Dyslexia and learning musical notation: A pilot study

Dyslexia and learning musical notation: A pilot study

Jaarsma, B S

Both the alphabet and our system of musical notation are largely based on arbitrary conventions and associations. Conforming to the Dyslexic Automatization Deficit Hypothesis, children suffering from dyslexia are supposed to have difficulty in automating these types of conventions and/or associations. Scientific research into the relation between dyslexia and learning musical notation is rare. Therefore, we developed a new intervention paradigm on learning musical notation. This program was followed by five dyslexic children and four children without dyslexia, in order to study and compare the learning processes of both groups. The program consisted of instruction, practice in the skills and knowledge related to musical notation, and test assignments. During each session, the problem-solving process of the individual child was observed and the amount of time required to complete each assignment in the program was recorded. In addition we analyzed the errors made by the two groups of children in each session. A pretest-posttest comparison revealed that dyslexic children needed more time to learn musical notation than did children without dyslexia. Dyslexic children also made more mistakes and produced more ‘third transpositions’. The implications for teaching musical notation are outlined.

INTRODUCTION

There is a growing body of evidence that children suffering from dyslexia might have a general deficit in automating a variety of cognitive and motor skills (Nicolson and Fawcett 1990, 1994a, 1994b, 1995; Yap and Van der Leij 1994; Fawcett, Nicolson, and Dean 1996). This Dyslexic Automatization Deficit (DAD) hypothesis may explain some behaviors not readily accounted for by the dominant explanation of dyslexia, the phonological deficit hypothesis (cf. Stanovich 1988; Brady and Shankweiler 1991). One of the intriguing elements of the DAD hypothesis is its prediction that people with dyslexia are likely to have problems in learning which does not involve reading per se. One such task, learning to read musical notes, is especially interesting because it is conceptually analogous to learning to read the alphabet. Both the alphabet and our system of musical notation are based largely on arbitrary conventions and associations. Moreover, there is empirical evidence that disabled readers have particular problems in learning the structural or rule-governed aspects of reading and language-related skills (Manis et al. 1987; Scarborough 1990). Manis et al. (1987) found that, in learning arbitrary paired associates, disabled readers performed at a level below that of nondisabled readers only when the associates conformed to specific underlying rules (see also Morrison 1993). Should the prediction that dyslexic children have difficulties in learning musical notation be confirmed, it will further support the DAD hypothesis, and from a practical point of view, it may improve techniques for teaching music to dyslexic children. In fact, scientific research into the relation between dyslexia and difficulty in learning musical notation is rare. This is surprising given that the problems that dyslexic persons have with the automatization of graphemephoneme association would seem a necessary obstacle to the process of associative learning, critical in musical instruction.

Research on the learning of musical notation by persons with dyslexia has been carried out by Atterbury (1983, 1984), Hubicki (1994), and Ganschow, Lloyd-Jones, and Miles (1994). In the article “Music teachers need your help,” Atterbury (1984) describes the regrettable lack of communication between the resource teachers and the music educators of learning disabled children and their resulting lack of knowledge regarding each other’s long- and short-term goals. Atterbury describes specific problems LD children may have with music notation because of their excessive visual material and confusing formats. Any child with a reading disability may have problems, not only in reading the music but also with the physical layout of the page. Although Atterbury’s article was published several years ago, only a few studies have been done to document a relation between dyslexia and difficulty in learning musical notation, or to identify difficulties children with dyslexia experience when confronted with music in school.

Hubicki (1994) also describes some of the difficulties people with dyslexia experience when learning musical notation. She focusses on awareness of implied information,1 that is, the gap dividing the abstract sound of music itself from both the facts of musical theory and its notation. She highlights the difficulties experienced by individuals with dyslexia in processing this information which includes words and names referring to the pitch or the lengths of notes, and symbols which represent pitch, lengths of note, phrase markers, and interpretation signs. She claims this notation poses particular difficulty to persons with dyslexia for whom words and symbols are problematic (Hubicki 1994). According to Hubicki, some of the music reading terms which can cause difficulty include “high”, “low”, “left”, and “right”. She describes some compensatory strategies that could heighten awareness. Based on research and on her own clinical experience, Hubicki proposes the use of colors for each pitch symbol, or of familiar object symbols (like fruits) to represent the corresponding notes. In our opinion, this type of mediation does not solve the problem because people with dyslexia would still have to automatize a system based on arbitrary associations. Even if easier, it is not a permanent solution. The exclusive learning of an alternative system would become limiting as a student’s musical ability increased. To play Mozart or Chopin, a gifted musician with dyslexia would either have to recode the composition in the familiar system, memorize it, or ultimately learn to read the traditional musical notation after all.

