Hand Movement

Training in Braille Reading

DIANE P. WORMSLEY, PH.D.

Dr. Wormsley was assistant professor of special education, program in Visually Handicapped, Illinois State University, when this article was written.

Abstract: Twenty-one children ages 6 though 13 were taught to use their hands independently when reading braille to determine how this pattern of hand movements affected reading variables, excluding character recognition. Although all the children learned this pattern of hand movements during the 20 days scheduled for training, only nine children exhibited a dramatic decrease in inefficient tracking movements such as pauses and scrubbing motions. Because these children were younger and more intelligent than the others, read braille more slowly, and had received less training in braille at school, the results strongly suggested that skill in tracking and use of an efficent hand movement pattern is closely tied to perceptual ability. Thus when teaching children to read braille, the motor aspects of the task should be combined with the perceptual aspects from the beginning.

The research described in this article was supported by a Thesis Incentive Award from the American Foundation for the Blind and a Student Research Award from the University of Pittsburgh. The author wishes to thank her dissertation adviser, Dr. Mary W. Moore, and committee members Dr. Fileda Hammermeister, Dr. June Mullins, Dr. Ralph Peabody and Dr. Louis Pingel.

One question that has plagued researchers for years involves the relationship between reading rate and independent use of the hands in braille reading. Readers who use their hands independently are invariably faster readers (Gray & Todd, 1968; Williams, 1971). The question is one of causality: that is, are faster readers faster because they use their hands independently or do they use their hands independently because they are faster readers?

Studies on the teaching of braille reading indicate that teachers do not always encourage or include independent use of the hands as part of the braille-reading curriculum (Cardinale, 1973; Lowenfeld, Abel, & Hatlen, 1969). Yet, in any task that involves a motor component, how the hands are used becomes extremely important when efficiency is the goal. If people do not receive specific instructions on efficient use of the hands when initially learning a task, they will develop their own techniques. Because generally speaking there are only a limited number of ways the hands can be used to accomplish a particular task, some individuals will develop the most efficient method by accident. Others will not, and once the inefficient method becomes habitual, changing it often requires extensive time and effort. Since initial instruction in most manual skills (e.g., touch typing) incorporates efficient hand use, why not-as some authors have suggested (Miller, 1955; Olson, 1976; Rodgers, 1961)-teach braille readers to use their hands efficiently from the beginning?

Review of the Literature

Braille reading is both a perceptual and a motor task. Curriculum materials developed to improve reading speeds through perceptual aspects such as character recognition have been successful (Brothers, 1974; NolaD & Kederis, 1969; Umsted, 1972). However, because perception can occur only by moving the hands over the braille stimuli (Foulke, 1974; Gibson, 1962; Karp, 1962), the motor aspects of tracking and hand movement pattern are also critically important. To track, readers must progress from left-to-right across the line of braille and make a return sweep to the next line. Thus the pattern of hand movements must be smooth to be efficient since regressions and pauses take additional time (Maxfield, 1928).

When Mangold (1978) compared the effectiveness of combined training in accurate character recognition and tracking skills with traditional methods of teaching recognition of braille letters, she found that the combined training resulted in smoother tracking.

Kusajima (1974) identified six patterns of hand movements used in reading braille:

1. Right hand only.

2. Left hand only.

3. Both hands with the right hand reading and the left hand serving as a marker.

4. Both hands in tandem in reading as well as in return movements.

5. Both hands in tandem until near the end of the line, when the left hand returns to find the next line while the right finishes reading. The right hand then returns to meet the left at the margin, and both hands read together again.

6. Each hand independently: the left hand reads from the beginning of the line to approximately the middle; the right hand then takes over the reading process while the left locates the next line. In other words, the hands meet in the middle of each line and then separate, alternating the reading process. According to Gray and Todd, (1968), the fastest readers used pattern six. Others (Maxfield, 1928; Mommers, 1976; Williams, 1971) also reported that hand movement characteristics such as single regressions, multiple regressions, scrubbing motions, searching motions, pauses, and erratic movements were typical of poor reading.

