Communication and Handicap: Aspects of
Psychological Compensation and Technical Aids, pp. 233-251
E. Hjelmquist and L.-G. Nilsson (editors)
(c) Elsevier Science Publishers B.V. (North-Holland), 1986

BLISSYMBOLICS, COGNITION, AND THE HANDICAPPED

PAUL MUTER

Psychology Department, University of Toronto, Toronto, Canada

This work was supported by Research Grant UO149 from the Natural Sciences and Engineering Research Council of Canada to the author. I thank Peter Graf, Peter Reich, and Amanda Walley for helpful comments.

Advantages and disadvantages of alphabetic and logographic writing systems are discussed, with particular emphasis on Blissymbolics as an example of a logographic system. It is concluded that logographic systems might be easier to learn than alphabetic systems and, in particular, that Blissymbols could be a useful communication device for people with various kinds of handicaps. This suggestion is substantiated by empirical research.

1 INTRODUCTION

There are tens of thousands of handicapped people in dozens of countries whose ability to communicate has apparently been enhanced by the use of Blissymbolics. Blissymbolics is an international laogography composed of pictographs, in which symbols represent words in a pictorial manner; ideographs, in which symbols suggest the meaning of word, though not necessarily in a pictorial manner; and a few arbitrary symbols such as "+ and "?". The purpose of the present paper is to briefly describe Blissymbols and their use by handicapped people; to review data bearing on the issue of whether Blissymbols are useful; and to discuss some theoretical issues relating to writing systems, cognitive processes, and the handicapped.

2 CHARLES BLISS AND BLISSYMBOLICS

Charles K. Bliss was born Karl Blitz in 1897 in Austria, where 20 different languages were spoken. He was incarcerated in a Nazi concentration camp during World War Two, and became convinced that major contributors to war and international conflict were differences and ambiguities in language. Later in the war, Bliss was incarcerated in Shanghai, and was profoundly influenced by Chinese characters. Since Chinese characters are essentially not sound based, Chinese people speaking many different languages and dialects can communicate easily with each other.

After World War Two, Bliss moved to Australia, where he worked in an automobile assembly plant. In his spare time, he developed a new nonalphabetic writing system, called Semantography, later Blissymbolics, based on approximately 100 basic shapes. In 1949, he completed a book Semantography, describing the writing system, but in spite of endorsements from Julian Huxley, Bertrand Russell, and others, he was unable to find a publisher. Finally, he published it himself (Bliss, 1965).

Bliss designed Blissymbolics for use among all people, and did not create them with handicapped people in mind. However, in 1971, the Ontario Crippled Children's Centre in Canada began using Blissymbols to help pre-reading children communicate. In 1975, the Blissymbolics Communication Foundation, now called the Blissymbolics Communication Institute, was established in Toronto, Ontario. In cooperation with Charles Bliss, the Institute has developed and disseminated Blissymbolics. Charles Bliss died on July 13, 1985. For further information on Charles Bliss, the Blissymbolics Communication Institute, and Blissymbols, see McDonald (1980).

The pictographic, ideographic, and arbitrary aspects of Blissymbols are all exemplified in Figure 1.

The symbols for table and tree are pictographic: The symbols visually resemble the referents. The symbol for wind is ideographic: The top horizontal line represents the sky; the diagonal and the bottom line together represent a nose; sky and nose together indicate air; the combination of air and forward (the arrow) denotes wind. The arbitrary elements 2 and 7 are included in the symbols for summer and week.

Blissymbols were not intended to be self-explanatory, though Bliss felt that the meanings of the symbols would be easy to remember once an explanation had been given. (It is probably unrealistic to expect that a reasonably comprehensive logography could be directly understandable without some learning; see Kolers, 1969.) Explanations for approximately 1400 Blissymbols are provided in Blissymbols for Use (Hehner, 1980). Following are explanations, based on those in Hehner, for the symbols in Figure 1.

son : protection + male: a small male born under the family roof

song: mouth + musical note

star: pictographic

store: room that opens onto the street + business (money plus to hold)

summer: sun + second (season) + hot (fire)

table: pictographic

time: pictographic: the symbol looks like a clock face

tree: pictographic

week: 7 + day (sun + earth: the sun appears over the horizon; it shines on the earth during the day)

weight: pictographic: the symbol looks like scales

wind: air + forward: air that moves forward

word: part of + language (mouth + ear: what is spoken and heard)

Note that some of the explanations are culture specific. For example, the explanation of the symbol for time would be unhelpful for someone who had never seen a clock.