Ganschow et al. (1994) investigated the difficulties dyslexic musicians face during the formalized study of music, particularly with musical notation. Through seven case studies based on self-reports, they present individual experiences with learning the musical notation system. In their article, Ganschow and her colleagues also discuss the possible correspondence between learning to read and learning to interpret musical notation. They see a similar pattern in the difficulties experienced with the notation system of music and that of written language: problems with the representation of time, rhythm, and sequencing might be compared to the difficulties some dyslexics have with identifying and representing phonological units of language.

Both the paucity of research on the relation between dyslexia and difficulty in learning musical notation, and the need for empirical information on this topic, prompted us to carry out this study. Our main goal was to gain more insight into the possible difficulties children with dyslexia experience when they learn musical notation. To this end, we developed a music program which was followed by a group of children with dyslexia (n = 5), as well as a group of children without dyslexia (n = 4) in order to study and compare the learning processes of both groups. This music program was structured as an intensive individual training, where special attention was given to a qualitative analysis of the learning process. The main principles to structure the training program were comparable to the adaptive strategies used in the tradition of dynamic assessment procedures: gradual structuring of item presentation and modeling (Ruijssenaars and Hamers 1989; Lidz 1991; Haywood and Tzuriel 1992).

Because of the intensity required to prepare and execute the training program, we opted for a small-size intensive pilot study to determine whether our approach was meaningful and feasible. We sought answers to the following questions:

What specific difficulties do children with dyslexia have when learning musical notation, and are these shared by children without dyslexia?

What specific mistakes do children with and without dyslexia make when learning musical notation?

How much learning gain in the naming of notes is evident over time within groups? Is there a difference in learning rate between children with and without dyslexia?

The results of the investigation pertain not only to differences in learning gain between the two groups, but also within each group. Even though our research is at a preliminary stage and presently consists of only a small pilot study, we believe this kind of research to be relevant to the debate on automaticity versus exclusively phonological deficits in reading disability. Looking at the acquisition of musical notes seems a very good test of the DAD hypothesis. To meet these research goals we compared-in a pretest-posttest design-a group of dyslexic children with a group of nondyslexic children based on their progress in learning musical notes.

METHOD

DESCRIPTION OF THIS SAMPLE

Nine children participated in the investigation (seven boys and two girls, aged nine years, two months to nine years eleven months). Five of the participants were dyslexic and received special education at a school for children with learning disabilities (LD). The other four attended standard primary school and were confirmed by their teachers to have no reading problems. The children were enrolled in the study by the head teachers of two primary schools and three LD schools from the Middle Holland region, and were selected according to explicit criteria.

In order for the children from the LD schools to qualify for the study, the specific learning disorder of dyslexia had to have been diagnosed by an expert. In the Netherlands, the criteria outlined in the DSM-IV (APA 1996) are used for this diagnosis. Experimenters administered standardized didactic tests on word identification (AVI-test 1996) to reaffirm that the reading development of these children was at least two years below age level. The children from standard primary school scored above average and showed no reading difficulties on these tests. All participating children had to be at least nine years old (the average age at which children start formalized music education in the Netherlands). In addition, the children had to be of normal intelligence (IQ > 90) and free of visual and auditory handicaps. Finally, the children did not have any previous musical instruction, nor have been in special contact with musical notation in any other way. These selection criteria were strictly applied in an effort to ensure integrity of the data.

