518GenevaWorld Health Organization(Leat, et al., 1999; Massof, et al., 1995; WHO-Study-Group, 1973).
Eligibility for funding for low vision aids, rehabilitative intervention, and support services intended to overcome handicaps, is dependent on the classification and definitions of visual impairment and visual disability ADDIN EN.CITE Leat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211Repeat(Leat, et al., 1999). In the United States, rehabilitative services to visually impaired and legally blind persons have been funded primarily through the U.S. Department of Education ADDIN EN.CITE Massof20013811Massof, R. W.Lidoff, L2001Issues in Low Vision Rehabilitation. Service Delivery, Policy, and FundingNew YorkAFB Press(Massof & Lidoff, 2001). Allocation of funding is determined by each state. In New York State the range of available services varies on a case by case basis. Persons who do not have educational or vocational goals are eligible for services only if they are legally blind, and the resources available for this group are very limited ADDIN EN.CITE Massof20013811Massof, R. W.Lidoff, L2001Issues in Low Vision Rehabilitation. Service Delivery, Policy, and FundingNew YorkAFB Press(Massof & Lidoff, 2001). The reported prevalence of legal blindness is variable, ranging from 3% to 8.2% for persons over the age of 85 years compared to less than 1% of the population under 75 years ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755(Brabyn, Haegerstrom-Portnoy, Schneck & Lott, 2000). There are many individuals who are not legally blind who have a disability due to a visual impairment ADDIN EN.CITE Leat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211Repeat(Leat, et al., 1999). The majority of patients who seek low vision care in order to overcome a disability are over age 65 ADDIN EN.CITE Elliott19971390Elliott, D.BTrukolo-Ilic, MStrong, J.GPace, RPlotkin, ABevers, P1997Demographic characteristics of the vision-disabled elderlyInvest Ophthalmol Vis Sci382566-2575repeat(Elliott, Trukolo-Ilic, Strong, Pace, Plotkin & Bevers, 1997) and do not have educational or vocational goals. For this population the optical devices and assistive technology that are necessary for overcoming potentially handicapping vision loss are an out-of-pocket expense.
In the United States there is no nationally agreed upon standard for defining levels of visual impairment other than legal blindness ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755(Brabyn, et al., 2000). Many states have a legal criterion of best-corrected visual acuity of 20/40 as the cutoff for an unrestricted drivers license ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755Tielsch20013787Tielsch, J.M2001Prevalence of Visual Impairment and Blindness in the United StatesMassof, R. W.Lidoff, LIssues in Low Vision Rehabilitation. Service Delivery, Policy, and FundingNew YorkAFB Press13-26(Brabyn, et al., 2000; Tielsch, 2001). In most states entitlement to educational services for the visually impaired requires best-corrected visual acuity equal to or worse than 20/70 ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755(Brabyn, et al., 2000).
A standard that defines visual impairment as best-corrected visual acuity worse than 20/70 may not be adequate ADDIN EN.CITE Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatTielsch20013787Tielsch, J.M2001Prevalence of Visual Impairment and Blindness in the United StatesMassof, R. W.Lidoff, LIssues in Low Vision Rehabilitation. Service Delivery, Policy, and FundingNew YorkAFB Press13-26(Legge, Pelli, Rubin & Schleske, 1985a; Legge, Rubin, Pelli & Schleske, 1985b; Tielsch, 2001). Several investigators argue that impairment be defined as best-corrected visual acuity worse than 20/40 but better than 20/200 in the better eye ADDIN EN.CITE Tielsch20013787Tielsch, J.M2001Prevalence of Visual Impairment and Blindness in the United StatesMassof, R. W.Lidoff, LIssues in Low Vision Rehabilitation. Service Delivery, Policy, and FundingNew YorkAFB Press13-26Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266(Legge, et al., 1985b; Tielsch, 2001). Low vision has been defined as the inability to read a newspaper, with best spectacle correction, at a normal reading distance (40 cm) ADDIN EN.CITE Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatLegge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266(Legge, et al., 1985a; Legge, et al., 1985b). Further, the use of high contrast visual acuity charts may not provide a true measure of the impact that vision loss has in everyday situations, particularly for the elderly ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755Charness1999250Charness, NDijkstra, K1999Age, luminance and print legibility in homes, offices and public placesHuman Factors41173-193Friedman19991650http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0010102282Friedman, S. M.Munoz, B.Rubin, G. S.West, S. K.Bandeen-Roche, K.Fried, L. P.Characteristics of discrepancies between self-reported visual function and measured reading speed. Salisbury Eye Evaluation Project TeamAgedAged, 80 and overContrast Sensitivity/physiologyFemaleGeriatric AssessmentHumanMaleMaryland/epidemiologyPrevalence*Reading*Self DisclosureSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision Disorders/*diagnosis/epidemiology/physiopathology*Vision TestsVisual Acuity/*physiologyDepartment of Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA.0010102282Invest Ophthalmol Vis Sci1999405858-64Haegerstrom-Portnoy199950G Haegerstrom-PortnoyM.E SchneckJ.A Brabyn1999Seeing into Old Age: Vision Function Beyond AcuityOptom Vision Sci76141-158Rubin19941780http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0007898882Rubin, G. S.Roche, K. B.Prasada-Rao, P.Fried, L. P.Visual impairment and disability in older adultsActivities of Daily LivingAgedAged, 80 and overAging/physiologyContrast Sensitivity/physiology*Disabled PersonsFemaleHumanMaleQuestionnairesRisk FactorsSelf DisclosureSupport, U.S. Gov't, P.H.S.Vision Disorders/*physiopathologyVision TestsVisual Acuity/physiologyWilmer Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland.0007898882Optom Vis Sci19947112750-60(Brabyn, et al., 2000; Charness & Dijkstra, 1999; Friedman, Munoz, Rubin, West, Bandeen-Roche & Fried, 1999; Haegerstrom-Portnoy, Schneck & Brabyn, 1999; Rubin, Roche, Prasada-Rao & Fried, 1994). Under conditions of low luminance and reduced contrast 57% of subjects 85 years of age and older had visual acuity worse than 20/200 compared to 4% when the measure was high contrast acuity ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755(Brabyn, et al., 2000). By comparison, the same conditions of low luminance and reduced contrast resulted in 24% of adults under age 55 failing to attain visual acuity of 20/40. No subjects in the younger age group were reduced to worse than 20/70 visual acuity under conditions of low luminance and reduced contrast ADDIN EN.CITE Brabyn20003270Brabyn, J.AHaegerstrom-Portnoy, GSchneck, M. E.Lott, L.A2000Visual impairments in elderly people under everyday viewing conditionsJ Vis Impair Blind94741-755(Brabyn, et al., 2000).
More than three million people in the United States suffer from impaired vision to the extent of being unable to read ordinary print with corrective lenses ADDIN EN.CITE Tielsch19903100Tielsch, J.MSommer, AWitt, KKatz, JRoyall, R.M1990Blindness and Visual Impairment in an American Urban PopulationArch Ophthalmol108286-290Lighthouse19953771Lighthouse, National Survey, on, Vision, Loss,1995The experience, attitude, and knowledge of middle-aged and older AmericansNew YorkLighthouse(Lighthouse, National, Survey, on, Vision & Loss, 1995; Tielsch, Sommer, Witt, Katz & Royall, 1990). Shifts in the age distribution of the population are expected to result in a doubling of the number of visually impaired and legally blind persons by the year 2030 ADDIN EN.CITE Bureau19963791Bureau of theCensus, U.S. Departmentof Commerce1996Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050.Current Population Reports (P25-1130)Washington, D.C.(Bureau, of, the, Census, U.S., Department, of & Commerce, 1996).
Age-related macular degeneration (AMD) is the leading cause of irreversible central vision loss in persons older than 65 years of age. Between 4 and 10 million people in the United States are estimated to have AMD ADDIN EN.CITE Kini19783070Kini, M.MLeibowitz, H.M Colton, T1978Prevalence of senile cataract, diabetic retinopathy, senile macular degeneration, and open-angle glaucoma in the Framingham Eye StudyAm J Ophthalmol8528-34Klein19921120Klein, RKlein, B.E.KLinton, K.L.P1992Prevalence of age-related maculopathy: the Beaver Dam Eye StudyOphthalmology99933-943Sommer19911210Sommer, ATielsch, J.MKatz, JQuigley, H. A.Gottsch, J.DJavitt, J.CMartone, J.FRoyall, R.MWitt, KEzrine, S1991Racial differences in the cause-specific prevalence of blindness is east BaltimoreN Engl J Med3251412-1417Berger19993081Berger, J.WFine, S.LMaguire, M.G1999Age-related Macular DegenerationSt. LouisMosby(Berger, Fine & Maguire, 1999; Kini, Leibowitz & Colton, 1978; Klein, Klein & Linton, 1992; Sommer, Tielsch, Katz, Quigley, Gottsch, Javitt, Martone, Royall, Witt & Ezrine, 1991). The prevalence of AMD in the United States is strongly age-related with rates of 1.6% in 52-64 years olds, 11% in 65-74 year olds, and 27.9% in those 75 years or older ADDIN EN.CITE Leibowitz19801410Leibowitz, H.MKrueger, D.EMaunder, L.RMilton, R.CKini, M.MKahn, H.ANickerson, R.JPool, JColton, T.LGanley, J.PLoewenstein, J.IDawber, T.R1980The Framingham Eye Study monographSurv Ophthalmol24(suppl)335-610(Leibowitz, Krueger, Maunder, Milton, Kini, Kahn, Nickerson, Pool, Colton, Ganley, Loewenstein & Dawber, 1980). Similar findings were reported in a recent population-based cross-sectional study from Australia. AMD increased from 0% in 40-49 year old participants to 27% of participants 90 years or older ADDIN EN.CITE VanNewkirk20002930VanNewkirk, M.RNanjan, M.BWang, J.JMitchell, PTaylor, H. R.McCarty, CA2000The prevalence of age-related maculopathy. The visual impairment projectOphthalmology1593-1600(VanNewkirk, Nanjan, Wang, Mitchell, Taylor & McCarty, 2000).
Early anatomical signs of AMD have been defined as the presence of soft distinct, soft indistinct, or reticular drusen, or the presence of retinal pigmentary abnormalities in the absence of late AMD lesions ADDIN EN.CITE VanNewkirk20002930VanNewkirk, M.RNanjan, M.BWang, J.JMitchell, PTaylor, H. R.McCarty, CA2000The prevalence of age-related maculopathy. The visual impairment projectOphthalmology1593-1600(VanNewkirk, et al., 2000). When early signs of AMD are included in prevalence studies almost three quarters of people who reach the age of 90 years can be expected to have age-related maculopathy ADDIN EN.CITE VanNewkirk20002930VanNewkirk, M.RNanjan, M.BWang, J.JMitchell, PTaylor, H. R.McCarty, CA2000The prevalence of age-related maculopathy. The visual impairment projectOphthalmology1593-1600(VanNewkirk, et al., 2000). One third of all cases have late AMD defined as wet AMD (neovascular) or dry AMD (geographic atrophy) ADDIN EN.CITE VanNewkirk20002930VanNewkirk, M.RNanjan, M.BWang, J.JMitchell, PTaylor, H. R.McCarty, CA2000The prevalence of age-related maculopathy. The visual impairment projectOphthalmology1593-1600(VanNewkirk, et al., 2000). Severe vision loss occurs whenever choroidal neovascularization or geographic atrophy of the retinal pigment epithelium involves the foveal center ADDIN EN.CITE Ferris19841090Ferris, F.L. IIIFine, S.LHyman, L.1984Age-related macular degeneration and blindness due to neovascular maculopathyArch Ophthalmol1021640-1612(Ferris, Fine & Hyman, 1984). In the United States approximately 540,000 people are estimated to have late AMD ADDIN EN.CITE Klein19991240Klein, RKlein, B.E.KJensen, S.CMares-Perlman, J.ACruickshanks, K.JPalta, M1999Age-related maculopathy in a multiracial United States population. The National Health and Nutrition Examination Survey IIIOphthalmology1061056-1065(Klein, Klein, Jensen, Mares-Perlman, Cruickshanks & Palta, 1999). Because individuals over 65 years old are one of the fastest growing sectors of the population, AMD is a major public health problem ADDIN EN.CITE Seddon19943420Seddon, J.MAjani, U.ASperduto, R.DHiller, RBlair, NBurton T.CFarber, MDGragoudas, ESHaller, JMiller, D.TYannuzzi, L.AWillett, WEyeDisease Case-ControlStudyGroup1994Dietary Carotenoids, Vitamins A, C, and E, and Advanced Age-Related Macular DegenerationJAMA2721413-1420(Seddon, Ajani, Sperduto, Hiller, Blair, T.C, Farber, Gragoudas, Haller, Miller, Yannuzzi, Willett, Eye, Disease, Case-Control, Study & Group, 1994).
There are currently no effective treatments for the vast majority, about 80%, of patients who have the non-neovascular (dry) form of AMD ADDIN EN.CITE Ferris19841090Ferris, F.L. IIIFine, S.LHyman, L.1984Age-related macular degeneration and blindness due to neovascular maculopathyArch Ophthalmol1021640-1612National19983447NationalAdvisory EyeCouncil1998Report of the Retinal Diseases PanelU.S. Department of Health and Human ServicesVision Research; A National Plan:1999-2003BethesdaNational Institutes of Health publication 98-412013-40(Ferris, et al., 1984; National, Advisory, Eye & Council, 1998). The neovascular form however may account for almost 90% of cases with severe visual loss and legal blindness ADDIN EN.CITE Ferris19841090Ferris, F.L. IIIFine, S.LHyman, L.1984Age-related macular degeneration and blindness due to neovascular maculopathyArch Ophthalmol1021640-1612Kahn19771200Kahn, H.ALeibowitz, H.MGanley, J.P1977The Framingham eye study, I: outline and major prevalence findingsAm J Epidemiol10617-32Hyman19913760Hyman, L.Lilienfeld, A.MFerris, F.LFine, S.L1991Senile macular degeneration: a case-control studyAm J Epidemiol118213-227(Ferris, et al., 1984; Hyman, Lilienfeld, Ferris & Fine, 1991; Kahn, Leibowitz & Ganley, 1977). Age-related macular degeneration (AMD) accounts for approximately 50% of annual blind and partially sighted registrations in Britain ADDIN EN.CITE Russell20013240Russell, W Harper, RReeves, BWaterman, HHenson, DMcLeod, D2001Randomised controlled trial of an integrated versus an optometric low vision rehabilitation service for patients with age-related macular degeneration: study design and methodologyOphthal Physiol Opt2136-44(Russell, Harper, Reeves, Waterman, Henson & McLeod, 2001). In the Beaver Dam eye study, 57 % of legally blind had late AMD ADDIN EN.CITE Klein19951100Klein, RWang, QKlein, B.EMoss, S.EMeuer, S.M1995The relationship of age-related maculopathy, cataract, and glaucoma to visual acuityInvest Ophthalmol Vis Sci36182-91(Klein, Wang, Klein, Moss & Meuer, 1995). In the Baltimore Eye Study late AMD was the cause of 30% of the white population who were legally blind ADDIN EN.CITE Sommer19911210Sommer, ATielsch, J.MKatz, JQuigley, H. A.Gottsch, J.DJavitt, J.CMartone, J.FRoyall, R.MWitt, KEzrine, S1991Racial differences in the cause-specific prevalence of blindness is east BaltimoreN Engl J Med3251412-1417(Sommer, et al., 1991). In the Framingham Eye Study exudative AMD was present in 79% of legally blind eyes ADDIN EN.CITE Kahn19771200Kahn, H.ALeibowitz, H.MGanley, J.P1977The Framingham eye study, I: outline and major prevalence findingsAm J Epidemiol10617-32(Kahn, et al., 1977). For some forms of choroidal neovascularization, secondary to AMD, laser treatment is more successful than observation in postponing or limiting severe vision loss but treated eyes also lose vision after laser treatment ADDIN EN.CITE Bressler19953380Bressler, N. M.Bressler, S. B.1995Preventive Ophthalmology. Age-related macular degenerationOphthalmology1021206-1211(Bressler & Bressler, 1995).
Currently, there is no treatment that restores visual function in the presence of the severe late dry form of AMD geographic atrophy ADDIN EN.CITE Sunness19971010Sunness, J. S.Rubin, G. S.Applegate, C. A.Bressler, N. M.Marsh, M. J.Hawkins, B. S.Haselwood, D1997Visual function abnormalities and prognosis in eyes with age-related geographic atrophy of the macula and good visual acuityOphthalmology1041677-1691(Sunness, Rubin, Applegate, Bressler, Marsh, Hawkins & Haselwood, 1997). Geographic atrophy (GA) is present in 3.5% of people 75 years of age or older and it is responsible for 20% of the cases of legal blindness due to AMD, and for a much larger percentage of cases of moderate visual loss ADDIN EN.CITE Sunness1999977Sunness, J.S1999The use of the scanning laser ophthalmoscope in the geographic atrophy form of advanced age-related macular degenerationVision '99: International Conference on Low Vision. Vision Rehabilitation for the 21st centuryNew YorkLighthouse International99(Sunness, 1999). Areas of GA are associated with absolute scotoms ADDIN EN.CITE Sunness19971010Sunness, J. S.Rubin, G. S.Applegate, C. A.Bressler, N. M.Marsh, M. J.Hawkins, B. S.Haselwood, D1997Visual function abnormalities and prognosis in eyes with age-related geographic atrophy of the macula and good visual acuityOphthalmology1041677-1691(Sunness, et al., 1997).
Patients with advanced forms of AMD tend to have bilateral involvement ADDIN EN.CITE Macular19933410MacularPhotocoagulationStudyGroup1993Five-year follow-up of fellow eyes of patients with age-related macular degeneration and unilateral extrafoveal choroidal neovascularizationArch Ophthalmol1111189-1199(Macular, Photocoagulation, Study & Group, 1993) and central visual field loss (CFL) ADDIN EN.CITE Bressler19953380Bressler, N. M.Bressler, S. B.1995Preventive Ophthalmology. Age-related macular degenerationOphthalmology1021206-1211Ferris19841090Ferris, F.L. IIIFine, S.LHyman, L.1984Age-related macular degeneration and blindness due to neovascular maculopathyArch Ophthalmol1021640-1612Macular19933410MacularPhotocoagulationStudyGroup1993Five-year follow-up of fellow eyes of patients with age-related macular degeneration and unilateral extrafoveal choroidal neovascularizationArch Ophthalmol1111189-1199Sunness19971010Sunness, J. S.Rubin, G. S.Applegate, C. A.Bressler, N. M.Marsh, M. J.Hawkins, B. S.Haselwood, D1997Visual function abnormalities and prognosis in eyes with age-related geographic atrophy of the macula and good visual acuityOphthalmology1041677-1691(Bressler & Bressler, 1995; Ferris, et al., 1984; Macular, et al., 1993; Sunness, et al., 1997). Research has demonstrated that CFL results in devastating reading disability in the form of very slow reading rates (Elliott, Bhavesh, & Whitaker 2001; Legge et al. 1985b; Rubin & Turano, 1994; ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatLegge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.Sunness19961000Sunness, J. S.Applegate, C. A.Haselwood, DRubin, G. S.1996Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt DiseaseOphthalmology1031458-1466(Fine & Peli, 1998; Legge, Ross, Luebker & LaMay, 1989; Sunness, Applegate, Haselwood & Rubin, 1996). Several factors may contribute to the slow reading rates of low vision patients. The visual span is reduced in low vision; fewer characters are recognized in a single glance, and it takes longer to recognize words that exceed the visual span because more than one fixation is required ADDIN EN.CITE Legge19971950http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009274784Legge, G. E.Ahn, S. J.Klitz, T. S.Luebker, A.Psychophysics of reading--XVI. The visual span in normal and low visionAdolescenceAdultContrast SensitivityEye MovementsFemaleFixation, OcularHumanMalePsychophysics*ReadingSaccadesSupport, U.S. Gov't, P.H.S.Time FactorsVision, Low/*physiopathologyDepartment of Psychology, University of Minnesota, Minneapolis 55455, USA. legge@eye.psych.umn.edu0009274784Vision Res199737141999-2010Legge20013820Legge, G. E.Mansfield, J. S.Chung, S. T.2001Psychophysics of reading. XX. Linking letter recognition to reading speed in central and peripheral visionVision Res41725-743(Legge, Ahn, Klitz & Luebker, 1997; Legge, Mansfield & Chung, 2001). For subjects with normal vision reading around artificial scotomas, reading rate decreases as the number of letters occluded by a scotoma increases ADDIN EN.CITE Fine19991520http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0010341961Fine, E. M.Rubin, G. S.Reading with central field loss: number of letters masked is more important than the size of the mask in degreesAdult*Eye MovementsHumanMasksPsychological Tests*ReadingScotoma/physiopathology/*psychologySupport, U.S. Gov't, P.H.S.*Visual FieldsLions Vision Center, Johns Hopkins University, Baltimore, MD 21205, USA. fine@lions.med.jhu.edu0010341961Vision Res1999394747-56(Fine & Rubin, 1999).