There is in principle no limit to the number of symbols that can be created from the basic shapes. The system is generative: Shapes and symbols can be combined and recombined indefinitely according to certain rules. All of the symbols in Figure I represent nouns, but of course other parts of speech are included in Blissymbolics, and some of these are achieved by the use of indicators. For example, the action indicator (^) above a symbol indicates action taking place in the present (often a present tense verb). A small right parenthesis, resembling a parabolic mirror, above a symbol indicates action in the past (past tense verb). A small left parenthesis above a symbol indicates action in the future (future tense verb). A small multiplication sign, X, above a symbol indicates the plural. Bliss also designed a simple syntax, somewhat different from the syntax of English; see McDonald (1980, pp. 60-68) for a description.

Blissymbols are standardized. Lack of standardization has been a problem with other logographies. For example, Chinese characters were originally much more pictographic and ideographic than they are now, but they have evolved over the centuries, are now very stylized, and have lost virtually all of their visual resemblance to the referents. To avoid a similar problem, strict standards for composing Blissymbols are prescribed by the Blissymbolics Communication Institute. With computer technology, presenting and standardizing Blissymbols is relatively easy.

There have been several computer implementations of Blissymbols. For example, Sawchuck and Bown (1977) delivered Blissymbols via Fortran and a DEC PDP11. Their software permitted selection of Blissymbols to display sentences, storing and retrieving of sentences, "turning the pages" of a symbol dictionary, and the creation of new symbols for addition to the dictionary. Giddings, Norton, Nelson, McNaughton, and Reich (1979) devised a Z80-based system which permitted interactive use of Blissymbols on a home television set. Carlson, Granstrom, and Hunnicutt (1982) devised a multi-language, portable microcomputer-based system which can transform input from a Bliss communication board into either alphabetic or synthesized speech output. Lexical prediction capability has recently been added to this system (Hunnicutt, this volume). Finally, Blissymbols can be delivered on the screen or printer of an Apple II+ or Apple IIe microcomputer by means of software created by the Minnesota Educational Computing Consortium (1983). These symbols can be displayed under program control (Applesoft Basic).

There are a variety of more traditional means of presenting and using Blissymbols. For example, Blissymbol stamps and flash cards (3 inches square) are available, as are templates for drawing Blissymbols. A popular method of using Blissymbols is by means of display boards containing from 30 to 512 Blissymbols to which people can point. Typically, the corresponding alphabetic word appears under the Blissymbol.

3 ARE BLISSYMBOLS EASY TO LEARN AND USE?

The issue of whether Blissymbols facilitate communication among the handicapped is a special case of the more general question of whether people are better adapted to learning to read alphabetic codes or logographies. Both the general and the specific question will be considered in this section. (It is possible that people are typically best suited for syllabaries, in which each symbol represents a syllable, but discussion of syllabaries is beyond the scope of this paper.)

It is estimated that human language originated approximately one million years ago. Writing systems originated five or ten thousand years ago. Approximately two-thirds of the world's languages have no writing system, and new writing systems are being introduced every year (Grimes & Gordon, 1980). It has been predicted (e.g., Thompson, 1979, 1983) that, because of advances in computer technology, logographies may become more widespread, and may even displace alphabetic systems. On the other hand, the Chinese may switch from a logographic system to an alphabetic system. In spite of these considerations, it is unknown what the optimal writing system is for human users, or which system is optimal under what circumstances.

