Reading without saccadic eye movements

Does print size matter for reading? A review of findings from vision science and typography

The size and shape of printed symbols determine the legibility of text. In this paper, we focus on print size because of its crucial role in understanding reading performance and its significance in the history and contemporary practice of typography. We present evidence supporting the hypothesis that the distribution of print sizes in historical and contemporary publications falls within the psychophysically defined range of fluent print size--the range over which text can be read at maximum speed. The fluent range extends over a factor of 10 in angular print size (x-height) from approximately 0.2° to 2°. Assuming a standard reading distance of 40 cm (16 inches), the corresponding physical x-heights are 1.4 mm (4 points) and 14 mm (40 points). We provide new data on the distributions of print sizes in published books and newspapers and in typefounders' specimens, and consider factors influencing these distributions. We discuss theoretical concepts from vision science concerning visual size coding that help inform our understanding of historical and modern typographical practices. While economic, social, technological, and artistic factors influence type design and selection, we conclude that properties of human visual processing play a dominant role in constraining the distribution of print sizes in common use.

Influence of background on image recognition in normal vision and age-related macular degeneration

PURPOSE

The influence of background attenuation on the spatial frequency bandwidth requirements for image recognition was assessed in normal young and older groups and in a group with age-related macular degeneration (AMD). Bandwidth requirements were also assessed in the visual periphery of young normal observers.

METHODS

In Experiment 1, each observer was presented with 20 series of images. Each series consisted of a sequence of progressively low-pass filtered images, presented in an order of increasing bandwidth, i.e., according to an ascending method of limits. For half of the series, the background of the base image was selectively darkened by 80% of its original luminance. Three measures were analyzed: (1) the critical bandwidth defined as the bandwidth in cycles/image (cpi) at which 50% of the images were recognized, (2) the minimal bandwidth, defined as the minimal bandwidth at which images were recognized and (3) the proportion of images recognized at full bandwidth. In Experiment 2, young normal observers were similarly tested in central vision and at 5.5° eccentricity (superior or inferior visual field). A third background attenuation condition was included, as well, in which the background was low-pass filtered.

RESULTS

The critical bandwidth for image recognition was significantly reduced by darkening the image background for normal young and old and the AMD groups. This improvement was found to be contrast dependent for the darkened background. In addition, AMD observers tended to recognize more images at full bandwidth if the background was darkened. For normal young observers, making the background low-pass was ineffective in lowering the critical bandwidth in the fovea. Fewer images were recognized at full bandwidth at 5.5° eccentricity for a low-pass background and marginally fewer for a darkened background.

CONCLUSIONS

Selective attenuation of the image background can lead to reductions in the bandwidth requirements for image recognition in AMD. However, performance of young normal observers for images presented in the periphery was unlike AMD performance under the conditions investigated. These results have interesting implications for the design of image enhancement algorithms to aid low vision observers.

Training to improve reading speed in patients with juvenile macular dystrophy: a randomized study comparing two training methods

PURPOSE

In this study, we examined the clinical application of two training methods for optimizing reading ability in patients with juvenile macular dystrophy with established eccentric preferred retinal locus and optimal use of low-vision aids.

METHOD

This randomized study included 36 patients with juvenile macular dystrophy (35 with Stargardt's disease and one with Best's disease). All patients have been using individually optimized low-vision aids. After careful ophthalmological examination, patients were randomized into two groups: Group 1: Training to read during rapid serial visual presentation (RSVP) with elimination of eye movements as far as possible (n = 20); Group 2: Training to optimize reading eye movements (SM, sensomotoric training) (n = 16). Only patients with magnification requirement up to sixfold were included in the study. Training was performed for 4 weeks with an intensity of ½ hr per day and 5 days a week. Reading speed during page reading was measured before and after training. Eye movements during silent reading were recorded before and after training using a video eye tracker in 11 patients (five patients of SM and six of RSVP training group) and using an infrared reflection system in five patients (three patients from the SM and two patients of RSVP training group).

RESULTS

Age, visual acuity and magnification requirement did not differ significantly between the two groups. The median reading speed was 83 words per minute (wpm) (interquartile range 74-105 wpm) in the RSVP training group and 102 (interquartile range 63-126 wpm) in the SM group before training and increased significantly to 104 (interquartile range 81-124 wpm) and 122, respectively (interquartile range 102-137 wpm; p = 0.01 and 0.006) after training, i.e. patients with RSVP training increased their reading speed by a median of 21 wpm, while it was 20 wpm in the SM group. There were individual patients, who benefited strongly from the training. Eye movement recordings before and after training showed that in the RSVP group, increasing reading speed correlated with decreasing fixation duration (r = -0.75, p = 0.03), whereas in the SM group, increasing reading speed correlated with a decreasing number of forward saccades (r = -0.9, p = 0.01).

CONCLUSION

Although the median effect of both training methods was limited, individual patients benefited well. Our results may indicate a difference in the training effect between both methods on the reading strategy: the RSVP method reduces fixation duration, the SM method decreases the number of forward saccades. Patients can apply their newly learned reading strategy in the natural reading situation, e.g. in page reading without special presentation of the text. These results can be used as a basis for further improvement in training methods for optimizing reading performance in patients with a central scotoma.

The effect of presentation paradigm on syntactic processing: An event-related fMRI study

An event-related fMRI study was conducted to investigate the effect of two different sentence presentation paradigms-rapid serial visual presentation (RSVP) and whole sentence presentation-on syntactic processing. During scanning, sentences were presented using one of the two presentation paradigms and were followed by a short delay and a probe to verify sentence comprehension. The delay was included in an attempt to separate sentence-related activity from probe-related activity. The behavioral data showed a main effect of syntactic complexity for reaction time and accuracy, and accuracy revealed an interaction between complexity and the presentation paradigm employed-RSVP produced many more errors for syntactically complex sentences than did whole sentence presentation. The imaging data revealed a syntactic complexity effect during the sentence phase in left BA 44 and during the probe phase in left BA 44 and the left posterior MTG. In addition, timecourse analysis revealed that these two regions also showed an interaction between complexity and presentation paradigm such that there was no complexity effect during RSVP but a significant effect during whole sentence presentation. In addition to finding that these two presentation paradigms differentially affected syntactic processing, there were main effects within the visual pathway (V1/V2 vs. V5) and the hippocampus that revealed significant differences in activation between the paradigms.

