AUTHOR: Flanigan, Patrick J. PH.D.
TITLE: Automated Training and Braille Reading
CATEGORY: Braille Training, Braille Reading
PUBLICATION: The New Outlook for the Blind, Vol. 60, No. 3, May 1966, 141-146
ABSTRACT: The author compares the reading rates for two groups of students,
one given instructions in reading on a variable speed programmable braille
display. The control group was given no automated instruction.
MANY FUNCTIONAL aspects of the blind are highly dependent upon the adequate development of braille as a tool for the acquisition of information and for the exchange of communication. As a useful tool, braille offers numerous resources through which blind children can participate in a variety of educational settings. In the educational process, braille is presently the basic mode used in establishing lasting academic learning patterns.
Blind individuals, even when classified as excellent braille readers, are at a distinct disadvantage in the amount of reading materials available to them and the excessive amount of time required to read braille materials.5 Despite these drawbacks braille libraries throughout the country report a steady increase in circulation and in borrowers over the past several years.
On the basis of a review of the literature dealing with the
problems of touch reading and related subjects, it can be concluded that: 1) a
considerable amount of literature on the problems of touch (tactual) reading
already exists1,2; 2) some of this literature can form the basis for
a continuing chain of data-gathering on many of the unknown factors involved in
this process; and 3) many sources recommended the need for research studies to
evaluate the reading process of braille when consecutive stimuli are provided
for the finger tips by a continuous moving line of braille.1,2,3
To assure the educational progress of blind children, it is extremely important to increase effectively their speed and comprehension abilities in reading braille.
Presently, many children, aside from the problems of rate, learn to read braille quite well despite the many problems associated with the instructional development of functional braille reading ability. In extending the process of functional braille reading ability, opportunities to significantly improve reading speed and comprehension should be investigated through the development of new techniques.
The purpose of this experiment was to learn whether several of the variables involved in the traditional braille reading process could be influenced by training on an automated self-learning device (braille tape reader). The investigation consisted of 2,100 hours of braille reading for the experimental subjects on the braille tape reader and 2,100 hours of traditional braille reading for the control subjects. Each group of subjects spent three fifty- minute periods per week for fourteen weeks.
The variables involved in the traditional braille reading process investigated were the number of words read and the errors in the reading process, including verbal errors, retracings, additions, skippings, and vertical movements. Utilizing these variables, measurements were undertaken to determine if the training effects on the braille tape reader would be generalized to traditional braille reading materials to which the subjects had not been previously exposed. Also evaluated were the treatment effects (of the braille tape reader) on the experimental subjects in their grade-level reading achievement and comprehension levels. A three-month follow-up evaluation of the tactual and verbal braille reading variables was also undertaken. During this three-month period neither the experimental nor control subjects received any experimental treatment.
For the purpose of this experiment, forty elementary and junior high school students living at the Wisconsin School for the Visually Handicapped, who were classified as functional braille readers were randomly assigned to an experimental or control group. The subjects ranged in chronological age from 146 months (12.1 years), to 232 months (19.3 years and in measured intelligence from IQ 65 to 141. There were no statistically significant differences between the groups in mental age, grade placement or chronological age.
An attrition rate of five subjects from the experimental group and five subjects from the control group occurred during the experiment. The experimental subjects were lost because of eye surgery, a hand injury, and scheduling problems. Three control subjects were lost because of prolonged illness, while the other two were randomly dropped to equalize the experimental and control group cells for statistical analyses of the treatment data. The data describing the subjects in each treatment group are presented in Table I.
(NOTE: TABLE I IS NOT AVAILABLE AT THIS SITE.)
Evaluation Tasks
The following measures were administered prior to the initiation of the experimental treatments, and were repeated at the conclusion of the treatment period and again after a three-month interval during which no instruction was given to either group of subjects:
1) Braille reading tasks of speed and comprehension developed from Science Research Associates reading materials, with the addition of mechanical counting devices for tabulation purposes. Prior to the braille reading task the following instructions were given verbally to each subject:
We would like to find out some things about how you read braille. We are going to give you a story to read [appropriate level material] and as you read out loud to us we are going to watch your finger movements and listen to how you read out loud. These are the things about how you read that we want to know. Read as fast as you can without making mistakes.
2) Wide range vocabulary test produced in braille to measure the changes in the grade level reading achievements of the subjects from pre-to post treatments.