Low vision patients with CFL have eye movement characteristics that are different from readers with normal vision ADDIN EN.CITE Bullimore1995710Bullimore, M.ABailey, I.L1995Reading and eye movements in age-related maculopathyOptom Vis Sci72125-138McMahon1991720McMahon, T.THansen, MViana, M1991Fixation characteristics in macular disease: Relationship between saccadic frequency, sequencing and reading rateInvest Ophthalmol Vis Sci32567-574Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat(Bullimore & Bailey, 1995; McMahon, Hansen & Viana, 1991; Rubin & Turano, 1994). Readers with AMD and CFL take longer to initiate saccades to simple targets ADDIN EN.CITE White1990760White, J.MBedell, H.E1990The oculomotor reference in humans with bilateral macular diseaseInvest Ophthalmol Vis Sci311149-1161Whittaker1991750Whittaker, S.GCummings, R.WSwieson, L.R1991Saccade control without a foveaVision Res312209-2218(White & Bedell, 1990; Whittaker, Cummings & Swieson, 1991), and traverse fewer letters per forward saccade when reading than subjects with normal vision ADDIN EN.CITE Bullimore1995710Bullimore, M.ABailey, I.L1995Reading and eye movements in age-related maculopathyOptom Vis Sci72125-138McMahon1991720McMahon, T.THansen, MViana, M1991Fixation characteristics in macular disease: Relationship between saccadic frequency, sequencing and reading rateInvest Ophthalmol Vis Sci32567-574(Bullimore & Bailey, 1995; McMahon, et al., 1991). Frequent small saccades may slow reading rates because small saccades have longer latencies and result in longer fixations between saccades ADDIN EN.CITE Kowler1987790Kowler, EAnton, S1987Reading twisted text: Implications for the role of saccadesVis Research2745-60(Kowler & Anton, 1987).
The main objective of most elderly patients who seek care from a low-vision clinic is to gain improvement in personal reading ADDIN EN.CITE Elliott19971390Elliott, D.BTrukolo-Ilic, MStrong, J.GPace, RPlotkin, ABevers, P1997Demographic characteristics of the vision-disabled elderlyInvest Ophthalmol Vis Sci382566-2575repeatFaye19702981Faye, E.E1970The Low Vision Patient: Clinical Experience with Adults and ChildrenNew YorkGrune and StrattonHall19872990Hall, A Sacks, S.Z Dornbusch, H Raasch, T1987A preliminary study to evaluate services in a low vision clinicJ Vis Rehab17-25Krieger19672970Krieger, A.A1967The partially sighted patient: A study of 917 casesTrans Am Ophthalmol Soc65544-590(Elliott, et al., 1997; Faye, 1970; Hall, Sacks, Dornbusch & Raasch, 1987; Krieger, 1967). Reading is critical to full participation in modern society and includes reading for survival (medicine labels, bills, and bank statements), and reading for vocation or pleasure (books, magazines, and newspapers). Loss of the ability to read, due to low vision, affects the ability to function independently for simple everyday tasks such as shopping ADDIN EN.CITE Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266(Legge, et al., 1985b). Even mild visual impairment (visual acuity worse than 20/40) is associated with difficulty in tasks necessary for independent living, including: house-work, paying bills, and shopping ADDIN EN.CITE West19971730http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009008632West, S. K.Munoz, B.Rubin, G. S.Schein, O. D.Bandeen-Roche, K.Zeger, S.German, S.Fried, L. P.Function and visual impairment in a population-based study of older adults. The SEE project. Salisbury Eye Evaluation*Activities of Daily LivingAgedAged, 80 and overAging/*physiologyCaucasoid RaceFemaleHealth StatusHumanMaleMaryland/epidemiologyNegroid RaceQuestionnairesRandom AllocationRural PopulationSelf DisclosureSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision Disorders/*epidemiology/physiopathologyVisual Acuity/*physiologyDana Center for Preventive Ophthalmology, Wilmer Institute, Baltimore, Maryland 21287-9019, USA.0009008632Invest Ophthalmol Vis Sci199738172-82(West, Munoz, Rubin, Schein, Bandeen-Roche, Zeger, German & Fried, 1997).
The goal of low vision rehabilitation is to restore function ADDIN EN.CITE Massof19953160Massof, R.W Dagnelie, G Deremeik, J.L Suleiman, S.A Glasner, N.M1995Low vision rehabilitation in the US health care system.Journal of Vision Rehabilitation93-25(Massof, et al., 1995). The successful outcome of rehabilitation for low vision patients often depends on the success or failure of optical devices prescribed for reading ADDIN EN.CITE Lovie-Kitchin19992310Lovie-Kitchin, J.E Whittaker, S.G1999Prescribing near magnification for low vision patientsClin Exp Optom82214-224(Lovie-Kitchin & Whittaker, 1999). Reading performance is strongly associated with vision-related quality of life ADDIN EN.CITE Hazel20001480Hazel, C.A Petre, K.L Armstrong, R.A Benson, M.TFrost, N.A2000Visual function and subjective quality of life compared in subjects with acquired macular diseaseInvest Ophthalmol Vis Sci411309-1315(Hazel, Petre, Armstrong, Benson & Frost, 2000). Individual patients perceive the benefits they attain from visiting a low vision clinic to be as high as 89.5% ADDIN EN.CITE Leat19941470Leat, S.J Fryer, A Rumney, N.J1994Outcome of low vision aid provision: The effectiveness of a low vision clinicOptom Vision Sci71199-206(Leat, Fryer & Rumney, 1994). In spite of the perceived benefit the ability to read normal-sized print with optical devices may be as high as 75% in the clinic but drop to only 35% for the patients at home ADDIN EN.CITE Leat19941470Leat, S.J Fryer, A Rumney, N.J1994Outcome of low vision aid provision: The effectiveness of a low vision clinicOptom Vision Sci71199-206(Leat, et al., 1994). The discrepancy between in-office performance with optical devices and reported home use may include differences in lighting, print quality and contrast, difficulty remembering instructions, and deteriorating vision ADDIN EN.CITE Leat19941470Leat, S.J Fryer, A Rumney, N.J1994Outcome of low vision aid provision: The effectiveness of a low vision clinicOptom Vision Sci71199-206(Leat, et al., 1994). To assure fluent reading after the patient leaves the clinic much higher magnification should be prescribed than the threshold size that just allows the patient to read standard sized print ADDIN EN.CITE Lovie-Kitchin19992310Lovie-Kitchin, J.E Whittaker, S.G1999Prescribing near magnification for low vision patientsClin Exp Optom82214-224(Lovie-Kitchin & Whittaker, 1999). Optical devices that provide high magnification have many limitations, including reading distances as close as one inch, a field of view that may be smaller than a single word, the need for considerable manual dexterity, and hours of practice for proficient use ADDIN EN.CITE Goodrich1977540Goodrich, G.L Mehr, E.BQuillman, R.D Shaw, H.KWiley, J.K1977Training and practice effects in performance with low-vision aids: A preliminary studyAm J Optom Physiol Optics54312-318Repeat(Goodrich, Mehr, Quillman, Shaw & Wiley, 1977). Each of these factors limits reading performance and causes frustration for patients.
For readers with severe vision loss, a closed-circuit television (CCTV) is often the visual aid of choice ADDIN EN.CITE Arditi1999620Arditi, A1999Elicited sequential presentation for low vision readingVision Res394412-4418(Arditi, 1999). A CCTV consists of a video monitor and a vertically mounted camera that is aimed at the reading material on a moveable platform. A CCTV can provide high magnification, high contrast, and a larger field of view than standard low vision devices. Patients prefer CCTVs for extended reading, and are reported to use them for longer durations compared to other low vision devices ADDIN EN.CITE Watson19974420Watson, G. R.De l'Aune, W.Long, S.Maino, J.Stelmack, J.A.1997Veterans' use of low vision devices for readingOptom Vis Sci745260-5.9219283Health SurveysHuman*ReadingReproducibility of Results*Sensory Aids/utilizationSupport, U.S. Gov't, Non-P.H.S.Treatment OutcomeUnited States*VeteransVision, Low/physiopathology/*therapyVisual Acuity/physiologyhttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=9219283Rehabilitation Research and Development Center, Atlanta VA Medical Center, Georgia, USA.Goodrich20013850Goodrich, G.LKirby, J2001A comparison of patient reading performance and preference: optical devices, handheld CCTV (Innoventions Magni-Cam), or stand-mounted CCTV (Optelec Clearview or TSI Genie)J Am Optom Assoc72519-528(Goodrich & Kirby, 2001; Watson, De l'Aune, Long, Maino & Stelmack, 1997). However, the navigation demands of using a CCTV significantly lengthens the time required for reading compared to computerized presentation methods ADDIN EN.CITE Beckmann1996580Beckmann, P.JLegge, G.E1996Psychophysics of reading XIV. The page navigation problem in using magnifiersVision Res363723-3733Harland19981940Harland, S.Legge, G. E.Luebker, A.1998Psychophysics of reading. XVII. Low-vision performance with four types of electronically magnified textOptom Vis Sci753183-900009547799*ReadingVision, Low/*physiopathology/rehabilitationRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009547799Minnesota Laboratory for Low-Vision Research, University of Minnesota, Minneapolis, USA.(Beckmann & Legge, 1996; Harland, Legge & Luebker, 1998).
Computer generated displays can alter text format, providing much greater magnification, higher contrast, and a less restricted field of view ADDIN EN.CITE Fine19951580http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0007608788Fine, E. M.Peli, E.Enhancement of text for the visually impairedAdultAgedAged, 80 and overContrast Sensitivity/physiologyHumanLightMiddle Age*ReadingSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/*physiopathologyVisual AcuityVisual FieldsVisual Perception/*physiologySchepens Eye Research Institute, Harvard Medical School, Boston, Massachusetts 02114, USA.0007608788J Opt Soc Am A19951271439-47Fine19961550http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008888864Fine, E. M.Peli, E.Visually impaired observers require a larger window than normally sighted observers to read from a scroll displayAgedComputer TerminalsHumanMiddle Age*ReadingSensory Aids/*standardsSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision, Low/physiopathology/*rehabilitationVisual AcuityWilmer Eye Institute, Johns Hopkins University, Baltimore, MD 21205, USA.0008888864J Am Optom Assoc1996677390-6(Fine & Peli, 1995; Fine & Peli, 1996b). When high magnification is needed a simple computer and television can provide access to print that cannot be read by other means.
Electronic magnification and presentation technology has been used to investigate reading in low vision ADDIN EN.CITE Fine19951580http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0007608788Fine, E. M.Peli, E.Enhancement of text for the visually impairedAdultAgedAged, 80 and overContrast Sensitivity/physiologyHumanLightMiddle Age*ReadingSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/*physiopathologyVisual AcuityVisual FieldsVisual Perception/*physiologySchepens Eye Research Institute, Harvard Medical School, Boston, Massachusetts 02114, USA.0007608788J Opt Soc Am A19951271439-47Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat(Fine & Peli, 1995; Legge, et al., 1985b; Rubin & Turano, 1994). There is a trend for reading speed to increase as the number of letters in the field of view increases ADDIN EN.CITE Fine19961550http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008888864Fine, E. M.Peli, E.Visually impaired observers require a larger window than normally sighted observers to read from a scroll displayAgedComputer TerminalsHumanMiddle Age*ReadingSensory Aids/*standardsSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision, Low/physiopathology/*rehabilitationVisual AcuityWilmer Eye Institute, Johns Hopkins University, Baltimore, MD 21205, USA.0008888864J Am Optom Assoc1996677390-6Harland19981940Harland, S.Legge, G. E.Luebker, A.1998Psychophysics of reading. XVII. Low-vision performance with four types of electronically magnified textOptom Vis Sci753183-900009547799*ReadingVision, Low/*physiopathology/rehabilitationRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009547799Minnesota Laboratory for Low-Vision Research, University of Minnesota, Minneapolis, USA.Lovie-Kitchin1988600Lovie-Kitchin, J.EWoo, G.C1988Effect of magnification and field of view on reading speed using a CCTVOphthalmic Physiological Optics8139-145(Fine & Peli, 1996b; Harland, et al., 1998; Lovie-Kitchin & Woo, 1988). Dynamic displays, such as text that scrolls from right to left (SCROLL) ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.(Legge, et al., 1989), and RSVP (Rapid Serial Visual Presentation) ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat(Rubin & Turano, 1992; Rubin & Turano, 1994), which presents one word at a time, result in faster reading rates for most low vision subjects compared to conventional page formats. RSVP is often the fastest method if the text size is very large, though some low vision patients have been found to read SCROLL faster than RSVP regardless of text size ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat(Fine & Peli, 1998).
While reading speed is often the dependent variable in reading research, and reading speed is important for general visual quality of life ADDIN EN.CITE Hazel20001480Hazel, C.A Petre, K.L Armstrong, R.A Benson, M.TFrost, N.A2000Visual function and subjective quality of life compared in subjects with acquired macular diseaseInvest Ophthalmol Vis Sci411309-1315(Hazel, et al., 2000), accuracy may be more important to low vision patients ADDIN EN.CITE Leat19941470Leat, S.J Fryer, A Rumney, N.J1994Outcome of low vision aid provision: The effectiveness of a low vision clinicOptom Vision Sci71199-206(Leat, et al., 1994). The ability to perform a variety of daily activities such as reading bank statements, personal letters, instructions, and labels is associated with perceived benefit from low vision rehabilitation ADDIN EN.CITE Leat19941470Leat, S.J Fryer, A Rumney, N.J1994Outcome of low vision aid provision: The effectiveness of a low vision clinicOptom Vision Sci71199-206(Leat, et al., 1994). Computer systems should be the recommended reading aid when standard optical devices do not provide adequate magnification and or a wide enough field of view to allow accurate reading at maximum rate. Computer systems, via the World Wide Web, can increase access to text for low vision patients for both extended reading (newspapers or books) and spot reading (bank statements or instructions) for personal independence and safety ADDIN EN.CITE Leat19991490Leat, S.JLegge, G.EBullimore, M.A1999What is low vision? A Re-evaluation of definitionsOptom Vis Sci76198-211Repeat(Leat, et al., 1999).
The benefits of using computer technology to alter text displays are not always clear-cut for low vision patients with CFL from AMD ADDIN EN.CITE Fine19961550http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008888864Fine, E. M.Peli, E.Visually impaired observers require a larger window than normally sighted observers to read from a scroll displayAgedComputer TerminalsHumanMiddle Age*ReadingSensory Aids/*standardsSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision, Low/physiopathology/*rehabilitationVisual AcuityWilmer Eye Institute, Johns Hopkins University, Baltimore, MD 21205, USA.0008888864J Am Optom Assoc1996677390-6Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatHarland19981940Harland, S.Legge, G. E.Luebker, A.1998Psychophysics of reading. XVII. Low-vision performance with four types of electronically magnified textOptom Vis Sci753183-900009547799*ReadingVision, Low/*physiopathology/rehabilitationRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009547799Minnesota Laboratory for Low-Vision Research, University of Minnesota, Minneapolis, USA.Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat(Fine & Peli, 1996b; Fine & Peli, 1998; Harland, et al., 1998; Legge, et al., 1989; Rubin & Turano, 1994). While clear guidelines to maximize reading for patients with CFL have been difficult to make, advances in video technology are likely to benefit this group ADDIN EN.CITE Arditi1999620Arditi, A1999Elicited sequential presentation for low vision readingVision Res394412-4418Fine19961540http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008869984Fine, E. M.Peli, E.The role of context in reading with central field lossAgedComparative StudyHumanMiddle Age*ReadingSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision, Low/*physiopathologyVisual Acuity*Visual FieldsWilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland, USA.0008869984Optom Vis Sci1996738533-9(Arditi, 1999; Fine & Peli, 1996a).
References
QUOTE EN.REFLIST Arditi A.(1999).Elicited sequential presentation for low vision reading.Vision Res.39:4412-4418.
Beckmann PJ, Legge GE.(1996).Psychophysics of reading XIV. The page navigation problem in using magnifiers.Vision Res.36:3723-3733.
Berger JW, Fine SL, Maguire MG.(1999)Age-related Macular Degeneration.Age-related Macular Degeneration.St. Louis:Mosby.Pages.
Brabyn JA, Haegerstrom-Portnoy G, Schneck ME, Lott LA.(2000).Visual impairments in elderly people under everyday viewing conditions.J Vis Impair Blind.94:741-755.
Bressler NM, Bressler SB.(1995).Preventive Ophthalmology. Age-related macular degeneration.Ophthalmology.102:1206-1211.
Bullimore MA, Bailey IL.(1995).Reading and eye movements in age-related maculopathy.Optom Vis Sci.72:125-138.
Bureau, of, the, Census, U.S., Department, of, Commerce.(1996)Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050.Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050.Washington, D.C.:
Charness N, Dijkstra K.(1999).Age, luminance and print legibility in homes, offices and public places.Human Factors.41:173-193.
Dickinson C.(1998)Low Vision. Principles and Practice.Low Vision. Principles and Practice.Oxford:Butterworth Heinemann.Pages.
Elliott DB, Trukolo-Ilic M, Strong JG, Pace R, Plotkin A, Bevers P.(1997).Demographic characteristics of the vision-disabled elderly.Invest Ophthalmol Vis Sci.38:2566-2575.
Faye EE.(1970)The Low Vision Patient: Clinical Experience with Adults and Children.The Low Vision Patient: Clinical Experience with Adults and Children.New York:Grune and Stratton.Pages.
Ferris FLI, Fine SL, Hyman L.(1984).Age-related macular degeneration and blindness due to neovascular maculopathy.Arch Ophthalmol.102:1640-1612.
Fine EM, Peli E.(1995).Enhancement of text for the visually impaired.J Opt Soc Am A.12:1439-1447.
Fine EM, Peli E.(1996a).The role of context in reading with central field loss.Optom Vis Sci.73:533-539.
Fine EM, Peli E.(1996b).Visually impaired observers require a larger window than normally sighted observers to read from a scroll display.J Am Optom Assoc.67:390-396.
Fine EM, Peli E.(1998).Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter size.Optom Vision Sci.75:191-196.
Fine EM, Rubin GS.(1999).Reading with central field loss: number of letters masked is more important than the size of the mask in degrees.Vision Res.39:747-756.
Friedman SM, Munoz B, Rubin GS, West SK, Bandeen-Roche K, Fried LP.(1999).Characteristics of discrepancies between self-reported visual function and measured reading speed. Salisbury Eye Evaluation Project Team.Invest Ophthalmol Vis Sci.40:858-864.
Goodrich GL, Kirby J.(2001).A comparison of patient reading performance and preference: optical devices, handheld CCTV (Innoventions Magni-Cam), or stand-mounted CCTV (Optelec Clearview or TSI Genie).J Am Optom Assoc.72:519-528.
Goodrich GL, Mehr EB, Quillman RD, Shaw HK, Wiley JK.(1977).Training and practice effects in performance with low-vision aids: A preliminary study.Am J Optom Physiol Optics.54:312-318.
