Short Reports

Ease of Learning the Braille and Fishburne Alphabets

S.E. Newman; A.D. Hall

Abstract: Subjects (N= 72) studied the braille or Fishburne alphabet for eight minutes and were then tested for recall. More of the Fishburne than of the braille items were recalled, independent of the arrangement of the items. An implication for instructing visually impaired people is discussed.

At the present time, there is little research based information about the learning of braille or of other touch systems used for communication with blind people (Caton, Pestor, & Goldblatt, 1979; Foulke, 1982; Harley, Henderson, & Truan, 1979; Loomis & Lederman, 1986; Lorimer, 1982; Olson, 1981). The present experiment compared the amount learned from the braille alphabet (Figure 1) and from the Fishburne alphabet (Figure 2) within a fixed period of time. In addition, the effect of differences in arrangement of the items was studied. It was expected that the learning of braille would be facilitated if the symbols were arranged to enhance the likelihood that the relationship between items in the A-J set and their corresponding items in the K-T and U-Z sets would be identified (e.g., that adding dot 3 to 'A' gives 'K,'and that adding dots 3 and 6 to 'A' gives 'U'). It was expected, also, that the learning of the Fishburne alphabet would be facilitated if the symbols were arranged to enhance the likelihood that the left-to-right progression within each of the four six-item sets (i.e., A-F, G-L, M-R, and S-X) would be identified (i.e., that within each set the left-to-right progression is 1-item top, 1-item bottom, 1-item top and bottom, 2-items top, 2-items bottom, 2-items top and bottom).

Subjects

Subjects (N= 72) were sighted male students enrolled in the introductory psychology course at North Carolina State University. They were assigned to conditions through the use of a counterbalancing procedure. All subjects were run individually.

A B C D E F

G H I J KL

M N 0 P Q R

S T U V W X

Y Z

Figure 2. The Fishburne alphabet

Procedure

Each subject was given a study sheet with either the braille symbols and their letter names, or the Fishburne symbols and their letter names, arranged either in the three rows, A-J, K-T, and U-Z or in the five rows, A-F, G-L, M-R, S-X, and Y-Z. Each subject was also given a blank sheet of paper and a pencil to use while studying the symbols. Subjects were told that they had eight minutes to study the symbol-letter pairs and that they would then be tested. On the test, each symbol was presented for ten seconds, and the subject was to say its letter name. Two different test orders were used; thus, the items were presented on the test in one random order for half of the subjects in each of the four treatments, and in a different random order for the remaining subjects. All subjects studied the items visually and were tested visually. Following the test, subjects were asked how they had tried to learn the items and to recall them. They were then debriefed and dismissed.

Results

A 2 x 2 between-subjects analysis of variance of the number of items correct indicates a significant effect for type of alphabet (F[1,68] = 12.55,P<.0l, but not for arrangement or for the interaction [p >.05] ). Performance was better for those who studied the Fishburne alphabet than for those who studied braille. Table 1 shows the mean percent correct for the four conditions.

A B C D E F G H I J

K L M N 0 P Q R S T

U V W X Y Z

Figure 1. The braille alphabet

The number correct for each of the 26 symbols was determined for each condition. Rank-order correlations were then done for each pair of conditions. The correlations are substantially higher (.731 and .789 [p<.0l for each] ) for groups which studied the same alphabet (i.e., braille 3-rows with braille 5-rows, and Fishburne 3-rows with Fishburne 5-rows), than for groups which studied different alphabets (e.g., braille 3-rows with Fishburne 3-rows).

Table 1. Mean percent correct in each condition.

Alphabet Three rows Five rows

Braille 45.1 45.7

Fishburne 57.9 66.9

Of the latter correlations, two.386 and .373-are significant (P< .05) and two-. 306 and .289-are not. Thus, as would be expected, the order of difficulty of the items is more similar for subjects who studied the same alphabet than for those who studied different alphabets.

A significant correlation, .789 (p <.0l), was also obtained for the number correct for the combined braille groups and the number correct on an identification task in which the braille symbols were visually examined (Loomis, 1982). Correlations for combined braille group performance with that on three other haptic tasks-immediate memory (Newman, Hall, & Gupta, 1983), judgment of dot numerosity (Newman & Hall, 1985), and identification (Loomis, 1982)-were .479(p<.0l), .408(p<.05), and .405 (P<.05), respectively. These results suggest that the order of difficulty of braille symbols is similar across a variety of tasks, both visual and haptic.