Enhancing visual performance for people with central vision loss

People with central vision loss must use peripheral vision for visual tasks. It is well known that performance for almost all spatial tasks is worse in the normal periphery than in the normal fovea. The primary goal of my ongoing research is to understand the limiting factors and the potential for enhancing vision for people with central vision loss. Here I review my previous work related to understanding the limiting factors on reading, a task that is the primary complaint of many patients with age-related macular degeneration, the leading cause of visual impairment in the elderly. I also review my work related to enhancing visual functions in the normal periphery and how it may be applied to people with central vision loss.

Contrast polarity differences reduce crowding but do not benefit reading performance in peripheral vision

Previous studies have shown that the spatial extent of crowding in peripheral vision is reduced when a target letter and its flanking letters have opposite contrast polarity. We have examined if this reduction in crowding leads to improved reading performance. We compared the spatial extent of crowding, visual-span profiles (plots of letter-recognition accuracy versus letter position), and reading speed at 10 degrees inferior visual field, using white letters, black letters, or mixtures of white and black letters, presented on a mid-gray background. Consistent with previous studies, the spatial extent of crowding was reduced when the target and flanking letters had opposite contrast polarity. However, using mixed contrast polarity did not lead to improvements in visual-span profiles or reading speed.

Reading speed does not benefit from increased line spacing in AMD patients

PURPOSE

Crowding, the adverse spatial interaction due to the proximity of adjacent targets, has been suggested as an explanation for slow reading in peripheral vision. Previously, we showed that increased line spacing, which presumably reduces crowding between adjacent lines of text, improved reading speed in the normal periphery (Chung, Optom Vis Sci 2004;81:525-35). The purpose of this study was to examine whether or not individuals with age-related macular degeneration (AMD) would benefit from increased line spacing for reading.

METHODS

Experiment 1: Eight subjects with AMD read aloud 100-word passages rendered at five line spacings: the standard single spacing, 1.5x, 2x, 3x, and 4x the standard spacing. Print sizes were 1x and 2x of the critical print size. Reading time and number of reading errors for each passage were measured to compute the reading speed. Experiment 2: Four subjects with AMD read aloud sequences of six 4-letter words, presented on a computer monitor using the rapid serial visual presentation (RSVP) paradigm. Target words were presented singly, or flanked above and below by two other words that changed in synchrony with the target word, at various vertical word separations. Print size was 2x the critical print size. Reading speed was calculated based on the RSVP exposure duration that yielded 80% of the words read correctly.

RESULTS

Averaged across subjects, reading speeds for passages were virtually constant for the range of line spacings tested. For sequences of unrelated words, reading speeds were also virtually constant for the range of vertical word separations tested, except at the smallest (standard) separation at which reading speed was lower.

CONCLUSIONS

Contrary to the previous finding that reading speed improved in normal peripheral vision, increased line spacing in passages, or increased vertical separation between words in RSVP, did not lead to improved reading speed in people with AMD.

Two semantic systems in the brain for rapid and slow differentiation of abstract and concrete words

Most studies of semantic processing address changes in the late (300-800 msec) components of evoked potentials. However, recent years have seen the appearance of data showing that humans can perceive the sense of stimuli presented to them in significantly shorter periods of time. We report here studies of the mechanism of semantic analysis of written abstract and concrete words in four series of experiments: 1) reading of words on a monitor screen; 2) simple classification of all presented words into the categories "abstract" and "concrete;" 3) complex, i.e., selective classification of words written only in a specified color with a prompt as to which color would be used for the word; 4) complex classification of words of only a specified color without a prior prompt. Early (40-100 msec) differences in evoked brain potentials were seen on comparison of responses to abstract and concrete words, predominantly in the frontal areas in the case of simple reading of words and in the more dorsal areas in the case of tasks with simple classification. All cases of explicit classification of words were characterized by differences in late (450-700 msec) components in the left frontal zone. The results indicate the existence of two semantic systems: a rapid, implicit system associated with activation of the right frontal area, and a slow, explicit system of word classification which is predominantly associated with activity in the left frontal area. The relationship between the two systems is to a certain extent reciprocal: the rapid system can by inhibited by introduction of a word classification task.

An ERP study of category priming: evidence of early lexical semantic access

We investigated the effects of lexical semantic access on early components of high-density ERPs. Participants performed five blocks of a standard lexical decision (LD) task in which visually presented words were distinguished from nonwords (pseudowords) matched on word length. A second condition involved a lexical semantic (LS) version, in which the words (e.g., tiger) used in each block were selected from a single category (e.g., animals). ERP results indicated that (1) a main effect of lexicality occurred at 100 ms with P1 amplitude greater for words than for nonwords; (2) a task effect for both words and nonwords was also found in the P1 amplitude (larger for LS); (3) a lexicalityxtask interaction in the N1 amplitude indicated lexical semantic access by 168 ms; (4) semantic facilitation of response time to word-nonword judgment may be largely explained by P1 enhancement. Our findings of an early lexicality effect at 100 ms and of semantic access by 168 ms is in accordance with results of recent ERP and eye movement studies.

