{Towards evidence-based typography: Experiment design} {Boris Veytsman and Leyla Akhmadeeva} {Typography is both a science and an art with several hundred years of history\Dash or, if we count its ancestor, calligraphy, with several thousand years of history. A beginning typographer faces a large amount of knowledge and rules (see, for example, \cite{bringhurst2004elements}): for example, that serifed fonts improve readability of body texts, while sans serif is good for advertising and posters; we know the optimal number of words per line and the lines per page, etc. Some of these rules are {\ae}sthetic ones, while some are purported reflect the neurophysiology of reading. With respect to the latter ones, we can ask, how do we know what we know? The situation here may resemble the history of medical science (and art!). Centuries of practical medicine resulted in a vast number of rules and methods of cure. Some of them we now know to be reasonable, like the use of diuretics for lowering blood pressure. Some, like purging, have much narrower applicability than was assumed in the past. Some rules turned out to be ineffective or even harmful, like the unrestrained use of bloodletting. The modern evidence based medicine tries to use a more scientific approach to these rules, putting empirical knowledge in a more formal framework~\cite{Elstein04:EvidenceBasedMedicine}. In this talk we discuss the applicability of an evidence-based approach to typography. While it is difficult to measure the beauty of the book page, we can measure the readability and the understandability of the text and their dependence on the fonts, type area dimensions and other typographic parameters. This area has been actively developing in the last decade. The modern studies question the widespread notions of the classical typography like the use of serifed fonts~\cite{Arditi05, Bernard03:SerifSansSerif, Russell-Minda07:LegibilityLowVision}, the mix of minuscule and majuscule letters in body texts~\cite{Sheedy05:LegibilityMiniscules, Arditi07:LegibilityCase}, text layout~\cite{Wendt82:LegibilityTextBooks, Santos06:LegibilityExams} and other factors~\cite{Lewis89, Subbaram04:Legibility, Coronel-Beltran10:Optics}. This research was stimulated by the challenges presented by new technologies~\cite{Bernard03:SerifSansSerif, Dyson04:LegibilityScreen, Lee08:LegibilityElectronicPapers, Hasegawa08:ReadabilityMobile, Shen09:LegibilityDisplays}, the use of type in messages and signage~\cite{Wang03:MessageSigns, Garvey04:ParkGuides, Carlson05:LegibilityRailwaySigns, Ullman05:LegibilityRailways, Gabbard06:LegibilityAugmentedReality} and special situations like texts for low vision readers~\cite{Russell-Minda07:LegibilityLowVision, Arditi04:LowVision, Arditi07:LegibilityCase}, drug information leaflets and other medical data~\cite{Mackey09:FoodInfoLegibility, Chubaty09:SeniorsLegibility, Bix03:DrugLabelsLegibility}. An overwhelming majority of published studies deals with English texts, while there are some works on Arabic~\cite{Al-Harkan05:LegibilityArabic}, Chinese~\cite{Huang09:LegibilityChinese, Cai01:LigibilityChinese}, Japanese~\cite{Hasegawa08:ReadabilityMobile, Ayama07:LegibilityJapanese} and Korean~\cite{Kong11:KoreanLegibility} typography. There was no comparable research on Cyrillic scripts and text perception by Russian readers. Our group works on a a large scale study of the neurophysiology of reading for Russian subjects. We plan to collect a database of readability and understandability as dependent on typographic parameters for Cyrillic texts, In this talk we provide the literature review and discuss the setup of the experiments. This is a preliminary report of what is envisioned to be an ongoing project. At this stage we need the input of the practitioners of typography and would be grateful for advice and suggestions. } ,-------------------. | BIBTEX ENTRIES | `-------------------' @book{bringhurst2004elements, address = {Vancouver, BC, Canada}, author = {Bringhurst, R.}, publisher = {Hartley \& Marks, Publishers}, title = {The Elements of Typographic Style}, year = {2004}, isbn = {9780881792065}, url = {http://books.google.com/books?id=940sAAAAYAAJ}, } @article{Elstein04:EvidenceBasedMedicine, address = {VIADUKSTRASSE 40-44, PO BOX 133, CH-4010 BASEL, SWITZERLAND}, author = {Elstein, A. S.}, journal = {Inflamm. Res.}, month = {August}, number = {Suppl. 2}, pages = {S184--S189}, publisher = {BIRKHAUSER VERLAG AG}, type = {Article}, title = {On the Origins and Development of Evidence-Based Medicine and Medical Decision Making}, volume = {53}, year = {2004}, abstract = {The aims of this paper are to identify the issues and forces that were the impetus for two recent developments in academic medicine, evidence-based medicine (EBM) and medical decision making (MDM); to make explicit their underlying similarities and differences; and to relate them to the fates of these innovations. Both developments respond to concerns about practice variation; the rapid growth of medical technology, leading to a proliferation of diagnostic and treatment options; the patient empowerment movement; and psychological research that raised questions about the quality of human judgment and decision making. Their commonalities include: use of Bayesian principles in diagnostic reasoning, and the common structure embedded in an answerable clinical question and a decision tree. Major differences include: emphasis on knowledge or judgment as the fundamental problem; the status of formal models and utility assessment; and the spirit and tone of the innovation. These differences have led to broader acceptance of EBM within academic medicine, while decision analysis, the fundamental tool of MDM, has been less welcomed in clinical circles and has found its place in guideline development, cost-effectiveness analysis, and health policy.