Current Study

The study described in this article focused on the effect of training children to use pattern six-independent use of the hands. This tracking pattern is certainly the most efficient one since the reader loses no time during return sweeps and can use one hand as a contextual check while reading with the other. The curriculum materials developed for the study did not incorporate recognition training: i.e., focus was the ease of teaching pattern six and the pure effect of this training on reading variables. If independent use of the hands were a simple matter of training (rather than a skill related to the reader's perceptual ability), then this pattern should have transferred to the regular reading situation without the incorporation of recognition training. Thus the study was designed to examine the effect of the hand movement training program on hand movements and, if independent use of the hands or smoother tracking resulted, the effect on reading variables.

Since visually handicapped children represent a small, heterogeneous population (Warren, 1977), research that requires large numbers of subjects, matched pairs, or control groups is extremely difficult to conduct. In addition, different educational settings may constitute a problem variable when groups of subjects must be matched across settings. Therefore, the study was designed as single-subject research using the time series approach described by Campbell and Stanley (1963) which allowed for repeated measures of the dependent variables over intervals of time, before and after intervention. In other words, each subject's behavior before treatment served as a baseline against which to compare his or her post-treatment behavior (Hersen & Barlow, 1976; Kratochwill, 1978). By using many subjects at the same time, I was able to evaluate each subject separately as well as consider all of them together as a set of replications.

The subjects were 22 braille-reading children ages 6 through 13 who had no physical problems that prevented them from reading with both hands. All of them had completed one to five years of instruction in braille-reading before the study began and had parental permission to participate. Eleven were enrolled in the Program for Visually Handicapped, Philadelphia City Schools, Philadelphia, Pennsylvania, housed in Overbrook School for the Blind; the remainder were enrolled in the Western Pennsylvania School for Blind Children, Pittsburgh, Pennsylvania. The children were grouped according to reading level or class placement; each group contained no more than four students. One child did not receive the training, but data were retained for comparison purposes.

To facilitate communication during the training, I assessed the children's understanding of basic concepts such as "left," "right," "top of the page," and "bottom of the page." The training program was not designed to teach these concepts as such; the assessment simply provided information concerning which students understood the terms and which ones would need special directions or reminders when the terms were used.

In the discussion that follows, the study is divided into three phases: (1) training, (2) data collection, and (3) data analysis. In actuality, however, much of the data was collected before the training phase began.

Training

Before developing the lessons, I prepared a task analysis of pattern six (see Figure 1). The individual components of this analysis served as the basis for developing sequential objectives. Subsequently, I developed 11 lessons to be used over a 20-day period for 15 minutes each day.

Figure 1

Child uses hands independently in reading braille with the left hand reading to the middle of the line and the right hand reading from the middle to the end of the line as the left hand is returning to find the beginning of the next line, maintaining contact with the braille line with all four fingers.

Left hands reads to middle while right hand returns from end to previous line to meet left hand.

F-

Left hand reads from Right hand returns from

beginning to middle of end of previous line to

line. meet left.

Locates beginning of line. Locates middle of line.

Maintains contact with line

four fingers of left hand.

aintains contact

ith line with all four fingers of each hand.

Right hands reads to end of line while left hand locates the

begnning of the next line.

Brings hands

together in space.

Left hand locates begining of next line from middle of previous line.

end of line.

Maintains contact with four fingers of right hand.

Controls hands and fingers

Figure 1. Task analysis of moving the hands independently while reading braille.

During the first five lessons, the children learned to move their hands in the manner described in the task analysis. During lessons 6 through 10, they practiced using this pattern at increasing speeds and incorporated techniques for rapid movement from the left hand page to the right hand page. The materials developed for this practice eliminated the character-recognition component of reading yet incorporated formats such as different line lengths, paragraph indentations, double and single spacing of braille, and left hand as well as righthand page materials. Lesson 11 presented material brailled on both sides of the same age and thus incorporated rapid page-turning techniques into the pattern.

