Hand Skills for Reading Efficiency:
A Bibliography of Resources Pertaining to Reading Braille

Tessa Wright, Author
Eleanor Pester, Project Supervisor/Editor
Monica Coffey, Research Assistant
Inge Formenti, Research Librarian

The following bibliography strives to comprehensively present scientific findings pertaining to the most efficient ways for a braille reader to employ hand skills. As scientists and authors provide new information, or as we learn of other existing, pertinent articles, the new findings will be added. We welcome suggestions about material. Also, as we provide these resources, APH encourages that you use them as a launching pad. These annotations provide only a taste. If interested by a topic presented in our bibliography, please gather the information in its entirety by reading the full article.

Studies Using Blind Subjects

Baily, J. (2003). Surmounting the braille reading speed plateau. Braille Monitor. 46(5), 323-325.

This brief, non-scientific article outlines four techniques that can be used by braille readers to achieve faster speeds and applies the findings of many of the research based articles included within this bibliography.

Bertelson, P., Mousty, P., & D'Alimonte, G. (1984). A study of braille reading: 2. Patterns of hand activity in one-handed and two-handed reading.

In an effort to amend the lack of time and detailed study dedicated to patterns of finger exploration in reading braille, the three studies involved in this report examined finger preference, the benefits and disadvantages of one-handed reading versus two handed reading, and conjoint reading in comparison with disjoint reading during the use of two hands.

The experimentation involved 24 blind readers reading aloud while their hands were recorded on video from two angles. Through analysis of these recordings it was found that, though it is not prescribed, evidence asserts that braille readers rely on their index fingers for obtaining information. Little attention was given to one-handed reading, but the authors make note that the operating modes appear the same in both hands. A great deal of emphasis revolved around two-handed reading with a focus on comparing conjoint reading, a style where the index fingers travel side by side for the duration of each line, and dissociation of the hands to different degrees. Through this research, they discovered that most readers use a combination of conjoint and disjoint hand skills since disjoint reading provides for faster line transitions, and in some cases, parallel information intake, yet conjoint reading seems to allow for slightly faster forward motion as well as better collection of the text, probably due to different roles by the fingers with regard to the comprehension and intake of information.

Bradshaw, J. L., Nettleton, N. C., & Spehr, K. (1982). Braille reading and left and right hemispace. Neuropsychologia, 20(4), 493-502.

In an effort to determine the presence of left-hand superiority, to compare phonological and semantic processing and the effects thereof, and to compare the benefits of hand positions that crossed the body as well as a position that did not, 12 experienced, braille-reading adults, 6 of each gender, 10 with strong dextral preferences in other daily tasks, were given the task of scanning lists of names and letters to identify specific targets. Of the factors being tested, the only element that made a significant difference was the scanning task. The lists of names scanned were most easily and quickly completed while the lists of letters took more time. The superior hand depended on the individual, though some participants that purported to be dextral were not, and the superior hand in the uncrossed position was also the superior hand in the crossed position. Since, as those involved in this study believe, developmental age and familiarity may play a large role in determining laterality, those with long-term experience with braille should not be expected to show a left-handed bias unless the left hand is simply the preferred hand.

Davidson, P. W., Appelle, S., & Haber, R. N.(1992). Haptic scanning of braille cells by low- and high-proficiency blind readers. Research in Developmental Disabilities, 13(2), 99-111.

This study involved a controlled evaluation of 16 blind high school students to record and analyze the way that each braille cell is scanned and how this affects reading speed. Through the observation of these students who were all trained in braille reading by the same teacher and possess comparable visual impairments. Five methods were observed for cell scanning, one was a one-handed method, four involved both hands. For purposes of recording information four methods of cell scanning were categorized as right-finger methods and two were treated as left-finger methods; one method that involved the fingers side by side was included in both categories. After processing the data, the main conclusion was that the more proficient readers invoked one of the two-handed methods, and that while all readers paused and regressed, those with higher proficiency ratings, who scanned with both hands 31% more often than those with lower proficiency, usually paused with the left finger rather than the right. While this in no way refutes the previous assumption that two-handed reading proves faster because of less lag in line transitions, it indicates that since a two-handed reader can continue to scan with one hand while the other pauses or regresses, and the reader may save the majority of time in this manner.