}, doi = {10.1007/s00011-004-0357-2}, issn = {1023-3830}, language = {English}, } @article{Arditi05, author = {Arditi, A. and Cho, J.}, journal = {Vision Res.}, number = {23}, pages = {2926--2933}, title = {Serifs and Font Legibility}, volume = {45}, year = {2005}, } @article{Bernard03:SerifSansSerif, author = {Bernard, M. L. and Chaparro, B. S. and Mills, M. M. and Halcomb, C.G.}, journal = {Int. J. Hum.-Comput. Stud.}, month = {December}, number = {6}, pages = {823--835}, title = {Comparing the Effects of Text Size and Format on The Readibility of Computer-Displayed {T}imes {N}ew {R}oman and {A}rial Text}, volume = {59}, year = {2003}, abstract = {Times New Roman and Arial typefaces in 10- and 12-point, dot-matrix and anti-aliased format conditions were compared for readability (accuracy, reading speed, and accuracy/ reading speed), as well as perceptions of typeface legibility, sharpness, ease of reading, and general preference. In assessing readability, the 10-point anti-aliased Arial typeface was read slower than the other type conditions. Examining perceptions of typeface legibility, sharpness, and ease of reading detected significant effects for typeface, size, and format. Overall, the 12-point dot-matrix Arial typeface was preferred to the other typefaces. Recommendations for appropriate typeface combinations for computer-displayed text are discussed. (C) 2003 Elsevier Ltd. All rights reserved.}, doi = {10.1016/S1071-5819(03)00121-6}, issn = {1071-5819}, } @article{Russell-Minda07:LegibilityLowVision, author = {Russell-Minda, Elizabeth and Jutai, Jeffrey W. and Strong, J. Graham and Campbell, Kent A. and Gold, Deborah and Pretty, Lisa and Wilmot, Lesley}, journal = {J. Vis. Impair. Blind.}, month = {July}, number = {7}, pages = {402--415}, title = {The Legibility of Typefaces for Readers with Low Vision: A Research Review}, volume = {101}, year = {2007}, abstract = {This article presents a systematic review of the research evidence on the effects of the characteristics of typefaces on the legibility of text for adult readers with low vision. The review revealed that research has not produced consistent findings and thus that there is a need to develop standards and guidelines that are informed by evidence.}, issn = {0145-482X}, } @article{Sheedy05:LegibilityMiniscules, author = {Sheedy, J. E. and Subbaram, M. V. and Zimmerman, A. B. and Hayes, J. R.}, journal = {Hum. Factors}, month = {Winter}, number = {4}, pages = {797--815}, title = {Text Legibility and the Letter Superiority Effect}, volume = {47}, year = {2005}, abstract = {Effects of font design and electronic display parameters upon text legibility were determined using a threshold size method. Participants' visual acuity (inverse of the minimum detection size, representing the threshold legibility for each condition) was measured using upper- and lowercase letters and lowercase words in combinations of 6 fonts, 3 font-smoothing modes, 4 font sizes, 10 pixel heights, and 4 stroke widths. Individual lowercase letters were 10\% to 20\% more legible than lowercase words (i.e., lowercase words must be 10\%-20\% larger to have the same threshold legibility). This letter superiority effect suggests that individual letters play a large role and word shape plays a smaller role, if any, in word identification at threshold. Pixel height, font, stroke width, and font smoothing had significant main effects on threshold legibility. Optimal legibility was attained at 9 pixels (10 points). Verdana and Arial were the most legible fonts; Times New Roman and Franklin were least legible. Subpixel rendering (ClearType (TM)) improved threshold legibility for some fonts and, in combination with Verdana, was the most legible condition. Increased stroke width (bold) improved threshold legibility but only at the thinnest width tested. Potential applications of this research include optimization of font design for legibility and readability.}, issn = {0018-7208}, } @article{Arditi07:LegibilityCase, author = {Arditi, Aries and Cho, Jianna}, journal = {Vision Res.}, month = {September}, number = {19}, pages = {2499--2505}, title = {Letter Case and Text Legibility in Normal and Low Vision}, volume = {47}, year = {2007}, abstract = {It is thought by cognitive scientists and typographers alike, that lower-case text is more legible than upper-case. Yet lower-case letters are, on average, smaller in height and width than upper-case characters, which suggests an upper-case advantage. Using a single unaltered font and all upper-, all lower-, and mixed-case text, we assessed size thresholds for words and random strings, and reading speeds for text with normal and visually impaired participants. Lower-case thresholds were roughly 0.1 log unit higher than upper. Reading speeds were higher for upper- than for mixed-case text at sizes twice acuity size; at larger sizes, the upper-case advantage disappeared. Results suggest that upper-case is more legible than the other case styles, especially for visually-impaired readers, because smaller letter sizes can be used than with the other case styles, with no diminution of legibility. (C) 2007 Elsevier Ltd. All rights reserved.}, doi = {10.1016/j.visres.2007.06.010}, issn = {0042-6989}, } @article{Wendt82:LegibilityTextBooks, author = {Wendt, D.