Retinotopy of the face aftereffect

Physiological results for the size of face-specific units in inferotemporal cortex (IT) support an extraordinarily large range of possible sizes--from 2.5 degrees to 30 degrees or more. We use a behavioral test of face-specific aftereffects to measure the face analysis regions and find a coarse retinotopy consistent with receptive fields of intermediate size (10 degrees -12 degrees at 3 degrees eccentricity). In the first experiment, observers were adapted to a single face at 3 degrees from fixation. A test (a morph of the face and its anti-face) was then presented at different locations around fixation and subjects classified it as face or anti-face. The face aftereffect (FAE) was not constant at all test locations--it dropped to half its maximum value for tests 5 degrees from the adapting location. Simultaneous adaptation to both a face and its anti-face, placed at opposite locations across fixation, produced two separate regions of opposite aftereffects. However, with four stimuli, faces alternating with anti-faces equally spaced around fixation, the FAE was greatly reduced at all locations, implying a fairly coarse localization of the aftereffect. In the second experiment, observers adapted to a face and its anti-face presented either simultaneously or in alternation. Results showed that the simultaneous presentation of a face and its anti-face leads to stronger FAEs than sequential presentation, suggesting that face processing has a dynamic nature and its region of analysis is sharpened when there is more than one face in the scene. In the final experiment, a face and two anti-face flankers with different spatial offsets were presented during adaptation and the FAE was measured at the face location. Results showed that FAE at the face location was inhibited more as the distance of anti-face flankers to the face stimulus was reduced. This confirms the spatial extent of face analysis regions in a test with a fixed number of stimuli where only distance varied.

Page mode reading with simulated scotomas: a modest effect of interline spacing on reading speed

Crowding is thought to be one potent limiting factor of reading in peripheral vision. While several studies investigated how crowding between horizontally adjacent letters or words can influence eccentric reading, little attention has been paid to the influence of vertically adjacent lines of text. The goal of this study was to examine the dependence of page mode reading performance (speed and accuracy) on interline spacing. A gaze-contingent visual display was used to simulate a visual central scotoma while normally sighted observers read meaningful French sentences following MNREAD principles. The sensitivity of this new material to low-level factors was confirmed by showing strong effects of perceptual learning, print size and scotoma size on reading performance. In contrast, reading speed was only slightly modulated by interline spacing even for the largest range tested: a 26% gain for a 178% increase in spacing. This modest effect sharply contrasts with the dramatic influence of vertical word spacing found in a recent RSVP study. This discrepancy suggests either that vertical crowding is minimized when reading meaningful sentences, or that the interaction between crowding and other factors such as attention and/or visuo-motor control is dependent on the paradigm used to assess reading speed (page vs. RSVP mode).

Trade-off between object selectivity and tolerance in monkey inferotemporal cortex

Object recognition requires both selectivity among different objects and tolerance to vastly different retinal images of the same object, resulting from natural variation in (e.g.) position, size, illumination, and clutter. Thus, discovering neuronal responses that have object selectivity and tolerance to identity-preserving transformations is fundamental to understanding object recognition. Although selectivity and tolerance are found at the highest level of the primate ventral visual stream [the inferotemporal cortex (IT)], both properties are highly varied and poorly understood. If an IT neuron has very sharp selectivity for a unique combination of object features ("diagnostic features"), this might automatically endow it with high tolerance. However, this relationship cannot be taken as given; although some IT neurons are highly object selective and some are highly tolerant, the empirical connection of these key properties is unknown. In this study, we systematically measured both object selectivity and tolerance to different identity-preserving image transformations in the spiking responses of a population of monkey IT neurons. We found that IT neurons with high object selectivity typically have low tolerance (and vice versa), regardless of how object selectivity was quantified and the type of tolerance examined. The discovery of this trade-off illuminates object selectivity and tolerance in IT and unifies a range of previous, seemingly disparate results. This finding also argues against the idea that diagnostic conjunctions of features guarantee tolerance. Instead, it is naturally explained by object recognition models in which object selectivity is built through AND-like tuning mechanisms.

Rapid and slow brain systems of abstract and concrete words differentiation

Previous studies have repeatedly found that late (300-800 ms) components of event-related potentials (ERP) reflected semantic analysis, i.e. the differentiation between abstract and concrete words. However, the human brain may detect the meaning of the words much earlier. This study investigated the brain mechanisms of the processing of abstract and concrete written words in four experimental conditions: i) Simple Reading, during which volunteers were required to silently read words; ii) Simple Classification, during which volunteers were required to classify the presented word into the abstract and concrete categories; and iii) Cued and iv) Uncued Selective Classification conditions, during which subjects had to classify only the words typed in a particular colour. 19-channel EEG was recorded during the experiment from 13 subjects. The ERP to abstract and concrete words differed not only at the late but also at early (40-100 ms) latencies in the Simple Reading and Classification conditions, as well as for the words that should not be explicitly classified in the Cued Selective Classification condition. This means that semantic analysis can occur in a manner which is both very rapid and implicit. Moreover, increasing task demands can even suppress this rapid semantic analysis. The functional microstate analysis revealed a topographical difference in response to abstract and concrete words, which indicated that at least partly distinct brain networks are involved in the processing of words during both early (implicit differentiation) and late (explicit classification) latencies.

Letter case and text legibility in normal and low vision

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.

The case for the visual span as a sensory bottleneck in reading

The visual span for reading is the number of letters, arranged horizontally as in text, that can be recognized reliably without moving the eyes. The visual-span hypothesis states that the size of the visual span is an important factor that limits reading speed. From this hypothesis, we predict that changes in reading speed as a function of character size or contrast are determined by corresponding changes in the size of the visual span. We tested this prediction in two experiments in which we measured the size of the visual span and reading speed on groups of five subjects as a function of either character size or character contrast. We used a "trigram method" for characterizing the visual span as a profile of letter-recognition accuracy as a function of distance left and right of the midline (G. E. Legge, J. S. Mansfield, & S. T. L. Chung, 2001). The area under this profile was taken as an operational measure of the size of the visual span. Reading speed was measured with the Rapid Serial Visual Presentation (RSVP) method. We found that the size of the visual span and reading speed showed the same qualitative dependence on character size and contrast, reached maximum values at the same critical points, and exhibited high correlations at the level of individual subjects. Additional analysis of data from four studies provides evidence for an invariant relationship between the size of the visual span and RSVP reading speed; an increase in the visual span by one letter is associated with a 39% increase in reading speed. Our results confirm the visual-span hypothesis and provide a theoretical framework for understanding the impact of stimulus attributes, such as contrast and character size, on reading speed. Evidence for the visual span as a determinant of reading speed implies the existence of a bottom-up, sensory limitation on reading, distinct from attentional, motor, or linguistic influences.