Davidson, P. W., Wiles-Kettenmann, M., Haber, R. N., & Appelle, S. (1980). Relationship between hand movements, reading competence and passage difficulty in braille reading. Neuropsychologia, 13(4), 629-635.

Eighteen adolescents read four prose passages aloud and were tested for comprehension in the areas of free recall and recognition while their braille reading styles were recorded and analyzed in three areas: global search style; the frequency and duration of single and multiple regressions; and fixations, stoppage of the forward movement of braille scanning. In studying these areas, the researchers found that while no particular global scanning style significantly benefited the speed of the reader, all readers were very consistent in their use of a single style. There was also no notable number of fixations present. The most intriguing finding, however, was that in all students, in all reading passages regardless of difficulty or ease, there were a significant number of single regresses. Multiple regresses were also present but in much lower numbers, especially for the more proficient readers.

Eatman, P. F. (1942). An analytic study of braille reading. Unpublished doctoral dissertation, University of Texas, Austin.

Sixty-three braille reading students in grades 3-11 were screened for reading comprehension and intelligence and selected for this study that focused on the topics of tactual sensitivity, tactual perceptual time, muscular strength, and reading skills. To assess the students' reading skills, the researcher studied photographs taken at four exposures per second to determine hand movements during two practice selections and two experimental selections. One of each type was read aloud the other was read silently.

The researcher found that the whole group made over twice as many regressive movements when reading aloud than when reading silently. The majority of good readers made these regressions with the left hand while poor readers regressed more often with the right hand or both as the good readers tended to use the hands in a more independent fashion.

Return sweeps took up 6-7% of all reading time and were made in one of two ways: 1) One or both hands moved backward until the new line was found, or 2) the hands moved independently following the current line back to find the new one. The most efficient return method proved to be the latter when the left hand returns and holds the place at the beginning of the next line till the right hand joins it.

The study also focused on the role of the hands and found that 31.75% of students used both hands equally, 48.4% of the good readers and 13.33% of the poor readers. A group consisting of 39.68% of the readers were right hand dominant, 33.33% of the good readers and 46.48% of the poor readers. Students that depended on the left hand made up the smallest group of 28.75%. Only 18.18% of the good readers used the left-handed method of reading while 40% of the poor readers relied heavily on the left hand.

Fertsch, P. (1947). Hand dominance in reading braille. American Journal of Psychology, 60(70), 335-349.

In an effort to determine the primary hand in two-handed reading as well as the importance of hand dominance, the 63 subjects of varying levels of reading proficiency were recorded while reading with each hand separately, and the testers then evaluated comprehension. Students were also recorded reading in their usual manner, whether that involved both hands, only the right, or only the left.

Through this it was determined that handedness did play a part in that those that showed a strong left-handed preference read faster with their left hands. Those who prefer the right hand in other aspects read faster with the right. Of those that used two hands, especially those who used the hands independently during two-handed reading, most were considered stronger readers and read considerably faster than those who used the left hand only. Those considered right hand dominant read at about the same speed as those using a two-handed method. This group contained nearly equal numbers of more and less proficient readers. The left-hand dominant group contained mostly poor readers.

Funchess, L. V. (1934). The psychology of reading braille with eight fingers. Unpublished master's thesis, Louisiana State University, Baton Rouge.

In this thesis, Funchess hypothesizes that braille may be read more efficiently if read with eight fingers. He proposes that the ability to use all eight fingers simultaneously is a matter of training.

To test his theory, Funchess timed readers as they read lists of letters and words. Each list consisted of 10 columns and were arranged so that either all eight fingers were used simultaneously, or the untrained fingers were used. Words were not repeated to avoid the advantage of memorization and to prevent the teaching of reading during the experiment.

Funchess found the readers could read letters with all eight fingers. He found that readers usually called out common words presented to all eight fingers without spelling them first while uncommon words were spelled aloud before stating the whole word even though the readers indicated that this was a precaution to ensure the correct answer and their first impressions were almost always right. The other problem was that of the cells being rubbed by previous readers making letters hard to determine with the untrained fingers, and Funchess felt this would probably not have affected the trained fingers.

In conclusion, Funchess felt comfortable in stating that braille readers could read with eight fingers and the ability to do this simultaneously remains a matter of training and practice.