3.1 Armchair Considerations

A great deal of space in books and journals has been devoted to discussion of the advantages and disadvantages of alphabetic and logographic writing systems. (For a succinct summary of the main arguments, see Carroll, 1972). Indeed, some advantages and disadvantages are reasonably clear, and do not require empirical support. For example, with alphabets, a smaller number of symbols must be learned; a reader can, in principle at least, decode a word that he or she has heard but not seen before; and the use of a typewriter is easier, at least with present technology. (Developments in automatic speech recognition and other input media may soon render this last consideration irrelevant.) With logographies, the same system can be used with different dialects or even different languages; a reader may be able to guess the meaning of a word that he or she has neither seen nor heard before; and distinguishing among homophones is not a problem. These armchair considerations are not conclusive in support of either alphabetic codes or logographies. Neither is crosscultural evidence.

3.2 Crosscultural Evidence

Gray (1956) studied readers all over the world and concluded that "the general nature of the reading act is essentially the same among all mature readers," regardless of the writing system. However, it has frequently been stated that low literacy rates in some areas of Asia are due to the use of logographies (e.g., Goody, 1968). The opposite conclusion was reached by Makita (1968) and others. Makita's evidence suggested that in certain cultures in which logographies are prominent, for example, Japan, the incidence of serious reading problems is spectacularly low, less than one percent, compared to Canada and the United States, where the incidence of serious reading problems is estimated to be approximately 10 to 20 percent (Gibson & Levin, 1975). Stevenson, Stigler, Lucker, and Lee (1982) made a more concerted attempt to fairly compare rates of reading disability in logographic and alphabetic cultures, and they found essentially no difference.

Unfortunately, for present purposes crosscultural evidence is quite useless, in my opinion, because the cultures in question differ in so many ways in addition to their writing systems.

3.3 Clinical Evidence

Included among groups that have used Blissymbols to communicate are the following: physically handicapped, retarded, multiply handicapped, autistic, aphasic, and adult stroke patients (McNaughton, 1978). Several studies with these groups will now be briefly described.

McNaughton and Kates (1975) estimated that there are over 50,000 non-verbal cerebral palsied people in the United States. Speech is prevented or severely impaired in this group because of disturbances of the respiratory, phonotory, and articulatory systems. Blissymbols were first used with this population as a means of bridging the gap between pictures and words. That is, many non-verbal motorically impaired children can communicate by pointing to pictures, but cannot read, and therefore are unable to participate in complex communication. After three years of using Blissymbols with children of near normal or above average intelligence, the conclusions of McNaughton and Kates included the following: Blissymbols were effective both as a supplement and as a substitute for speech; Blissymbol use encouraged, rather than discouraged, speech and vocalization; the symbols allowed children to communicate with a wide range of people; the use of Blissymbols resulted in improvement of assessment in the areas of hearing, language, psychology, and education; and the children acquired greater self-confidence.

In another study involving the non-verbal motorically impaired, Hammond and Bailey (1976) introduced Blissymbols to four children: three quadriplegic athetoids, and a mild spastic with severe dysarthria. They were of average or low average intelligence. Hammond and Bailey observed that these children became more alert and aware of their surroundings after the introduction of Blissymbols. Their horizons appeared to have widened considerably. Two of the children began vocalising more freely.

The use of Blissymbols with mentally retarded nonverbal children with cerebral palsy was described by Harris-Vanderheiden, Brown, Reinen, MacKenzie, and Scheibel (1975). Results indicated that "Blissymbols were effectively implemented as a means of respondent and limited expressive communication for this population" (p. 36).

Song (1979) attempted to extend this result to four severely mentally retarded adolescents. Song concluded that Blissymbols were useful if and only if the person has the desire to communicate, can respond to the Peabody Picture Vocabulary Test, and learns Blissymbols quite easily in the early stages.

House, Hanley, and Magid (1980) taught 10 nonreading trainable mentally retarded adults 16 logographs (specially created pictographs and ideographs). The subjects were able to learn the logographs and construct sentences out of them.

A rare negative outcome was obtained by Calculator and Dollaghan (1982). They observed the use of Blissymbol communication boards by seven nonspeaking, nonambulatory, severely mentally retarded people interacting one-on-one with teachers. The results were that these people rarely used their Blissymbol boards in spontaneous interactions: 79% of messages were via a nonboard mode. Furthermore, use of the board did not increase the probability of message success and did not decrease the ambiguity of the message.