Haegerstrom-Portnoy G, Schneck ME, Brabyn JA.(1999).Seeing into Old Age: Vision Function Beyond Acuity.Optom Vision Sci.76:141-158.
Hall A, Sacks SZ, Dornbusch H, Raasch T.(1987).A preliminary study to evaluate services in a low vision clinic.J Vis Rehab.1:7-25.
Harland S, Legge GE, Luebker A.(1998).Psychophysics of reading. XVII. Low-vision performance with four types of electronically magnified text.Optom Vis Sci.75:183-190.
Hazel CA, Petre KL, Armstrong RA, Benson MT, Frost NA.(2000).Visual function and subjective quality of life compared in subjects with acquired macular disease.Invest Ophthalmol Vis Sci.41:1309-1315.
Hyman L, Lilienfeld AM, Ferris FL, Fine SL.(1991).Senile macular degeneration: a case-control study.Am J Epidemiol.118:213-227.
Kahn HA, Leibowitz HM, Ganley JP.(1977).The Framingham eye study, I: outline and major prevalence findings.Am J Epidemiol.106:17-32.
Kini MM, Leibowitz HM, Colton T.(1978).Prevalence of senile cataract, diabetic retinopathy, senile macular degeneration, and open-angle glaucoma in the Framingham Eye Study.Am J Ophthalmol.85:
Klein R, Klein BEK, Jensen SC, Mares-Perlman JA, Cruickshanks KJ, Palta M.(1999).Age-related maculopathy in a multiracial United States population. The National Health and Nutrition Examination Survey III.Ophthalmology.106:1056-1065.
Klein R, Klein BEK, Linton KLP.(1992).Prevalence of age-related maculopathy: the Beaver Dam Eye Study.Ophthalmology.99:933-943.
Klein R, Wang Q, Klein BE, Moss SE, Meuer SM.(1995).The relationship of age-related maculopathy, cataract, and glaucoma to visual acuity.Invest Ophthalmol Vis Sci.36:182-191.
Kowler E, Anton S.(1987).Reading twisted text: Implications for the role of saccades.Vis Research.27:45-60.
Krieger AA.(1967).The partially sighted patient: A study of 917 cases.Trans Am Ophthalmol Soc.65:544-590.
Leat SJ, Fryer A, Rumney NJ.(1994).Outcome of low vision aid provision: The effectiveness of a low vision clinic.Optom Vision Sci.71:199-206.
Leat SJ, Legge GE, Bullimore MA.(1999).What is low vision? A Re-evaluation of definitions.Optom Vis Sci.76:198-211.
Legge GE, Ahn SJ, Klitz TS, Luebker A.(1997).Psychophysics of reading--XVI. The visual span in normal and low vision.Vision Res.37:1999-2010.
Legge GE, Mansfield JS, Chung ST.(2001).Psychophysics of reading. XX. Linking letter recognition to reading speed in central and peripheral vision.Vision Res.41:725-743.
Legge GE, Pelli DG, Rubin GS, Schleske MM.(1985a).Psychophysics of reading I. Normal Vision.Vision Res.25:239-252.
Legge GE, Ross JA, Luebker A, LaMay JM.(1989).Psychophysics of reading. VIII. The Minnesota Low-Vision Reading Test.Optom Vis Sci.66:843-853.
Legge GE, Rubin GS, Pelli DG, Schleske MM.(1985b).Psychophysics of reading II. Low Vision.Vision Res.25:253-266.
Leibowitz HM, Krueger DE, Maunder LR, Milton RC, Kini MM, Kahn HA, Nickerson RJ, Pool J, Colton TL, Ganley JP, Loewenstein JI, Dawber TR.(1980).The Framingham Eye Study monograph.Surv Ophthalmol.24(suppl):335-610.
Lighthouse, National, Survey, on, Vision, Loss.(1995)The experience, attitude, and knowledge of middle-aged and older Americans.The experience, attitude, and knowledge of middle-aged and older Americans.New York:Lighthouse.Pages.
Lovie-Kitchin JE, Whittaker SG.(1999).Prescribing near magnification for low vision patients.Clin Exp Optom.82:214-224.
Lovie-Kitchin JE, Woo GC.(1988).Effect of magnification and field of view on reading speed using a CCTV.Ophthalmic Physiological Optics.8:139-145.
Macular, Photocoagulation, Study, Group.(1993).Five-year follow-up of fellow eyes of patients with age-related macular degeneration and unilateral extrafoveal choroidal neovascularization.Arch Ophthalmol.111:1189-1199.
Massof RW, Dagnelie G, Deremeik JL, Suleiman SA, Glasner NM.(1995).Low vision rehabilitation in the US health care system.Journal of Vision Rehabilitation.9:3-25.
Massof RW, Lidoff L.(2001)Issues in Low Vision Rehabilitation. Service Delivery, Policy, and Funding.Issues in Low Vision Rehabilitation. Service Delivery, Policy, and Funding.New York:AFB Press.Pages.
McMahon TT, Hansen M, Viana M.(1991).Fixation characteristics in macular disease: Relationship between saccadic frequency, sequencing and reading rate.Invest Ophthalmol Vis Sci.32:567-574.
National, Advisory, Eye, Council.(1998).Report of the Retinal Diseases Panel.In: Vision Research; A National Plan:1999-2003.Services USDoHaH editorBethesda:National Institutes of Health publication 98-4120.13-40.
Rubin GS, Roche KB, Prasada-Rao P, Fried LP.(1994).Visual impairment and disability in older adults.Optom Vis Sci.71:750-760.
Rubin GS, Turano K.(1992).Reading without saccadic eye movements.Vision Res.32:895-902.
Rubin GS, Turano K.(1994).Low vision reading with sequential word presentation.Vision Res.34:1723-1733.
Russell W, Harper R, Reeves B, Waterman H, Henson D, McLeod D.(2001).Randomised controlled trial of an integrated versus an optometric low vision rehabilitation service for patients with age-related macular degeneration: study design and methodology.Ophthal Physiol Opt.21:36-44.
Seddon JM, Ajani UA, Sperduto RD, Hiller R, Blair N, T.C B, Farber M, Gragoudas E, Haller J, Miller DT, Yannuzzi LA, Willett W, Eye, Disease, Case-Control, Study, Group.(1994).Dietary Carotenoids, Vitamins A, C, and E, and Advanced Age-Related Macular Degeneration.JAMA.272:1413-1420.
Sommer A, Tielsch JM, Katz J, Quigley HA, Gottsch JD, Javitt JC, Martone JF, Royall RM, Witt K, Ezrine S.(1991).Racial differences in the cause-specific prevalence of blindness is east Baltimore.N Engl J Med.325:1412-1417.
Sunness JS.(1999).The use of the scanning laser ophthalmoscope in the geographic atrophy form of advanced age-related macular degeneration.In: Vision '99: International Conference on Low Vision. Vision Rehabilitation for the 21st century.New York:Lighthouse International.99.
Sunness JS, Applegate CA, Haselwood D, Rubin GS.(1996).Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt Disease.Ophthalmology.103:1458-1466.
Sunness JS, Rubin GS, Applegate CA, Bressler NM, Marsh MJ, Hawkins BS, Haselwood D.(1997).Visual function abnormalities and prognosis in eyes with age-related geographic atrophy of the macula and good visual acuity.Ophthalmology.104:1677-1691.
Tielsch JM.(2001).Prevalence of Visual Impairment and Blindness in the United States.In: Issues in Low Vision Rehabilitation. Service Delivery, Policy, and Funding.Massof RW,Lidoff L editors.New York:AFB Press.13-26.
Tielsch JM, Sommer A, Witt K, Katz J, Royall RM.(1990).Blindness and Visual Impairment in an American Urban Population.Arch Ophthalmol.108:286-290.
VanNewkirk MR, Nanjan MB, Wang JJ, Mitchell P, Taylor HR, McCarty C.(2000).The prevalence of age-related maculopathy. The visual impairment project.Ophthalmology.1593-1600:
Watson GR, De l'Aune W, Long S, Maino J, Stelmack JA.(1997).Veterans' use of low vision devices for reading.Optom Vis Sci.74:260-265.
West SK, Munoz B, Rubin GS, Schein OD, Bandeen-Roche K, Zeger S, German S, Fried LP.(1997).Function and visual impairment in a population-based study of older adults. The SEE project. Salisbury Eye Evaluation.Invest Ophthalmol Vis Sci.38:72-82.
White JM, Bedell HE.(1990).The oculomotor reference in humans with bilateral macular disease.Invest Ophthalmol Vis Sci.31:1149-1161.
Whittaker SG, Cummings RW, Swieson LR.(1991).Saccade control without a fovea.Vision Res.31:2209-2218.
WHO-Study-Group.(1973)The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518.The Prevention of Blindness. World Health Organization Technical Report Series, Report No. 518.Geneva:World Health Organization.Pages.
Title: Within-session practice of computer-generated reading methods for low vision due to age-related macular degeneration
Authors: Kathy Aquilante, Dean Yager, Robert A Morris
Corresponding author: Kathy Aquilante, O.D., Ph.D., Assistant Professor, SUNY State College of Optometry, 33 West 42nd St., NY NY 10013
212-780-5152
212-780-5137 fax
aquilante@sunyopt.edu
Dean Yager, Ph.D., Distinguished Professor, SUNY College of Optometry
Robert A. Morris, Ph.D., Professor of Mathematics and Computer Science, University of Massachusetts, Boston MA
Summary. The purpose of this study was to investigate whether reading performance, measured in words per minute, improved during an hour of within-session practice. The reading methods were three computer-generated presentations including (1) MNREAD, a modified page format, (2) RSVP, which presents one word at a time, and (3) SCROLL, where text pans from right to left across a screen. Forty-five young readers with normal vision, forty-five elder readers with normal vision, and forty-five readers with low vision due to age-related macular degeneration read by one of these methods. None of the participants had previous experience reading with MNREAD, RSVP of SCROLL. There was little evidence that within-session practice improved performance. Only 10 of 135 participants had modest reading rate gains, and there was no statistical difference between reading method or subject group for this small subset of readers.
Keywords: Low vision, age-related macular degeneration (AMD), practice, reading, MNREAD, RSVP, SCROLL
Introduction
Low vision patients with central scotomas from age-related macular degeneration develop a preferred retinal locus(PRL) that falls at a peripheral retinal location compared to the fovea.(Timberlake, Mainster, Peli, Augliere, Essock & Arend, 1986) Reading with peripheral retina is slow compared to foveal reading.(Legge, Rubin, Pelli & Schleske, 1985b; Rubin, 1987; Legge, Ross, Isenberg & LaMay, 1992; Rubin & Turano, 1994; Sunness, Applegate, Haselwood & Rubin, 1996; Chung, Mansfield & Legge, 1998; Elliott, Bhavesh & Whitaker, 2001) Training may improve eccentric fixation and use of PRLs for reading in patients who have macular scotomas.(Holcomb & Goodrich, 1976; Goodrich, Mehr, Quillman, Shaw & Wiley, 1977; Backman & Inde, 1979; Goodrich & Mehr, 1986; Peli, 1986; Freeman, Jose & Goodrich, 1995; O'Connell, RG Cole, BP Rosenthal, 1996) However, there are no clinical trials to demonstrate that training works better than simple practice.(Rubin, 2001)
Patients with low vision due to macular degeneration need magnification in order to read. Some patients can use prescribed magnification after brief instructions(Nilsson & Nilsson, 1986) though training may further improve performance with low vision devices.(Goodrich et al., 1977; Nilsson & Nilsson, 1986; Watson, De l'Aune, Stelmack, Maino & Long, 1997) In a retrospective study of patients with macular degeneration, rehabilitation with optical aids and training increased the percentage of participants who could read newsprint. Each training session lasted one hour with the number of training sessions varying between one to twenty-three however, the authors did not delineate the change in newsprint reading ability due to improved visual acuity through magnification versus the change due to training.(Nilsson & Nilsson, 1986) It is likely that some portion of the need for extensive training in the use of low vision devices (Goodrich et al., 1977; Nilsson & Nilsson, 1986; Goodrich, Kirby, Keswick, Oros, Wagstaff, Donald, Hazan, Peters, 2000) depends on acquiring the necessary motor skills.(Legge, Pelli, Rubin & Schleske, 1985a; Legge et al., 1985b; Karni & Sagi, 1993; Karni, Meyer, Jezzard, Adams, Turner & Ungerleider, 1995)
Gordon Legge and colleagues have removed the confounds of using low vision devices in the study of low vision reading by developing a method of automatically presenting sentences in a moving text display (SCROLL).(Legge et al., 1985a; Legge et al., 1985b). Developments in computer technology and text presentation eliminate several of the traditional reasons for training low vision patients since there is no optical device to focus at a specific distance, and page navigation can be controlled by software.(Fine & Peli, 1995; Rubin, 2001) The use of computer generated reading methods would be advantageous to patients with low vision if minimal training is required.
Rehabilitation of some low vision patients with macular degeneration can occur within an hour. (Nilsson & Nilsson, 1986) Learning for visual tasks, such as vernier acutiy, occur with less than an hour of training. (Poggio, Fahle & Edelman, 1992; Fahle, Edelman & Poggio, 1995) Within-session learning might occur for novel reading methods such as MNREAD(Legge, Ross, Luebker & LaMay, 1989), RSVP(Rubin & Turano, 1992; Rubin & Turano, 1994) and SCROLL.(Legge et al., 1985a; Legge et al., 1985b) The purpose of this study was to determine whether reading performance measured in words per minute (wpm) would improve after an hour of practice with these reading methods.
METHODS
Research participants
Three groups of paid volunteers participated in the reading sessions. Forty-five patients with low vision due to age-related macular degeneration, forty-five elder subjects with normal vision and forty-five younger subjects with normal vision took part in the study. The elder and AMD groups were not matched for age. Fifteen individuals from each group of participants completed one of the reading methods. Summary data and statistics for age and logMAR visual acuity are presented in Table 1. The young and elder subjects with normal vision were included if their medical eye report indicated no history of eye disease. Participants with age-related macular degeneration were selected if their visual acuity had been stable for a minimum of six months and excluded if they were receiving medical treatment (other than vitamins) for macular degeneration.
Low vision patients with macular degeneration and elders with normal vision were recruited from the State University of New York, University Optometric Center, Low Vision and Primary Care Services. Younger subjects were primarily students from the State University of New York State College of Optometry. None of the participants had previous experience reading text in MNREAD, RSVP, or SCROLL formats.
Binocular best-corrected distance visual acuity was measured for each participant prior to the reading session using modified ETDRS chart.(Ferris, Kassof, Bresnick & Bailey, 1982)(Lighthouse International: professional catalog@lighthouse.org) The three subsets of subjects within each group, for each reading method, were approximately matched for visual acuity. Corrected visual acuity was different between macular degeneration, elder, and younger groups (means 1.11, 0.04 & -0.07; F2,132 = 441.32, p < 0.0001) with the difference between each group confirmed using a Tukey-Kramer HSD test (p = 0.05).
The procedures used in this study conformed to the tenets of the Declaration of Helsinki. Before testing began all subjects read (or had read to them) and then signed an informed consent form. None of the participants were aware of the specific aim of the study. All study protocols were approved by the Institutional Review Board on human research at the State University of New York, State College of Optometry.
Display and text presentation
Text was displayed with a 640 x 480 pixel Eiki LCD projector and rear-projected onto a four-foot by three-foot rear projection screen. The text characters were white on a dark background. Average luminance was 940 cd/m2 for the white letters and 3.3 cd/m2 for the background with a Weber contrast of 0.99. Room lights were off during the session. A Macintosh computer with a frame rate of 67 Hz controlled the projection system.
Sentences were presented in three reading formats. For MNREAD the entire sentence appeared on the screen at once. The MNREAD sentences had 13 character spaces per line, with the words of the sentence stacked in four left-justified lines.(Legge et al., 1989) For RSVP a sentence was displayed one word at a time, horizontally centered on the screen. The image onset and offset of the projection system was approximately 4 msec so that each word was rapidly replaced by the next. In SCROLL the first letter of the first word of each sentence began at the right edge of a display window and moved horizontally from right to left across the screen. For SCROLL the maximum number of characters visible at one time was set to 10 by specifying a window boundary within the software. This assured that all participants who read SCROLL saw the same number of characters regardless of the angular size of the text.
Each sentence had 9 to 13 words and was selected from a library of about 1650 sentences. No subject read a single sentence more than once. Five hundred and forty of the sentences came from the Minnesota Laboratory for Low-Vision Research.(Legge, et al., 1989) Additional sentences came from the Gates-MacGinity Reading Tests(Rubin & Turano, 1992; Rubin & Turano, 1994) and from a variety of books downloaded from Project Gutenberg (www.promo.net/pg/). All sentences were edited so that they were similar in length and complexity to the MNREAD sentences. Sentence order was counterbalanced across subjects and reading methods so that subtle differences in the difficulty of sentences were distributed across conditions.
Text size was defined by lower case "x" height.(NAS-NRC-Committee-on-Vision, 1980) Visual angle was controlled by selecting font size with the Macintosh computer and by varying the distance of the subject from the projection screen. Letter size was approximately five times larger than the best-corrected distance visual acuity that was measured at the start of the session. This size was used in order to assure that reading rates approached the maximum possible for each method (Legge et al., 1985a; Legge et al., 1985b; Rubin & Turano, 1992; Rubin & Turano, 1994; Fine & Peli, 1998) although participants with macular degeneration may need additional magnification to read at their fastest.(Fine & Peli, 1998; Aquilante, Yager & Morris, 2003)
Testing procedure
A verbal description of the reading method was given before the first trial. An audible prompt alerted the subject prior to the onset of the first word in each sentence. There was no practice prior to the first sentence trial, and no feedback was provided during the session. Subjects read aloud, and continued to read after the sentence was off the screen suggesting that the display rate was faster than their speaking rate. Reading rate (words per minute) was calculated and stored by the software for each sentence and was based on the number of words in the sentence divided by the amount of time that each sentence appeared on the screen. This gave a measure of display duration rather than time needed to verbalize the sentence.(Rubin & Turano, 1992) All of the words of a sentence had to be read, and read in order for the sentence to be counted correct. A period marked the end of the sentence in all three reading methods. The experimenter saw the entire sentence in a persistent display on a separate monitor and recorded with a keystroke whether the sentence was read correctly or if mistakes were made. If the experimenters keystroke indicated that the sentence was read perfectly the software reduced the duration of the next sentence by 19%, (0.075 log unit), or increased the sentence duration by the same fraction if a mistake is made. This staircase procedure continued until there were either 4 or 8 reversals in reading rate to complete one block of sentence reading trials. Four reversals were used when the subjects reading rate was 60 words per minute (wpm) or slower, and 8 reversals were used for faster rates. This made it possible to keep the total session time to about an hour for all participants. Each session included 10 blocks of trials with each sentence acting as a single trial. The software terminated each block after the specified number of reading rate reversals. Rest breaks were allowed as needed.
The number of sentences read in a single block of trials varied between blocks and participants, however adjusting the initial presentation rate minimized this difference. Available starting rates were 15, 30, 60, 120, and 300 wpm based on the level of best-corrected visual acuity,(Aquilante, Yager & Morris, 2000) and the starting rate for the second block was adjusted up or down as needed after completing the first block of trials. This allowed each individual to begin reading at a rate close to their limit, and reduced the total number of attempts, either reading sentences correctly or with errors, before there was a reversal in the staircase procedure.
The software created a session data file for each participant. Each data file included a record of the reversal reading rates for each staircase, and the geometric mean of the staircase reversals (in wpm) for each block of trials.