Gray, P. G., & Todd, J. E. (1968). Mobility and reading habits of the blind: An inquiry made for the Ministry of Health, covering the registered blind of England and Wales in 1965. (Publication No. 1968-00-00). London, England: Her Majesty's Stationary Office.

While this study focused on the large spanning topic of reading and mobility based on a large and varied group of registered blind persons, notes were made about the hand movements of blind people of varied reading speeds. The results were reported in terms of three groups of reading speed: below 60 words per minute, 60-90 words per minutes, and 100 words per minute and above.

Hermelin, B. & O'Connor, N. (1971). Functional asymmetry in the reading of Braille. Neuropsychologia, 9(4), 431-435.

In an effort to discover whether Braille is analyzed as a spatial item in the right hemisphere of the brain rather than or before it is coded with corresponding letter meanings by the left hemisphere, as well as if body placement in relation to the material being gathered had an effect, 16 blind children and 15 blind adults were observed and recorded as they read passages of braille aloud. Through this, the observers found that most children read faster with the left middle finger than the right, and the number of errors drops for both adults and children when using the left. Though the majority of readers normally read with the index finger, the difference between the hands was greatest with the middle finger. The results led the observers to infer that braille information is analyzed in the right hemisphere.

Hislop, D. W. (1984). Characteristics of tactual reading by blind optacon and braille readers (Doctoral dissertation, University of Illinois at Chicago, 1984). Dissertation Abstracts International, 45(02B), 0799.

In his dissertation, Hislop provides a general overview of tactile reading, citing many previous studies regarding reading speed and methods in Braille reading as well as Optacon usage. He goes on to describe two studies of his own. The first involved six college-educated subjects; those with the best vision only had a degree of light perception. Each subject read five texts in their usual two-handed manner and five with the use of only one hand. He found that texts read with only one hand were completed in a negligibly slower amount of time with the subjects' average reading rate equaling 126 words per minute when using their usual methods and 116.8 words per minute when restricted to one hand. The second experiment went on to examine the limitations presented by the relative massiveness of the forearm, hand, and finger in comparison with the eyes and how those skills affect reading rate. Four experienced tactile readers were given the task of scanning and tracking, in both single and multiple cycles, three types of information laid out in a 6 inch format to resemble a text line that would normally be read. While a noticeable difference in reading rate for the three types of input surfaced, the corresponding reading rates to the scanning tasks resulted in around 756 words per minute as the best, most conservative estimate. Presumably this contradicts the idea that the motor skills involved in tactual reading severely limit the potential speed of the process.

Holland, B.F. (1934). Speed and pressure factors in braille reading. The Teacher's Forum for Instructors of Blind Children, 7, 13-17.

This experiment used a timing device and a balance system to measure pressure and a kymograph to calculate the amount of pressure used by 17 readers in grades 4 through 10. The students were chosen so as to have two distinct groups, slow readers and fast readers. Each student read a practice section and a section that was used for evaluation.

As a general rule, it was found that the faster readers applied less pressure to the braille that they were reading, but exceptions were prevalent in individuals outside of the trend. Another consistent finding was that both slow and fast readers start with less pressure at the beginning of the line and apply increasing pressure throughout the line. Still, it is noted that the pressure suffers frequent interruptions such as brakes between words.

Holland, B.F. and Eatman, P.F. (1933). The silent reading habits of blind children. The Teacher's Forum for Instructors of Blind Children, 6, 1,4-11, 19.

In this study, two good readers and two poor readers were taken from grades 3, 4, 6, 7, 8, 9, and 11 for a total of 28 subjects. The subjects were given material to read and were photographed every fifth of a second. The pictures were used to analyze reading techniques and the number of exposures was used to calculate the reading time.

The study found the good readers generally:

Kozel, R. J. (1995). Consideration of hand in the reading of braille. RE:view, 27(2), 78-82.

This article examines the findings of several others who have studied the subjects of speed and accuracy in reading braille with the left hand versus the right hand in boys and girls separately. The discussion that ensues seems to agree with the theory that the left hand is superior in reading braille for accuracy and suggests that environment possibly affects the lateralization process that takes place suggesting the girls' pattern decoding skills and left hand use may improve with more attention to regular left-hand use.

Kusajima, T. (1974). Visual reading and braille reading: An experimental investigation of the physiology of visual and tactual reading. New York: American Foundation for the Blind.