}, journal = {Vis. Lang.}, number = {1}, pages = {88--93}, title = {Improving the Legibility of Textbooks---Effects of Wording and Typographic Design}, volume = {16}, year = {1982}, issn = {0022-2224}, } @article{Santos06:LegibilityExams, author = {dos Santos Lonsdale, Maria and Dyson, Mary C. and Reynolds, Linda}, journal = {J. Res. Read.}, month = {November}, number = {4}, pages = {433--453}, title = {Reading in Examination-Type Situations: The Effects of Text Layout on Performance}, volume = {29}, year = {2006}, abstract = {Examinations are conventionally used to measure candidates' achievement in a limited time period. However, the influence of text layout on performance may compromise the construct validity of the examination. An experimental study looked at the effects of the text layout on the speed and accuracy of a reading task in an examination-type situation. A survey of the reading strategies used in examinations was conducted to help in defining the reading context in which text layout may have an effect. A set of guidelines was also derived from research on typographic features of text and these were used to select three text layouts (intended to be more or less legible) from the layouts used in English language reading examinations. Results of the experiment showed that task time was significantly shorter and the number of correct answers per second was significantly higher with the layout conforming to legibility guidelines. Participants' judgements indicated that this layout was also the easiest in which to find answers and the most attractive. The main conclusion of the study is that text layout affects performance in a task that involves reading text to search for specific information in order to answer questions on it under time pressure. Consequently, the construct validity of examinations may be compromised by confounding legibility with reading skills.}, doi = {10.1111/j.1467-9817.2006.00317.x}, issn = {0141-0423}, } @article{Lewis89, author = {Lewis, C. and Walker, P.}, journal = {British J. Psychol.}, pages = {241--257}, title = {Typographic Influences on Reading}, volume = {80}, year = {1989}, } @article{Subbaram04:Legibility, author = {Subbaram, M. V. and Sheedy, J. E. and Hayes, J. R.}, journal = {Invest. Ophthalmol. Vis. Sci.}, month = {April}, number = {Suppl. 2}, pages = {4354}, title = {Effects of Font Type, Smoothing, and Stroke Width on Legibility}, volume = {45}, year = {2004}, issn = {0146-0404}, } @article{Coronel-Beltran10:Optics, author = {Coronel-Beltran, Angel and Alvarez-Borrego, Josue}, journal = {J. Modern Optics}, number = {1}, pages = {58--64}, title = {Comparative Analysis Between Different Font Types and Letter Styles Using a Nonlinear Invariant Digital Correlation}, volume = {57}, year = {2010}, abstract = {We present, in this paper, a comparative analysis of the letters in Times New Roman (TNR), Courier New (CN) and Arial (Ar) font types in plain and italic style and the effects of five foreground/background color combinations using an invariant digital correlation system with a nonlinear filter with k = 0.3. The evaluation of the output plane with this filter is given by the peak-to-correlation energy (PCE) metric. The results show that the letters in TNR font have a better mean PCE value when compared with the CN and Ar fonts. This result is in agreement with some studies on text legibility and for readability where the reaction time (RT) of some participant individuals reading a text is measured. We conclude that the PCE metric is proportional to 1/RT.}, doi = {10.1080/09500340903511695}, issn = {0950-0340}, } @article{Dyson04:LegibilityScreen, author = {Dyson, M. C.}, journal = {Behav. Inf. Technol.}, month = {November-December}, number = {6}, pages = {377--393}, title = {How Physical Text Layout Affects Reading From Screen}, volume = {23}, year = {2004}, abstract = {The primary objective of this paper is to critically evaluate empirical research on some variables relating to the configuration of text on screen to consolidate our current knowledge in these areas. The text layout variables are line length, columns, window size and interlinear spacing, with an emphasis on line length due to the larger number of studies related to this variable. Methodological issues arising from individual studies and from comparisons among studies are identified. A synthesis of results is offered which provides alternative interpretations of some findings and identifies the number of characters per line as the critical variable in looking at line length. Further studies are needed to explore the interactions between characters per line and eye movements, scrolling movements, reading patterns and familiarity with formats.}, doi = {10.1080/01449290410001715714}, issn = {0144-929X}, } @article{Lee08:LegibilityElectronicPapers, author = {Lee, Der-Song and Shieh, Kong-King and Jeng, Shie-Chang and Shen, I-Hsuan}, journal = {Displays}, month = {January}, number = {1}, pages = {10--17}, title = {Effect of Character Size and Lighting on Legibility of Electronic Papers}, volume = {29}, year = {2008}, abstract = {Effects of character size under ambient illuminances and light sources on legibility of electronic paper displays (electrophoretic display and cholesteric liquid crystal display) were studied and compared with paper. Sixty subjects participated in a letter-search task in the experiment. The results showed that search speed depends on the illuminance but not light source. Search speed increased as the illuminance increase from 300, 700 to 1500 1x. Search speed also increased with the increase of character size, from character height of 1.4 min (9.6 min visual angle), 2.2 min (15.1 min visual angle) to 3.3 min (22.7 min visual angle), and the increase leveled off at 4.3 min (29.6 min visual angle). The effect of character size on accuracy was also significant. Accuracy increased with the increase of character size. However, the effect of illuminance and light source on accuracy was not statistically significant. Based on the results of this study, it seems that E-paper displays may need greater illumination (700 1x or higher), greater character size (3.3 mm or 22 min of visual angle). (C) 2007 Elsevier B.V. All rights reserved.