Data analysis
The ten geometric means of reading rate in words per minute (wpm) for each subject were converted to log reading rates and log reading rate ratios. The reading rate ratio was calculated by comparing the reading rate of each block with the reading rate for the first block [log wpm (block N) - log wpm (block 1)]. The reading rate ratio describes the change in reading rate over the session, and has the additional advantage of allowing direct comparisons of change across subjects who read at very different rates. In general the wpm data, and log wpm data across groups had binomial or trinomial distributions for each of the reading methods. Reading rate across groups does not follow a normal distribution for wpm (Shapiro-Wilk W = 0.92, p < 0.0001) or log wpm (W = 0.84, p < 0.0001). The log ratio transformation of MNREAD reading rates approaches a normal distribution (W = 0.98, p = 0.03), and for RSVP and SCROLL the log ratio transformations of the data are compatible with a normal distribution (RSVP: W = 0.99, p = 0.62; SCROLL: W = 0.99, p = 0.77).(Altman, 1991) Log ratio data were entered into SAS JMP, Version 4 software, (SAS Institue, Cary NC 27513) the distribution was tested for each group by method with a Shapiro-Wilk W, and then subjected to one-way repeated measures ANOVA across blocks of trials and across subjects. The planned ad hoc, when indicated, was a Dunnett's test with the first block, or the mean log ratio from the subject data selected as the control. A non-parametric one way analysis of variance, the Kruskal-Wallis test, was also used when a Shapiro-Wilk W test indicated that the data failed to fit a normal distribution.
RESULTS
Figure 1 shows data from all subjects who read with MNREAD giving the difference in reading rates for blocks 2-10 compared to the reading rates for the first block. Visual inspection of the wpm data in Figure 1A demonstrates that the 15 individuals with age-related macular degeneration (subject numbers 1-15) read more slowly than the elder readers (subjects number 16 30), and young readers (subject numbers 31-45). Also by visual inspection of the wpm data in Figure 1A, it can be seen that there is at least a bimodal distribution across subjects for the change in reading rate.(Altman & Bland, 1983) Figure 1B shows the same data converted to log ratios, and suggests that the proportional change in reading rate during the session is similar between all three groups of readers.(Altman & Bland, 1983)
A one-way repeated measures ANOVA on log ratios for MNREAD reveals no significant difference between the ten session blocks for the young readers (F9,140= 1.07, p = 0.39). There were no significant differences between blocks for the elder readers (F9,140= 0.41, p = 0.93; Kruskal-Wallis Chi square = 3.80, df = 9, p = 0.92). Readers with macular degeneration (F9,140= 0.39, p = 0.94) showed no significant differences between log ratios for the ten MNREAD reading blocks within the session. The mean log ratios of the three subject groups who read with MNREAD were not different (F2, 42= 1.84, p = 0.17). Variability of the log ratios, described by the standard deviation, was larger for the AMD group compared to young and elder readers but this difference was not statistically significant (means of SD: 0.07, 0.06, 0.05; F2, 42= 2.35, p = 0.11).
For RSVP, young readers (F9,140= 0.40, p = 0.93), elder readers (F9,140= 0.22, p = 0.99), and readers with macular degeneration (F9,140= 0.29, p = 0.98; Kruskal-Wallis Chi square = 4.23, df = 9, p = 0.90) showed no significant difference in log ratios across blocks for the session. The mean log ratios were not different between groups (F2, 42= 0.22, p = 0.81), and the standard deviation of the log ratios for the three groups who read with RSVP were not different (F2, 42= 0.64, p < 0.53).
For SCROLL, there were no differences in log ratio measures across blocks for young readers (F9,140= 0.44, p = 0.91), elder readers (F9,140= 1.32, p = 0.23), or participants who had macular degeneration (F9,140= 1.23, p = 0.28). Log ratios did not differ between the three groups (F2, 42= 0.16, p = 0.86), and the standard deviation of the log ratios for SCROLL were not different between groups (F2, 42= 1.58, p = 0.22).
A comparison of mean log ratios across methods and groups did not yield a difference between either the reading methods (Kruskal-Wallis: Chi square = 1.54, df = 2, p = 0.46), or the subject groups (Kruskal-Wallis: Chi square = 1.47, df = 2, p = 0.48). The standard deviations of the mean log ratios across the three reading methods were not different (Kruskal-Wallis: Chi square = 1.47, df = 2, p = 0.48). The variability indicated by the standard deviations of the three subject groups, across all three reading methods, were not different (Kruskal-Wallis: Chi square = 1.47, df = 2, p = 0.48). Based on the results by block, across methods, an effect size of 0.26 was calculated using a Kruskal-Wallis Chi square (9.04, df = 9, p = 0.43).(Portney & Watkins, 2000) This effect size combined with the standard deviation of the log ratios (0.074) and the sample size (n = 135), gives a power of 0.89 (alpha = 0.05) adding credence to the finding that reading rates did not change during the practice session.(Portney & Watkins, 2000)
Figure 2 is a plot of mean log ratios by method for each subject by group. Maximum and minimum log ratios and the standard deviation of the log ratio for each subject are included with the mean. The horizontal line at 0.00 marks the log ratio for the first block of reading trials. Inspection of Figure 2 reveals that, within each method and group, the mean performance of some individuals exceeded their performance for the first block of trials. The mean log ratio for other individuals indicates that their average reading rate was slower during the session than it was for the first block of trials. In Figure 2, data sets with the minimum rate for the session on or above the zero horizontal line are the subjects who read slowest on the first block of trials and maintained their initial rate, or read more rapidly, on subsequent blocks during the session.
In order to account for the possibility that there can be large inter-individual differences in visual task learning (Fendick & Westheimer, 1983; Fahle & Edelman, 1993; Beard, Levi & Reich, 1995; Fahle et al., 1995; Saarinen & Levi, 1995; Fahle & Henke-Fahle, 1996; Levi, Polat & Hu, 1997; Westheimer, 2001) we analyzed subjects individually. A Kruskal-Wallis one-way ANOVA was used to test log ratios across all subjects without regard for reading method (Chi square = 624.37, df = 134, p < 0.001). A Dunnetts test (p = 0.5) was run using the mean log ratio of the entire subject pool (n = 135) as the control. From this analysis 10 of the 135 subjects had mean log ratios (mean = 0.14; range 0.11 0.19) greater than the group mean (-0.01). Seven of the subjects had mean log ratios (mean = -0.15; range 0.12 - -0.20) that were less than the group mean. We evaluated these 10 individuals whose gains were greater than the group separately. There was no statistical difference between reading methods among the 10 individuals (Kruskal-Wallis Chi square = 2.81, df = 2, p = 0.24), and there was no statistical difference between subject type (Chi square = 1.09, df = 2, p = 0.58). In other words, these 10 individuals had a statistically significant difference in their mean log reading rate ratio for the session compared to the cohort. For this small sample, the difference cannot be attributed to a particular reading method or a particular subject group.
DISCUSSION
We did not find compelling evidence of with-in session learning after practice for any of the three reading methods or subject groups. The ten individuals with mean log reading rate ratios greater than the sample could not be classified as belonging to one of the three subject groups or be classified by method of reading. This subset of ten readers reached their fastest rate, on average, at 6.3 blocks (range 2-9) during a session that took about 1 hour. The average performance of these 10 individuals increased by a factor of 1.38 which is greater than the average within-session variation that occurs for visual tasks such as vernier acuity, but is less than the three-fold improvements that have been reported for that task after practice.(Westheimer, 2001)
There are factors that could have contributed to the lack of learning in our study. No feedback was provided during the reading session and while learning can occur without feedback, (Fahle et al., 1995) the learning rate can be slower, (Fahle & Edelman, 1993; Fahle & Henke-Fahle, 1996) and benefits of feedback can be greatest for participants with the worst initial performance.(Ball & Sekuler, 1987; Fahle & Henke-Fahle, 1996) All readers in this study were nave to the purpose and may not have approached the task with the intent of trying to improve their reading rates.(Ahissar & Hochstein, 1993) Motivation can be a driving force, and lack of feedback could have contributed to the lack of a learning effect.(Hovland, SS Stevens, 1951; Beard et al., 1995)
All of the low vision patients with AMD who participated had stable vision over the previous six months to 2 years and all of them were experienced low-vision device users. Reading with low vision aids has been likened to reading SCROLL(Legge et al., 1985b) because the text moves through the field of a magnifier, or over the video monitor of a closed-circuit television, in a manner that resembles the moving text in a SCROLL display. It is possible that the low vision readers already had experience that was similar to SCROLL reading. Similarly, the MNREAD format resembles reading from a page of written text, or reading text with standard formatting on a computer monitor so MNREAD may not be a truly novel reading experience. However, no within-session learning was apparent for RSVP reading and RSVP would not have been encountered for sustained reading in everyday experience. In addition, the statistical comparisons of the 10 individuals whose mean log ratios were greater than the group were not more likely to have read with RSVP compared to the other two methods. For this reason it is not likely that the general lack of improvement with practice can be attributed entirely to previous experience.
Other investigators have reported large inter-individual differences in the magnitude, time-course and pattern of learning for visual perceptual tasks.(Fiorentini & Berardi, 1980; Fendick & Westheimer, 1983; Fahle & Edelman, 1993; Karni & Sagi, 1993; Beard et al., 1995; Fahle et al., 1995; Saarinen & Levi, 1995; Levi et al., 1997; Westheimer, 2001) Fast within-session learning occurs for vernier acuity after about 100 trials in less than one hour of training.(Poggio et al., 1992; Fahle et al., 1995) Learning could have occurred within the first block of reading trials although previous investigators reported no learning effect during testing with four unique sentences on printed MNREAD cards.(Ahn, Legge & Luebker, 1995)
It is also possible that improvement in performance may require extended practice over many days,(Goodrich et al., 1977; Karni & Sagi, 1991; Fahle & Edelman, 1993; Karni & Sagi, 1993) or that improvement occurs between sessions and not within a single session(Beard et al., 1995). We do not know if reading performance would improve for these methods if practice were provided in the form of additional sessions.
Computer generated text can be modified in ways that are not possible with standard printed text and can provide alternative reading methods for low vision patients. (Legge et al., 1985a; Legge et al., 1985b; Fine & Peli, 1995; Rubin, 2001) Our goal was to test whether simple within-session practice would improve performance. The participants in this study did not have reading rate gains after an hour of practice even though they were reading with a method that was novel for them. If the reading method can be used efficiently at first exposure it is likely to benefit readers with low vision. We believe that this result adds to the argument for the continued development of computer controlled reading software for visually impaired readers.
Acknowledgments: Supported by National Institutes of Health grants K23-EY0366 to Kathy Aquilante and R01-EY11617 to Dean Yager. We thank William H. Swanson for his thoughtful comments on an earlier version and three anonymous reviewers.
Short title: Within-session reading practice and low vision
References
Ahissar M, Hochstein S. (1993). Attentional control of early perceptual learning. Proc Natl Acad Sci USA.90:5718-5722.
Ahn SJ, Legge GE, Luebker A. (1995). Printed Cards for measuring low-vision reading speed. Vision Res.35:1939-1944.
Altman D. (1991). Practical Statistics for Medical Research Boca Raton: Chapman & Hall/CRC.
Altman D, Bland JM. (1983). Measurement in medicine: Analysis of method comparison studies. The Statistician.32:307-317.
Aquilante K, Yager D, Morris RA. (2000). Can visual acuity predict the size of text that low vision readers need to read at maximum rates? In C Stuen, A Arditi, A Horowitz, MA Lang, B Rosenthal, KR Seidman eds. Vision rehabilitation. Assessment, intervention and outcomes; Lisse: Swets & Zeitlinger.288-292.
Aquilante K, Yager D, Morris RA. (2003). The effect of size and reading method for patients with low vision due to age-related macular degeneration. Ophthal Physiol Opt.In press.
Backman O, Inde K. (1979). Low Vision Training Malmo: LiberHermods.
Ball K, Sekuler R. (1987). Direction-specific improvement in motion discrimination. Vision Res.27:953-965.
Beard BL, Levi DM, Reich LN. (1995). Perceptual learning in parafoveal vision. Vision Res.35:1679-1690.
Chung STL, Mansfield JS, Legge GE. (1998). Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral vision. Vision Res.38:2949-2962.
Elliott DB, Bhavesh P, Whitaker D. (2001). Development of a Reading Speed Test for Potential-Vision Measurements. Invest Ophthalmol Vis Sci.42:19451949.
Fahle M, Edelman S. (1993). Long-term learning in vernier acuity: effects of stimulus orientation, range and of feedback. Vision Res.33:397-412.
Fahle M, Edelman S, Poggio T. (1995). Fast perceptual learning in hyperacuity. Vision Res.35:3003-3013.
Fahle M, Henke-Fahle S. (1996). Interobserver variance in perceptual performance and learning. Invest Ophthalmol Vis Sci.37:869-877.
Fendick M, Westheimer G. (1983). Effects of practice and the separation of test targets on foveal and peripheral stereoacuity. Vision Res.23:145-150.
Ferris FL, Kassof A, Bresnick GH, Bailey I. (1982). New visual acuity charts for clinical research. Am J Ophthalmol.94:91-96.
Fine EM, Peli E. (1995). Enhancement of text for the visually impaired. J Opt Soc Am A.12:1439-1447.
Fine EM, Peli E. (1998). Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter size. Optom Vision Sci.75:191-196.
Fiorentini A, Berardi N. (1980). Perceptual learning specific for orientation and spatial frequency. Nature.287:43-44.
Freeman PB, Jose RT, Goodrich GL. (1995). The Art and Practice of Low Vision: Butterworth-Heinemann.
Goodrich GL, Kirby J, Keswick C, Oros T, Wagstaff P, Donald B, Hazan J, Peters LJ (2000). Training the patient with low vision to read: does it significantly improve function? In: Vision Rehabilitation. Assessment, Intervention and Outcomes. C Stuen, A Arditi, A Horowitz, MA Lang, B Rosenthal and KR Seidman editors. New York: Swets & Zeitlinger Publishers, pp. 230-236.
Goodrich GL, Mehr EB. (1986). Eccentric viewing training and low vision aids: current practice and implications of peripheral retinal research. Am J Optom Physiol Optics.63:119-126.
Goodrich GL, Mehr EB, Quillman RD, Shaw HK, Wiley JK. (1977). Training and practice effects in performance with low-vision aids: A preliminary study. Am J Optom Physiol Optics.54:312-318.
Holcomb JG, Goodrich GL. (1976). Eccentric viewing training. J Amer Optom Assoc.47:1438-1443.
Hovland CI. (1951). Human Learning and Retention. In SS Stevens ed. Handbook of Experimental Psychology; New York: John Wiley & Sons, Inc.613-689.
Karni A, Meyer G, Jezzard P, Adams MM, Turner R, Ungerleider LG. (1995). Functional MRI evidence for adult motor cortex plasticity during motor skill learning. Nature.377:155-158.
Karni A, Sagi D. (1991). Where practice makes perfect in texture discrimination: evidence for primary visual cortex plasticity. Proc Natl Acad Sci U S A.88:4966-4970.
Karni A, Sagi D. (1993). The time course of learning a visual skill. Nature.365:250-252.
Legge GE, Pelli DG, Rubin GS, Schleske MM. (1985a). Psychophysics of reading I. Normal Vision. Vision Res.25:239-252.
Legge GE, Ross JA, Isenberg LM, LaMay JM. (1992). Psychophysics of reading. Clinical predictors of low-vision reading speed. Invest Ophthalmol Vis Sci.33:677-687.
Legge GE, Ross JA, Luebker A, LaMay JM. (1989). Psychophysics of reading. VIII. The Minnesota Low-Vision Reading Test. Optom Vis Sci.66:843-853.
Legge GE, Rubin GS, Pelli DG, Schleske MM. (1985b). Psychophysics of reading II. Low Vision. Vision Res.25:253-266.
Levi DM, Polat U, Hu Y-S. (1997). Improvement in vernier acuity in adults with amblyopia. Practice makes better. Invest Ophthalmol Vis Sci.38:1493-1510.
NAS-NRC-Committee-on-Vision. (1980). Recommended standard procedures for the clinical measurement and specification of visual acuity Adv. Ophthalmol.103-148.
Nilsson UL, Nilsson SRG. (1986). Rehabilitation of the visually handicapped with advanced macular degeneration. Doc Ophthalmol.62:345-367.
O'Connell WF. (1996). Eccentric viewing. In RG Cole, BP Rosenthal eds. Remediation and management of low vision; St. Louis: Mosby.27-57.
Peli E. (1986). Control of eye movement with peripheral vision: implications for training of eccentric viewing. Am J Optom Physiol Opt.63:113-118.
Poggio T, Fahle M, Edelman S. (1992). Fast perceptual learning in visual hyperacuity. Science.256:1018-1021.
Portney LG, Watkins MP. (2000). Foundations of clinical research. Applications to practice. Upper Saddle River: Prentice Hall Health.
Rubin GS. (1987). Predicting reading performance in low vision observers with age-related maculopathy (ARM). In GC Woo ed. Low vision. Principles and applications. Proceedings of the International Symposium on Low Vision, University of Waterloo, June 25-27, 1986; New York: Springer-Verlag.323-333.
Rubin GS. (2001). Vision rehabilibation for patients with age-related macular degeneration. Eye.15:430-435.
Rubin GS, Turano K. (1992). Reading without saccadic eye movements. Vision Res.32:895-902.
Rubin GS, Turano K. (1994). Low vision reading with sequential word presentation. Vision Res.34:1723-1733.
Saarinen J, Levi DM. (1995). Perceptual learning in vernier acuity: What is learned? Vision Res.35:519-527.
Sunness JS, Applegate CA, Haselwood D, Rubin GS. (1996). Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt Disease. Ophthalmology.103:1458-1466.
Timberlake GT, Mainster MA, Peli E, Augliere RA, Essock EA, Arend LE. (1986). Reading with a macular scotoma. I. Retinal location of scotoma and fixation area. Invest Ophthalmol Vis Sci.27:1137-1147.
Watson GR, De l'Aune W, Stelmack JA, Maino J, Long S. (1997). National survey of the impact of low vision device use among veterans. Optom Vis Sci.74:249-259.
Westheimer G. (2001). Is peripheral visual acuity susceptible to perceptual learning in the adult? Vision Res.41:47-52.
Table 1: Summary Statistics of Participants. Mean, standard deviation and range for age and log MAR visual acuity for each age group and each subset of readers for each of the three reading methods.
Mean; SD;
range
(n = 45 in each group)MNREAD Mean; SD;
range
(n = 15) RSVP
Mean; SD;
range
(n = 15) SCROLL Mean; SD;
range
(n = 15)
ANOVA
F2,42; pYounger
Age
log MAR VA24.9; + 3.0; 18 - 3923.8; + 1.9; 18 - 27
-0.04; + 0.07;
-0.18 to 0.0225.5; + 2.6; 22 - 32
-0.10; + 0.08;
-0.30 to 0.0025.5; + 4.2; 23 - 39
-0.07; + 0.09;
-0.22 to 0.1
1.56; 0.22
1.92; 0.16Elder
Age
log MAR VA75.8; + 7.4; 60 - 9175.5; + 7.2; 60 - 86
0.03; + 0.08;
-0.1 to 0.1874.5; + 6.6; 60 - 91
0.05; + 0.6;
-0.02 to 0.1677.6; + 8.5; 60 - 91
0.05; + 0.07;
-0.08 to 0.16
0.69; 0.51
0.26; 0.77AMD
Age
log MAR VA84.5; + 7.4;
67 - 9884.5; + 7.1; 67 - 98
1.05; + 0.33; 0.54 - 1.983.8; + 8.0;
74 - 95
1.06; + 0.40
0.4 1.687.8; + 6.1;
76 98
1.21; + 0.29
0.6 - 1.5
2.97; 0.06
1.00; 0.38
Figure 1. The change in reading rate in words per minute (A) and log ratio (B) of all blocks of reading trials compared to the first block for all subjects who read with the MNREAD method. Subject numbers 1-15 are data for the readers with macular degeneration. The log transformation (B) gives the proportional change in reading rate and allows comparisons across all three subject groups without regard to reading rate.
Figure 2. Mean log ratio ( ), minimum and maximum log ratios ( ), and standard deviation (vertical line) for each method and subject in each group. The horizontal line at zero marks block one. A few participants read as fast or faster for each block after the first (minimum on or above zero line).