Acuity, crowding, reading and fixation stability

People with age-related macular disease frequently experience reading difficulty that could be attributed to poor acuity, elevated crowding or unstable fixation associated with peripheral visual field dependence. We examine how the size, location, spacing and instability of retinal images affect the visibility of letters and words at different eccentricities. Fixation instability was simulated in normally sighted observers by randomly jittering single or crowded letters or words along a circular arc of fixed eccentricity. Visual performance was assessed at different levels of instability with forced choice measurements of acuity, crowding and reading speed in a rapid serial visual presentation paradigm. In the periphery: (1) acuity declined; (2) crowding increased for acuity- and eccentricity-corrected targets; and (3), the rate of reading fell with acuity-, crowding- and eccentricity-corrected targets. Acuity and crowding were unaffected by even high levels of image instability. However, reading speed decreased with image instability, even though the visibility of the component letters was unaffected. The results show that reading performance cannot be standardised across the visual field by correcting the size, spacing and eccentricity of letters or words. The results suggest that unstable fixation may contribute to reading difficulties in people with low vision and therefore that rehabilitation may benefit from fixation training.

Where people look when watching movies: do all viewers look at the same place?

Magnification around the most important point of a movie scene (center of interest-COI) might aid people with visual impairments that cause resolution loss. This will be effective only if most people look at the same place when watching a movie. We recorded the eye movements of 20 normally sighted subjects as each watched six movie clips, totaling 37.5 min. More than half of the time the distribution of subject gaze points fell within an area statistic that was less than 12% of the movie scene. Male and older subjects were more likely to look in the same place than female and younger subjects, respectively. We conclude that the between-subject agreement is sufficient to make the approach practical.

Beating the beat: reading can be faster than the frequency of eye movements in persons with congenital nystagmus

PURPOSE

Rapid serial visual presentation (RSVP) of text has been reported to foster higher reading rates than presentation in a continuous text (CT) format, possibly because scanning eye movements are minimized. We investigated how this might be relevant for persons with congenital nystagmus (CN).

METHODS

We evaluated whether reading rates differ in persons with CN for RSVP versus CT presentation of single sentences under otherwise similar conditions. In a second experiment, we presented unrelated words to observers with CN in RSVP format while measuring their eye movements to determine whether reading can occur during the high-velocity, nonfoveating periods of the CN wave form. Both sentences and random words were selected from the MNRead corpus and displayed at 2x, 4x, or 8x the threshold word size on a 21-inch computer monitor.

RESULTS

Subjects with CN have virtually equivalent maximum reading speeds of 449 and 448 words per minute, respectively, for RSVP and CT presentation of sentences. Typically, reading rates were faster than the frequency of CN, which suggests that subjects could read during the nonfoveating periods of the nystagmus waveform. This finding was confirmed using random words that, unlike those in sentences, cannot be inferred from contextual cues. Examination of eye movements recorded during reading indicated that random words are read correctly with 47% to 65% accuracy (depending on word size) during the nonfoveating periods of the CN waveform.

CONCLUSION

A clinical implication of these results is that reading performance in persons with CN should be facilitated by large text sizes that remain legible during a greater fraction of the CN waveform.

Multiple object response normalization in monkey inferotemporal cortex

The highest stages of the visual ventral pathway are commonly assumed to provide robust representation of object identity by disregarding confounding factors such as object position, size, illumination, and the presence of other objects (clutter). However, whereas neuronal responses in monkey inferotemporal cortex (IT) can show robust tolerance to position and size changes, previous work shows that responses to preferred objects are usually reduced by the presence of nonpreferred objects. More broadly, we do not yet understand multiple object representation in IT. In this study, we systematically examined IT responses to pairs and triplets of objects in three passively viewing monkeys across a broad range of object effectiveness. We found that, at least under these limited clutter conditions, a large fraction of the response of each IT neuron to multiple objects is reliably predicted as the average of its responses to the constituent objects in isolation. That is, multiple object responses depend primarily on the relative effectiveness of the constituent objects, regardless of object identity. This average effect becomes virtually perfect when populations of IT neurons are pooled. Furthermore, the average effect cannot simply be explained by attentional shifts but behaves as a primarily feedforward response property. Together, our observations are most consistent with mechanistic models in which IT neuronal outputs are normalized by summed synaptic drive into IT or spiking activity within IT and suggest that normalization mechanisms previously revealed at earlier visual areas are operating throughout the ventral visual stream.

Serifs and font legibility

Using lower-case fonts varying only in serif size (0%, 5%, and 10% cap height), we assessed legibility using size thresholds and reading speed. Five percentage serif fonts were slightly more legible than sans serif, but the average inter-letter spacing increase that serifs themselves impose, predicts greater enhancement than we observed. RSVP and continuous reading speeds showed no effect of serifs. When text is small or distant, serifs may, then, produce a tiny legibility increase due to the concomitant increase in spacing. However, our data exhibited no difference in legibility between typefaces that differ only in the presence or absence of serifs.

Word Processing differences between dyslexic and control children

BACKGROUND

The aim of this study was to investigate brain responses triggered by different wordclasses in dyslexic and control children. The majority of dyslexic children have difficulties to phonologically assemble a word from sublexical parts following grapheme-to-phoneme correspondences. Therefore, we hypothesised that dyslexic children should mainly differ from controls processing low frequent words that are unfamiliar to the reader.

METHODS

We presented different wordclasses (high and low frequent words, pseudowords) in a rapid serial visual word (RSVP) design and performed wavelet analysis on the evoked activity.

RESULTS

Dyslexic children had lower evoked power amplitudes and a higher spectral frequency for low frequent words compared to control children. No group differences were found for high frequent words and pseudowords. Control children had higher evoked power amplitudes and a lower spectral frequency for low frequent words compared to high frequent words and pseudowords. This pattern was not present in the dyslexic group.