Two former special education teachers were hired to conduct the actual training. Each received a teacher's manual containing lessons, objectives, and suggested methods and materials for use in training hand movements. The lessons supplemented the children's regular reading instruction. All the children learned to use hand movement pattern six within 20 days.

Data Collection

Observations; To collect data on reading rates, reading accuracy, hand movement patterns, and hand movement characteristics, each child was observed individually eight times: five times before and three times after the hand movement training program was introduced and completed. During these observations, the children were videotaped while reading passages selected by their teachers from their basal readers. All passages were chosen according to the following guidelines: (1) they were long enough to allow the child to read orally for five minutes, (2) had been read by the children at least two weeks before the observations, and (3) were different for each observation. These guidelines assured that the passages were appropriate for each child's reading level (i.e., they contained familiar braille forms) and incorporated the same approach (phonies or wholeword) as the child's reading instruction.

Reading comprehension test. To determine whether the children's ability to learn or retain information from the passages decreased because of the training program, I administered the Gilmore Oral Reading Test (Forms C and D) to measure oral reading comprehension. To compensate for possible differences between forms, the test was administered to each child three times before and once after the training program. Children were randomly assigned, within schools, to receive either Form C or D at the first administration. The forms were then alternated for each child during subsequent testing as follows: CDCD or DCDC. Each child's raw scores were reported previously(Wormsley, 1979).

The observations and tests were conducted at different times at the two schools, and the schedules were staggered. During the five pretraining sessions, the children were videotaped for five minutes while reading aloud from the passages selected by their teachers. During sessions three through five, the Gilmore Oral Reading test was also administered. The 20-day training program was then introduced. Immediately afterward, the children were again videotaped for five minutes while reading passages aloud, and the Gilmore test was administered. During the final two sessions, the children were videotaped only. Each child's videotapes were then randomly assigned letter codes to eliminate the problem of observer drift (Kazdin, 1977).

Observers. Two University of Pittsburgh graduate students in education of the visually handicapped were trained to observe the videotapes and code data on hand movement patterns, hand movement characteristics, and reading errors. The training consisted of two parts: in the first part, the observers learned to code hand movement patterns and characteristics; in the second, they learned to code reading errors. A Kappa coefficient was calculated to determine interobserver reliability (Cohen, 1960). Observers were required to reach a Kappa coefficient of .80 for hand movement patterns and for characteristics. The coefficients obtained after approximately 12 hours of training were .85 for hand movement patterns and .84 for hand movement characteristics, and the percentage agreements were 88.4 and 92.5 percent, respectively.

Because reading errors were used to calculate reading accuracy, defined in this study as the number of words read correctly per minute, the observers were trained to mark each child's errors using techniques drawn from informal reading inventories (May, 1973; Silvaroli, 1969). Only six types of errors were counted: insertions, omissions, words that had to be pronounced by the examiner, substitutions, mispronounciations, and reversals of word order. The observers coded each word read on three tapes prepared by the investigator that were coded for reading errors and were used as standards. When coefficients were computed for agreement of each observer with the standards and for the observers with each other, one observer attained coefficients of .91, .91, and .98 with the standard; the other attained coefficients of .96, .96, and .92; and the coefficients of interobserver agreement were .84, .95, and .90.

The observers then viewed and coded the videotapes. Each child's reading rate was computed by counting the number of words read during the middle three minutes of each five-minute passage and dividing by three.

Each child's accuracy was determined by counting the errors he or she made during the same middle three minutes of each passage. While listening to the tape, the observers marked each error on a typewritten copy of the passage. The number of correct words per minute was then calculated by subtracting total errors from total words read and dividing by three.

The observers also coded each child's hand movement patterns during the first four minutes of each videotape. In addition, they coded the pattern in use at each six-second interval throughout the four-minute period (total number of intervals 40).