3.4 Experimental Evidence

While the clinical evidence suggests that Blissymbols can be useful, it is by no means conclusive. In none of the above studies were control groups employed. Fortunately, there is some experimental evidence available.

Regarding the more general issue of whether logographies are easier to learn than alphabetic codes, Rozin, Poritsky, and Sotsky (1971) taught eight second-grade children with "clear reading disability" 30 Chinese characters in a few hours. Parallel tutoring in English reading yielded little progress. The control condition was not rigorous, and only simple concrete words were used in the study, but the results may have been attributable to the nature of the writing systems.

Clark (1981) directly compared learning of Blissymbols and learning of alphabetic codes in preschool, nonreading, English-speaking children, ages 4.3 to 5.4, with no apparent handicaps. One group of nine children attempted to learn 15 Blissymbols, and another group attempted to learn 15 words written alphabetically in English. On the final test, mean performance was 81% in the Bliss condition, and 23% in the alphabet condition, a statistically significant difference. In two other logographic conditions, Carrier symbols (noniconic) and pictographic symbols were also learned better than alphabetic words.

Muter and Johns (1985) conducted four experiments addressing this issue. English-speaking university students learned to identify or to extract meaning from various kinds of symbols over several sessions. In the alphabet condition, subjects learned to read English words written in an unfamiliar alphabet (using Devanagari or Shaw characters), much as students in English-speaking countries do in grade 1. Subjects were explicitly told that, in this alphabet condition, a particular unfamiliar character would always represent the same sound. In Experiment 1, subjects had one training session. In subsequent experimental sessions, on each trial subjects saw a question, e.g. "WHICH REQUIRES MORE PHYSICAL WORK". Then two symbols from the same condition, e.g., two Blissymbols, appeared side by side, and the subject pressed a key indicating which of the two symbols provided a better answer to the question. Feedback was presented, and all of this information then remained on the screen for several seconds. Thus, each trial was a study trial as well as a test trial. Subjects gradually learned 30 words in each condition. In 28 50-minute experimental sessions, the results were very clear for all four subjects: Both speed and accuracy of performance were dramatically and reliably superior in the Bliss condition than in the alphabet condition. (Performance in a Chinese condition was better than in the alphabet condition and worse than in the Bliss condition.) Furthermore, transfer to a new vocabulary was worse in the alphabet condition than in the Bliss condition, in spite of the fact that in the alphabet condition, the grapheme-phoneme correspondences remained identical to the ones that subjects had experienced for 28 sessions.

The investment of learning an alphabetic code may not pay off until vocabulary size reaches a certain criterion. In Experiment 2 of Muter and Johns (1985), subjects learned 240 words per condition. In addition, a naming task was used instead of a meaning extraction task. Pronouncing a word without recognizing it was thus possible in the alphabet condition, but not in the Bliss condition. In spite of this, performance was again much superior in the Bliss condition. In Experiment 3 of Muter and Johns, the results of Experiment 1 were replicated with no training session.

In Experiments 1 to 3 of Muter and Johns, the mapping between graphemes and phonemes was perfectly consistent, in marked contrast to English. (Berdiansky, Cronell, and Koehler, 1969, concluded that 166 rules for grapheme-phoneme correspondence were necessary to account for 90 percent of words in some children's books written in English.) In Experiment 4, an attempt was made to approximate the inconsistency of grapheme-to-phoneme mapping that exists in English. Consistent with some crosscultural evidence (e.g., Kyostio, 1980), the inconsistent condition produced dramatically worse performance than the consistent condition. Apparently, the results of Experiments 1 to 3 of Muter and Johns would have been even more dramatic if the mapping between graphemes and phonemes had been inconsistent, as it is in English.