CONCLUSION

Dyslexic children differed from control children only in their brain responses to low frequent words while showing no modulated brain activity in response to the three word types. This might support the hypothesis that dyslexic children are selectively impaired reading words that require sublexical processing. However, the lacking differences between word types raise the question if dyslexic children were able to process the words presented in rapid serial fashion in an adequate way. Therefore the present results should only be interpreted as evidence for a specific sublexical processing deficit with caution.

Effect of dynamic display and speed of display movement on reading Chinese text presented on a small screen

Automatic dynamic displays, e.g., scrolling displays, are frequently used to present text information on small screens. This study examined the effects of three dynamic displays [leading, scrolling, and rapid serial visual presentation (RSVP)] and three presentations speeds [171, 250, and 305 wpm (words per minute)] on subjects' reading comprehension for different types of small screens for laptops, personal digital assistants (PDAs), and mobile phones. 12 college students who were native speakers of Chinese (4 men and 8 women between the ages of 19 and 36 years) participated. Scores for reading comprehension indicated (1) for laptops, both leading displays and RSVP, at speeds of about 250 wpm, comprehension was highest; (2) for PDAs and mobile phones, scrolling displays and RSVP, speeds of about 305 wpm were more suitable for presenting information in Chinese text; (3) consequently, RSVP appeared to be the most viable dynamic display on small screens and led to the highest reading comprehension for these Chinese readers.

Preferred retinal loci and macular scotoma characteristics in patients with age-related macular degeneration

Many patients with macular scotomas due to age-related macular degeneration do not perceive black spots in the visual field where the scotomas are located. Rather, they describe objects as "vanishing," "jumping out of nowhere" or "having blurry parts," or a combination of features. In addition, when the macular scotoma affects the fovea, the visual system uses 1 or more preferred retinal loci (PRLs) as a "pseudofovea" to perform visual tasks. Visual function testing with the scanning laser ophthalmoscope has provided a wealth of information regarding how patients perceive the visual world and how the oculomotor system directs eye movements. This article describes 2 specific functions of the oculomotor system, fixation stability and refixation precision, with data collected from normally sighted people and patients with visual field loss. The implications of the characteristics of PRLs and macular scotomas for clinical testing are discussed.

The recognition potential during sentence presentation: stimulus probability, background stimuli, and SOA

Recognition potential (RP) is an electrical brain response that has proved its usefulness for studying semantic processing of isolated words, and appears when subjects view meaningful stimuli embedded in a stream of background images at a high rate of presentation: the rapid stream stimulation paradigm (RSS). The present technical study is aimed at testing the validity of this procedure in the study of words within sentences. For this purpose, we varied word and background probability of appearance, the number of background stimuli preceding each word, and stimulus onset asynchrony. Probability did not have significant effects on RP, but it was found that a minimum number of two background stimuli preceding each word and a high rate (250 ms) of presentation are preferable for enhancing RP amplitude. The RSS paradigm would therefore improve the visibility-and, hence, refine the analysis-of a component that can nevertheless be obtained with more standard paradigms, such as rapid serial visual presentation, devoid of interspersed background stimuli.

Preferred Retinal Locus and Reading Rate with Four Dynamic Text Presentation Formats

BACKGROUND

Electronic display devices hold the potential to improve access to written material by people with low vision. For those with central field loss, the optimal form of electronic text presentation may vary according to the location of the preferred retinal locus, but this has never been investigated. In this study, we examined the relationship between preferred retinal locus location and reading rate for four electronic display formats (rapid serial visual presentation, horizontal scroll, vertical scroll, and page).

METHODS

Short sentences were presented in each format to 35 low-vision (most with central field loss) and 14 age-matched control subjects. Subjects read aloud to determine maximum oral reading rate and read silently to determine preferred silent reading rate.

RESULTS

With the exception of page format, maximum oral reading rates were faster than silent preferred reading rates for both groups of subjects. For the low-vision group, there was no significant difference in maximum oral reading rates between the four display formats; and when reading at a preferred silent rate, page format was faster than the other three formats. Though page format was read more quickly, half of the low-vision subjects preferred the horizontal-scroll format. Contrary to our predictions, there was no significant effect of preferred retinal locus location (vertical vs. lateral) on reading rate and no significant interaction between preferred retinal locus location and display format.

CONCLUSIONS

The differences between maximum oral and preferred silent reading rates and the lack of a relationship between reading rates and preferred display format reinforce the importance of patient preference in the evaluation and selection of a device or display format during rehabilitation.

The effect of text orientation, visual meridian, and inter-character spacing on word identification in the retinal periphery

Previous research has demonstrated that the masking effects of flankers about a target in the peripheral retina are not isotropic. Rather, regions of lateral interaction are ellipsoid in shape with the major axis oriented radially along a meridian through the fovea. This finding leads to the counterintuitive prediction that horizontal text positioned to the right of fixation might be read more slowly than similarly positioned text oriented diagonally or vertically. Similarly, vertically oriented text above fixation might be read more slowly than horizontally or diagonally oriented text above fixation. We investigated the effect of text orientation and inter-character spacing on word identification in the retinal periphery. Text was presented by rapid serial visual presentation. Words were centered 3 degrees from fixation along four visual field meridians (VM) (right horizontal, upper-right diagonal, vertical, and upper-left diagonal). Regardless of VM identification, performance was best for horizontal text, declining slightly for orientations between +60 degrees and -60 degrees and declining more quickly for acute orientations. A weak effect of VM was observed for text with normal inter-character spacing. Performance was best for text centered along the horizontal meridian and declined slightly along the other VM. Finally, identification rates increased by approximately 33 words min(-1) with the addition of one character space between adjacent letters. The word-recognition processes are very tolerant of text orientation, exhibiting a modest decline for orientations within +/- 60 of horizontal regardless of VM.