Finally, they noted the inefficient hand movement characteristics that each child was exhibiting at the end of each six second interval on the videotapes. Because each hand was coded separately, as many as 80 characteristics could occur during one four-minute interval.

The data were then rearranged in chronological order by child for each variable in preparation for analysis.

Data Analysis

This section describes the trends evident in the results of individual case studies taken together as a set of replications. The analytic procedures differ for each dependent variable. (The results obtained from each individual child have been reported in Wormsley, 1979).

The number of intervals in which each hand movement pattem was employed by each child at each observation was used to determine the prevailing pattern (i.e., the one the child used for the largest number of intervals during each observation). The results of all eight observations were then examined to determine whether the child had shifted to pattern six after the training.

Even if a shift in pattern did not occur, the other dependent variables were examined for possible effects of the training program. Because the training emphasized smooth tracking, smooth return sweeps, and efficient page turning, the number of inefficient hand movement characteristics might have decreased as a result of training, whether or not a child switched to pattern six. If a decrease did occur, the reading process should have become more efficient and the child's reading rate might have increased. Therefore, each characteristic was graphed for each child, and the graphs were inspected visually to determine whether any characteristics had declined in frequency after the training.

Although all 21 children learned pattern six easily in the 20 lessons, none of them exhibited any changes in hand movement pattern that could be attributed to the training. Furthermore, only nine children demonstrated a reduction in any one inefficient characteristic. No characteristic was decreased by more than seven children. Thus the training program apparently had no effect on inefficient characteristics. It did, however, seem to affect individual children differently since some exhibited fewer inefficient characteristics after training while others did not. The nine children whose inefficient characteristics declined were younger, were more intelligent than the others, were slower readers, and had received fewer years of braille instruction (see Wormsley, 1979).

It seemed conceptually unsound to try to attribute any change in reading variables to a program that did not affect the hand movement variables it was specifically designed to change. Therefore, the data on reading rates were not explored.

Discussion

This study examined training in only one aspect of the braille reading task, the motor aspect, and the results indicated that the training had no appreciable effect on the children's hand movements and thus no effect on their reading.

One could say that although the training period was long enough for the children to learn the hand movement pattern, it was too short for the pattern to become habitual and ensure a transfer effect. (The basic pattern of hand movements was taught in the first five lessons; in the remaining six lessons, the children practiced on materials with different formats.) Because the children completed the first five lessons at different times, practice time was, at most, 3 hours and 45 minutes. Further research will be needed to determine whether, with extended practice, children would incorporate hand movement pattern six into their reading. The few children who used their hands independently during some pretraining observations were the fastest readers, and their perceptual abilities were well-developed. If the training had affected their hand movements at all, it should have reinforced their use of this pattern. However, the results indicate that this did not happen, which strongly supports the suggestion that hand movement pattern is closely tied to perceptual factors of recognition. Therefore, rather than having children spend extended periods practicing an efficient pattern that might not transfer automatically, it might be more beneficial to develop curriculum materials that incorporate this pattern into recognition training.

In addition to its lack of effect on hand movement pattern, the training program also did not cause any dramatic improvement in hand movement characteristics: few children decreased any tracking movements associated with inefficient reading. Therefore, tracking skills also seem to be tied to perceptual factors, and inefficient movements may develop because of perceptual difficulties. Because the children who exhibited fewer inefficient characteristics after training were younger and more intelligent, were slower readers, and had received fewer years of braille instruction, they could have been expected to make more gains in perceptual ability during the school year than would the older children, who would be more proficient in recognizing braille characters and more familiar with the code. The decline in inefficient characteristics among the younger group that apparently occurred because of the training program may have been a result of the cumulative effects of improved perceptual abilities as the year progressed.

Conclusion

Apparently, the hand movements used by braille readers are important from the viewpoint of proficiency in reading in that some patterns are simply not as efficient as others. In addition, tracking skills and use of a particular pattern of hand movements are tied to perceptual ability.