PROCEDURE

For the collection of data in this pilot study, a chronological– age-match design was used. We chose this design both because it was appropriate for use with our selection criteria, and because many nine-year olds without previous musical training were available to participate. An acknowledged drawback of this design is that a posttest learning difference between the two groups can be interpreted as the result of the initial differences in reading levels; a reading-level-match design should also be applied in future research. Both groups were offered the music reading program in which measurements were taken before and after the assignments. The music reading program consisted of individual sessions, each lasting a maximum of forty-five minutes, which took place once a week over a period of five weeks. The training consisted of instruction and practice in the skills and knowledge related to musical notation (completed with assistance from the experimenter) and test assignments (completed without assistance). During each session, the problem-solving process of the individual child was observed. In this way, it was possible to work out how the child made a mistake. Afterwards, the mistakes were discussed with the child in order to verify whether the investigator’s conclusions about their origins were correct. In addition, at a later stage, we analyzed the errors made by the two groups of children in each session. Using these two approaches (on-line and post hoc), we were able to gain more insight into the difficulties experienced by the children with dyslexia in comparison with those experienced by the children without dyslexia. By comparing the errors in the two samples, it was possible to investigate whether the errors made by the children with dyslexia were qualitatively unique.

In addition to observing errors made, the amount of time required to complete each assignment in the program was recorded. To this end, each of the first four training sessions was followed by a test in which the children had to name, as quickly as possible, the notes taught so far. These four posttests were repeated in the final session, so each was presented twice. Next, we compared the difference between the time to name the notes at initial and final posttest. If less time was needed for the second note-naming exercise, this might indicate that a learning or automatization process had taken place. By carrying out further intergroup comparisons, we were able to investigate whether dyslexic children had more trouble automating musical notation than did children without dyslexia.

THE INDIVIDUAL TRAINING PROGRAM FOR READING MUSIC

The program’s assignments were specifically designed to cultivate the ability to “read music”: to give the name of a particular note. The program consisted of five series of notes, which the child was taught over five sessions. As shown in figure 1, each series contained four different notes2. Series A contained gl, al, bl, and c2; Series B contained c1, dl, el, and fl; and Series C included d2, e2, f2, and g2. In Series D and E, the child learned flats (6) and sharps (#). In these series, the notes which were taught in series A through C were used, with the addition of b(natural) and e(natural), and f#1 and c#2, respectively. We chose to teach the flats and sharps because these symbols imply additional information which must be interpreted by the child. In the Dutch language, these symbols change not only the pitch of a previously learned note, but its name as well; for example, a “b” becomes a “bes” (b flat) and an “f” becomes a “fis” (f sharp). The phenomenon of a sound changing because of cumulative factors is also common in alphabetic orthographies; for example the letter “o” represents /a/ in hot and the letter “e” represents /e/ in pet. But “oe” together represents /u/ (as in shoe). Dyslexic children often have difficulty understanding this concept when learning to read words. The addition of flats and sharps to certain series in our program was intended to test whether similar difficulty would arise when learning to read notes.

Each training session was based on one or more of these note series, and included a maximum of eight different exercises. The same exercises were repeated in the next session, but with a different series of notes in order to allow the children to familiarize themselves with the task. When carrying out the exercises, the children were allowed to make use of a learning aid, a chart showing all the notes with their names. The order of the assignments was always the same (with the exception of training sessions 4 and 5):

1. Learning the new series (week 1 = Series A, week 2 = Series B, and so forth). The four notes from a series were graphically presented and offered to the child on separate small cards, four times in a random order. The child received four small cards containing the names of the notes and was asked to place the name of the note with its corresponding picture.

2. Recognizing notes within composition. Within four different compositions consisting of eight notes divided in two bars, the child was asked each time to point to a different requested note.

3. Reviewing the notes from a previous series. The child was either asked to draw the requested notes on the staff or to complete a task as described in assignment 1. However, this time, the child had to identify a greater number of notes, in five different orders.

4. Combining notes from previously learned series. In each session, the child was shown a four-bar composition consisting of 16 notes, and was asked to point to notes named, to name notes, or to find errors in bars where notes had been named.

5. Drawing notes whose names were under the staff, or were read out loud.

6. Naming notes shown on a large chart, or finding (from a collection of small cards laying on the table) the correct corresponding cards to the notes presented on the large chart.