The effect of size and reading method for patients with low vision due to age-related macular degeneration
Authors: Kathy Aquilante, Dean Yager, Robert A. Morris
Corresponding author: Kathy Aquilante, SUNY State College of Optometry, 33 West 42nd St., NY NY 10013
212-780-5152
212-780-5137 fax
HYPERLINK mailto:aquilante@sunyopt.edu aquilante@sunyopt.edu
Dean Yager, SUNY College of Optometry
Robert A. Morris, Ph.D., Professor of Mathematics and Computer Science, University of Massachusetts, Boston MA
Summary
We examined change in reading performance with change in character size for three reading methods including MNREAD, RSVP, and SCROLL. A video projection system with a four-foot by three-foot rear projection screen was used to display characters between approximately one-tenth of a degree to over 60 degrees of visual angle. Participants had low vision due to age-related macular degeneration (AMD) (n =36), or were age-matched with normal vision (n=34). Reading rates increased rapidly from the smallest (2x) acuity reserve, remained constant over a range of larger sizes, and slowed for the largest sizes. In contrast to previous studies, we found that reading rate functions of patients with AMD had an inverted U shape for each reading method. This confirms previous reports that low vision patients with AMD have reading rate gains when acuity reserve increases, and demonstrates that their reading rate functions are qualitatively similar to readers with normal vision.
Keywords: Low vision, age-related macular degeneration, reading, MNREAD, RSVP, SCROLL
Introduction
Most of the patients who seek low vision care are elderly and have age-related macular degeneration. ADDIN EN.CITE Elliott19971390Elliott, D.BTrukolo-Ilic, MStrong, J.GPace, RPlotkin, ABevers, P1997Demographic characteristics of the vision-disabled elderlyInvest Ophthalmol Vis Sci382566-2575repeatRussell20013240Russell, W Harper, RReeves, BWaterman, HHenson, DMcLeod, D2001Randomised controlled trial of an integrated versus an optometric low vision rehabilitation service for patients with age-related macular degeneration: study design and methodologyOphthal Physiol Opt2136-44{Elliott, Trukolo-Ilic, Strong, Pace, Plotkin & Bevers, 1997; Russell, Harper, Reeves, Waterman, Henson & McLeod, 2001} Even a moderate reduction in visual acuity to 20/40 can reduce reading rates to a level associated with disability. ADDIN EN.CITE West20024500West, S .K.Rubin, G. S.Broman, A. TMunoz, BBandeen-Roche, KTurano, K2002How does visual impairment affect performance on tasks of everyday life?Arch Ophthalmol120774-780{West, Rubin, Broman, Munoz, Bandeen-Roche & Turano, 2002} Vision loss is associated with increased difficulty with the visual tasks necessary for independent living such as paying bills. ADDIN EN.CITE Friedman19991650http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0010102282Friedman, S. M.Munoz, B.Rubin, G. S.West, S. K.Bandeen-Roche, K.Fried, L. P.Characteristics of discrepancies between self-reported visual function and measured reading speed. Salisbury Eye Evaluation Project TeamAgedAged, 80 and overContrast Sensitivity/physiologyFemaleGeriatric AssessmentHumanMaleMaryland/epidemiologyPrevalence*Reading*Self DisclosureSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision Disorders/*diagnosis/epidemiology/physiopathology*Vision TestsVisual Acuity/*physiologyDepartment of Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA.0010102282Invest Ophthalmol Vis Sci1999405858-64{Friedman, Munoz, Rubin, West, Bandeen-Roche & Fried, 1999}
Improvement in reading is the primary goal for most elderly low vision patients. ADDIN EN.CITE Elliott19971390Elliott, D.BTrukolo-Ilic, MStrong, J.GPace, RPlotkin, ABevers, P1997Demographic characteristics of the vision-disabled elderlyInvest Ophthalmol Vis Sci382566-2575repeat{Elliott et al., 1997} Low vision rehabilitation, by providing a means for improving reading, subsequently leads to improved measures of quality of life. ADDIN EN.CITE Wolffsohn20003220Wolffsohn, J.SCochrane, S.L2000Design of the low vision quality-of life questionnaire (LVQOL) and measuring the outcome of low-vision rehabilitationAm J Ophthalmol130793-802{Wolffsohn & Cochrane, 2000} Numerous researchers have reported that persons with macular degeneration read slowly, but not all people who read slowly report that reading is difficult, possibly because they use magnification to compensate for their vision loss. ADDIN EN.CITE West20024500West, S .K.Rubin, G. S.Broman, A. TMunoz, BBandeen-Roche, KTurano, K2002How does visual impairment affect performance on tasks of everyday life?Arch Ophthalmol120774-780Friedman19991650http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0010102282Friedman, S. M.Munoz, B.Rubin, G. S.West, S. K.Bandeen-Roche, K.Fried, L. P.Characteristics of discrepancies between self-reported visual function and measured reading speed. Salisbury Eye Evaluation Project TeamAgedAged, 80 and overContrast Sensitivity/physiologyFemaleGeriatric AssessmentHumanMaleMaryland/epidemiologyPrevalence*Reading*Self DisclosureSupport, Non-U.S. Gov'tSupport, U.S. Gov't, P.H.S.Vision Disorders/*diagnosis/epidemiology/physiopathology*Vision TestsVisual Acuity/*physiologyDepartment of Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland, USA.0010102282Invest Ophthalmol Vis Sci1999405858-64{Friedman et al., 1999; West et al., 2002} Reading activities become less difficult after low vision rehabilitation, in part because text size can be magnified with optical or electronic low vision devices making it possible to read ordinary print in a newspaper, read small print in a telephone book or on a medicine label, or read a bill to know if it is accurate. ADDIN EN.CITE Stelmack20024510Stelmack, J.A.Stelmack, T.R.Massof, R. W.2002Measuring low-vision rehabilitation outcomes with the NEI VFQ-25Invest Ophthalmol Vis Sci432859-2868{Stelmack, Stelmack & Massof, 2002}
Acuity reserve, a metric used to describe magnification, compares the size of letters used for reading to the size of letters at visual acuity threshold. ADDIN EN.CITE Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.{Whittaker & Lovie-Kitchin, 1993} The advantage of describing character size by acuity reserve is that readers with very different threshold visual acuity can be compared on the same scale. ADDIN EN.CITE Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.{Whittaker & Lovie-Kitchin, 1993} A specific multiple of acuity reserve has not been able to describe the amount of magnification that best serves all low vision readers. ADDIN EN.CITE Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatLovie-Kitchin20003560http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=11045244Lovie-Kitchin, J. E.Bowers, A. R.Woods, R. L.Oral and silent reading performance with macular degenerationAdultAge FactorsAgedContrast SensitivityFemaleHumanLanguage TestsLensesMacular Degeneration/*physiopathologyMaleMiddle AgePredictive Value of Tests*ReadingRegression AnalysisSupport, Non-U.S. Gov'tTime FactorsVisual AcuityCentre for Eye Research, School of Optometry, Queensland University of Technology, Australia. j.lovie-kitchin@qut.edu.au11045244Ophthalmic Physiol Opt2000205360-70.{Whittaker & Lovie-Kitchin, 1993; Fine & Peli, 1998; Lovie-Kitchin, Bowers & Woods, 2000} Some low vision patients with defects of their central field require print sizes more than 10 times larger than acuity threshold in order to attain maximum reading rates. ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Fine & Peli, 1998} Standard optical devices can limit acuity reserves to less than 3:1 in patients who have severe vision loss ADDIN EN.CITE Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.{Whittaker & Lovie-Kitchin, 1993}, and even electronic devices such as a closed-circuit television can reduce the reading field of view to a few words or characters when very high magnification is required. ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Fine & Peli, 1998}
Maximum reading rates for normal vision readers and low vision readers occur over a range of character sizes, ADDIN EN.CITE Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatLegge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733RepeatWhittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.{Legge, Pelli, Rubin & Schleske, 1985a; Legge, Rubin, Pelli & Schleske, 1985b; Rubin & Turano, 1992; Whittaker & Lovie-Kitchin, 1993; Rubin & Turano, 1994} and the character size that results in the fastest reading rate depends on the method of text presentation ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxFine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Legge, Ross, Luebker & LaMay, 1989; Rubin & Turano, 1992; Fine & Peli, 1998}. Text is read very slowly when it subtends a visual angle equal to visual acuity. ADDIN EN.CITE Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.{Whittaker & Lovie-Kitchin, 1993} As print size increases reading rates increase rapidly. The smallest size that is read at maximum has been labeled critical print size. ADDIN EN.CITE Mansfield19961980http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008675391Mansfield, J. S.Legge, G. E.Bane, M. C.Psychophysics of reading. XV: Font effects in normal and low visionAdultAgedAged, 80 and overAging/physiologyContrast SensitivityForm Perception/*physiologyHumanMiddle AgePrintingPsychophysics*ReadingSize Perception/*physiologySupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/etiology/*physiopathologyVisual Acuity/physiologyMinnesota Laboratory for Low-Vision Research, Department of Psychology, University of Minnesota, Minneapolis 55455, USA.0008675391Invest Ophthalmol Vis Sci19963781492-501{Mansfield, Legge & Bane, 1996}. In normal vision, reading rates increase from threshold size up to critical print size for both foveal and, with larger text, peripheral retinal locations, although the maximum achievable rate decreases as the retinal location used for reading moves further out into peripheral retina. ADDIN EN.CITE Chung1998160Chung, S.T.LMansfield, J.SLegge, G.E1998Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral visionVision Res382949-2962{Chung, Mansfield & Legge, 1998}
If character size continues to increase, reading rates slow for subjects with normal vision, and slow for some subjects with low vision. The decline in reading rate at large sizes occurs for text that moves right to left continuously across a visual display (SCROLL), ADDIN EN.CITE Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatLegge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266{Legge et al., 1985a; Legge et al., 1985b} for reading rates of text presented one word at a time by rapid serial visual presentation (RSVP), ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat{Rubin & Turano, 1992; Rubin & Turano, 1994} and for MNREAD, where all the words of a sentence appear at once in three or four lines ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1989}. Reading rates for text presented in a conventional page format may be less dependent on character size than reading rates for RSVP. ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sx{Rubin & Turano, 1992} In a study comparing fonts for MNREAD, reading rates increased rapidly for sizes between threshold and critical print size but did not decline for large sizes. ADDIN EN.CITE Mansfield19961980http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008675391Mansfield, J. S.Legge, G. E.Bane, M. C.Psychophysics of reading. XV: Font effects in normal and low visionAdultAgedAged, 80 and overAging/physiologyContrast SensitivityForm Perception/*physiologyHumanMiddle AgePrintingPsychophysics*ReadingSize Perception/*physiologySupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/etiology/*physiopathologyVisual Acuity/physiologyMinnesota Laboratory for Low-Vision Research, Department of Psychology, University of Minnesota, Minneapolis 55455, USA.0008675391Invest Ophthalmol Vis Sci19963781492-501{Mansfield et al., 1996}
The effect of size on reading rates for low vision readers with heterogeneous pathologies is less predictable than it is for normal vision readers. ADDIN EN.CITE Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Whittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733RepeatFine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Legge et al., 1985b; Whittaker & Lovie-Kitchin, 1993; Rubin & Turano, 1994; Fine & Peli, 1998} The fastest reading rate measured for some low vision readers with central field loss has been for the largest size used in the study. ADDIN EN.CITE Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1985b; Legge et al., 1989} To date there is no study that compared MNREAD, RSVP, and SCROLL for a range of sizes that produced an inverted U function for each reading method in readers with AMD.
The purpose of this study was to compare reading rates over a large enough range of character sizes to determine if patients with low vision due to AMD had reading rate functions in the shape of an inverted U for MNREAD, RSVP, and SCROLL methods. Text was rear-projected onto a four-foot by three-foot screen. The projection system made it possible to display characters that subtended a visual angle less than 0.1 degree to visual angles larger than 60 degrees. With the projection system we were able to include low-vision patients with AMD who met the definition of legal blindness in the United States (visual acuity equal to and worse than 20/200) while measuring reading rates for acuity reserves between approximately 2x to 16x. The video projection system magnified print to sizes that were equal to and larger than the maximum size used in other studies and made it possible to measure reading rates across the entire range of sizes for low vision patients with severe visual acuity loss. ADDIN EN.CITE Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatLegge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733RepeatWhittaker19933620Whittaker, S. G.Lovie-Kitchin, J. E.1993Visual requirements for readingOptom Vis Sci70154-65.8430009AgedCase ReportContrast SensitivityHumanMacular Degeneration/physiopathologyMalePsychophysics*ReadingScotoma/physiopathologySupport, Non-U.S. Gov'tVision/*physiologyVision Disorders/physiopathologyVisual AcuityVisual Fieldshttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=8430009Pennsylvania College of Optometry, Philadelphia.Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Legge et al., 1985a; Legge et al., 1985b; Whittaker & Lovie-Kitchin, 1993; Rubin & Turano, 1994; Fine & Peli, 1998}
Methods
Display of Text:
The text was displayed with a 640 x 480 pixel Eiki LCD projector and rear-projected onto a four-foot by three-foot rear projection screen. Room lights were off during the session. Text characters were white on a dark background. Luminance of the text was 940 cd/m2 for the white letters and 3.3 cd/m2 for the background with a Weber contrast of 0.99. A Macintosh computer with a frame rate of 67 Hz controlled the projection system. The display dynamics of the Eiki LCD projector were measured using a photodiode from United Detector Technology (www.udt.com). The image onset reached 100 percent in approximately 4 msec, and the offset fell to zero within 4 msec (making phosphor persistence inappreciable after the sentence display terminated for MNREAD, or between words for RSVP.
Text was presented in three sentence formats. For MNREAD, the entire sentence appeared on the screen at once displayed as a box of four lines with 13 characters each. The first word of each line was left-justified and the center of the box of text was set at, or just below, the line-of-sight for each subject. For RSVP sentences, one center-justified word was presented for a fixed number of frames and replaced by the next word in the next complete frame. Horizontal and vertical coordinates specified in the software placed each word at the same screen location and individual words were set to appear at or just below the subjects line-of-sight. For SCROLL, a sentence moved horizontally from right to left across the screen with the approximate vertical center of the characters set at or just below eye level for each subject.
Text size was defined by lower case x height and controlled by selecting font size with the Macintosh computer and/or by varying the distance of the subject from the projection screen. The sans serif proportionally spaced font Geneva was used for MNREAD and RSVP. Courier was used for the SCROLL because the fixed-width font can be more accurately timed as it moves through the display window. The fixed width font also allows precise control of the number of characters visible on the screen at a given time. The sentence for SCROLL appeared within in a window whose size was controlled by specifying the right and left margins in the software. The window kept the maximum number of characters at 10 regardless of the font size. There were no visual features that marked the boundaries of the SCROLL window other than the appearance and disappearance of the white letters at a specific location on the dark background. The font x-height of Geneva and Courier were matched. Each font height was verified by taking an average of 5 measurements of the character image on the projection screen with a 0.5 millimeter increment ruler. At sizes close to reading acuity limits, (the smallest text size that can be read ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1989}), sentences displayed with fixed-width fonts are read at faster rates than sentences composed of proportionally spaced fonts ADDIN EN.CITE Mansfield19961980http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008675391Mansfield, J. S.Legge, G. E.Bane, M. C.Psychophysics of reading. XV: Font effects in normal and low visionAdultAgedAged, 80 and overAging/physiologyContrast SensitivityForm Perception/*physiologyHumanMiddle AgePrintingPsychophysics*ReadingSize Perception/*physiologySupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/etiology/*physiopathologyVisual Acuity/physiologyMinnesota Laboratory for Low-Vision Research, Department of Psychology, University of Minnesota, Minneapolis 55455, USA.0008675391Invest Ophthalmol Vis Sci19963781492-501Arditi19904020Arditi, AKnoblauch, KGrunwald, I1990Reading with fixed and variable pitchJ Opt Soc Am72011-2015{Arditi, Knoblauch & Grunwald, 1990; Mansfield et al., 1996}. At critical print size (the smallest size that can be read at maximum rates) ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1989} and larger, there is no difference in reading rates for fixed-width and proportionally spaced fonts ADDIN EN.CITE Morris19914037Morris, RABerry, SHargreaves, K.ALiarokapisk, D.1991How typeface variation and typographic scaling affect readability at small sizesProceedings of the 7th International Congress on Advances in non-impact printing technologiesSoc for Imaging Sci and TechMansfield19961980http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0008675391Mansfield, J. S.Legge, G. E.Bane, M. C.Psychophysics of reading. XV: Font effects in normal and low visionAdultAgedAged, 80 and overAging/physiologyContrast SensitivityForm Perception/*physiologyHumanMiddle AgePrintingPsychophysics*ReadingSize Perception/*physiologySupport, U.S. Gov't, P.H.S.Vision TestsVision, Low/etiology/*physiopathologyVisual Acuity/physiologyMinnesota Laboratory for Low-Vision Research, Department of Psychology, University of Minnesota, Minneapolis 55455, USA.0008675391Invest Ophthalmol Vis Sci19963781492-501{Morris, Berry, Hargreaves & Liarokapisk, 1991; Mansfield et al., 1996}. The smallest size used in this experiment was twice as large as visual acuity threshold and differences in reading rates due to font effects are expected be minimal. ADDIN EN.CITE Morris19914037Morris, RABerry, SHargreaves, K.ALiarokapisk, D.1991How typeface variation and typographic scaling affect readability at small sizesProceedings of the 7th International Congress on Advances in non-impact printing technologiesSoc for Imaging Sci and Tech{Morris et al., 1991}
Seven different character sizes were used. Each size differed from its neighbor by a factor of 20.5. Calculated sizes expressed as acuity reserves were 2.00, 2.83, 4.00, 5.66, 8.00, 11.31, and 16.00. The stroke thickness of the lower case characters at the smallest size were, at minimum, two pixels wide with a height of 13 pixels. The stroke-width of the largest size was 16 pixels and letter height was 104 pixels. The actual x-height, and change between sizes corresponded to acuity reserves of 2, 3, 4, 6, 8, 11, and 16. Two blocks of trials were run at each letter size. The size order either began with the smallest size, ascended to the largest size, and was followed by the descending order, or began with the largest size and followed the reverse pattern, for different subjects.
Sentences with between 9 and 15 words each were selected from a library of about 1650 sentences. Five hundred and forty of the sentences came from the Minnesota Laboratory for Low-Vision Research. The MNREAD sentences have a range of 9 to 13 words per sentence. ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1989} Additional sentences came from the Gates-MacGinity Reading Tests ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat{Rubin & Turano, 1992; Rubin & Turano, 1994}, and from sentences edited from a variety of books downloaded from Project Gutenberg (www.promo.net/pg/). These additional sentences were edited so that they were similar in length (9 to 13 words per sentence) and complexity to the MNREAD sentences. Each sentence began with a capital letter, and the last word of the sentence in each format was marked with a period.
The sentence order was counterbalanced across subjects, reading methods, and sizes so that subtle differences in the difficulty of sentences were distributed across conditions. MNREAD sentences were used in RSVP and SCROLL displays, however the unique formatting of the MNREAD display precluded use of the entire sentence library for that method. No individual subject read any given sentence more than once.
Task
Prior to the reading trials, binocular best-corrected visual acuity was measured for each participant using an ETDRS distance visual acuity chart ADDIN EN.CITE Ferris19824 0F.L Ferris A KassofG.H BresnickI Bailey1982New visual acuity charts for clinical researchAm J Ophthalmol9491-96Repeat{Ferris, Kassof, Bresnick & Bailey, 1982 }(Lighthouse International: professional catalog@lighthouse.org). Binocular contrast sensitivity was measured with a Pelli-Robson contrast sensitivity chart ADDIN EN.CITE Pelli19883310Pelli, D. G.Robson, J.G.Wilkins, A.J.1988The design of a new letter chart for measuring contrast sensitivityClin Vision Sci2187-99{Pelli, Robson & Wilkins, 1988}(Haag-Streit; haag-streit-uk.com/ophthalmic/catalogue).