}, doi = {10.1016/j.displa.2007.06.007}, issn = {0141-9382}, } @article{Hasegawa08:ReadabilityMobile, author = {Hasegawa, Satoshi and Fujikake, Kazuhiro and Omori, Masako and Miyao, Masaru}, journal = {Int. J. Occup. Saf. Ergon.}, number = {3}, pages = {293--304}, title = {Readability of Characters On Mobile Phone Liquid Crystal Displays}, volume = {14}, year = {2008}, abstract = {Subject performance in reading characters on mobile phone liquid crystal displays was researched by using (a) English sentences with 3 or 4 different sizes of characters on 2 types of displays with different resolutions (n = 23; age = 31.0 +/- 6.0 years), (b) Japanese characters in 3 different sizes and 2 types of font (n = 98; age = 44.5 +/- 18.5 Years), and (c) Japanese characters vertically enlarged 1-4 times (n = 120; age = 46.9 +/- 18.6 years). Subjective; evaluation, reading speed, number of reading errors, and viewing distance were recorded. Readability was higher with higher resolution displays, and with Gothic than with Mincho font in Japanese. Young subjects shortened the viewing distance as characters became smaller, whereas elderly subjects increased the viewing distance irrespective of the size of characters. Characters of 3-5 min are appropriate for the young but inadequate for the elderly. Readability of Japanese characters improved when they were vertically enlarged to approximately twice the width.}, issn = {1080-3548}, } @article{Shen09:LegibilityDisplays, author = {Shen, I-Hsuan and Shieh, Kong-King and Chao, Chih-Yu and Lee, Der-Song}, journal = {Displays}, month = {April}, number = {2}, pages = {53--58}, title = {Lighting, Font Style, and Polarity on Visual Performance and Visual Fatigue with Electronic Paper Displays}, volume = {30}, year = {2009}, abstract = {Visual performance and visual fatigue of electronic paper displays (electrophoretic display and cholesteric liquid crystal display) under ambient illuminances and light sources were studied and compared with paper. Forty-eight participants participated in a character-search task in the experiment. The results showed that search speed depends on the illuminance but not light source. Search speed increased as illumination increased from 300 lx (45.6 sec), 700 lx (44.18 sec) to 1500 lx (43.24 sec). The effect of medium display and polarity on accuracy was also significant. Accuracy was greater for electrophoretic display and positive polarity. However, the effect of illuminance and light source on visual fatigue was not statistically significant. Based on the results of this study, it seems that E-paper displays may need greater illumination (700 lx or higher). (c) 2008 Elsevier B.V. All rights reserved.}, doi = {10.1016/j.displa.2008.12.001}, issn = {0141-9382}, } @article{Wang03:MessageSigns, author = {Wang, J. H. and Cao, Y.}, journal = {Int. J. Ind. Eng.-Theory Appl. Pract.}, month = {December}, number = {4}, pages = {339--344}, title = {A Human Factors Study on Message Design of Variable Message Sign}, volume = {10}, year = {2003}, abstract = {Variable message sign (VMS) is a key component of intelligent transportation system (ITS). This paper describes a human factors study focusing on the message design of VMS. It examined within-subject factors, message font size and color; and between-subject factors, subjects' age and gender; and their impacts on human/ITS interface. Experiments were conducted in a motorist performance laboratory where an environment of real video scenes integrated with computer-generated VMS was created to simulate a virtual driving experience. The study found that font color, subjects' age, and gender significantly affected the response time. Green font color resulted in the shortest response time. Older subjects responded faster but with lower accuracy. Female subjects also responded faster with lower accuracy. No meaningful correlations were found between subjects' response time and accuracy. These findings might help promote safe and efficient driving through effective and accurate delivery of information to motorists via VMS. Significance: Proper message design is critical for effective and accurate delivery of information to motorists. The human factors issues in VMS addressed herein can significantly affect the message legibility, and human/ITS interface.}, issn = {1072-4761}, } @incollection{Garvey04:ParkGuides, author = {Garvey, P. M. and Chirwa, K. N. and Meeker, D. T. and Pietrucha, M. T. and Zineddin, A. Z. and Ghebrial, R. S. and Montalbano, J.