Thus, under a reasonably wide range of conditions, Muter and Johns found that logographic writing systems, particularly Blissymbolics, were substantially easier to learn to read than alphabetic writing systems. This result was obtained in spite of the fact that the subjects were literate, English-speaking adults, and had long since attained "linguistic awareness" (Liberman, Liberman, Mattingly, and Shankweiler, 1980; Mattingly, 1972): They were intimately familiar with the general ways in which alphabets work.

4 WRITING SYSTEMS AND COGNITION

Downing (1973) has argued that comparative reading, i.e., the study of people reading different languages and writing systems, is likely to increase understanding of the reading process and cognition.

We have seen that there is evidence that human performance seems to be different with different writing systems. There is also evidence that the underlying cognitive processes may be different with different writing systems. For example, Biederman and Tsao (1979) found a larger Stroop effect (1935) with logographs than with nonlogographs. According to results of Park and Arbuckle (1977), logographs produce better recall and recognition, whereas a sound-based writing system produced better performance in paired-associate learning and in serial learning. Biederman and Tsao (1979) and Park and Arbuckle (1977) concluded that the underlying cognitive processes are fundamentally different with different writing systems. After reviewing the literature, Hung and Tzeng (1981) also concluded that processing varies as a function of the writing system, though only for lower-level cognition. Other tentative evidence indicates that under certain circumstances, sound-based words yield a right visual field advantage, whereas logographs yield a left visual field advantage (Hatta, 1977) and that lesions in different areas of the brain tend to affect the use of sound based systems and logographies differentially (Sasanuma, 1974).

Blissymbolics and alphabetic codes differ in many ways, including the following:

Most Blissymbols are pictographic or ideographic; alphabetic words are not.

Alphabets are sound based; Blissymbols are not.

The general visual configuration (Brooks, 1977) is different in the two systems.

Blissymbols are glyphic; the elements of alphabetic words are arrangedlinearly.

The elements of Blissymbols are simple; alphabetic characters are complex.

Typically, literate adults in the west will have already learned an alphabetic code, but they will not have learned a logography.

Blissymbols may be more likely to induce holistic processing; alphabets may be more likely to induce analytic, rule-based processing.

The first step in determining what the obtained performance differences tell us about the reading process, is to establish which of the above differences account for the performance differences. Some of the possibilities are discussed in the remainder of this section.

4.1 Iconicity

Experiments by Clark (1981) and Muter (1985) suggest that iconicity (pictographic and ideographic properties) is important. In the Clark study, briefly described earlier, some children learned 15 Carrier symbols, which are logographic but visually meaningless, some learned 15 Blissymbols, and some learned 15 pictographs. Performance was correlated with iconicity: Percentages correct on the criterion measure were 48.1, 81.5, and 96.3 in the Carrier, Bliss, and pictograph conditions respectively.

Muter (1985) compared the learning of Blissymbols to the learning of specially created Pseudo-Blissymbols, which were similar to Blissymbols in most respects, but were not iconic. Pseudo-Blissymbols were constructed by means of a computer program using the same shapes used as building blocks for the Blissymbols, but combining them randomly according to the probability distributions used in creating Blissymbols. Performance with the noniconic PseudoBlissymbols was substantially and reliably inferior to performance with iconic Blissymbols.

Perhaps iconic symbols can be remembered better than noniconic symbols from one trial to the next because they are more readily processed to the semantic level (Craik & Lockhart, 1972). A second possibility is in terms of transfer in associative learning. To the extent that iconic symbols are in some sense similar to the corresponding visual stimuli in the real world, the iconic conditions could be regarded as the transfer task in an A-B, A'-B paradigm. This paradigm typically produces greater positive transfer than the A-B, C-B paradigm (Kausler, 1974), which corresponds to the noniconic conditions.

4.2 Holistic Versus Analytic Processing

According to Allport (1979), there are two possible mechanisms in alphabetic word recognition: analytic, rule-based translation from graphemes to sounds, and access to the lexicon via a more holistic visual process. Some evidence (e.g., Barron, 1980; Bryant and Bradley, 1980, Frith, 1979) suggests that people typically read by means of a holistic method, even when reading an alphabetic code. If word recognition is holistic, a logography should be more optimal for reading than an alphabetic code.