3) Intelligence tests (WISC) (WAIS)
All measures were administered on an individual basis to both the experimental and control subjects in a private testing situation which was free from extraneous distortions. Two examiners were required to administer the braille reading measures of speed and comprehension. One examiner measured the oral responses while the other examiner measured the tactual responses.
The control subjects began, at the first treatment period to
read the traditional braille material, while the experimental subjects received
approximately one hour of instruction on the braille tape reader. The treatment
periods were scheduled at variable times spanning the regular school day to
distribute the number of subjects, both experimental and control, who were going
through the programmed and traditional braille materials at any given time.
Materials
The printed materials programmed* into braille tapes** and traditional braille materials encompassed reading ability levels which extended from beginning reading ability, grade level 2.0, to the proficient reading ability level 13.0. The materials were categorized into graded vocabulary and interest levels for all subjects in the study.
*(Programmed learning as used throughout this paper is defined as "instruction characterized by the controlled presentation of materials, the elicitation of proper response guidance with respect to subject matter, and control of the way in which learning proceeds." Stolurow8 )
** Science Research Associates materials were programmed by the experimenters.
The control group materials were prepared in traditional braille form, while the experimental group materials were cut into strip form, butted together into tapes, and placed on a reel for use with the experimental subjects on the braille tape reader.
This device introduces braille on a tape which moves from right to left across an exposed presentation window. The presentation through which the tapes pass is a 7" X 3/4;" cut 2" X 12" (hand) finger rest plate. The presentation rate of the braille tape may be varied from zero to approximately five hundred words per minute. The device is designed so that the subject's fingers are able to remain stationary.
An endless self-tightening rubber belt pulls the braille tape through the presentation area. On each end of the (hand) finger rest plate is mounted a six-inch reel support aimed at a 45 degree angle. A reel filled with braille tape is fed into the right side, across the presentation area, and wound on the left reel. which is operated by the drive shaft rotating the presentation belt. The drive shaft is operated by a variable speed zeromax transmission which is powered by a quarter-horsepowered electric motor. The entire drive mechanism is mounted in rubber on a 14" X 1/4;" base plate and an aluminum cover that extends 8 inches above the mounting plate and is mounted on the base plate.
The variable speed selector extends up through the top of the aluminum cover and is easily accessible to the operator. Movement of the selector lever from right to left increases the rate at which the braille characters are exposed to the presentation area. A toggle switch is used for starting and stopping the machine and is placed in the upper right end of the machine cover.
In comparing the performance of the experimental and control groups in this experiment before, immediately following the experiment and in a three-month follow-up of traditional braille reading, variables were used.
For the analysis of pre- post- and follow-up comparisons tests of the significance of the differences between the means of the experimental and control groups were performed using a one tailed "t" test. Except for the verbal error variable in which an initial difference necessitated the use of analysis of co variance, analysis of variance was also used to measure the weekly mean correct responses to five verbal questions following each reading exposure by the experimental and control subjects.
Table II presents a summary of the pre treatment, post-treatment, and follow-up comparisons obtained on all variables investigated in this experiment.
The results of the analysis of the treatment data on the speed of traditional braille reading during the course of the experiment indicated significant differences in favor of the experimental subjects, both during the post-treatment comparison (t = 4.15, df 28, P < .05) and follow-up comparison (t = 7.36, df 28, P < .05). A graphic presentation of these results are presented in Figure I.
Errors in the Traditional Braille Reading Process
The data analysis in this regard was concerned with verbal errors, retracings, additions, skippings, and vertical movements utilized in the traditional braille reading process.
Inasmuch as there were initial differences between the groups in the number of errors in the oral reading process, an analysis of covariance was used to equate the groups statistically. From pre-to post treatment this analysis was nonsignificant (f = 3.90, df 1, 28, P > .05). Significant differences were found when the follow-up comparison was made (t = 2.24, df 28, P < .05).
The results of the treatment data on the variable of retracings produced significant differences in favor of the experimental subjects (t = -30.50, df 28, P < .05) at the post-treatment comparison. The follow-up tasks also indicated significant differences in favor of the experimental subjects, (t = -8.12, df 28, P < .05). These data are presented graphically in Figure II.
Nonsignificant treatment effects were noted on the variables of additions and skippings, both at the post-treatment and follow-up comparisons.
Vertical movements were significant (t = 2.45 df 28, P < .05) at the post-treatment comparison but were non-significant at the follow-up comparison (t = 1.06, df 28, P <.05).
Analysis of the treatment data resulting from the effects on reading achievement levels as a function of the method of reading (tape vs. traditional braille reading) produced nonsignificant results (f = less than 1, df 1, 28, P> .05).