Although several authors have suggested that independent use of the hands should be taught from the beginning of braille reading instruction (Miller, 1955; Olson, 1976; Rodgers, 1961), this study demonstrated that a curriculum to train hand movements may not be effective unless the motor and perceptual tasks involved in braille reading are combined.

One additional factor should be considered when developing a curriculum to teach independent use of the hands: children may use different hand-movement patterns because the perceptual abilities of each hand are different. More research is needed to determine whether children whose hands differ in this respect can be trained to perceive equally well with both hands so that they can learn to use their hands independently when reading braille.

References

Brothers, B. J. Classroom use of the braille code recognition materials. Education of the Visually Handicapped, 1974, 6, 6-13.

Campbell, D. T., & Stanley, J. C. Experimental and quasi-experimental designs for research. Chicago: Rand McNally, 1963.

Cardinale, J. F. Methods and procedures of braille reading. American Foundation for the Blind, Research Bulletin, 1973,26, 171-183.

Cohen, J. A. A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 1960, 20, 37-46.

Foulke, E. Report of the Perceptual Alternatives Laboratory for the Period July l,1972 to June 30, l973. America Foundation for the Blind, Research Bulletin, 1974, 27, 277, 301.

Gibson, J. J. Observations on active touch. Psychological Review, 1962, 69, 477-491.

Gray, P. G., & Todd, J. E. Government social survey, mobility and reading habits of the blind. London, U.K.: Her Majesty's Stationery Office, SS386, 1967.

Hersen, M., & Barlow, D. (Eds.) Single case experimental designs; Strategies for studying behavior change. New York: Pergamon Press, 1976.

Karp, S. An experiment using revised stimulus presentation. American Foundation for the Blind, Research Bulletin, 1962, 2, 12-14.

Kazdin, A. E. Statistical analysis for single-case experimental designs. In M. Hersen & D. Barlow (Eds.), Single case experimental designs: Strategies for studying behavior change. New York: Pergamon Press, 1976, 265-316.

Kratochwill, T. R. Single subject research; strategies for evaluating change. New York: Academic Press, 1978.

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

Lowenfeld, B., Abel, G. L., & Hatlen, P. H. Blind children learn to read. Springfield, Ill. Charles C Thomas, 1969.

Mangold, S. S. Tactile perception and braille letter recognition: Effects of developmental teaching. Journal of Visual Impairment and Blindness, 1978, 72, 259-266.

Maxfield, K. E. The blind child and his reading. New York: American Foundation for the Blind, 1928.

May, F. B. To help children read. Columbus, Ohio: Charles E. Merrill, 1973.

Miller, G. The teaching of braille reading. International Journal for the Education of the Blind, 1955, 4, 69-71.

Mommers, M. J. C. Braille reading: Factors affecting achievement of Dutch elementary school children. New Outlook for the Blind, 1969.

Nolan, C. Y, & Kederis, C. J. Perceptual factors in braille word recognition. New York: American Foundation for the Blind, 1969.

Olson, M. R. Faster braille reading: Preparation at the reading readiness level. New Outlook for the Blind, 1976, 70, 341-343.

Rodgers, C. T. Recommendations for scientific investigations into the problems of touch reading and related subjects. In Report of Proceedings of Conference on Research Needs in Braille. New York: American Foundation for the Blind, 1961.

Silvaroli, N. J. Classroom reading inventory. Dubuque, Iowa: Wm. C. Brown, 1969.

Umsted, R. G. Improving braille reading. New Outlook for the Blind, 1972, 66, 169-177.

Warren, D. H. Blindness and early childhood development. New York: American Foundation for the Blind, 1977.

Williams, M. Braille reading. Teacher of the Blind, 1971, 59, 103-116.

Wormsley, D. P. The effects of a hand movement training program on the hand movements and reading rates of young braille readers. Ann Arbor, Mich.: University Microfilms International, 1980. (Doctoral Dissertation, University of Pittsburgh, 1979).