Eye movements reveal the on-line computation of lexical probabilities during reading

Skilled readers are able to derive meaning from a stream of visual input with remarkable efficiency. In this article, we present the first evidence that statistical information latent in the linguistic environment can contribute to an account of reading behavior. In two eye-tracking studies, we demonstrate that the transitional probabilities between words have a measurable influence on fixation durations, and using a simple Bayesian statistical model, we show that lexical probabilities derived by combining transitional probability with the prior probability of a word's occurrence provide the most parsimonious account of the eye movement data. We suggest that the brain is able to draw upon statistical information in order to rapidly estimate the lexical probabilities of upcoming words: a computationally inexpensive mechanism that may underlie proficient reading.

The shape and size of crowding for moving targets

Our ability to identify alphanumeric characters can be impaired by the presence of nearby features, especially when the target is presented in the peripheral visual field, a phenomenon is known as crowding. We measured the effects of motion on acuity and on the spatial extent of crowding. In line with many previous studies, acuity decreased and crowding increased with eccentricity. Acuity also decreased for moving targets, but the absolute size of crowding zones remained relatively invariant of speed at each eccentricity. The two-dimensional shape of crowding zones was measured with a single flanking element on each side of the target. Crowding zones were elongated radially about central vision, relative to tangential zones, and were also asymmetrical: a more peripheral flanking element crowded more effectively than a more foveal one; and a flanking element that moved ahead of the target crowded more effectively than one that trailed behind it. These results reveal asymmetrical space-time dependent regions of visual integration that are radially organised about central vision.

Dimensions of affective semantic meaning--behavioral and evoked potential correlates in Chinese subjects

The affective meaning of words can be quantified statistically by the 'semantic differential technique'. We studied a total of 55 Chinese adults in two experiments: first, 210 nouns were rated by 32 subjects. Factor analysis on these data yielded three independent semantic dimensions. Semantically unique words were used in electrophysiological experiments in another group of 23 healthy right-handed adults. Words of similar physical appearance belonging to different semantic classes were presented visually in random order. The electroencephalogram [corrected] was recorded in 32 channels. Evoked activity was computed for each semantic class. Significant differences in electrical brain activation between semantic word classes were seen as early as 80 ms after stimulus onset confirming related findings in German subjects. These results illustrate similar early neural activation in subject groups of different language and culture.

Low-vision patients with age-related maculopathy read RSVP faster when word duration varies according to word length

Normally sighted younger and elder subjects as well as subjects with central visual field loss (CFL) from age-related maculopathy read rapid serial visual presentation (RSVP) text with words presented at a constant rate and at three different rates where word presentation duration varied according to word length. The elder subjects reading sentences foveally read fastest when word duration was constant. The younger group reading random words peripherally read faster at a variable word duration rate. The subjects with CFL read sentences an average of 33% faster when the presentation rate varied with word length. There was a trend for slow readers with CFL to benefit more than fast readers with CFL. We conclude that varying word duration based on word length in rapid serial visual presentation reading would improve reading rates for low-vision patients with CFL.

Vision multiplexing: an engineering approach to vision rehabilitation device development

Multiplexing is the transmission of two or more messages simultaneously over the same communication channel in a way that enables them to be separated and used at the receiving end. The normal visual system provides us with a very wide field of view at an apparent high resolution. The wide field of view is continuously monitored at a low resolution, providing information for navigation and detection of objects of interest. These objects of interest are sampled over time using the high-resolution fovea. Most disabling visual conditions impact only one of the components, the peripheral low-resolution wide field or the central high-resolution fovea. The loss of one of these components prevents the interplay of central and peripheral vision needed for normal function and causes disability. Traditionally low-vision aids replace or supplement the missing function, but usually at a cost of a significant loss in the surviving function. For example, magnifying devices increase resolution but reduce the field of view, whereas minifying devices increase the field of view but reduce resolution. A proposal to resolve many of the problems of current visual aids by exploring a general engineering approach--vision multiplexing--that takes advantage of the dynamic nature of human vision is presented. Vision multiplexing seeks to provide both the wide field of view and the high-resolution information in ways that could be accessed and interpreted by the visual system. This paper describes the use of optical methods and computer technologies in the development of a number of new visual aids, all of which apply vision multiplexing to restore the interplay of central and peripheral vision using eye movements in a natural way.

Reading with eccentric fixation is faster in inferior visual field than in left visual field

PURPOSE

People with central field loss tend to fixate so that information falls in the inferior or left visual field. Studies of reading from a page of text suggest that using inferior field is advantageous relative to using left visual field. In this study, we investigated whether reading without eye movements in normal peripheral vision is better when text is presented in inferior or left visual field.

METHODS

Reading rates were determined for retinally stabilized rapid serial visual presentation sentences of seven letter sizes, presented at 5 degrees in inferior and left visual field of six normal observers.

RESULTS

When print size is appropriately magnified for peripheral viewing, reading speed in inferior field is faster than in left visual field. There is no significant difference between inferior and left visual field in the print size required to reach maximum reading speed.

CONCLUSIONS

For reading tasks not involving eye movements, there is an advantage in eccentrically fixating such that text falls in inferior rather than left visual field.

The recognition potential: An ERP index of lexical access

Recognition potential (RP) is a brain electrical response that appears when a subject views recognizable images of words. However, it has yet to be determined whether the processes reflected by RP are related to orthographic or to semantic analysis. This study aimed to resolve this question by studying the RP evoked by orthographically correct stimuli that were devoid of meaning. Results showed RP not only to this type of stimuli, but also to others achieving lower levels in the reading process. Strikingly, however, the RP amplitude significantly differed in parallel with the levels of the reading processes attained by the stimuli, the amplitude of the RP progressively increasing as the level approached the semantic one, which showed the highest amplitude. These results not only confirm the replicability of RP, but also its promise of potential usefulness in the study and assessment of language perception.