Experiments in this work test several different aspects of Braille reading. The first experiment focuses around the Wundt/Heller theory which proposes that in bimanual braille reading, each hand serves a different purpose, the left analyzes while the right functions synthetically. The experiment based off of this theory, however, found that proficient braille readers could comparably read with only the preferred hand; therefore, seemingly disputing the theory.

Another experiment tested the effects of forcing the braille reader to read more quickly. As had been expected, the conductors found that, depending on the reader and the task, reading comprehension diminishes at certain levels of speed. They also asserted that proficient readers tailor finger tracking speed to the task at hand.

The authors continue to draw assumptions concerning some of the characteristics of bimanual reading and the movements involved when returning to the next line of text based upon the experiements conducted.

Lowenfeld, B., Abel, G. L., Hatlen, P. H. (1969). Blind Children Learn to Read. (pp. 39-120). Springfield: Charles C Thomas Publisher.

The authors conducted a survey of many aspects of braille reading in an effort to determine the methods employed at the time in local and residential schools. Several of the questions pertained to hand skills and reading efficiency. The following information resulted from the 362 surveys completed:

Maxfield, K. (1928). The blind child and his reading: A handbook for teachers of primary braille reading. (pp. 44-50). New York: American Foundation for the Blind.

In this handbook the results of several studies performed by the Uniform Type Committee appear. The studies performed by this group involved 1,200 blind students. Their research found:

Millar, S. (1984). Is there a 'best hand' for Braille? Cortex, 20, 75-87.

Many studies over the years have yielded great debate over which hand produced the best results when reading braille: the left, right, or both, and if this preference links directly with braille proficiency and experience. No previous study had studied this correlation in isolation.

The first hypothesis suggested that the advantage of employing one hand over the other depends upon task demands at different reading proficiency levels which develop hierarchically from letter discrimination through word reading. The first experiment tested congenitally blind students', whom had been divided into two groups, retarded (students that were at least 18 months behind their I.Q. test indicated mental age) and adequate readers, speed in letter identification with first one hand, then the other. Retarded readers produced better results with the left hand while adequate readers enjoyed more success with the right.

A second hypothesis asserted that the strategy preferences that lie in association with relative reading proficiency rather than absolute determine hand preference. This experiment with the same group of students involved a similar timed task of simple word reading with each hand individually as well as both together. Two-handed reading proved advantageous in this task which required lateral movement, and neither individual hand proved advantageous to a recognizable group.

The study produced no clear answer to the question of which hand is better for braille other than there may be no such thing as a blanket answer.

Millar, S. (1987). The perceptual "window" in two-handed braille: Do the left and right hands process text simultaneously. Cortex, 23, 111-122.

This study of two-handed reading skills involved two experiments with ten fluent braille readers in high school. In both studies, the students read prose selections (two practice and one actual trial) aloud, and the reading was recorded in a way so that, when analyzed frame by frame, each movement could be detected as well as what part of the braille or space was touching each finger at all times.

The first study sought to prove or disprove the theory that fluent readers simultaneously read with both hands. Regressions and line changes were disregarded and finger placements were subcategorized. When analyzed, the researchers found that simultaneous processing was more rare than any other movement. Most frequently, when one finger was on a letter, the other finger was on the gap between letters or a space between words.

The second study focused on whether or not, when the left hand returned to a new line as the right hand finished a line, the left hand started reading as the right hand was still reading the previous line. Once again analysis proved that simultaneous processing is extremely rare. The vast majority of the time the left hand did not touch the first letter until the right hand had left the last letter of the previous line and was transitioning to the next line.

Mommers, M. J. (1980). Braille reading: Effects of different hand and finger usage. Journal of Visual Impairment and Blindness, 74(9), 338-343.

Created as a modified and expanded recreation of Hermelin and O'Connor's 1971 experiment, student's were timed and errors were tracked as they read word and number lists with the right index finger, the left index finger, the right middle finger, the left middle finger, and spontaneous reading, which for the majority of students involved both hands. The greatest difference was when reading the word list where the left hand produced greater speeds and fewer errors. The index finger produced better results than the middle finger, but the differences between the left and right middle fingers were comparable to the results of the index fingers.