7. Testing ready knowledge of note names with five flash cards. The cards were presented in ascending order of difficulty. The first card contained two to five different notes (depending on the week), the second card contained three to six different notes, the third card four to seven, and so on.

8. Evaluation of what the child considered difficult/easy or fun/boring.

During all assignments except the seventh, extra help was given when needed. The assignments were designed to utilize different learning processes such as visual discrimination, recognizing associations, knowing associations, reproduction, and applying knowledge. In the study, no use was made of a clef sign at the beginning of a staff. This was a conscious choice, intended to ensure that any error made in the execution of assignments could be attributed to problems with the naming of the notes rather than with a possible failure in understanding the clef and the conventions associated with it.

RESULTS

By analyzing the errors made by the children of both groups, and observing the problem-solving processes of each child, we tried to find an answer to our first research question: What specific difficulties do the children with dyslexia encounter in the formalized study of musical notation? By comparing errors and observations between the groups, we hoped to be able to discover whether there was a difference between the learning processes of children with and without dyslexia.

On one assignment in particular, the dyslexic children performed considerably below the level of the nondyslexic children. During this, the sixth assignment of the second session, children had to choose two to five cards whose graphic representation matched the notes pictured on a large chart (see individual training program). Although other assignments (in which the children had to compare two notes which were presented on different charts) demonstrated that the visual aspect of note matching did not cause any problems, the dyslexic children performed markedly worse on this assignment. During 70 trials, they made a total of 18 mistakes (26 percent), of which about 30 percent were third transpositions.3 The nondyslexic children, on the other hand, did not make a single mistake and required far less time to complete the assignment.

As the study progressed, the dyslexic children gradually made fewer mistakes, corrected themselves more often, and experienced less difficulty in the naming of some notes. Over the course of the investigation, some children also made less use of the learning aid, a long chart divided into five series, showing all the notes and their names, which was always on the table. Those who continued to use it showed an increasingly more efficient routine, knowing exactly where to look for each note.

The goal of our second research question was to determine which specific mistakes are most likely to be made by children with dyslexia when learning musical notation. Our error analysis, summarized in table I, showed that the dyslexic children not only made more mistakes, but also were particularly prone to making third transposition errors. That is, rather than confusing two notes next to each other (al, gl), they confused notes that were either both on a line or on a space (al, fl).

During the training and test sessions, the five dyslexic children made a total of 262 errors, out of which 160 (61 percent) were third transpositions. In comparison, the four nondyslexic children made a total of 146 errors, of which 56 were third transpositions (38 percent). On average, the dyslexic children made twice as many third transpositions as did the nondyslexic children (t = 1.70, 7df, p = 0.05 by randomization test; cf. Edgington 1987), while making only equally as many other mistakes as the nondyslexic children. The most frequent mistake made by the nondyslexic children was to identify the note in question with the name of a note just above or below it.

One of the nondyslexic children (No. 3) also made a relatively large number of third transpositions. Upon inquiry, we found that this child had difficulties with arithmetic, and was receiving remedial teaching in that subject. The nondyslexic children were also found to have more problems with b(natural)/e(natural) and f#/c#.

Not all the thirds that were taught in the program were transposed with equal frequency. In the error analysis, we found that of the possible 20 third transpositions, only 13 were made by any of the children. Except for the transposition “al named as c2” (see figure 2), six transpositions were made to either side (that is upward as well as downward in the scale). The transpositions bl/dl, el/gl, fl/al, gl/el, d2/bl, d2/f2, and al/fl were the most common made by the dyslexic children. The most frequent and intriguing transposition, in both groups of children, was bl/di. In figure 2, the possible third transpositions, and the number of times these transpositions were made during the music program by the dyslexic and nondyslexic children, are shown.

Notably, transpositions of thirds did not occur in the naming of notes whose graphic representation was located above or below the five lines of the staff (cl/el, dl/fl, e2/g2) (see figure 3).

The third research question asked how much time was required by the two groups of children for the naming of the notes. Each training session was followed by a test of the material learned at that session. This test consisted of five graded cards, based on one or more note series, presented in ascending order of difficulty. The first measurement took place at the end of the relevant practice session, and the second took place in the fifth and final session, when all cards were tested consecutively. The purpose of these two tests was to ascertain whether dyslexic children have more problems with this learning process than do nondyslexic children. First, we compared the average times for the dyslexic children (group I) and the nondyslexic children (group II) at the time of the first measurement. Figure 4 shows the results.