Subjects sat in an armchair in an unconstrained natural posture and read sentences from the rear-projection screen binocularly, with best correction for the reading distance. Viewing distance ranged between 50 cm to 3 meters from the rear-projection screen for all but the largest acuity reserve. Acuity reserve was based on best-corrected binocular ETDRS visual acuity. For the largest acuity reserve (16x) sixteen of the thirty-six subjects with AMD sat 25 cm from the rear projection screen. Smaller sizes were read at 50 cm or further from the display. Clip-on near correction for the viewing distance was provided as needed. For the 50 cm and 25 cm viewing distances lens correction was selected by having each subject view appropriately sized letters in a sentence and compare lenses with a focal length for straight ahead (+2.00 to +4.00), and far edge (+1.25 to +1.50) distances. Because the letter size was well above threshold the effect of blur, due to the difference in viewing distance between the center of the screen compared to the edges of the screen, was expected to be minimal. ADDIN EN.CITE Pelli19883310Pelli, D. G.Robson, J.G.Wilkins, A.J.1988The design of a new letter chart for measuring contrast sensitivityClin Vision Sci2187-99Legge19872090http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0003598739Legge, G. E.Mullen, K. T.Woo, G. C.Campbell, F. W.Tolerance to visual defocusHumanPupilSupport, U.S. Gov't, P.H.S.Vision Disorders/*physiopathology*Visual AcuityVisual Perception0003598739J Opt Soc Am [A]198745851-63{Legge, Mullen, Woo & Campbell, 1987; Pelli et al., 1988}
Subjects were observed during each trial, and reminded not to move closer or farther from the screen in order to minimize variation in visual angle due to shifts in position. If the subject moved, the block of trials was stopped, and the chair was repositioned so that they could sit comfortably at the correct distance.
An audible prompt alerted subjects when each sentence was about to begin. Subjects read aloud and could continue to read after the display of the sentence ceased. The reading rate was based on the presentation rate rather than on how quickly the words were pronounced. ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sx{Rubin & Turano, 1992} Each subject was given the option of reading the sentence when it appeared on the screen, waiting to begin to read after the sentence display ended, or using a strategy that combined approaches. Most of the participants began reading when the first word of the sentence appeared on the screen, and continued to read aloud after the last word of the sentence disappeared, suggesting that the display rate was faster than their speaking rate. Subjects were instructed that they should read as rapidly as possible but that all words in a sentence had be read, and read in the correct order, for the sentence to be counted as correct. Subjects were informed that the presentation rate would increase when the sentence was repeated verbatim, and decrease if mistakes were made, and were also told that the presentation rate would sometimes be too rapid for them to read without error. Each subject had prior experience (approximately an hour in a previous session) reading by one of the three methods at a fixed size (5 times larger than best-corrected visual acuity).
For MNREAD, the entire sentence appeared for a fixed duration. For RSVP, each word in the sentence appeared for a fixed duration and was immediately followed by the next word. For SCROLL, duration was measured from the appearance of the first letter at the right edge of the display window to the disappearance of the last letter out of the display window on the left. The experimenter saw the entire sentence in a persistent display on a separate monitor and recorded with a keystroke whether the sentence was read correctly or if any mistake was made. If the sentence was read perfectly the duration of the next sentence was reduced by 19%, (0.075 log unit), and the duration was increased by the same fraction if a mistake was made. An estimate of reading rate was made after 4 or 8 reversals in this staircase procedure. In an effort to minimize differences across subjects in the time necessary to complete the experimental sessions, 4 reversals were used when the subjects reading rate was slower than 120 words per minute (wpm), and 8 reversals were used for faster rates. While some accuracy may have been lost by using fewer reversals, this kept the entire session time to about an hour and reduced potential fatigue during the session. The starting rate for each subject was based on his or her reading rate from the previous fixed-size reading session. The starting rates were 15, 30, 60, 120, and 300 wpm.
Each participant read with each reading method. They were given the option of completing the three different methods over multiple visits in order to avoid fatigue. Readers with normal vision were able to complete all of the trials and each of three reading methods in about an hour. Most of the participants with low vision from AMD completed all of the trials for one method in a single session and returned for two additional sessions (one for each of the remaining reading methods). The slowest readers needed an hour to complete a single method. The reading method order was counterbalanced across subjects.
The software calculated the geometric mean of the staircase reversal points in the staircase procedure. The accuracy of the softwares timing calculation was confirmed manually by running a continuous 288-word sentence for the RSVP and SCROLL methods. A digital timer was started at the onset of the first word and stopped at the offset of the last word over a range of presentation rates between 15 wpm to 800 wpm. The total duration for the 288-word sentence measured by the digital timer was converted to words per minute (wpm), and the average of three measurements for each presentation rate was compared to the wpm rate recorded by the software. The accuracy of the softwares timing calculation for MNREAD was confirmed by averaging a series of manual measurements with the digital timer for sentence duration/words and comparing this to the softwares stored wpm rate. The reading rates at the staircase reversals, and the geometric mean of the reversals in wpm, for each block of trials were stored in a unique session data file for each subject. The final estimate of each subjects reading rate was calculated as the geometric mean from the two completed blocks at each size after the session finished.
Subjects
Adult paid volunteers were recruited from the State University of New York College of Optometry, University Optometric Center Low Vision Rehabilitation Service and Primary Care Service. Low vision participants with a diagnosis of age-related macular degeneration (AMD), and age-matched elder subjects with log MAR visual acuity of 0.2 (20/30) or better, and no diagnosed ocular pathology, took part in the study. Thirty-six subjects with AMD and thirty-four subjects with normal vision participated in the study. The mean age of the AMD group was 82.3 years (median 81.5; range 67 - 98) and the mean age of the elder group was 79.5 years (median 80.0; range 60 91) with no statistical age difference between groups (Median test: Chi Sq = 0.52, df = 1, p = 0.47). The mean log MAR visual acuity for the elder group with normal vision (0.05; median 0.04; range 0.08 to 0.16) was statistically different than the group with AMD (1.06; median 1.14; range 0.40 to 1.70; Median test Chi Sq = 65.17, df = 1, p < 0.0001). Mean Pelli-Robson contrast was also statistically different between the two groups, (Median test Chi Sq = 54.12, df = 1, p < 0.0001) being 1.71 log units (median 1.70; range 1.2 2.1) for the elder group with normal vision, and 1.00 (median 0.98; range 0.40 1.65) for the group with AMD.
The procedures used in this study conformed to the tenets of the Declaration of Helsinki. All subjects read (or had read to them) and then signed an informed consent form before testing began. All study protocols were approved by the Institutional Review Board on human research at the State University of New York, State College of Optometry.
Data analysis:
Data were analyzed with SAS JMP software, (Version 4 (SAS Institue, Cary NC 27513)). A Shapiro-Wilk test was used to test for a normal distribution and nonparametic tests were used for analysis when the data did not conform to a normal distribution. Mean wpm data were converted to log reading rates and a three-way repeated measures analysis of variance (ANOVA) was used to compare reading rate for each acuity reserve, and method, between groups. Two multiple comparisons tests were used. A one-tailed Dunnetts test ADDIN EN.CITE Keppel19914581Keppel, G1991Design and Anallysis. A researcher's handbook.Upper Saddle RiverPrentice Hall{Keppel, 1991} was used to compare the mean reading rate for the smallest size (control) of each method to each of the larger sizes within that method. Hsus MCB (multiple comparisons with the best) ADDIN EN.CITE Hsu19844590Hsu, J.C1984Constrained simultaneous confidence intervals for multiple comparisons with the bestAnn Statistics121136-1144{Hsu, 1984} was used to determine if there was a discrete maximum reading rate at a given acuity reserve within each method, or if the maximum rate was maintained over a range of acuity reserves. Mean reading rates within a method were averaged across the corresponding acuity reserves if the rates were not statistically different.
Acuity reserve for optimum reading is reported to vary among individuals who have macular degeneration. ADDIN EN.CITE Lovie-Kitchin20003560http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=11045244Lovie-Kitchin, J. E.Bowers, A. R.Woods, R. L.Oral and silent reading performance with macular degenerationAdultAge FactorsAgedContrast SensitivityFemaleHumanLanguage TestsLensesMacular Degeneration/*physiopathologyMaleMiddle AgePredictive Value of Tests*ReadingRegression AnalysisSupport, Non-U.S. Gov'tTime FactorsVisual AcuityCentre for Eye Research, School of Optometry, Queensland University of Technology, Australia. j.lovie-kitchin@qut.edu.au11045244Ophthalmic Physiol Opt2000205360-70.{Lovie-Kitchin et al., 2000} Reading rate data were sorted across subjects with AMD in order to compare the maximum rate for each method across individuals. A nonparametric Median test was used for between group comparisons ADDIN EN.CITE Siegel19562621Siegel, S1956Nonparametric statistics for the behavioral sciencesNew YorkMcGraw-Hill166-173RepeatAltman19914011Altman, DG1991Practical Statistics for Medical ResearchBoca RatonChapman & Hall/CRCPortney20003251Portney, L.GWatkins, M.P2000Foundations of clinical research. Applications to practice.Upper Saddle RiverPrentice Hall Health{Siegel, 1956; Altman, 1991; Portney & Watkins, 2000}, as was Friedman's statistic, a ranking for subjects who receive multiple treatments. ADDIN EN.CITE Siegel19562621Siegel, S1956Nonparametric statistics for the behavioral sciencesNew YorkMcGraw-Hill166-173RepeatAltman19914011Altman, DG1991Practical Statistics for Medical ResearchBoca RatonChapman & Hall/CRCPortney20003251Portney, L.GWatkins, M.P2000Foundations of clinical research. Applications to practice.Upper Saddle RiverPrentice Hall Health{Siegel, 1956; Altman, 1991; Portney & Watkins, 2000} Stepwise multiple regression was used to analyze predictor variables age, visual acuity, and peak contrast sensitivity. Probability levels were set as less than 0.05 to indicate statistical significance for all analyses.
Results
Figure 1 demonstrates the inverted U shape of the reading rate functions for low vision patients with AMD, for each of the reading methods. About half (n = 17) of the group with AMD either said the sizes corresponding to 2x and 3x acuity reserves were too small to read, or were not able to read without error until the acuity reserve reached 4x. Because of this, readers with AMD were separated into sub-groups for graphing of their reading rate functions. A repeated measures ANOVA comparing log reading rates for elders with normal vision and low vision patients with AMD, by method, and sizes 4x to 16x, showed a significant main effect for subject group (F1,68 = 105.73, p < 0.0001)), a significant main effect for size (F4,272 = 63.92, p < 0.0001), a significant main effect for method (F2,136 = 9.49 p < 0.0001), and significant interactions between group and size(F = 46.74, p < 0.0001), group and method (F =19.00, p < 0.0001) and group, size, and method (F = 10.82, p < 0.0001).
For all three reading methods the elder readers with normal vision read the 2x acuity reserve more slowly than other sizes within the method (Dunnetts method with 2x as the control, p < 0.05). For the sizes used in this study elders read the 16x acuity reserve more slowly than peak rates for MNREAD and RSVP, but the reading rate for the16x size in SCROLL was not statistically different than the peak rate for that method (Hsus MCB, p < 0.05). The peak reading rates for MNREAD and RSVP occurred with text sized between 4x and 11x acuity reserves. Reading rates with SCROLL approached maximum, for the elder group with normal vision, when character size ranged from 6x to 16x (Hsus MCB, p < 0.05).
As expected, Figure 1 demonstrates that readers with AMD read more slowly than elders with normal vision. When the text size based on acuity reserve is compared within each method, reading rate depends on size [Friedmans two-way analysis of variance (Chi Sq = 32.00, df = 2, p < 0.001)]. The 4x acuity reserve was set as the control for the Dunnetts test of all subjects with AMD. The Dunnetts confirmed (p < 0.05) that the acuity reserve of 4x size was read more slowly than larger sizes for each of three reading methods. Reading rates fell off for the 16x acuity reserve (Hsus MCB, p < 0.05) with the fastest rates being between acuity reserves of 6x-8x for MNREAD, and 8x to 11x for RSVP and SCROLL. When Hsus MCB is applied separately to the faster group and slower group of subjects with AMD the size range for maximum reading rates are different. The fastest readers with AMD had maximum rates between 4x and 11x for MNREAD and SCROLL (ps < 0.05), and between 6x and 11x for RSVP (p < 0.05). The group that read more slowly had maximum reading rates when the character size was between 6x to 11x for MNREAD (p < 0.05), and between 8x and 11x for RSVP and SCROLL (ps < 0.05).
Figure 2 plots log reading rate against character size in degrees of visual angle on a log scale. Each subject is represented reading by each of the three methods. The elder subjects with normal vision are graphed with the solid symbols. Six of the elders with normal vision read by each method for character sizes that were as large as the largest sizes read by the low vision patients with AMD. These pilot data (degrees > 2 log units) are included in the graph. At these large sizes low vision patients with AMD read more slowly than elders with normal vision.
The reading rates of each participant with AMD were sorted from fastest to slowest, for each method, in order to make additional comparisons between the reading methods. The difference in maximum log reading rates between methods for each subject with AMD are shown in Figure 3. For this comparison the log of the maximum rate for MNREAD was subtracted from the log of the maximum reading rates for RSVP and SCROLL for each subject. The horizontal line at zero is the equality line along which reading rates for the three methods would be identical. The subjects are arranged in order, by ascending reading rates, with the slowest at the left (1 on the abscissa). As reading rates increase across individual subjects a few of the data points for the maximum rates with SCROLL minus maximum rates with MNREAD, cross the equality line. The fastest readers with AMD read MNREAD more rapidly than SCROLL. All of the participants with macular degeneration read RSVP more rapidly than MNREAD, and most read SCROLL faster than MNREAD (within-subject ANOVA F(2, 78) = 25.81, p < 0.005). A Tukey-Kramer HSD test confirms (p < 0.05) that RSVP and SCROLL differ from MNREAD. This pair-wise comparison did not indicate a statistical difference between RSVP and SCROLL.
The lack of difference between RSVP and SCROLL reading rates was unexpected since we varied size across methods. ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196Repeat{Fine & Peli, 1998} In order to compare these methods further we sorted reading rates for all of the participants with AMD and selected either the maximum rate for RSVP, or the maximum rate for SCROLL, for each subject depending on which was faster. These data are shown in Figure 4 as the difference between the log of the maximum reading rate of RSVP minus the log of the maximum reading rate of SCROLL. Each datum represents one subject. The horizontal line at zero is the equality line (RSVP minus SCROLL = 0). The slowest readers (N = 17) read SCROLL more rapidly (data below zero equality line) than RSVP, but the rate benefit shifted to RSVP (N = 19) for faster readers (Median test: Chi Sq = 24.38, df = 1, p < 0.0001). The median log difference between RSVP and SCROLL, for the subjects who read SCROLL more rapidly, falls along the horizontal line at 0.11 in Figure 4. The median log difference (RSVP minus SCROLL) for the faster readers falls along the horizontal line at 0.20). There were differences besides reading rate between the groups with AMD who read either RSVP or SCROLL faster. The SCROLL faster group had a higher mean age although this did not reach statistical significance (85.0 vs 80.37: Median test: Chi Sq = 0.98, df = 1, p = 0.32). The SCROLL faster group had more severe visual acuity loss (mean logMAR 1.30 vs. 0.84: Median test: Chi Sq = 13.11, df = 1, p = 0.0003). The SCROLL faster group also had lower peak contrast sensitivity but this was not statistically different between groups (mean 0.81 vs. 1.16: Median test: Chi Sq = 2.70, df = 1, p = 0.10). Multiple stepwise regression was used to evaluate the predictive value of age (p = 0.33), peak contrast sensitivity (p = 0.64), and log MAR distance visual acuity on maximum reading rate. Log MAR visual acuity was the only variable that was statistically significant in the regression model. The Spearman Rho rank correlation coefficient was 0.74 (p < 0.0001).
Discussion
The video projection system allowed us to measure reading rates over a large enough range of acuity reserves to demonstrate that low vision readers with macular degeneration have reading rate functions qualitatively similar to readers with normal vision. The reading rate functions for each of the reading methods increases to a maximum as character size increases, remains at the maximum rate for a range of sizes, and then falls for large sizes.
We used Hus MCB to determine if there was a discrete maximum reading rate among the acuity reserves for each method across subjects. The result of this comparison for the group with AMD who read rapidly was that there was no statistical difference in reading rates for 4,6,8 and 11x acuity reserves for MNREAD and SCROLL, or 6,8 and 11x acuity reserves for RSVP. This is in agreement with previous reports that RSVP requires larger text sizes to be read at maximum rates. ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat{Rubin & Turano, 1994; Fine & Peli, 1998} The maximum for the group that read more slowly was between 6 and 11x for MNREAD, and between 8 and 11x for RSVP and SCROLL also in agreement with previous studies reporting that reading rates are slower for readers who require large letters to reach maximum. ADDIN EN.CITE Fine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733RepeatLegge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266{Legge et al., 1985b; Rubin & Turano, 1994; Fine & Peli, 1998}
We found a Spearman correlation of -0.74 between logMAR distance visual acuity and the maximum reading rate for the readers with AMD. The predictive value of distance visual acuity has been variable ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.Legge19921290Legge, G. E. Ross, J. A. Isenberg, L. M.LaMay, J. M.1992Psychophysics of reading. Clinical predictors of low-vision reading speedInvest Ophthalmol Vis Sci33677-687Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Ahn19952000http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0007660598Ahn, S. J.Legge, G. E.Psychophysics of reading--XIII. Predictors of magnifier-aided reading speed in low visionAdultAge FactorsAgedAged, 80 and overFemaleHumanLensesMaleMiddle AgePredictive Value of TestsPsychophysics*ReadingSupport, U.S. Gov't, P.H.S.Time FactorsVision TestsVision, Low/*physiopathologyVisual AcuityMinnesota Laboratory for Low-vision Research, Department of Psychology, University of Minnesota, Minneapolis 55455, USA.0007660598Vision Res199535131931-8Ahn1995570Ahn, S.J Legge, G.E Luebker, A1995Printed Cards for measuring low-vision reading speedVision Res351939-1944Repeat{Legge et al., 1985b; Legge et al., 1989; Legge, Ross, Isenberg & LaMay, 1992; Ahn & Legge, 1995; Ahn, Legge & Luebker, 1995} with a number of researchers arguing that near word reading acuity has a higher correlation to reading rate than distance acuity ADDIN EN.CITE Bullimore1995710Bullimore, M.ABailey, I.L1995Reading and eye movements in age-related maculopathyOptom Vis Sci72125-138Lovie-Kitchin20003560http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=11045244Lovie-Kitchin, J. E.Bowers, A. R.Woods, R. L.Oral and silent reading performance with macular degenerationAdultAge FactorsAgedContrast SensitivityFemaleHumanLanguage TestsLensesMacular Degeneration/*physiopathologyMaleMiddle AgePredictive Value of Tests*ReadingRegression AnalysisSupport, Non-U.S. Gov'tTime FactorsVisual AcuityCentre for Eye Research, School of Optometry, Queensland University of Technology, Australia. j.lovie-kitchin@qut.edu.au11045244Ophthalmic Physiol Opt2000205360-70.{Bullimore & Bailey, 1995; Lovie-Kitchin et al., 2000}. The high correlation in our study could be a range effect ADDIN EN.CITE Altman19832600Altman, DGBland, J.M1983Measurement in medicine: Analysis of method comparison studiesThe Statistician32307-317Repeat{Altman & Bland, 1983} since visual acuity in our sample of low vision patients with macular degeneration ranged from 0.40 to 1.70 logMAR (20/50 20/1000). In normal vision, reading rates slow, and visual acuity worsens, as the target moves farther away from the fovea into peripheral retina. Slower reading rates for subjects with low vision due to AMD have been associated with worsening visual acuity, and increasing area of central scotomas. ADDIN EN.CITE Sunness19961000Sunness, J. S.Applegate, C. A.Haselwood, DRubin, G. S.1996Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt DiseaseOphthalmology1031458-1466Sunness1999990Sunness, J.SGonalez-Baron, JApplegate, C.ABressler, N.MTian, YanHawkins, BBarron, YBergman, A1999Enlargement of atrophy and visual acuity loss in the geographic atrophy form of age-related macular degenerationOphthalmology1061768-1779{Sunness, Applegate, Haselwood & Rubin, 1996; Sunness, Gonalez-Baron, Applegate, Bressler, Tian, Hawkins, Barron & Bergman, 1999}
We have confirmed previous reports that reading rates for RSVP slow for print sized at 16x acuity reserves and larger for readers with normal vision. ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sx{Rubin & Turano, 1992} The decline in reading rate at the 16x acuity reserve for MNREAD in subjects with normal vision has not been previously reported. Static text is not as dependent on size as RSVP reading ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sx{Rubin & Turano, 1992}, and previous studies with the MNREAD format did not find declining rates for normal vision readers until text size was 3 degrees or larger. ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.{Legge et al., 1989} The readers with normal vision in our study read at their maximum rates, on average, for sizes that were 4 to11 times (mean 0.40 to1.10 degrees, range 0.28 to 3.8 degrees visual angle) larger than threshold acuity for MNREAD and RSVP, and between 4x to16x for SCROLL (mean 0.40 to 1.61 degrees, range 0.28 to 5.6 degrees of visual angle).