}, booktitle = {Traffic Control Devices, Visibility, and Rail-Highway Grade Crossings 2004}, number = {1862}, pages = {1--9}, series = {Transportation Research Record}, title = {New Font and Arrow for National Park Service Guide Signs}, year = {2004}, abstract = {Although highly legible, the National Park Service's (NPS) Clarendon font produces sign legends that can be 10\% to 20\% longer than those depicted in conventional sans serif fonts. Placing these wider signs on narrow park roads and in urban historic districts can be prohibitive. To address this problem, a project was initiated to create a new NPS Roman-style (i.e., serif) font that requires less horizontal sign space than Clarendon while improving sign readability and retaining Clarendon's unique ``signature quality.{''} The present study also evaluated a set of guide sign arrows to select the most legible for use on NPS guide signs. Three candidate typefaces were developed for daytime and nighttime field evaluation with 72 older and younger subjects. From the results of the evaluation, a fourth font was created and field tested with 12 additional subjects. Words created with the fourth font (NPS Roadway) were 5\% to 11.5\% shorter than those created with Clarendon. Further, subjects were able to read these words at 10.5\% longer average threshold legibility distances than the same words composed in Clarendon. The relative legibility of 12 candidate guide sign arrows was evaluated in an outdoor field study. Forty-eight subjects participated in the daytime, and 32 subjects viewed the arrows at night. There were statistically significant differences in legibility distance among the various arrow shapes. The arrow ultimately recommended for use on NPS road guide signs performed 18\% better than the FHWA ``standard arrow{''}.}, isbn = {0-309-09454-2}, issn = {0361-1981}, } @incollection{Carlson05:LegibilityRailwaySigns, author = {Carlson, P. J. and Holick, A.}, booktitle = {Traffic Control Devices, Visibility, and Rail-Highway Grade Crossings 2005}, crossref = {national2005traffic}, number = {1918}, pages = {26--34}, publisher = {National Academy Press}, series = {Transportation Research Record}, title = {Maximizing Legibility of Unlit Freeway Guide Signs With {C}learview Font and Combinations of Retroreflective Sheeting Materials}, year = {2005}, abstract = {The legibility of full-scale guide signs was evaluated in a controlled field study. The objective was to determine how the legibility of unlit guide signs could be maximized with combinations of font and retroreflective sheeting. Two fonts were tested: Series E (Modified) and Clearview 5WR, a modified version of Clearview 5W. Five combinations of retroreffective sheeting were also evaluated. The study included 30 subjects split into three age groups. Without informing the subjects, the researchers varied headlamp output to evaluate the legibility of various sign lighting levels. The results of the study clearly show that the most legible combination of font and retroreflective sheeting (of those tested) is the Clearview 5WR font and microprismatic legends on beaded, high-intensity backgrounds. Although the combination of microprismatic legends and backgrounds performed statistically equivalently to microprismatic legends on beaded, high-intensity backgrounds (in term of legibility distances), the research shows that the higher luminance contrast and lower costs associated with the latter combination make it a much more attractive application.}, isbn = {0-309-09391-0}, issn = {0361-1981}, url = {http://books.google.com/books?id=FA6ZAAAACAAJ}, } @incollection{Ullman05:LegibilityRailways, author = {Ullman, B. R. and Ullman, G. L. and Dudek, C. L. and Ramirez, E. A.}, booktitle = {Traffic Control Devices, Visibility, and Rail-Highway Grade Crossings 2005}, crossref = {national2005traffic}, number = {1918}, pages = {56--62}, publisher = {National Academy Press}, series = {Transportation Research Record}, title = {Legibility Distances of Smaller Letters in Changeable Message Signs with Light-Emitting Diodes}, year = {2005}, abstract = {This paper presents the results of a legibility study of 9-in. and 10.6-in. letters on changeable message signs (CMSs) with light-emitting diodes. The study, conducted in Dallas, Texas, consisted of 60 subjects, demographically balanced with respect to age, gender, and education. The subjects drove a test vehicle as they approached CMSs with one of the above letter heights. Study administrators recorded the distance from the sign at which the participant could correctly read a three-letter word. Data were recorded for three trials on each of the letter heights for each participant. Data were collected during daylight (sun overhead) and nighttime conditions. The 85th percentile legibility distance for the 9-in. letter height was 228ft for daytime conditions and 114ft for nighttime conditions. For the 10.6-in. letter height, the 85th percentile legibility distances were 324ft for daytime conditions and 203ft for nighttime conditions, respectively. The 85th percentile legibility distances for each letter height were used to estimate available viewing times under various approach speeds. These available viewing times dictate the units of information that can be presented on a CMS of a particular letter size.}, isbn = {0-309-09391-0}, issn = {0361-1981}, url = {http://books.google.com/books?id=FA6ZAAAACAAJ}, } @article{Gabbard06:LegibilityAugmentedReality, author = {Gabbard, J. L. and Swan, J. E. and Hix, D.}, journal = {Presence-Teleoper. Virtual Env.