Design

Design principles for reading-acuity charts and their implementation in the MNREAD Charts

OVERVIEW

a) Definition of print size The characteristic of the print used to define print size should be clearly specified. For Latin alphabets the height of a lowercase letter such as 'x' or 'o' should be used. For non-latin alphabets the height of a common, well-known character should be used, one that has an unambiguous height. The goal is to design continuous-text reading-acuity charts for normal and low vision. The text passages should closely resemble 'normal everyday reading', and should have very simple linguistic content. The visual characteristics of the chart should conform with design principles for modern visual-acuity charts. In the following list, we state generic design principles. Where appropriate, these statements are followed by a * giving the English MNREAD implementation.

1. PRINT SIZE

a) Definition of print size

The characteristic of the print used to define print size should be clearly specified. For Latin alphabets the height of a lowercase letter such as 'x' or 'o' should be used. For non-latin alphabets the height of a common, well-known character should be used, one that has an unambiguous height.

* MNREAD uses 'x' height.

b) Range of print sizes and "Standard Reading Distance":

The smallest print size should be smaller than the acuity limit of people with normal vision when viewed from the standard reading distance. The largest print size should be as great as practical, to accommodate the widest range of low-vision subjects.

* MNREAD has print sizes from logMAR -.5 to 1.3 (corresponding to Snellen 20/6.3 to 20/400), for a standard reading distance of 40 cm.

 

c) Print size steps

The print size steps between successive sentences should follow a logarithmic progression.

* MNREAD uses 0.1 logMAR steps.

d) Terminology for Print size:

Print size should be specified on the chart in LogMAR, and in any other well-defined units that are important for the likely users. Jaeger and point size should be avoided because no unambiguous definition is universally accepted for them.

* MNREAD uses Snellen and M notation in addition to logMAR.

2. TYPEFACE

a) Choice of font

The font should be similar to fonts that are commonly used for newspapers or other everyday printed material in the locale of the chart.

* MNREAD uses Times Roman.

 

b) Character and Line Spacing

The character-to-character spacing, and line-to-line spacing should match that used in typical everyday print.

c) Special characters, ligatures, accents:

Accents on letters should be used where they naturally occur in text. (Intentionally avoiding accented letters might unduly distort the representativeness of the text.) Special characters (ie '$') or ligatures (ie Æ in Æsop, or fi in Mansfield) should be avoided where possible.

d) Typographical scaling

While it is conventional typographical practice for proportionally broader and bolder letters to be used for smaller print sizes, this should be avoided for a reading-acuity chart. Instead, every effort should be used to simply scale the size of the characters.

e) Rendering Resolution:

Care must be taken so that the smallest letters on the chart are rendered with sufficient resolution so that coarse sampling does not compromise visual acuity assessment.

* The MNREAD charts produced by the Lighthouse have a resolution of 3600 dots/inch. For the bottom line on the chart (logMAR = -.5, Snellen = 20/6.3), this corresponds to about 20 dots per x-height. The MNREAD charts were produced with a photo typesetting method.

3. SENTENCE LAYOUT

a) Sentence length

All sentences must be matched for length, measured as the number of characters. The character count includes the spaces between words, the at the the end of each line, and an implied period at the end of the sentence.

The selection of the number of characters per sentence represents a compromise between a) sufficient text to estimate reading performance; b) too much text to fit on the chart at the large-print end; and c) time required by low-vision subjects to read through the sentences.

* MNREAD uses 60 characters per sentence. This turns out to be convenient for scoring if we define a "standard-length word" to have 6 characters (a definition already extant in the reading literature.) A 60-character sentence consists of 10 standard-length words.

b) Sentence layout

All sentences should fit snugly into a rectangular box of fixed aspect ratio. The size scaling of the sentences on successive lines is equivalent to scaling the size of this bounding box. This type of layout can be achieved as follows.

The sentences should be printed onto 3 or more lines of text. (It is important to have at least one line with text above and below it to include the impact of vertical crowding on reading.) The lines of text should be left and right justified so that each sentence fits into the bounding rectangle. For proportionally spaced fonts, to allow for complete justification white space can be added or removed to the space between words. However, to ensure uniformity of word-to-word spacing the total amount of padding should not exceed 0.5 mean character widths. For fixed-width fonts no padding is allowed.

* MNREAD uses 60 characters on 3 lines. Because Times Roman is a proportional font, the spatial constraints are sometimes met by sentences that have unequal numbers of characters on the 3 lines (e.g.., 21, 19, 20). A major challenge in constructing MNREAD sentences is to simultaneously meet the 60-character constraint and the bounding-box spatial constraint.

4. SENTENCE COMPOSITION

a) Content

The sentences should be simple declarative sentences. The sentences should be independent in their semantic content. There should be no continuity in story line or theme between the sentences.

b) vocabulary

The vocabulary should be selected from high-frequency words in reading material for 8-year-old children (3rd grade students in the United States). Word-frequency lists may be available to guide the choice of vocabulary. * MNREAD was based on a restricted vocabulary of this type. The majority of the words were in the 1000 most frequent words in third grade school books.

c) Dialect/cross cultural considerations

Care should be taken to avoid use of words with regional spellings (such as American "color" vs. British "colour"), or with regional meanings (e.g., American pants vs. British trousers). This is especially important for languages that are spoken in several countries.

* MNREAD is designed to comply with both American and British English.

d) Punctuation

Use sentences with no punctuation.

5. SENTENCE SETS

a) How many sets of sentences? There should be at least two versions of the chart, for left and right eye testing, or pre- and post treatment testing. Additional versions may be useful for applications in which the same subject is tested repeatedly.

* MNREAD has two versions. We have an additional pool of sentences for research purposes. The MNREAD is printed as both a black-on-white chart, and a white-on-black chart.

6. PHYSICAL PROPERTIES OF THE CHARTS

a) Contrast The text on the chart should have high contrast (> 80% Michelson definition).

* The MNREAD charts have contrast exceeding 85%.

b) Luminance

The white background of the chart should be uniformly lit without shadows to a luminance of about 100 cd/m2, usually achievable using ordinary room lighting or desk lamps.

c) Surface

The chart should have a matte surface, avoiding specular reflections.