With regard to some handicapped people, the holistic-analytic dimension may be particularly relevant. There is some evidence (Kemler, 1983) to suggest that retarded people tend to perceive objects as wholes to a greater extent than people of normal intelligence. If this is true, then logographies may be particularly advantageous for this population. Similarly, children appear to be more inclined than adults to process stimuli holistically (e.g., Smith & Kemler, 1978), and writing systems are normally learned in childhood.

4.3 Phoneme Processing in Reading

Why are alphabetic codes difficult to learn? According to Rozin and Gleitman (1977), it is because phonemes are difficult to isolate. Rozin and Gleitman agree that the brain extracts phonemes from the sound stream for the purposes of speech processing, but they argue that phoneme-processing mechanisms are not easily available for the purposes of reading. A problem with this line of reasoning is that the subjects in the experiments of Muter and Johns (1985) were literate English-speaking adults, who had already demonstrated that they could access the phoneme-processing mechanisms. (Even if literate English-speaking adults typically use a holistic method in reading, they generally are able to use a rule-based method; e.g., they can correctly pronounce legal nonwords.) If gaining access to the phoneme-processing machinery is an impediment to learning to read an alphabetic code, apparently it is not the only impediment.

Similarly, the theory of Rozin and Gleitman has difficulty explaining the results of Brooks (1977). Brooks' subjects (also literate English-speaking adults) attempted to learn a new alphabetic code, and could often correctly name all of the letters of a word without being able to recognize the word, could name all of the letters of a word faster than they could recognize the word, and could sometimes pronounce the word correctly without recognizing it. In other words, these people had difficulty learning to read words written in an unfamiliar alphabetic code despite the fact that they were able to isolate phonemes.

What if the sound based properties of an alphabet are dispensed with, and alphabetic words are treated by the reader as arbitrary logographs? Brooks and Miller (1979) conducted an experiment in which subjects learned words written in an artificial alphabet, and at the same time learned to read words written essentially with random strings of characters. When subjects were not informed of the grapheme-phoneme mappings, and later reported no awareness of them, they performed better in the alphabet condition than in the random condition, but when subjects were informed of the grapheme-phoneme mappings, they performed worse in the alphabet condition than in the random condition. Thus a conscious attempt to use mapping rules was maladaptive, but the mapping was useful if used only implicitly.

Brooks and Miller used a very small vocabulary size: six words per condition. Muter (1985) conducted a similar experiment with 120 words per condition over nine 50-minute sessions. Subjects performed consistently better in the alphabet condition than in the random condition, despite the fact that they were fully informed of the natures of the conditions. Again, the investment entailed in learning grapheme-phoneme correspondence rules may reliably pay off only if vocabulary size surpasses a certain criterion. At least with a reasonably large vocabulary size, alphabetic strings regarded as arbitrary logographs are apparently the most difficult words to learn of any kind discussed in the present paper. Even though isolating phonemes in learning to read alphabetic codes is difficult, bypassing the sound based property of alphabetic codes apparently does not render them easier to learn.

5 IN CONCLUSION

The relative utility of logographies and alphabetic codes may depend on the language under consideration. For example, the desirability of an alphabetic system is probably dependent on the number of phonemes in the language in question. The desirability of a logography is probably dependent on the number of homophones in the language in question. (Logographs disambiguate homophones.) In addition, the optimal writing system for input may not be the optimal system for output. However, alphabetic codes seem to be extremely difficult to learn to read. Millions of people attempt to learn to read alphabetic codes and fail. Millions of others succeed, but at the cost of a large amount of work. The weight of the evidence suggests that logographies are easier to learn to read than alphabetic codes, and that Blissymbols are a useful communications device, particularly for people with various kinds of handicaps.

This evidence comes from studies in which comparisons are made between conditions that differ in many ways (e.g., the alphabet condition and the Bliss condition in Muter & Johns, 1985). These differences by and large reflect intrinsic differences among writing systems in general. Iconicity apparently accounts for some of the advantage of Blissymbols, but more research is needed to explicate the reasons for different performance and processing with different writing systems.