}, month = {February}, number = {1}, pages = {16--32}, title = {The Effects of Text Drawing Styles, Background Textures, and Natural Lighting on Text Legibility in Outdoor Augmented Reality}, volume = {15}, year = {2006}, abstract = {A challenge in presenting augmenting information in outdoor augmented reality (AR) settings lies in the broad range of uncontrollable environmental conditions that may be present, specifically large-scale fluctuations in natural lighting and wide variations in likely backgrounds or objects in the scene. In this paper, we motivate the need for research on the effects of text drawing styles, outdoor background textures, and natural lighting on user performance in outdoor AIR. We present a pilot study and a follow-on user-based study that examined the effects on user performance of outdoor background textures, changing outdoor illuminance values, and text drawing styles in a text identification task using an optical, see-through AR system. We report significant effects for all these variables, and discuss user interface design guidelines and ideas for future work.}, issn = {1054-7460}, } @article{Arditi04:LowVision, author = {Arditi, A.}, journal = {Ergonomics}, month = {April}, number = {5}, pages = {469--482}, title = {Adjustable Typography: an Approach to Enhancing Low Vision Text Accessibility}, volume = {47}, year = {2004}, abstract = {Millions of people have low vision, a disability condition caused by uncorrectable or partially correctable disorders of the eye. The primary goal of low vision rehabilitation is increasing access to printed material. This paper describes how adjustable typography, a computer graphic approach to enhancing text accessibility, can play a role in this process, by allowing visually-impaired users to customize fonts to maximize legibility according to their own visual needs. Prototype software and initial testing of the concept is described. The results show that visually-impaired users tend to produce a variety of very distinct fonts, and that the adjustment process results in greatly enhanced legibility. But this initial testing has not yet demonstrated increases in legibility over and above the legibility of highly legible standard fonts such as Times New Roman.}, doi = {10.1080/0014013031000085680}, issn = {0014-0139}, } @article{Mackey09:FoodInfoLegibility, author = {Mackey, Mary Alton and Metz, Marilyn}, journal = {Int. J. Consumer Studies}, month = {July}, number = {4}, pages = {369--381}, title = {Ease of Reading of Mandatory Information on {C}anadian Food Product Labels}, volume = {33}, year = {2009}, abstract = {Food product labels present individual product information, safety, nutrition, electronic inventory, container and environmental information, in various formats, languages and images. Some information is mandatory; much is promotional. The food label is an essential tool for regulators of safe food handling, nutrition policy and fair competition. Mandatory information on food labels in Canada is required to be presented in both English and French, readily discernable, prominently displayed and legible. This study examines the ease of finding and reading of mandatory label components on selected Canadian food products. A validated typographical scoring system assessed the lists of ingredients on a purposive sample of 100 food labels representing foods in all groups in Canada's Food Guide. Seven percent of the ingredient lists were easy to read; 26\% were difficult to read and 67\% were very difficult to read. Well-educated resourceful readers in consumer focus groups examined food labels for key elements that influence ease of finding and reading information. Focus groups and typographical scoring identified: colour contrast, case, print style, print size, space between the lines, reverse print, organization, justification, type of surface, hyphenation and print reproduction as factors that affect ease of reading. Print that curves around a container, lack of paragraphing or point form organization make reading difficult; text blocks at right angles to each other make comparisons difficult; separation of the nutrition facts table from the list of ingredients makes decision making tedious. Inadequate spacing between lines of print creates problems for readers of English and exacerbates problems for readers of French. Words placed over illustrations, busy backgrounds or watermarks increase reading difficulty. Hazard statements, instructions and storage information imbedded in other information without added space or appropriate heading is difficult to find and read. Canadian consumers echo consumers in 28 European countries who find label information difficult to find and to read and want clear guidelines/regulations on the placement and the typography of mandatory food label components}, doi = {10.1111/j.1470-6431.2009.00787.x}, issn = {1470-6423}, } @article{Chubaty09:SeniorsLegibility, author = {Chubaty, Adriana and Sadowski, Cheryl A. and Carrie, Anita G.}, journal = {Age \& Ageing}, month = {July}, number = {4}, pages = {441--447}, title = {Typeface Legibility of Patient Information Leaflets Intended for Community-Dwelling Seniors}, volume = {38}, year = {2009}, abstract = {Objective: to use current typeface guidelines to describe the design of health information leaflets. Design: this was a cross-sectional study of leaflets from pharmacies and seniors' clinics. Setting: community pharmacies, seniors' clinics in Edmonton, Canada. Methods: health information leaflets and hydrochlorthiazide information sheets were collected. The body of each was evaluated, based on guidelines (from Canada, UK and USA). Adherence to recommendations was assessed descriptively. Results: a total of 388 unique leaflets and 10 hydrochlorthiazide sheets were collected from 21 pharmacies and 3 clinics. Most leaflets were produced by pharmaceutical companies (42.8\%) and contained disease information (43.8\%). Only one-third of all leaflets used the minimum recommended point size (12 point), 18.6\% followed American guidelines for line spacing (1.5 lines), but 77.1\% had appropriate contrast. Conclusions: although guidelines are available, most leaflets did not meet recommendations. Improvements in the leaflet design should be considered to aid seniors in the uptake of information.