The reading rate gain for RSVP compared to static text for our group of elders with normal vision was smaller than previously reported. ADDIN EN.CITE Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat{Rubin & Turano, 1994} Instructing subjects to read for understanding results in measured reading rates that are slower than instructing them to read as fast as possible. ADDIN EN.CITE Lovie-Kitchin20003560http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=11045244Lovie-Kitchin, J. E.Bowers, A. R.Woods, R. L.Oral and silent reading performance with macular degenerationAdultAge FactorsAgedContrast SensitivityFemaleHumanLanguage TestsLensesMacular Degeneration/*physiopathologyMaleMiddle AgePredictive Value of Tests*ReadingRegression AnalysisSupport, Non-U.S. Gov'tTime FactorsVisual AcuityCentre for Eye Research, School of Optometry, Queensland University of Technology, Australia. j.lovie-kitchin@qut.edu.au11045244Ophthalmic Physiol Opt2000205360-70.{Lovie-Kitchin et al., 2000} Although we did not instruct our subjects to read for understanding, they were instructed to repeat the sentence exactly as it appeared, and they were told that the sentence may sometimes appear too quickly to be read accurately. In addition, all of the subjects had prior experience with the staircase procedure and they may have been more likely to respond with an error knowing that the following sentence would appear at a slower rate and be easier to read without error. ADDIN EN.CITE Woods19963010Woods, R.L1996Spatial frequency dependent observer bias in the measurement of contrast sensitivityOphthal Physiol Opt16513-519{Woods, 1996}
In previous studies low vision readers have not had the same reading rate advantage for increasing print size as readers with normal vision ADDIN EN.CITE Legge19892050Legge, G. E.Ross, J. A.Luebker, A.LaMay, J. M.1989Psychophysics of reading. VIII. The Minnesota Low-Vision Reading TestOptom Vis Sci6612843-530002626251*ReadingVision Tests/*methodsVision, Low/*physiopathologyRepeathttp://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0002626251Department of Psychology, University of Minnesota, Minneapolis.Legge1985190Legge, G. E.Rubin, G. S.Pelli, D. G.Schleske, M. M.1985Psychophysics of reading II. Low VisionVision Res25253-266Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatFine1998170E.M FineE Peli1998Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter sizeOptom Vision Sci75191-196RepeatLovie-Kitchin19871347Lovie-Kitchin, J.EWoo, G.C1987Effect of magnification and field of view on reading speed using a CCTVWoo, G.CLow vision. Principles and applications. Proceedings of the International Symposium on Low Vision, University of Waterloo, June 25-27, 1986New YorkSpringer-Verlag308-322RepeatRubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxRubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat{Legge et al., 1985a; Legge et al., 1985b; Lovie-Kitchin & Woo, GC Woo, 1987; Legge et al., 1989; Rubin & Turano, 1992; Rubin & Turano, 1994; Fine & Peli, 1998}. Our readers with normal vision and vision loss from AMD, on average, had a doubling of reading rate for each of the reading methods when the smallest size read was compared to the averaged maximum.
In conclusion, the projection system accommodated a large range of character sizes, and made it possible to include low vision patients with severe vision loss and legal blindness due to age-related macular degeneration. The data demonstrate that reading rate functions for subjects with AMD had an inverted U shape for all three reading methods. The 8x acuity reserve was included within the range of sizes that resulted in maximum reading rates for each of the reading methods. We recommend that magnification should encompass acuity reserves of 8x for these reading methods in order to assure that reading rates are close to maximum.
Acknowledgments
This work was supported by National Institutes of Health grants K23-EY-00366 (KA) and R01-EY-11617 (DY). We thank William H. Swanson for his thoughtful comments.
References
QUOTE EN.REFLIST
Ahn SJ, Legge GE. (1995). Psychophysics of reading-XIII. Predictors of magnifier-aided reading speed in low vision. Vision Res.35:1931-1938.
Ahn SJ, Legge GE, Luebker A. (1995). Printed Cards for measuring low-vision reading speed. Vision Res.35:1939-1944.
Altman D. (1991). Practical Statistics for Medical Research Boca Raton: Chapman & Hall/CRC.
Altman D, Bland JM. (1983). Measurement in medicine: Analysis of method comparison studies. The Statistician.32:307-317.
Arditi A, Knoblauch K, Grunwald I. (1990). Reading with fixed and variable pitch. J Opt Soc Am.7:2011-2015.
Bullimore MA, Bailey IL. (1995). Reading and eye movements in age-related maculopathy. Optom Vis Sci.72:125-138.
Chung STL, Mansfield JS, Legge GE. (1998). Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral vision. Vision Res.38:2949-2962.
Elliott DB, Trukolo-Ilic M, Strong JG, Pace R, Plotkin A, Bevers P. (1997). Demographic characteristics of the vision-disabled elderly. Invest Ophthalmol Vis Sci.38:2566-2575.
Ferris FL, Kassof A, Bresnick GH, Bailey I. (1982). New visual acuity charts for clinical research. Am J Ophthalmol.94:91-96.
Fine EM, Peli E. (1998). Benefits of rapid serial visual presentation (RSVP) over scrolled text vary with letter size. Optom Vision Sci.75:191-196.
Friedman SM, Munoz B, Rubin GS, West SK, Bandeen-Roche K, Fried LP. (1999). Characteristics of discrepancies between self-reported visual function and measured reading speed. Salisbury Eye Evaluation Project Team. Invest Ophthalmol Vis Sci.40:858-864.
Hsu JC. (1984). Constrained simultaneous confidence intervals for multiple comparisons with the best. Ann Statistics.12:1136-1144.
Keppel G. (1991). Design and Anallysis. A researcher's handbook. Upper Saddle River: Prentice Hall.
Legge GE, Mullen KT, Woo GC, Campbell FW. (1987). Tolerance to visual defocus. J Opt Soc Am [A].4:851-863.
Legge GE, Pelli DG, Rubin GS, Schleske MM. (1985a). Psychophysics of reading I. Normal Vision. Vision Res.25:239-252.
Legge GE, Ross JA, Isenberg LM, LaMay JM. (1992). Psychophysics of reading. Clinical predictors of low-vision reading speed. Invest Ophthalmol Vis Sci.33:677-687.
Legge GE, Ross JA, Luebker A, LaMay JM. (1989). Psychophysics of reading. VIII. The Minnesota Low-Vision Reading Test. Optom Vis Sci.66:843-853.
Legge GE, Rubin GS, Pelli DG, Schleske MM. (1985b). Psychophysics of reading II. Low Vision. Vision Res.25:253-266.
Lovie-Kitchin JE, Bowers AR, Woods RL. (2000). Oral and silent reading performance with macular degeneration. Ophthalmic Physiol Opt.20:360-370.
Lovie-Kitchin JE, Woo GC. (1987). Effect of magnification and field of view on reading speed using a CCTV. In GC Woo ed. Low vision. Principles and applications. Proceedings of the International Symposium on Low Vision, University of Waterloo, June 25-27, 1986; New York: Springer-Verlag.308-322.
Mansfield JS, Legge GE, Bane MC. (1996). Psychophysics of reading. XV: Font effects in normal and low vision. Invest Ophthalmol Vis Sci.37:1492-1501.
Morris R, Berry S, Hargreaves KA, Liarokapisk D. (1991). How typeface variation and typographic scaling affect readability at small sizes Proceedings of the 7th International Congress on Advances in non-impact printing technologies: Soc for Imaging Sci and Tech.
Pelli DG, Robson JG, Wilkins AJ. (1988). The design of a new letter chart for measuring contrast sensitivity. Clin Vision Sci.2:187-199.
Portney LG, Watkins MP. (2000). Foundations of clinical research. Applications to practice. Upper Saddle River: Prentice Hall Health.
Rubin GS, Turano K. (1992). Reading without saccadic eye movements. Vision Res.32:895-902.
Rubin GS, Turano K. (1994). Low vision reading with sequential word presentation. Vision Res.34:1723-1733.
Russell W, Harper R, Reeves B, Waterman H, Henson D, McLeod D. (2001). Randomised controlled trial of an integrated versus an optometric low vision rehabilitation service for patients with age-related macular degeneration: study design and methodology. Ophthal Physiol Opt.21:36-44.
Siegel S. (1956). Nonparametric statistics for the behavioral sciences New York: McGraw-Hill.
Stelmack JA, Stelmack TR, Massof RW. (2002). Measuring low-vision rehabilitation outcomes with the NEI VFQ-25. Invest Ophthalmol Vis Sci.43:2859-2868.
Sunness JS, Applegate CA, Haselwood D, Rubin GS. (1996). Fixation patterns and reading rates in eyes with central scotomas from advanced atrophic age-related macular degeneration and Stargardt Disease. Ophthalmology.103:1458-1466.
Sunness JS, Gonalez-Baron J, Applegate CA, Bressler NM, Tian Y, Hawkins B, Barron Y, Bergman A. (1999). Enlargement of atrophy and visual acuity loss in the geographic atrophy form of age-related macular degeneration. Ophthalmology.106:1768-1779.
West SK, Rubin GS, Broman AT, Munoz B, Bandeen-Roche K, Turano K. (2002). How does visual impairment affect performance on tasks of everyday life? Arch Ophthalmol.120:774-780.
Whittaker SG, Lovie-Kitchin JE. (1993). Visual requirements for reading. Optom Vis Sci.70:54-65.
Wolffsohn JS, Cochrane SL. (2000). Design of the low vision quality-of life questionnaire (LVQOL) and measuring the outcome of low-vision rehabilitation. Am J Ophthalmol.130:793-802.
Woods RL. (1996). Spatial frequency dependent observer bias in the measurement of contrast sensitivity. Ophthal Physiol Opt.16:513-519.
Short title:
Text size and reading in AMD
Figure 1. Reading rate as a function of acuity reserve averaged across 34 elder subjects with normal vision and 36 subjects with AMD. The top 3 functions are MNREAD ( ), RSVP ( ), and SCROLL ( ) for the elder readers. The center three functions are for readers with AMD who could read at the smallest acuity reserve, and the lower three functions are readers with AMD who began reading without error at an acuity reserve of 4x for MNREAD ( ), RSVP ( ), and SCROLL ( ).
Figure 2. Reading rate as a function of visual angle in degrees. Elder subjects with normal vision are represented by the solid symbols, and subjects with AMD are represented by the open symbols.
Figure 3. Open circles show the log difference between the maximum RSVP reading rate minus the maximum MNREAD reading rate for each subject with AMD. Triangles show the log difference between the maximum SCROLL reading rate minus the maximum MNREAD reading rate for the same subjects. Data above the zero equality line indicate that RSVP and SCROLL rates were faster than MNREAD.
Figure 4. The log difference between RSVP reading rates minus SCROLL reading rates for each subject with AMD. Data below the zero equality line (open triangles) show the subjects who read SCROLL faster than RSVP (median line at 0.11 log units). Data above the zero equality line (open circles) show the subjects who read RSVP faster than SCROLL (median line at 0.2 log units).
Figure 1
Figure 2
Figure 3
Figure 4
Title: RSVP reading with variable word duration based on word length and a minimum duration imposed on short words
Author 1: Kathy Aquilante, O.D., MS, Assistant Professor, SUNY College of Optometry, 33 W 42nd St., NY, NY 10036
Author 2: Dean Yager, Ph.D., Distinguished Professor, SUNY College of Optometry
Author 3: Robert A. Morris, Ph.D., Professor of Mathematics and Computer Science, University of Massachusetts, Boston MA 02125
Number of Figures: 6
Submitted Optometry and Vision Science
ABSTRACT: Young and elder participants with normal vision, and patients with impaired vision from age related macular degeneration (AMD) read RSVP sentences that were presented at four different rates. Each word in a sentence appeared for the same duration, or the duration of each word varied according the number of characters in the word. The three variable word duration conditions differed from one another the amount of
with words presented at a constant rate and read RSVP sentences at three rates where word duration varied according to word length. Half of the blocks of trials also included a minimum duration that set a ceiling on the most rapid presentation rate. Readers with normal vision read more slowly when presentation duration varied with word length but read a similar rates for the blocks of trials that included a minimum duration for all of the timing conditions except the most extreme. The readers with impaired vision read more rapidly when presentation rate varied with word length and lost some of that advantage when the blocks of trials included a minimum duration. This study confirms our previous report that variable word duration based on word length improves reading rates for low vision patients with AMD. Imposing a minimum reduced the reading rate loss in the variable presentation rate for participants with normal vision. The minimum did not provide an advantage for readers with AMD and did not compensate for the loss of reading rate in the condition where word length had the largest effect on presentation duration.
We reported in a previous study that low vision patients with central visual field loss from age related macular degeneration (AMD) had faster reading rates when RSVP (rapid serial visual presentation) text was displayed such that the duration of words on the screen varied according to word length. ADDIN EN.CITE Aquilante20013340Aquilante, KYager, DMorris, R.AKhmelnitsky, F2001Low vision patients with age-related maculopathy read RSVP faster when word duration varies according to word lengthOptom Vis Sci78290-296(Aquilante, Yager, Morris & Khmelnitsky, 2001) RSVP text presents words in sequence, one at a time, at the same location on the display. Words can be presented at a constant rate or at a variable rate related to the time needed to process words of varying difficulty. ADDIN EN.CITE Ward1982800Ward, N.JJoula, J.F1982Reading with and without eye movements: Reply to Just, Carpenter and WoolleyJ Experim Psychol: Gen111239-241(Ward & Joula, 1982) Readers with normal vision, reading standard text, may pause longer on long words or less familiar words than they do on short words ADDIN EN.CITE Just1982650Just, M.ACarpenter, P.AWoolley, J.D1982Paradigms and processes in reading comprehensionJ Experim Psychol: Gen111228-238(Just, Carpenter & Woolley, 1982) and they may not stop at all to fixate short words such as I, a or at. ADDIN EN.CITE Just1980640Just, M.ACarpenter, P.A1980A theory of reading: From eye fixations to comprehensionPsychol Review87329-354(Just & Carpenter, 1980)
Unlike subjects with normal vision, low vision readers make multiple fixations within words during RSVP reading. ADDIN EN.CITE Rubin1994730Rubin, G. S. Turano, K1994Low vision reading with sequential word presentationVision Res341723-1733Repeat(Rubin & Turano, 1994) Previous studies have demonstrated that word recognition times are dependent on word length. ADDIN EN.CITE Legge19971950http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009274784Legge, G. E.Ahn, S. J.Klitz, T. S.Luebker, A.Psychophysics of reading--XVI. The visual span in normal and low visionAdolescenceAdultContrast SensitivityEye MovementsFemaleFixation, OcularHumanMalePsychophysics*ReadingSaccadesSupport, U.S. Gov't, P.H.S.Time FactorsVision, Low/*physiopathologyDepartment of Psychology, University of Minnesota, Minneapolis 55455, USA. legge@eye.psych.umn.edu0009274784Vision Res199737141999-2010(Legge, Ahn, Klitz & Luebker, 1997) And several investigators have suggested that visual span is reduced in low vision. ADDIN EN.CITE Bullimore1995710Bullimore, M.ABailey, I.L1995Reading and eye movements in age-related maculopathyOptom Vis Sci72125-138Legge19971950http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009274784Legge, G. E.Ahn, S. J.Klitz, T. S.Luebker, A.Psychophysics of reading--XVI. The visual span in normal and low visionAdolescenceAdultContrast SensitivityEye MovementsFemaleFixation, OcularHumanMalePsychophysics*ReadingSaccadesSupport, U.S. Gov't, P.H.S.Time FactorsVision, Low/*physiopathologyDepartment of Psychology, University of Minnesota, Minneapolis 55455, USA. legge@eye.psych.umn.edu0009274784Vision Res199737141999-2010(Bullimore & Bailey, 1995; Legge, et al., 1997) It takes longer for low vision readers to recognize words that exceed their visual span because multiple fixations are required. ADDIN EN.CITE Legge19971950http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=m&form=6&dopt=r&uid=0009274784Legge, G. E.Ahn, S. J.Klitz, T. S.Luebker, A.Psychophysics of reading--XVI. The visual span in normal and low visionAdolescenceAdultContrast SensitivityEye MovementsFemaleFixation, OcularHumanMalePsychophysics*ReadingSaccadesSupport, U.S. Gov't, P.H.S.Time FactorsVision, Low/*physiopathologyDepartment of Psychology, University of Minnesota, Minneapolis 55455, USA. legge@eye.psych.umn.edu0009274784Vision Res199737141999-2010(Legge, et al., 1997)
In self-paced variant of RSVP, readers were free to pause for as long as they needed on each word and there was a large reading rate advantage for the low vision group in that study. ADDIN EN.CITE Arditi1999620Arditi, A1999Elicited sequential presentation for low vision readingVision Res394412-4418(Arditi, 1999) While it has been suggested that subjects with normal vision can benefit from RSVP text presented at a variable rate ADDIN EN.CITE Ward1982800Ward, N.JJoula, J.F1982Reading with and without eye movements: Reply to Just, Carpenter and WoolleyJ Experim Psychol: Gen111239-241(Ward & Joula, 1982) we found that reading rates slowed for normal vision readers when the presentation duration was based on word length compared to a constant duration condition. ADDIN EN.CITE Aquilante20013340Aquilante, KYager, DMorris, R.AKhmelnitsky, F2001Low vision patients with age-related maculopathy read RSVP faster when word duration varies according to word lengthOptom Vis Sci78290-296(Aquilante, et al., 2001) A possible explanation for the slower reading rates was that short words were on the screen too briefly to be recognized. ADDIN EN.CITE Forster1970660Forster, K.I1970Visual perception of rapidly presented word sequences of varying complexityPercept Psychophys8215-222Rayner1981830Rayner, KInoff, A.WMorrison, R.ESlowiaczek, M.LBertera, J.H1981Masking of foveal and parafoveal vision during eye fixations in readingJ Experim Psychol: Human Percept7167-179(Forster, 1970; Rayner, Inoff, Morrison, Slowiaczek & Bertera, 1981) We also found that low vision patients lost their reading rate advantage for the variable word duration presentation when the difference in duration of short vs. long words was greatest. We hypothesize that a minimum duration should have been included in the software in order to set a ceiling for the display rate of short words. This study compares reading rates when there is a minimum duration set for short words to the same timing conditions without the limit of the minimum duration for short words.