There are two other findings of some interest. When subjects were asked to describe the method they had used to learn the items, there was a marked difference between alphabet treatments, in the number of subjects who referred to the structure of the alphabet they had studied (e.g., one of the braille subjects mentioned that the symbols for A-J did not have dots in the bottom row whereas the symbols for the other letters did; and some of the Fishburne subjects mentioned learning the key symbol for each of the symbol sets). Only 4 of the 36 who studied braille referred to the structure, for Fishburne, 32 of the 36 did. Chi-square for those frequencies is highly significant (p<.001). Contrary to our expectation, however, the tendency to note the difference was not a function of the way in which the items were arranged on the study sheet.

A similar finding emerged in response to the question about how subjects had tried to recall the items. Again, there was a marked difference between alphabet treatments (3 for braille and 22 for Fishburne), chi-square for which was again highly significant (p<.001). Again, also, the arrangement of items during the study trials was found to have no effect.

Discussion

The amount learned from the Fishburne alphabet was greater than from the braille alphabet, independent of the arrangement of the items. Since the items were studied visually, these results suggest that the comparatively easier learning of the haptically examined Fishburne as contrasted with the braille items (M. Dixon, personal communication, May 3, 1985; Shafrath, 1986; Young, 1979) may be a function not only of greater haptic discriminability of the Fishburne items, but of differences in the structure of the two alphabets. Whether differences due to structure would occur were the symbols to be examined haptically is, however, still to be determined.

There are at least two ways that the Fishburne system may prove useful to the visually impaired population (Shafrath, 1986; Young, 1979). They are: 1) as an alternative to braille for labeling items for personal use, specially for persons with poor finger sensitivity and 2) as a preliminary task for those who find the learning of braille too difficult. In more than one instance, persons who had become discouraged with their slow progress in learning braille were switched to Fishburne, which they then learned quite easily (M. Dixon, personal communication, May 3, 1985). When switched back to braille, they were more successful than they had been before. The success enjoyed in learning the Fishburne system appears to demonstrate that the subjects were able to learn haptically a symbol system that could be of value to them. Thus, they were more willing to try braille again and to persist at it. Since these reports are anecdotal, more systematic study of this use of the Fishburne system appears to be indicated.

References

Caton, H., Pester, E., & Goldblatt, S. (1979). Specificationsfor selecting vocabulary and teaching methods for beginning readers. New York: American Foundation for the Blind.

Foulke, E. (1982). Reading braille. In W. Schiff & E. Foulke (eds.), Tactual perception: A sourcebook. Cambridge: Cambridge University Press.

Harley, R.K., Henderson, E.M., & Truan, M.B. (1979). The teaching of braille reading. Springfield, IL: Thomas.

Loomis, J.M. (1982). Analysis of tactile and visual confusion matrices. Perception and Psychophysics, 31, 41-52.

Loomis, J.M. & Lederman, S.J. (1986). Tactual perception. In K. Boff, L. Kaufman, & J. Thomas (eds.), Handbook of Perception and performance. New York: Wiley.

Lorimer, P (1982). The braille code and the teaching of braille reading: An annotated bibliography. London: Royal National Institute for the Blind.

Newman, S.E. & Hall, A.D. (1985). Some factors affecting numerosity judgments in haptically-examined braille symbols. Paper presented at the annual meeting of the Eastern Psychological Association, Boston.

Newman, S.E., Hall, A.D., & Gupta, V. (1983). Immediate memory for visually- and haptically-examined braille symbols. Paper presented at the annual meeting of the Psychonomic Society, San Diego.

Olson, M.R. (1981). Guidelines and games for teaching efficient braille reading. New York: American Foundation for the Blind.

Shafrath, M. R. (1986). An alternative to braille. Journal of Visual Impairment & Blindness, 11, 955-956.

Young, P.S. (1979). A new alphabet for the blind. The Record, 42.

We appreciate the help of Miriam Dixon, instructor at the North Carolina Rehabilitation Center for the Blind, who introduced us to the Fishburne system and educated us about it, and her comments on an earlier version of this paper. We appreciate also the help of George Dexter in conducting an exploratory study and of Nevine El-Shiekh, Darryl Belle, and Maxine Shaw in gathering the data for the experiment reported here. A paper describing the results of this study was presented at the annual meeting of the Eastern Psychological Association in New York City, April, 1986.

Slater E. Newman, Ph. D., professor, Department of Psychology, North Carolina State University, Raleigh, NC 27695; Anthony D. Hall, Ph.D., development manager, LAN information department, IBM Corporation, PO. Box 12195, Department E 33, Building 656, Research Triangle, NC 27709.