}, doi = {10.1093/ageing/afp065}, issn = {0002-0729}, } @article{Bix03:DrugLabelsLegibility, author = {Bix, L and Lockhart, H. and Selke, S. and Cardoso, F. and Olejnik, M.}, journal = {Packag. Technol. Sci.}, month = {September-October}, number = {5}, pages = {199--207}, title = {Is {x}-Height a Better Indicator of Legibility Than Type Size For Drug Labels?}, volume = {16}, year = {2003}, abstract = {In 1999 the US Food and Drug Administration published a regulation in an attempt to ensure the legibility of OTC drugs, specifying, among other things, a minimum type size of 6 points. This is problematic because different typefaces of the same size vary widely in type heights and, presumably, legibility. We hypothesized that specifying a minimum x-height, the height of the lowercase x, would produce more consistent legibility than the minimum type size specified within the regulation. Twenty-six subjects viewed two groups of typefaces using the Lockhart Legibility Instrument to quantify legibility. The first group contained typefaces that were all 6 points, but, by nature of their design, varied greatly in their x-heights. The second group was made from the same set of typefaces, but these were manipulated so that their x-heights were equal to the average x-height of group 1. A likelihood ratio test indicated that the group that varied in x-height, group 1, produced significantly more variable results than the group with equal x-heights' group 2. This indicates that specifying a minimum type size may not be the best approach for producing consistent legibility. Copyright (C) 2003 John Wiley Sons, Ltd.}, issn = {0894-3214}, } @article{Al-Harkan05:LegibilityArabic, author = {Al-Harkan, I. M. and Ramadan, M. Z.}, journal = {Int. J. Ind. Ergon.}, month = {July}, number = {7}, pages = {652--664}, title = {Effects of Pixel Shape and Color, and Matrix Pixel Density of {A}rabic Digital Typeface on Characters' Legibility}, volume = {35}, year = {2005}, abstract = {Two experiments were conducted to determine: (1) the possibility of fitting Arabic characters on the common rectangular resolutions; (2) the best resolution for Arabic characters to be formed in dot matrix formats; and (3) the possibility of developing a legible low resolution for Arabic characters. The second experiment was a replication of the first one except that some of the Arabic characters that were hard to be identified by the subjects in the first experiment were modified for the second one. Recognition speed, correct response, and preference ranking were dependent variables in both experiments. The independent variables were matrix pixel density, pixel shape, and pixel color. The levels of the matrix pixel density were 5 x 7, 7 x 9, 8 x 8, 9 x 14, 16 x 16, and 32 x 32 resolutions. Square and circular pixels were pixel shape. The pixel colors were red and green. Twenty-four treatment combinations were randomly presented to all subjects, where within subject design was employed to analyze the collected data. In these experiments, fifteen and six subjects exposed to 100 slides of Arabic letters presented on video display terminal for each matrix format. The results indicated that readability was improved significantly with increasing matrix pixel density. The results revealed that Arabic letters were more readable in square matrices (e.g., 8 x 8) rather than rectangular matrices (e.g., 7 x 9) in terms of correct responses and performance subjective rating. In addition, readability was significantly improved with square pixel shape. The subjects rated green pixels as more comfortable than the red ones. Finally, this work offers engineers with a legible low-resolution of Arabic typeface, which is essential for designing electronic displays. Relevance to industry: Displayed Arabic typeface in visual displays that are ready made in USA or European countries is not well technically prepared. This work presents a digital typeface that will be professionally applicable to be implemented in different types of digital matrix resolutions. In addition, it offers the industrial engineers with a legible low-resolution of Arabic typeface. (c) 2005 Elsevier B.V. All rights reserved.}, doi = {10.1016/j.ergon.2005.01.009}, issn = {0169-8141}, } @article{Huang09:LegibilityChinese, author = {Huang, Ding-Long and Rau, Pei-Luen Patrick and Liu, Ying}, journal = {Int. J. Ind. Ergonomics}, month = {January}, number = {1}, pages = {81--89}, title = {Effects of Font Size, Display Resolution and Task Type on Reading {C}hinese Fonts From Mobile Devices}, volume = {39}, year = {2009}, abstract = {The objective of this study was to investigate the impacts of font size, display resolution and task type on the readability of Chinese fonts on mobile devices. In the experiment, participants conducted text reading tasks and Visual searching tasks. Both objective and subjective measures were used to determine the readability of Chinese fonts. Significant differences were found in participants' reading performance and preference for different font sizes and display resolutions. Subjective measures were found to be more sensitive than objective measures. Optimum font sizes that can provide good readability of Chinese characters for four different resolutions were elicited. Recommendations for enhancing readability of Chinese fonts on mobile devices were discussed. Relevance to industry: The results can be used in determining the optimal font sizes for presenting Chinese characters on mobile devices with different display resolutions. (C) 2008 Elsevier B.V. All rights reserved.