Invariance between subjects of brain wave representations of language

In three experiments, electric brain waves of 19 subjects were recorded under several different experimental conditions for two purposes. One was to test how well we could recognize which sentence, from a set of 24 or 48 sentences, was being processed in the cortex. The other was to study the invariance of brain waves between subjects. As in our earlier work, the analysis consisted of averaging over trials to create prototypes and test samples, to both of which Fourier transforms were applied, followed by filtering and an inverse transformation to the time domain. A least-squares criterion of fit between prototypes and test samples was used for classification. In all three experiments, averaging over subjects improved the recognition rates. The most significant finding was the following. When brain waves were averaged separately for two nonoverlapping groups of subjects, one for prototypes and the other for test samples, we were able to recognize correctly 90% of the brain waves generated by 48 different sentences about European geography.

Elicited sequential presentation for low vision reading

This paper reports on a variant of the rapid serial visual presentation (RSVP) technique for low vision reading called elicited sequential presentation (ESP). In both techniques, words are presented sequentially at a constant screen location, but with ESP, the reader elicits presentation of each new word by means of a button press, rather than (as with RSVP) being presented with it automatically at fixed intervals. An experiment comparing reading speeds using, ESP, RSVP and a conventional closed-circuit television (CCTV) reading aid showed that for 15 slow readers who were customary CCTV users with low vision, ESP is superior to RSVP and yields reading speeds averaging 47% faster than RSVP--about the same as CCTV reading speed. The log of the ratio of ESP to RSVP reading speeds was significantly negatively correlated with the log of RSVP reading speed, showing that slower readers benefit more than faster readers; regression predicted no benefit for readers who read with RSVP at 133 wpm or greater. Finally, word length and word presentation duration chosen by subjects reading with ESP were significantly correlated, suggesting that part of the benefit of ESP is due to reader's ability to allocate time based on word length and difficulty.

Idiosyncratic characteristics of saccadic eye movements when viewing different visual environments

Eye position was recorded in different viewing conditions to assess whether the temporal and spatial characteristics of saccadic eye movements in different individuals are idiosyncratic. Our aim was to determine the degree to which oculomotor control is based on endogenous factors. A total of 15 naive subjects viewed five visual environments: (1) The absence of visual stimulation (i.e. a dark room); (2) a repetitive visual environment (i.e. simple textured patterns); (3) a complex natural scene; (4) a visual search task; and (5) reading text. Although differences in visual environment had significant effects on eye movements, idiosyncrasies were also apparent. For example, the mean fixation duration and size of an individual's saccadic eye movements when passively viewing a complex natural scene covaried significantly with those same parameters in the absence of visual stimulation and in a repetitive visual environment. In contrast, an individual's spatio-temporal characteristics of eye movements during active tasks such as reading text or visual search covaried together, but did not correlate with the pattern of eye movements detected when viewing a natural scene, simple patterns or in the dark. These idiosyncratic patterns of eye movements in normal viewing reveal an endogenous influence on oculomotor control. The independent covariance of eye movements during different visual tasks shows that saccadic eye movements during active tasks like reading or visual search differ from those engaged during the passive inspection of visual scenes.

Reading with central field loss: number of letters masked is more important than the size of the mask in degrees

When the center of a readers, visual field is blocked from view, reading rates decline and eye movement patterns change. This is true whether the central visual field is blocked artificially (i.e. a mask) or through disease (e.g. a retinal scotoma due to macular degeneration). In past studies, when mask size was defined in terms of the number of letters masked from view, reading rates declined sharply as number of letters masked increased. Patients with larger central scotomas (in degrees of visual angle) also read slower. We sought to determine whether number of letters masked or size of the mask in degrees is the predominant factor affecting reading rates and eye movement behavior. By matching number of letters masked across several mask sizes (and compensating for reduced acuity in the periphery), we found that number of letters masked is the more important factor until mask size is quite large (> or = -7.5 degrees) and number of letters masked from view is more than seven.

Psychophysics of reading. XVIII. The effect of print size on reading speed in normal peripheral vision

Reading in peripheral vision is slow and requires large print, posing substantial difficulty for patients with central scotomata. The purpose of this study was to evaluate the effect of print size on reading speed at different eccentricities in normal peripheral vision. We hypothesized that reading speeds should remain invariant with eccentricity, as long as the print is appropriately scaled in size--the scaling hypothesis. The scaling hypothesis predicts that log-log plots of reading speed versus print size exhibit the same shape at all eccentricities, but shift along the print-size axis. Six normal observers read aloud single sentences (approximately 11 words in length) presented on a computer monitor, one word at a time, using rapid serial visual presentation (RSVP). We measured reading speeds (based on RSVP exposure durations yielding 80% correct) for eight print sizes at each of six retinal eccentricities, from 0 (foveal) to 20 deg in the inferior visual field. Consistent with the scaling hypothesis, plots of reading speed versus print size had the same shape at different eccentricities: reading speed increased with print size, up to a critical print size and was then constant at a maximum reading speed for larger print sizes. Also consistent with the scaling hypothesis, the plots shifted horizontally such that average values of the critical print size increased from 0.16 deg (fovea) to 2.22 deg (20 deg peripheral). Inconsistent with the scaling hypothesis, the plots also exhibited vertical shifts so that average values of the maximum reading speed decreased from 807 w.p.m. (fovea) to 135 w.p.m. (20 deg peripheral). Because the maximum reading speed is not invariant with eccentricity even when the print size was scaled, we reject the scaling hypothesis and conclude that print size is not the only factor limiting maximum reading speed in normal peripheral vision.

High and low luminance letters, acuity reserve, and font effects on reading speed

We compared reading speed with two fonts, Dutch (serif) and Swiss (sans serif). Text was displayed on a computer monitor, white letters on black, with the RSVP method. Luminance of the letters was either 146.0 or 0.146 cd m-2. Lower-case x-height of the fonts was approximately 5.5 times as large as letter acuity. At the high luminance, there was no difference between reading rates. There was a significant advantage for the Swiss font at the low luminance. The acuity reserve for Swiss was higher than for Dutch at the low luminance, which may account for the difference in reading speeds.

Detecting saccades with jerk

In studies of smooth eye movements, saccades are often detected and removed from eye movement records during analysis. A simple and effective method for saccade detection is described; the method uses jerk (the third derivative of eye position with respect to time).