Data pertaining to measuring the differences be tween the groups in comprehension abilities as a function of the method of braille reading on a weekly basis also produced nonsignificant treatment effects (f = 1.68, df 1, 28, P > .05).
Since there is a paucity of relevant research available in this area, an attempt was made in this study to look into the broad spectrum of possibilities for adopting these techniques for use by blind children.
The results of this investigation indicate that training in braille reading on an automated self- learning device (braille tape reader) does seem to transfer to traditional braille reading materials geared to individual reading achievement levels and which are self-paced. Although there was no research basis for predicting these differences at the onset of this experiment, it was anticipated that the principles of programmed instruction applied in this manner would contribute towards elevating the experimental subjects' rate of reading traditional braille materials.
Analysis of the results did indicate significant differences ferences in the rate of traditional braille reading speed, both at post-treatment and follow-up comparisons between the groups. The positive gains the experimental subjects made during the treatment periods remained constant during the noninstructional period.
The results obtained may be attributable to several factors. Namely that the tactual or anticipation scanning ability4 and certain finger movements involved in the braille reading process were modified through training on the braille tape reader. The tactual scanning ability did transfer to the traditional method of reading braille. Increasing these aspects of braille reading may contribute to the prevention, elimination, or amelioration of either existing conditioned responses in reading or the development of negative habits in reading braille.
Mention must be made of the uncontrolled variable of motivation which may partially account for the decrement in performance on the part of the control subjects from pre-to-post comparisons. During the noninstructional period their performance remained fairly constant. Future study in this area is certainly warranted.
Both the experimental and control subjects made significant gains in reading achievement levels during the course of the treatment periods. This result substantiates the recommendations of Nolan,7 Ashcroft,3 Meyers, 6 and numerous other authors in regard to the effectiveness of providing appropriate level material for braille readers. In this study a significant relationship was obtained for both the experimental and control subjects and the braille reading materials transcribed and used in this experiment. The materials were geared to the individual reading level of each subject, regardless of grouping. Progression through the reading materials was self-paced and evaluations took place at the completion of each program frame. We should all take a critical look at the kinds of reading materials presently used in all educational settings for blind children.
The results obtained in this experiment lead to many possible suggestions for future research purposes. Inasmuch as this study covered the broad spectrum of possibilities for the adoption of programmed techniques to the blind, numerous specifics of the braille system, programmed reading materials, and subject adaptation to learning devices were not focused on. These areas provide the basis for future research.
The nature and size of the sample is much too limited to attempt to make generalizations, yet the results may present some answers to previously unknown factors in the braille reading process.
The braille reading behavior of fifteen experimental and fifteen control subjects was compared as a function of training by the experimental subjects on an automated self-learning device and control subjects reading traditional braille. Statistical analyses of the data indicated that 1) the performance of the experimental subjects was significantly superior to that of the control subjects in the comparison of certain variables involved in the traditional braille reading process following treatment exposure; 2) following a noninstructional period the experimental gains, with minor exceptions, remained constant; 3) nonsignificant treatment effects were noted on comprehension and reading achievement levels effects as a function of the method of reading; and 4) the appropriateness of sequenced grade level reading materials for reading instruction was pointed out. Suggestions for future research in this area were given.
1. American Foundation for the Blind: The Pine Brook Report Revised Edition, National Work Sessions of the Education of the Blind with the Sighted in Public and Private Schools, Group Reports, No.2, New York: The Foundation, 1957.
2. _________________: The Report of Proceedings of Conference on Research Needs in Braille, 1961.
3. Ashcroft, S. C.: "Errors in Oral Reading of Braille at Elementary Grade Levels," Unpublished doctoral thesis, University of Illinois, 1960.
4. Hildreth, Gertrude Howell: Teaching Reading: A Guide to Basic Principles and Modern Practices. New York: Henry Holt and Company, 1958.
5. Lowenfeld, B. Braille and Talking Book Reading, New York: American Foundation for the Blind, 1945.
6. Myers, E., Ethington, D., and Ashcroft, S. C.: "Readability of Braille as a Function of Three Spacing Variables," Applied Psychology 42: 163-165, 1958.
7. Nolan, C. Y.: "A Comparison of Learning Achieved Through Braille Reading and Listening," Louisville: American Printing House for the Blind, 1962, (unpublished report).
8. Stolurow, L. M.: "Teaching Machines and Special Education," Educational and Psychological Measurement. 20: 429-448,1960.