Mousty, P. & Bertelson, P. (1985). A study of braille reading: 1. Reading speed as a function of hand usage and context. The Quarterly Journal of Experimental Psychology, 37A, 217-233.

This study focused on the differences of speed with the right or left hand versus two-handed braille reading. Those conducting the experiments were also interested in the effects of different types of literature on reading rates with various hand patterns.

The study involved 24 blind adults reading prose, approximations, and scrambled words aloud while timed and videotaped. Virtually no errors were made in reading, so these were not used as a measure, only an indicator of speed. The main findings of this study included the definite speed advantage of two-handed reading and that the type of material clearly affected the reading rate. The latter finding provoked further investigation into using different hand patterns for the different types of literature in order to examine a correlation and strategy; however, hand combination did not have a significant effect in this situation. The findings did nothing to further the common hypothesis that the reader's hands fulfill different roles in braille reading.

Wormsley, D. M. (1981). Hand movement training in braille reading. Journal of Visual Impairment and Blindness, 75(10), 327-331.

After reviewing literature which includes Kusajima's 1974 study that outlined six different hand patterns used when reading braille, the author designed and experimented with a training program that taught Kusajima's sixth hand pattern which is believed by many to be the most efficient and fastest since it uses two hands independently.

The training program involved 22 braille reading children. The training took place, 15 minutes a day over twenty days. All children were reported to have easily learned the method within 20 days. Subjects had been videotaped and their reading methods analyzed before the training occurred. The same process of videotaping and analysis occurred after the training to see if any of the students switched to the method as their primary reading method, and if not, if the training had still improved hand efficiency. While analysis proved different results for individual students, mainly the younger, more intelligent, slower readers, none of the students switched to this method and there was no significant improvement in efficiency. The author suggests that this lack of effect might be a result of the study's focus only on motor efficiency and suggests this efficient motor emphasis should be combined with recognition training from the beginning of instruction.

Wormsley, D. M. (1996). Reading rates of young braille-reading children. Journal of Visual Impairment and Blindness, 90(3), 278-282.

This research note addresses an earlier long-term study done by the author to assess the impact of hand technique training while also considering factors such as IQ, age, and the number of years of braille instruction. In this note, the author takes the information from that study and reanalyzes it as an evaluation of reading speed and the effects of the other factors. The findings of this analysis indicate that the fastest method of reading may involve both hands with the left hand reading as the right hand returns to meet the left in the middle at which time the left hand returns to the next line. Wormsley encourages teachers to set up in-classroom studies using this as a pattern.

Studies With Sighted Subjects with Blocked Vision

Lederman, S. J., Jones, B., & Segalowitz, S. J. (1984). Lateral symmetry in the tactual perception of roughness. Canadian Journal of Psychology, 38(4), 599-609. (French)

The article discusses four experiments conducted with sighted, blindfolded subjects to find the extent and effects of lateral symmetry when perceiving items tactually.

The first experiment involved right-handed students consistently using the right or left hand (determined by the experimenter) to tactually explore a symbol. Then the symbol was removed and a group of three symbols was presented. The student was to choose the explored symbol from the three options. The second involved a similar process, but the student was given two symbols to explore simultaneously.

The third experiment was similar in nature. Right-handed students were assigned to use either the right or left hand. Similar to the second experiment, the subject examined two textures. However, half of the students were to state whether or not the textures were the same. The other half had to choose which was rougher. Likewise, the fourth experiment had participants numerically rate the roughness of textures when blindfolded.

While the experimenters found a strong suggestion of functional symmetry between the hands, they did not find a significant advantage for the left hand as they had expected given that most braille readers report a left hand advantage in single handed reading.

Myers, D. H. (1976). Right- and left-handed counting of Braille dots in subjects unaccustomed to Braille. British Journal of Psychology, 67(3), 407-412.

Three experiments were conducted with right-handed individuals who had no previous experience with braille. In all trials the subjects received groups of varying braille cells in varying orders and at varying speeds on strips that were run by the subject using a drum. In each instance the subject was to call out the number of dots they perceived.

The results pointed to a significant advantage when both hands were used as opposed to one. In two of the experiments there was a tendancy toward a left hand advantage, but the difference was not statistically relavant. In the third experiment, when the tape with the braille cells was not paused for exploration, no significant difference was found between the index and middle fingers as had been noted in the previous experiment, and, once again, there was not a significant difference between the left and right hands.