}, doi = {10.1016/j.ergon.2008.09.004}, issn = {0169-8141}, } @article{Cai01:LigibilityChinese, author = {Cai, D. C. and Chi, C. F. and You, M. L.}, journal = {Int. J. Ind. Ergon.}, month = {January}, number = {1}, pages = {9--17}, title = {The Legibility Threshold of {C}hinese Characters in Three Type Styles}, volume = {27}, year = {2001}, abstract = {A recognition test was conducted on the most commonly used Chinese characters in the Ming, Kai, and Li styles, measuring the minimum visible size of each character in each style. A total of 120 college students participated in the experiment. Each character was initially displayed on a PC screen at its minimum size and enlarged gradually until the participant could recognize it. Analysis of variance showed that character style and number of strokes both have a significant impact on the legibility threshold. Ming is the most legible among the three styles, and Kai is also significantly more legible than the Li style. A regression equation for predicting the legibility threshold based on the number of strokes in a character indicates that an inclusion of an additional stroke in a character must be compensated for by an enlargement of the character size of approximately 0.25 min of are to maintain the same legibility.}, issn = {0169-8141}, } @article{Ayama07:LegibilityJapanese, author = {Ayama, Miyoshi and Ujike, Hiroyasu and Iwai, Wataru and Funakawa, Masami and Okajima, Katsunori}, journal = {Opt. Rev.}, month = {January-February}, number = {1}, pages = {48--56}, title = {Effects of Contrast and Character Size Upon Legibility of {J}apanese text Stimuli Presented on Visual Display Terminal}, volume = {14}, year = {2007}, abstract = {We examined effects of contrast and character size upon legibility of Japanese text stimuli presented on visual display terminal (VDT). In the experiment, three different character sizes were employed and the text stimulus was presented under a variety of conditions where contrast between the text and the background changed. Reading speed and the rate of readable characters were measured. Subjective rating for legibility was also evaluated. Results showed that legibility increases with luminance contrast for all character sizes examined here. A strong correlation was found between the subjective rating index and reading speed. (c) 2007 The Optical Society of Japan.}, issn = {1340-6000}, } @article{Kong11:KoreanLegibility, author = {Kong, Yong-Ku and Lee, Inseok and Jung, Myung-Chul and Song, Young-Woong}, journal = {Ergonomics}, number = {5}, pages = {453--465}, title = {The Effects of Age, Viewing Distance, Display Type, Font Type, Colour Contrast and Number of Syllables on the Legibility of {K}orean Characters}, volume = {54}, year = {2011}, abstract = {This study evaluated the effects of age (20s and 60s), viewing distance (50 cm, 200 cm), display type (paper, monitor), font type (Gothic, Ming), colour contrast (black letters on white background, white letters on black background) and number of syllables (one, two) on the legibility of Korean characters by using the four legibility measures (minimum letter size for 100\% correctness, maximum letter size for 0\% correctness, minimum letter size for the least discomfort and maximum letter size for the most discomfort). Ten subjects in each age group read the four letters presented on a slide (letter size varied from 80 pt to 2 pt). Subjects also subjectively rated the reading discomfort of the letters on a 4-point scale (1 = no discomfort, 4 = most discomfort). According to the ANOVA procedure, age, viewing distance and font type significantly affected the four dependent variables (p < 0.05), while the main effect of colour contrast was not statistically significant for any measures. Two-syllable letters had smaller letters than one-syllable letters in the two correctness measures. The younger group could see letter sizes two times smaller than the old group could and the viewing distance of 50 cm showed letters about three times smaller than those at a 200 cm viewing distance. The Gothic fonts were smaller than the Ming fonts. Monitors were smaller than paper for correctness and maximum letter size for the most discomfort. From a comparison of the results for correctness and discomfort, people generally preferred larger letter sizes to those that they could read. The findings of this study may provide basic information for setting a global standard of letter size or font type to improve the legibility of characters written in Korean.}, doi = {10.1080/00140139.2011.568635}, issn = {0014-0139}, } @book{national2005traffic, author = {{National Research Council (U. S.). Transportation Research Board and National Research Council (U. S.) Transportation Research Board Meeting}}, booktitle = {Traffic Control Devices, Visibility, and Rail-Highway Grade Crossings, 2005}, number = {1918}, publisher = {National Academy Press}, series = {Transportation research record}, title = {Traffic Control Devices, Visibility, and Rail-Highway Grade Crossings, 2005}, year = {2005}, isbn = {9780309093910}, url = {http://books.google.com/books?id=FA6ZAAAACAAJ}, }