Evoked potential correlates of semantic meaning--A brain mapping study

According to the 'semantic differential technique' the affective meaning of words can be quantified in statistically defined, independent dimensions where every word is uniquely located on the three dimensions evaluation ('good-bad'), potency ('strong-weak'), and activity ('active-passive'). Two experiments were performed on a total of 52 adults: first, 162 nouns were rated by 30 subjects. All words had a comparable number of letters and frequency of occurrence in the German language. A factor analysis followed by varimax rotation on the ratings yielded three semantic dimensions, and for each dimension up to 20 words were selected which scored highly positive or highly negative on one of the three dimensions, and had small scores on the others. This resulted in six semantic word classes which were then used in electrophysiological experiments performed on another group of 22 healthy right-handed adults. Stimuli were presented sequentially on a computer monitor in a randomized order, and the EEG was recorded in 30 channels and continuously stored on hard disk. A checkerboard reversal stimulus was used in a control condition. Evoked potentials were computed off-line for each semantic class. Comparison of the factor structure revealed highly similar semantic dimensions and classification of all words used. In the electrophysiological data, specific brain activity occurred that was related to semantic processing. These components, however, showed distinctive differences to brain activity elicited by contrast reversing checkerboard patterns as was evident from significant differences in component latency, amplitude, and scalp topography. Significant differences in scalp topography, latency and field strength between semantic word classes were not restricted to late 'cognitive' components, but brain activity at small latencies was affected by semantic meaning of the stimuli. Our data show how visually evoked brain activity is modulated by the meaning of the stimuli at early processing stages without reflecting hemispheric differences.

Fillers and spaces in text: the importance of word recognition during reading

Current theories of reading eye movements claim that reading saccades are programmed primarily on the basis of information about the length of the upcoming word, determined by low-level visual processes that detect spaces to the right of fixation. Many studies attempted to test this claim by filling spaces between words with various non-space symbols (fillers). This manipulation, however, confounds the effect of inserting extraneous characters into text with the effect of obscuring word boundaries by filling spaces. We performed the control conditions necessary to unconfound these effects. Skilled readers read continuous stories aloud and silently. Three factors were varied: (i) position of the fillers in the text (at the beginning, the end, or surrounding each word); (ii) the presence or absence of spaces in the text; and (iii) the effect of the type of filler on word recognition (from greatest effect to least effect: Latin letters, Greek letters, digits and shaded boxes). The effect of fillers on reading depended more on the type of filler than on the presence of spaces. The greater effect the fillers had on word recognition, the more they showed reading. Surrounding each word with digits or Greek letters slowed reading as much as filling spaces with these symbols. Surrounding each word with randomly chosen letters, while preserving spaces, slowed reading by 44-75%--as much as, or more than, removing spaces from normal text. Removing spaces from text with Latin-letter fillers slowed reading by only 10-20% more. We conclude that fillers in text disrupt reading by affecting word recognition directly, without necessarily affecting the eye movement pattern.

Simulated cataract does not reduce the benefit of RSVP

Normally sighted younger and older (mean age 71 years) observers read sentences and random lists of words from a rapid serial visual presentation (RSVP) display and a scroll display using their normal vision and through two levels of cataract simulators. Unlike patients with central field loss (CFL), there was no decrease in the benefit of RSVP with reduced vision due to the cataract simulators. However, the usefulness of sentence-level context was reduced as visual acuity was reduced. In addition, older readers did not benefit as much from RSVP as younger readers, and many in the older group were unable to read using the more severe cataract simulators from either display format. From these data we conclude that the benefits of RSVP are not reduced with reduced acuity and contrast sensitivity, and that there are age-related changes in reading rates from dynamic text displays not related to acuity.

Psychophysics of reading--XVI. The visual span in normal and low vision

The visual span in reading is the number of characters that can be recognized at a glance. The shrinking visual span hypothesis attributes reading deficits in low vision, and slow reading in normal vision at low contrast, to a reduction in the visual span. This hypothesis predicts that reading time (msec/word) becomes increasingly dependent on word length as text contrast decreases. We tested and confirmed this prediction using the rapid serial visual presentation (RSVP) method. Estimates of the visual span ranged from about 10 characters for high-contrast text to less than two characters for low-contrast text. Eye-movement recordings showed that longer reading times at low contrast are partitioned about equally between prolonged fixation times and an increased number of saccades (presumably related to a reduced visual span). RSVP measurements for six out of seven low-vision subjects revealed a strong dependence of reading time on word length, as expected from reduced visual spans.

A comparison of word recognition and reading performance in foveal and peripheral vision

Word recognition thresholds and rapid serial visual presentation (RSVP) reading rates for both unrelated words and meaningful sentences were examined across the visual field. Both word recognition thresholds and RSVP reading rates for unrelated words can be equated across the visual field by an increase in size scale. RSVP reading rates for meaningful sentences cannot be equated across the visual field, with the fovea demonstrating a qualitative superiority over the periphery, irrespective of scale. The results suggest that the early visual machinery which underlies word recognition is the same across the visual field apart from a change in scale, whilst the periphery is qualitatively inferior to the fovea at interpreting sentences with meaning.

Scrolled and rapid serial visual presentation texts are read at similar rates by the visually impaired

Visually impaired observers read dynamically displayed text faster than text displayed in a normal page view. The goal of this study was to compare reading rates from two dynamic-presentation methods that have been proposed to facilitate reading from computer-based displays. Prior research has shown that both normally sighted and low-vision observers read text displayed to the same location, one word at a time [known as rapid serial visual presentation (RSVP)], faster than a page of text. A similar comparison with text scrolled continuously across the screen also shows faster reading for low-vision patients, but the relative change from a standard page view is substantially less (15% faster for the scroll display versus 80% faster for RSVP). In this study we directly compared these techniques. For those with normal vision, reading from the RSVP display was 1.3 times faster than reading from the scroll display [t(9) = 3.32, P = 0.009]. Although the difference in reading rates for the visually impaired group did not reach statistical significance, as a group they read 13% slower from the RSVP than from the scroll display.