Newman, S. E., Hall, A. D., Amein, J. C., & White, J. E. (1986). Rate of learning braille with the left and right hand. Washington, DC: Paper presented at the annual meeting of the American Psychological Association.

Two experiments were conducted involving sighted subjects tactually learning either the first ten characters of the braille alphabet or the second ten characters without the aid of vision. In the first experiment the subjects then read and named the character, half using the left index finger, half using the right index finger. The second experiment involved half of the participants learning and testing with the same hand (half with the left, half with the right), the other participants learned the characters with one hand and were tested with the other hand.

While a significant difference was found in those who learned and tested with the first ten characters rather than the second ten, females were found to perform notably better than males in the first experiment, and those who used the same hand in learning and testing showed heightened ability, no significance could be noted indicating a dominant hand as the preferred reader of braille, those using the right hand performed as well as those using the left.

Rudel, R. G. (1974). The functional asymmetry of Braille letter learning in normal sighted children. Neurology, 24(8), 733-738.

This experiment consisted of 80 right-handed, sighted students, divided equally by sex and age (7-14 years), with no previous exposure to braille. In an effort to determine the effects of age, sex, left versus right hand usage, and the hand order in which the braille is presented. The students were given lists of braille to learn tactually and then were asked to name the letters when they were presented in a different order.

The study found that age is a significant factor; the older students performed better. Age also played a role in hand preference. Sex also seems to be a significant factor since the girls 7-8 performed better with the right hand with performance of the left hand becoming stronger by the age 13-14. However, the boys tended to perform equally well with the left and right at an early age with the left hand steadily improving by the older ages. Order effects could not be determined alone, but when these effects were analyzed with hand preference and age all subjects except for girls age 7-8 performed better when the braille was presented to the right hand first. When order was grouped with hand preference and sex a greater significance appeared in girls.

Rudel, R. G., Denkla, M. B., & Hirsch, S. (1977). The development of left hand superiority for discriminating braille configurations. Neurology, 27(2), 160-164.

Sixty males and sixty females, all sighted, were given 40 cards comparing two braille cells. Half of the cards had two of the same characters, the other half had different characters. The cards were presented in a box so that the subject could not see the characters and had to discern if the characters were the same or different by touch. Half were scanned with the left hand, half with the right.

Age proved a significant factor; the number correctly identified increased with the age of the subject. Younger subjects showed a right hand advantage while subjects aged 11 years and older performed better with the left. Gender and hand preference differed with the age of the subject rather than remaining consistant.

Schmidt, J. M., & Lechelt, E. C. (1981). Hemispheric differences in tactile and visual recognition of braille-like stimulus patterns with static and dynamic modes of inspections. Psychological Research, 43(3), 293-305.

Thirty sighted, right-handed adults were divided into three groups of ten and tested on one of the following variables: reading a 2 x 3 dot pattern resembling a braille cell in a static situation where the pattern was presented to the middle finger and did not move, reading a 2x3 dot pattern that was presented to the middle finger and was scanned back and forth across the fingertip, and visually reading a white card that contained black dots in patterns that matched the aforementioned braille-like dot patterns. 40 trials were run, half on each hand in tactual testing and subjects were timed.

The primary statistically relevant difference occurred between the passive scanning tactile reading and the static tactile reading with the greatest accuracy occurring with static reading. The results favored this method with the right hand. The accuracy of differing hands appeared to be significantly divided by gender in the process of passive scanning tactile reading.

Smith, J. M. (1929). Which hand is the eye of the blind?. Genetic Psychology Monographs: Child behavior, Differential and Genetic Psychology, 3, 213-252.

In order to determine if there is a distinct advantage of one hand over the other when reading braille, Smith set up numerous trials of a sighted person with no previous experience of braille to read in a number of manners varying the hand(s) used and the direction of reading. The subject wore "cataract glasses" that obscured vision greatly and was taught braille. During the experiments, the subject was timed and errors were recorded. Through this method, Smith sought to determine which had proved the fastest in reading braille and whether extension or flexion provided the best method.

Through these trials, Smith states that she definitely concluded that the left hand was superior, and extension proved to be slightly more efficient than flexion.

Visit Our House!