Letter recognition at low contrast levels: effects of letter size

The Effects of Letter Design Features and Aging on Legibility

We examined how letter legibility is affected by letter design features (size, contrast, font-type, and complexity), and how this changes with age (by comparing younger [ N = 54] and older participants [ N = 54]). Types of letters in the study were Japanese characters, Latin alphabets, and digits. Participants verbally reported the letter presented on a Cathode Ray Tube (CRT) display in natural viewing conditions. The legible-contrast threshold was calculated as the contrast boundary of 50% correct answer. We found that letter complexity (i.e., number of strokes) did not affect legible contrast threshold, except for small (<0.35°) letters. Moreover, letter size even affected the threshold with a function similar but not identical to contrast sensitivity function. This supports the previous hypothesis proposed by Majaj et al. that optimal spatial frequency band exists for letter identification and shifts with its size. Another novel finding was the effect of aging on the legible-contrast threshold and how it related to other letter features. Constant aging effect on legible-contrast was found across letter sizes. Among older adults, the legible-contrast thresholds were 1.2 to 2 times those of younger adults. Moreover, the aging effects were greater for positive-contrast than for negative.

Normal values for a clinical test of letter-recognition contrast thresholds

PURPOSE

To investigate the contrast thresholds (CTs) in normal subjects using a high-luminance, letter-recognition task under clinically relevant testing conditions.

SETTING

Texas Tech University Health Sciences System, Lubbock, Texas, USA.

METHODS

Sixty normal subjects aged 20 to 49 years with a best corrected visual acuity of 20/20 or better in both eyes participated. M & S Technologies software was used to display black-on-white Sloan letters at contrast levels of 25%, 20%, 15%, 12%, and 10% through 1% in 1% decrements. The effects of age, sex, optotype size, eye dominance, ambient illumination level (bright = 625 - 630 lux; dim = <3 lux), and direction of approach to threshold were analyzed using a multivariate, ordinary, least-squares analysis.

RESULTS

Age and sex did not influence CTs. Ascending versus descending testing was not statistically significant (P>.5). The effects of room illumination and eye dominance were significant (P<.01). Significant differences were found between 20/30 and 20/50, 20/30 and 20/70, and 20/50 and 20/70 optotype sizes (P<.01 for all comparisons).

CONCLUSIONS

A commercially available, computer-based test of CTs was easy to administer and apparently easy for inexperienced subjects to perform. The results suggest criteria for detecting visual problems concerned with familiar but complex spatial-image shapes. This information might be used to assess the effects of treatments such as laser refractive surgery on recognition contrast. Further study is warranted.

Invariance of the psychometric function for character recognition across the visual field

The psychometric function for recognition of singly presented digits as a function of digit contrast was measured at 2 degrees steps across the horizontal meridian of the visual field, under monocular and binocular viewing conditions. A maximum-likelihood staircase procedure was used in a 10-alternative forced-choice recognition paradigm to gather the data Both the Weibull and the logistic psychometric functions provide excellent fits to the observed data. The slopes of these functions at their point of inflection ranged from 4.0 to 5.0 proportion-correct/log10-unit contrast, for both monocular and binocular viewing and for all loci in the visual field. These slope values correspond to short-term measurements (around 30 trials, or 1 min) and do not include performance variations of longer duration; the latter are estimated to increase slope by a factor of about 1.5. A single psychometric function shape, centered around a threshold value, therefore describes recognition performance at all retinal loci and binocularity. An empirical comparison of slope results across the literature shows that the function's slope is about twice that reported for a number of detection tasks. The comparison of recognition contrast thresholds, percentage correct values, and other performance measures across studies requires the knowledge of the psychometric function's slope, and our results thus provide a firm basis for the study of low-contrast character recognition.

Visual acuity and contrast sensitivity for individual Sloan letters

Sloan letter optotypes are used frequently to evaluate visual impairment, and scoring procedures have been developed that are based on the numbers of letters that are identified correctly. However, previous studies have presented conflicting evidence regarding the relative identifiability of the individual Sloan letters. To investigate this issue further, we measured psychometric functions for the identification of each of the 10 Sloan letters, with individual letters presented in random order on the gray-scale display of a Macintosh computer-based testing system. Data were obtained from three visually normal subjects under each of three conditions: (1) as a function of log contrast at a relatively large letter size; (2) as a function of log contrast at a letter size near the acuity limit; and (3) as a function of log MAR (minimum angle of resolution) at maximum letter contrast. Estimates of threshold log contrast and threshold log MAR were derived from best-fitting Weibull functions. Threshold log contrast for small letters showed the greatest interletter variability. There was relatively little interletter variability in either threshold log contrast for large letters or threshold log MAR for high-contrast letters. However, due to the relatively steep psychometric functions under these latter two conditions, the different Sloan letters had considerably different percent correct values near threshold. The overall pattern of results suggests that the contrast sensitivity functions for individual Sloan letters are displaced laterally along a log MAR axis, while their vertical positions are essentially equivalent.

Reading with magnifiers

The design of low vision aids for partially sighted people, such as magnifiers, is evaluated theoretically in order to identify meaningful directions of research for improving these aids. Both reading with and without a magnifier involves alternating sequences of locating and recognizing textual information. Little is known, however, about these processes in magnifier reading. On the basis of an extensive review of the literature, two topics are identified that stand out as being in need of experimental investigation: (1) the relationship between the (typo)graphical characteristics of printed text and the location and recognition of textual information, and (2) the interplay of location and recognition processes in magnifier reading, including the role of non-visual factors, such as movements of the hand, trunk, head and eyes in this interplay. With regard to the first topic, it is expected that the visibility and, hence, the recognition of textual information by partially sighted people can be improved by matching the fundamental spatial frequencies of graphical structures with the spatial contrast sensitivity of partially sighted people. With regard to the second topic, it is argued that persistent problems in the design of the magnifiers, such as the optimal window size, can only be resolved by studying reading with a magnifier as a (multimodal) perceptual-motor activity.

What children see affects how they read

Children of mixed abilities were given three single-word lists to read, matched for linguistic complexity. The visual component of the task was made harder by reducing the print size with each new list. The reading errors made by children who did and who did not have a visual impairment were compared. The visually impaired children's pattern of reading errors changed as their vision was stressed by the reduction in print size; their errors became non-words (neologisms). This finding suggests a link between the efficiency of visual processing and the accuracy of reading of these children.