METHODS
Subjects
Twenty-four State University of New York, State College of Optometry students participated in the study (mean age: 24.5; mean log MAR visual acuity: 0.067). Twenty-four elder subjects [(mean age: 80.1 years, range (68 95); mean log MAR VA: 0.04, range (-0.06 0.2)] participated. Elder subjects were from the University Optometric Center Primary Care Service. Subjects with ARD were recruited from the Low Vision Service. Twenty-four subjects with AMD participated [mean age: 83.4, (range 74 95), mean log MAR VA 1.09, (range 0.4 1.9). Elder subjects were eligible if they reported no history of ocular disease and had visual acuity of 20/30 or better. AMD subjects were included if their medical record included a history of laser treatment for neovascular membrane or a history of geographic atrophy but no additional ocular disease. Each subject had previous experience with RSVP reading. A median test on age indicates that the groups were different (Chi square = 30.95, p < 0.0001) with a Tukey-Kramer HSD (p = 0.05) indicating that the student group was different from each of the other groups but that the age of the elder and AMD groups were not statistically different. Log MAR visual acuity was also different between groups (Medial test Chi square = 47.33, p < 0.0001, however a Tukey Kramer HSD (p = 0.05) indicated that the visual acuity was not statistically different between elder and student participants. The procedures used in this study conformed to the tenets of the Declaration of Helsinki. All subjects read (or had read to them) and then signed an informed consent form before testing began and all study protocols were approved by the Institutional Review Board on human research at the State University of New York, State College of Optometry.
Timing for word duration
Word duration was controlled by six parameters specified in the software. A general timing parameter set the base starting rate at approximately 15, 30, 60 120 or 300 words per minute (wpm).( The base rate changed relative to each participants performance increasing by a factor of 2.0^0.25 for the next sentence after a sentence was read perfectly or decreasing by the same factor on the following sentence when a mistake was made. This staircase procedure continued for four reversals in reading rate for each block of trials.
Word duration also changed between blocks of trials depending the value of one of four exponents in the following formula: Duration*(word length/5)^X. The exponent in the formula was set to 0.0, 0.25, 0.5 or 1.00. When the exponent was zero each word in the sentence appeared for the same duration regardless of the number of characters in the word. Figure 1 (word duration vs. characters per word) illustrates several characteristics about the timing of words specified by the formula for a base reading rate of approximately 300 words per minute (wpm). The horizontal square symbols at 0.2 second represent words when the exponent in the word duration formula is zero. Each word in a given sentence stays on the screen for 13 frames. When the exponent increases the word duration decreases for words with fewer than 5 characters and increases for words with more than 5 characters. For the exponent equal to 1.00, at this base reading rate, a single character word remains on the screen for 3 frames (0.045 seconds). Presentations this rapid during reading are not expected to lead to word recognition. ADDIN EN.CITE Rayner1981830Rayner, KInoff, A.WMorrison, R.ESlowiaczek, M.LBertera, J.H1981Masking of foveal and parafoveal vision during eye fixations in readingJ Experim Psychol: Human Percept7167-179(Rayner, et al., 1981) Blocks of trials were run with and without the minimum in order to test how this rapid presentation affected reading rate. The minimum specified for the 300 wpm base rate was set to 8 frames based on pilot data, and appears in Figure 1 as the solid circles connected by a line at 0.12 second. For the blocks of trials that included the minimum, word duration could not be less than the specified minimum duration regardless of word duration calculated by the formula. At the base reading rate illustrated in Figure 1, one and 2 character words for the exponent equal to 1.0, and single character words for the exponent equal to 0.5, were displayed for 8 frames during the blocks of trials that included minimum.
A minimum duration was used for half of the blocks of trials for each of the starting rates based on a simple proportion: 300/120 = 2.5; 8 frames * 2.5 = 20 frames for the 120 wpm starting rate. In the same fashion the minimum was set to 40 frames for 60 wpm, 80 frames for 30 wpm, and 160 frames for 15 wpm. Figure 2 illustrates how the effect of the minimum changes as a result of an increase in reading rate with the staircase procedure for a base reading rate of 15 wpm. Figure 2 is the inverse of Figure 1, plotting log rate in words per minute vs. the number of characters per word. Each word in a given sentence remained on the screen for the same amount of time (4 seconds or 268 frames) when the exponent in the word duration formula was zero. As was the case in the 300 wpm rate one and two character words for an exponent of 1.0 and single character words for and exponent of 0.5 were displayed for the minimum of 160 frames in the blocks of trials that specified a minimum. This converts to a maximum rate of 25 wpm (1.39 log wpm) for these words. For the blocks of trials where no minimum was specified a single character word appears at a rate equivalent to 75 wpm when the exponent was equal to one and the base rate was 15 wpm. When the first sentence in the block is read perfectly the base presentation rate increases for the next sentence to about 18 wpm, and increases to about 21 wpm on the third sentence if there were no errors. This condition is represented in Figure 2b, the base rate has increased and each word in the sentence would be affected depending on the exponent that was set for that block of trials. When the blocks included the minimum one to three character words were displayed for the duration set by the minimum rather than the word duration formula. If reading continued perfectly through two additional staircase changes the base rate increased to 30 wpm with single character words appearing at a rate equivalent to 150 wpm for blocks where no minimum was specified but were limited to the 25 wpm rate for blocks with the minimum included. This condition is represented in Figure 2c indicating that the base rate after four staircase increases was controlled by the minimum setting for all words with between one to six characters. The duration of longer words continued to be controlled by the word duration formula depending on the specified exponent.
Procedure
Text was controlled by a Macintosh computer and projected, with a 640 x 480 pixel LCD projector, onto a 4-foot by 3-foot rear-projection screen. The projection system was used in order to produce large enough letters to allow subjects with low visual acuity to sit at a comfortable distance from the screen while reading. Text was displayed as white characters on a dark background. Luminance of the text and background were measured with a Spectra Scan 650 photometer (940 cd/m2, 3.3 cd/m2) and had a Weber contrast of 0.99.
Sentences were constructed from frequently used English words such that every sentence had four or five words with a range between 8 to 12 characters.[Francis, 1982 #77]The remainder of the sentence used words from lists of words with 1 to 7 characters. Sentences ranged from 50 to 65 characters in length with an average word length of 6.1 characters (range: 4.7-7.8). No sentences were repeated within a session and text files were randomly assigned to timing conditions for each subject.
Subjects with normal vision read text that was 5 times larger than the size of threshold visual acuity.[Rubin, 1992 #74]Distance from the rear projection screen was determined for each CFL patient so that the angular size of the text matched the size that allowed them to read RSVP text at their fastest rate in a previous study.[Aquilante, 1999 #78]
The starting rate for all students and elder subjects with normal vision was 300 wpm. The starting rate was set for patients with CFL so that it was close to the maximum rate they achieved in a previous study. The slowest starting rate was 15 wpm. Subjects sat comfortably in an armchair and read binocularly with best correction for the reading distance.
Subjects read four blocks of trials for each of the exponent values with and without the minimum completing a total of 32 blocks of trials within a session. The presentation order of exponent blocks were randomized for with and without minimum blocks and presented in a different order for each subject. Half of the subjects within each group began the session reading blocks with the minimum the other half began reading without the minimum.
Room lights were off during the testing. The experimenter used a keystroke to present an audible prompt to cue the subject, and then started each trial with a mouse click. Subjects were instructed to read aloud as rapidly as possible and to repeat words exactly as they appeared. Most of the young and elder subjects who read a relatively rapid rates continued to read the sentence after the last word disappeared from the screen while a few waited for the display to stop prior to beginning the sentence. The experimenter saw each sentence in a persistent display on a second monitor and recorded whether the trial was read correctly or if a mistake was made by hitting a key on the keyboard. A trial was counted as correct when all of the words were repeated verbatim in the correct order.
The software calculated the geometric mean of the reading rate for the staircase reversal points by counting the number of words in the trial and the total duration of the sentence. The reading rate at the reversals and the geometric mean of each block of trials were stored in a session data file.
Data analysis
Geometric means were calculated for each block of trials at a given exponent and converted to log reading rates and log ratios (rate @ exponent = 0.0, 0.25, 0.5 and 1.0 minus the rate @ exponent = 0.0) for the with and without minimum conditions. After the log ratio transformation data were subjected to a Shapiro-Wilk test for a normal distribution which confirmed that parametric tests were appropriate (W = 0.98, p = 0.10). A repeated measures ANOVA was to examine the effects of group, exponent, and interaction between group and exponent. A repeated measures ANOVA was also used to compare the mean difference between log ratios and standard deviations for the log ratios between groups for exponent = 0.0 with minimum minus exponent = 0.0 without minimum. This comparison gives some indication about the repeatability of the within session reading rates. A Dunnetts test with the log ratio for the exponent = 0.0 condition set as the control was used to test for differences between the reading rates in each exponent condition.
RESULTS
The repeated measures ANOVA on log ratios confirms a significant effect of subject group (F(2,69) = 83.37, p < 0.0001), a significant effect of the word timing exponent (F = 116.20, p < 0.0001) and a significant interaction between the subject group and exponent (F = 37.31, p = 0.0005). Figure 3 gives the mean log ratios for each subject group at each of the timing conditions. The solid symbols give the mean log ratio for the blocks of trials that included the minimum duration while the open symbols give the mean for the blocks of trials that did not include a minimum. The means for the readers with normal vision fall on or below the horizontal line at zero indicating that they read more slowly when word duration varied with word length. That drop in reading rate was less for the blocks of trials that included the minimum. The mean log ratios for the group with AMD fall above the horizontal line at zero when the exponent was 0.25 and 0.5 but fall below the line when the exponent was zero. The graph suggests that the AMD group read more slowly at exponent 0.25 and 0.5 when the minimum was included.
Figure 4 compares the reading rates when the exponent was zero for the with and without minimum blocks of trials (log reading rate without the minimum duration minus the log reading rate with the minimum). If all of the data points fell along a horizontal line at zero repeatability would be perfect for the session. ADDIN EN.CITE Altman19832600Altman, DGBland, J.M1983Measurement in medicine: Analysis of method comparison studiesThe Statistician32307-317Repeat(Altman & Bland, 1983) Figure 4 illustrates that, for each subject group, data fall about equally above and below the theoretical zero line. The horizontal lines near zero in Figure 4 give the mean log ratio for each group. The mean for the exponent = 0.0 log ratio for the student group falls just below zero (-0.003). The mean log ratio for the elder group falls just above zero (0.0008) and the mean log ratio for the group with AMD falls below zero (0.01). These means were not statistically different (F(2,69) = 1.75, p = 0.18) but the standard deviation between groups is statistically different (F(2,69) = 6.26, p = 0.0032). A Tukey-Kramer HSD test (p = 0.05) confirmed that the variability between the exponent = 0.0 blocks was less for the students compared to the elder and AMD groups. The Tukey-Kramer test did not indicate that the elder and AMD groups differed even though the 95% confidence interval is largest for the readers with AMD.
Figure 5 gives log ratio data for the student and elder groups relative to their mean log ratio and 95% confidence interval for the exponent = 0.0 condition. Open symbols are log ratios for the with-minimum condition and solid symbols are for the no minimum condition. Most of the open symbols (with-minimum) fall near the mean difference for the zero condition while solid symbols fall farther from the line. A Dunnetts test (p = 0.5), with the exponent = 0.0 condition set as the control, indicates that student group read more slowly when the exponent was 0.5 and 1.0 for the no minimum condition. When the minimum was included only the exponent = 1.0 condition resulted in slower reading rates. The results were similar for the elder group except that in the no minimum group all of the exponent conditions were read more slowly than the zero condition. For the blocks where the minimum duration was included the exponent = 1.0 was the read more slowly but the conditions with the exponent = 0.25 and 0.5 were not statistically different that the exponent = 0.0 condition (Dunnetts, p = 0.05).
Figure 6 shows the log ratios for the group with AMD. For this graph the log ratio data are arranged along the abscissa from the slowest to the fastest readers. When no minimum was specified in the software a Dunnetts test (p= 0.5) indicates that reading rates were not different for the exponent = 0.0 and the exponent = 1.0 conditions. Reading rates were faster when the exponent was 0.25 and 0.5. For the blocks of trials where a minimum word duration was specified the AMD group read more rapidly for the blocks when the exponent was equal to 0.25 and 0.5 than for the exponent = 0.0 blocks. In figure 6b the reading rate data were sorted for each subject and the maximum rate for the no minimum (solid symbols) and the minimum (open symbols conditions are plotted. This group read blocks of trials more rapidly when there was no minimum specified in the software (Wilcoxon rank sums Chi square = 14.71, p < 0.0001).
DISCUSSION
This experiment suggests that at least part of the explanation for the drop in reading rates by subjects with normal vision can be explained by the fact that the very rapid exposures for short words resulted in errors and subsequently slower presentation rates. The drop in reading rate by all three groups of subjects when the exponent in the word duration formula was set to 1.0 cannot be explained fully by the fact that short words appeared to rapidly. Reading rates dropped in all blocks of trials when the exponent was set to 1.0 whether there was a minimum or not.
The average reading rate increase for the readers with AMD was about 50% in the no minimum condition when the fastest reading rate for each subject was compared to their reading rate when the word duration was constant. This is a somewhat larger increase than what we reported previously ADDIN EN.CITE Aquilante20013340Aquilante, KYager, DMorris, R.AKhmelnitsky, F2001Low vision patients with age-related maculopathy read RSVP faster when word duration varies according to word lengthOptom Vis Sci78290-296(Aquilante, et al., 2001) but is very close to the 47% average that was reported for the self-paced variant of RSVP. ADDIN EN.CITE Arditi1999620Arditi, A1999Elicited sequential presentation for low vision readingVision Res394412-4418(Arditi, 1999) The reading rate advantage for the no minimum variable word duration condition for the readers with AMD in this study ranged from a low of about 14% to a maximum of over 200%. When the minimum was included that rate advantage had a mean of 20% and a range between less that 1% to a maximum of about 60%. We dont know why this group of subjects had a greater reading rate increase compared to the previous study but performance for elders in general and elders with eye disease is more variable than it is for subjects with normal vision. ADDIN EN.CITE Johnson198770M.A JohnsonD Choy1987On the definition of age-related norms for visual function testingAppl Optics261449-1454(Johnson & Choy, 1987)
Because our sentences were constructed with frequently used English words it is possible that variable rate duration could still benefit readers with normal vision if the text was very difficult. ADDIN EN.CITE Ward1982800Ward, N.JJoula, J.F1982Reading with and without eye movements: Reply to Just, Carpenter and WoolleyJ Experim Psychol: Gen111239-241(Ward & Joula, 1982) The reading rate measures may be slightly underestimated for the normal vision group because they have to verbalize each word. However verbal and silent reading have been shown to be highly correlated. ADDIN EN.CITE Legge1985180Legge, G. E.Pelli, D. G.Rubin, G. S.Schleske, M. M.1985Psychophysics of reading I. Normal VisionVision Res25239-252RepeatRubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sx(Legge, Pelli, Rubin & Schleske, 1985; Rubin & Turano, 1992) The normal vision groups in our study read the RSVP text more slowly that subjects with normal vision from previous studies. ADDIN EN.CITE Rubin1992740Rubin, G. S.Turano, K1992Reading without saccadic eye movementsVision Res32895-902RepeatMacintosh HD:Desktop Folder:G3+EIKI/Oct14/4 and 8 \piucv sxChung1998160Chung, S.T.LMansfield, J.SLegge, G.E1998Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral visionVision Res382949-2962(Chung, Mansfield & Legge, 1998; Rubin & Turano, 1992) Factors that could have contributed to the relatively slow reading rates include the criteria that every word be repeated exactly and in the correct order, and that the sentences had a longer average word length.
In conclusion, presenting RSVP at a variable rate based on word length is a beneficial addition to this reading method for low vision patients who must read with peripheral retina. Although the minimum exposure duration reduced the reading rate decrement for the participants with normal vision there continued to be a drop in rate for the condition where the difference in exposure duration based on word length was greatest.
QUOTE EN.REFLIST Altman D, Bland JM.(1983).Measurement in medicine: Analysis of method comparison studies.The Statistician.32:307-317.
Aquilante K, Yager D, Morris RA, Khmelnitsky F.(2001).Low vision patients with age-related maculopathy read RSVP faster when word duration varies according to word length.Optom Vis Sci.78:290-296.
Arditi A.(1999).Elicited sequential presentation for low vision reading.Vision Res.39:4412-4418.
Bullimore MA, Bailey IL.(1995).Reading and eye movements in age-related maculopathy.Optom Vis Sci.72:125-138.
Chung STL, Mansfield JS, Legge GE.(1998).Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral vision.Vision Res.38:2949-2962.
Forster KI.(1970).Visual perception of rapidly presented word sequences of varying complexity.Percept Psychophys.8:215-222.
Johnson MA, Choy D.(1987).On the definition of age-related norms for visual function testing.Appl Optics.26:1449-1454.
Just MA, Carpenter PA.(1980).A theory of reading: From eye fixations to comprehension.Psychol Review.87:329-354.
Just MA, Carpenter PA, Woolley JD.(1982).Paradigms and processes in reading comprehension.J Experim Psychol: Gen.111:228-238.
Legge GE, Ahn SJ, Klitz TS, Luebker A.(1997).Psychophysics of reading--XVI. The visual span in normal and low vision.Vision Res.37:1999-2010.
Legge GE, Pelli DG, Rubin GS, Schleske MM.(1985).Psychophysics of reading I. Normal Vision.Vision Res.25:239-252.
Rayner K, Inoff AW, Morrison RE, Slowiaczek ML, Bertera JH.(1981).Masking of foveal and parafoveal vision during eye fixations in reading.J Experim Psychol: Human Percept.7:167-179.
Rubin GS, Turano K.(1992).Reading without saccadic eye movements.Vision Res.32:895-902.
Rubin GS, Turano K.(1994).Low vision reading with sequential word presentation.Vision Res.34:1723-1733.
Ward NJ, Joula JF.(1982).Reading with and without eye movements: Reply to Just, Carpenter and Woolley.J Experim Psychol: Gen.111:239-241.
Figure 1. Word duration in fractions of a second vs. the number of characters per word for a base reading rate of 300 wpm. Word duration was specified in the software according to the formula: Duration *(word length/5)^x (x = 0, 0.25, 0.5 or 1.0). When the exponent was 1.0 (open diamonds) a single character word was displayed for less that 0.05 sec while a 14-character word stayed on the screen for about 0.56 sec. When the exponent was 0.0 all words stayed on the screen for the same amount of time. Half of the blocks of trials included a minimum duration (line with solid circles) that set a limit for duration in those blocks.
Figure 2. Duration converted to log reading rate vs. characters per word for a base rate of 15 wpm. a). Initial condition with reading rate indicated for each word length relative to the exponent in the word duration formula. The solid line gives the minimum duration (equivalent 25 wpm reading rate) for the with-minimum blocks of trials. b). Staircase specified increase in base rate if no errors were made after two sentences were read. The minimum does not change. c). Base duration after 4 staircase increases. In blocks of trials that include the minimum even 5 and 6 character words appear at a rate based on the minimum rather than the word duration formula.
Figure 3. Mean log ratio for each of the reading subject groups vs. the exponent in the word duration formula. Solid symbols give the log ratio for the with minimum duration conditions. The open symbols are the log ratio for the no minimum condition.
Figure 4. Log difference between reading rates for blocks of trials when the word duration formula was set to 0.0 (word duration constant) for each of the subject groups. The centerline gives the mean difference between exponent zero blocks. The upper and lower lines define the 95% confidence interval.
Figure 5. Log ratios (log wpm @ x = 0.25, 0.5 or 1.0 log wpm @ x = 0) for with minimum (open symbols) and without minimum (closed symbols) plotted on the mean difference and confidence interval for the zero exponent blocks a) student and b) elder subjects.
Figure 6. Log ratios for minimum and no minimum blocks at a) each exponent value and b) for the fastest reading rate by each subject with AMD. The centerline is the mean difference between exponent = 0.0 blocks with upper and lower horizontal lines defining the 95% confidence interval for that condition.
( For RSVP reading a single word cannot be displayed on the screen for a fraction of the displays frame rate. With our 67 Hz system a word displayed for 13 frames would stay on the screen for about 0.19 second translating to 5.15 words per second or 309 wpm.
PAGE 910
PAGE 2
M
N O f g A B F G @ A ?" @" X% Y% ~% % &