REFERENCES

Allport, A. (1979). Word recognition in reading. In P.A. Kolers, M.E. Wrolstad & H. Bouma (Eds.), Processing of visible language 1. New York: Plenum Press.

Barron, R.W. (1980). Visual and phonological strategies in reading and spelling. In U. Frith (Ed.), Cognitive processes in spelling. New York: Academic Press.

Berdiansky, B., Cronnel, B., & Koehler, J. (1969). Spelling-sound relations and primary form-class descriptions for speech-comprehension vocabularies of 6-9-year-olds. Southwest Regional Laboratory for Educational Research and Development, Technical Report, No. 15.

Biederman, I. & Tsao, Y. (1979). On processing Chinese ideographs and English words: Some implications from Strooptest results. Cognitive Psychology, 11, 125-132.

Bliss, C.K. (1965). Semantography (2nd. edn). Sydney, Australia: Semantography.

Brooks, L., (1977). Visual pattern in fluent word identification. In A.S. Reber & D.L. Scarborough (Eds.), Toward a psychology of reading. New York: Erlbaum.

Brooks, L. & Miller, A. (1979). A comparison of explicit and implicit knowledge of an alphabet. In P.A. Kolers, M.E. Wrolstad & H. Bouma (Eds.), Processing of visible language 1. New York: Plenum.

Bryant, P.E., & Bradley, L. (1980). Why children sometimes write down words which they do not read. In U. Frith (Ed.), Cognitive processes in spelling. New York: Academic Press.

Calculator, S., & Dollaghan, C. (1982). The use of communication boards in a residential setting: an evaluation. Journal of Speech and Hearing Disorders, 47, 281-287.

Carlson, R., Granstrom, B., & Hunnicutt, S. (1982). Bliss communication with speech or text output. Conference Record IEEE-ICASSP, Paris, France.

Carroll, J.B. (1972). The case for ideographic writing. In J.K. Kavanagh & I.G. Mattingly (Eds.), Language by ear and by eye. Cambridge: MIT.

Clark, C.R. (1981). Learning words using traditional orthography and the symbols of Rebus, Bliss, and Carrier. Journal of Speech and Hearing Disorders, 46, 191-196.

Craik, F.I.M., & Lockhart, R.S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671-684.

Downing, J. (1973). Is literacy acquisition easier in some languages than in others? Visible Language, 7, 145-154.

Frith, U. (1979). Reading by eye and writing by ear. In P.A. Kolers, M.E. Wrolstad & H. Bouma, (Eds.), Processing of visible language 1. New York: Plenum Press.

Gibson, E.J., & Levin, H. (1975). The psychology of reading. Cambridge, Mass.: M.I.T. Press.

Giddings, W., Norton, J., Nelson, P., McNaughton, S., & Reich, P. (1979). Development of a Blissymbol terminal. Proceedings of the 6th Man-Computer Communications Conference, 63-71, Ottawa.

Goody, J. (1968). Literacy in traditional societies. London: Cambridge University Press.

Gray, W.S. (1956). The teaching of reading and writing, an international survey. Paris: UNESCO.

Grimes, J.E., & Gordon, R.G. (1980). Design of new orthographies. In J.F. Kavanagh & R.L. Venezky (Eds.), Orthography, reading, and dyslexia. Baltimore: University Park Press.

Hammond, J., & Bailey, P. (1976). An experiment in Blissymbolics. Special Education: Forward Trends, 3, 21-22.

Harris-Vanderheiden, D., Brown, W.P., Reinen, S., MacKenzie, P., & Scheibel, C. (1975). Symbol communication for the mentally handicapped. Mental Retardation, 13, 34-37.

Hatta, T. (1977). Recognition of Japanese Kanji in the left and right visual fields. Neuropsychologia, 15, 685-688.

Helmer, B. (1980). Blissymbols for use. Toronto: Blissymbolics Communication Institute.

House, B.J., Hanley, M.J., & Magid, D.F. (1980). Logographic reading by TMR adults. American Journal of Mental Deficiency, 85, 161-170.

Hung, D. L., & Tzeng, 0. J. L. (1981). Orthographic variations and visual information processing. Psychological Bulletin, 90, 377-414.

Hunnicutt, S. (1986). Lexical prediction for a text-to-speech system. This volume.

Kates, B., & McNaughton, S. (1975). The first application of Blissymbolics as a communication medium for nonspeaking children: history and development, 1971-1974. Toronto: Blissymbolics Communication Institute.

Kausler, D.H. (1974). Psychology of verbal learning and memory. New York: Academic Press.

Kemler, D.G. (1983). Wholistic and analytic modes in perceptual and cognitive development. In T. Tighe & B.E. Shepp (Eds.), Perception, cognition, and development: Interactional analyses. Hillsdale, N.J.: Erlbaum.

Kolers, P.A. (1969). Some formal characteristics of pictograms. American Scientist, 57, 348-363.

Kyostio, O.K. (1980). Is learning to read easy in a language in which the grapheme-phoneme correspondences are regular? In J.F. Kavanagh & R.L. Venezky (Eds.), Orthography, reading, and dyslexia. Baltimore: University Park Press.

Liberman, I., Liberman, A.M., Mattingly, I., & Shankweiler, D. (1980). Orthography and the beginning reader. In J.F. Kavanagh & R.L. Venezky (Eds.), Orthography, reading, and dyslexia. Baltimore: University Park Press.

Makita, K. (1968). The rarity of reading disability in Japanese children. American Journal of Orthopsychiatry, 38, 599-614.

Mattingly, I.G. (1972). Reading, the linguistic process, and linguistic awareness. In J.P. Kavanagh & I.G. Mattingly (Eds.), Language by ear and by eye. Cambridge, Mass.: M.I.T. Press.

McDonald, E.T. (1980). Teaching and using Blissymbolics. Toronto: Blissymbolics Communication Institute.

McNaughton, S. (1978). Blissymbolics. 3rd National Congress, Council for Exceptional Children, Winnipeg, Manitoba.

Minnesota Educational Computing Consortium (1983). Blissymbolics, Bliss Library. St. Paul, MN.

Muter, P. (1985). Orthography and reading: the role of iconicity and grapheme-phoneme correspondence. Manuscript submitted for publication.

Muter, P., & Johns, E.E. (1985). Learning logographies and alphabetic codes. Human Learning (now Applied Cognitive Psychology), 4, 105-125.

Park, S., & Arbuckle, T.Y. (1977). Ideograms versus alphabets: Effects of script on memory in "biscriptual" Korean subjects. Journal of Experimental Psychology: Human Learning and Memory, 3, 631-642.

Rozin, P., & Gleitman, L.R. (1977). The structure and acquisition of reading II: the reading process and the acquisition of the alphabetic principle. In A.S. Reber & D.L. Scarborough (Eds.), Toward a psychology of reading. New York: Erlbaum.

Rozin, P., Poritsky, S., & Sotsky, R. (1971). American children with reading problems can easily learn to read English represented by Chinese characters. Science, 171, 1264-1267.

Sasanuma, S. (1974). Kanji versus Kana processing in alexia with transient agraphia: A case report. Cortex, 10, 89-97.

Sawchuck, W., & Bown, H.G. (1977). Interactive graphics applied to symbol communication for non-speaking children. Computers and Graphics, 2, 201-204.

Smith, L.B., & Kemler, D.G. (1978). Levels of experienced dimensionality in children and adults. Cognitive Psychology, 10, 502-532.

Song, A. (1979). Acquisition and use of Blissymbols by severely mentally retarded adolescents. Mental Retardation, 253-255.

Stevenson, H.W., Stigler, J.W., Lucker, G.W., & Lee, S. (1982). Reading disabilities: The case of Chinese, Japanese, and English. Child Development, 53, 1164-1181.

Stroop, J.L. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662.

Thompson, G.B. (1979). Memo from Mercury: Information technology is different. Montreal: Institute for Research on Public Policy.

Thompson, G.B. (1983). Visual literacy and the time-varying icon. IEEE Journal on Selected Areas in Communications, SAC-.1, 304-305.