This chart shows that on all but five cards, the nondyslexic children needed less time to complete an assignment than the dyslexic children (t = 3.13, 24 df, p = 0.0012 by randomization test with 50,000 random permutations). For four of the cards (A2/A4/C4/D2), the time differences between groups were caused primarily by the long latencies of two children with dyslexia. Table II contains the average times scored on the cards by group I (dyslexic) and group II (nondyslexic). Times were scored without regard to errors in note naming.

Next, we compared the average times scored on the final posttest. By this time, the children had been taught all the notes. The less automated this knowledge, the more time a child would need in order to name the notes correctly. Figure 5 shows the average scores of group I and group II at the time of the final posttests.

Figure 5 shows that, at the final posttest, both groups generally needed less time for the assignments than in the initial posttests. The difference between the groups here was smaller compared with the first measurement, but the dyslexic children almost always needed more time than the nondyslexic children (t = 2.37, 24 df, p = 0.015 by randomization test with 50,000 random permutations).

Finally, we looked at the gain scores within each group. Table II contains the average times on both measurements and the learning gain scores in seconds. The figures show an average learning gain, for both groups, of 4.8 seconds. We considered an average learning gain of over 5 seconds per card to be substantial (shaded gray). For almost all cards, the children needed less time in the final posttest, with a pronounced learning gain on cards A2/A4 and C2 through D2 for the dyslexic children; their average gain in time is 7.1 seconds. If we compare the first and second measurements for the nondyslexic children, we notice that they made only modest gains between the first and second posttests. This phenomenon is likely due to a “ceiling” effect; their average gain in time is 3.4 seconds. The most noticeable gains were scored on the more difficult D and E series.

DISCUSSION

As we mentioned in the introduction, we hypothesize that dyslexic children will have notable difficulty learning and automating the connection between a symbol and a label (and vice versa). One, therefore, expects these children to have more problems in a training program for musical notation than would nondyslexic children. This difficulty may be reflected in the number and type of mistakes dyslexic children make, and the time they need to complete the assignments. The results of the investigation described here bear out this hypothesis.

This study demonstrates that, during the entire learning process, the dyslexic children needed more time for the assignments, making almost twice as many mistakes as the nondyslexic children. In addition, they showed a specific error pattern which included frequent third transpositions, suggesting that the specific patterns of the lines and the labels connected to them had not yet been sufficiently internalized. Nondyslexic children made, in a manner of speaking, more precise errors by mistaking a note for the one directly above or below it (the so-called second). These findings suggest that dyslexic children are less sensitive to the crucial position of the notes on the lines. They preferentially give their attention to more superficial cues (whether a specific note is on the line or between lines).

There was one type of assignment with which the dyslexic children had noticeably more problems regarding both numbers of mistakes and time. In this assignment, the children were asked to match notes on a large chart to notes on separate small cards. As these cards were spread randomly over the table, the children had to be able to memorize the notation before starting to look for it. It was an advantage here to be able to name the note, but when this knowledge was less automated, the task became very difficult. The assignment at that point specifically drew on the ability to transfer the material learned.

Learning musical notation caused dyslexic children more trouble, even though learning benefits did occur. Given the time scores of the nondyslexic children on the first and second measurements, we may conclude that with them the automatization process is completed sooner.

All our conclusions, however, are conditional on the assumption the dyslexic-nondyslexic distinction was the only factor significant enough to account for the observed differences between groups. Although we suspect that neither differences in reading ability nor intellectual discrepancies between the two groups were sufficient to explain our data, the only way to rule out these possibilities is to replicate this study on a larger scale, using appropriate designs and controls. This pilot study was mainly intended to explore the feasibility of this newly developed paradigm for the study of learning processes in dyslexia.

CONCLUSION

The results of our investigation show that dyslexic children are able to learn musical notation, but that they experience considerable difficulty with the automatization of this system of conventions. This conclusion paves the way for further research. The process of associative learning seems to be central to learning to read and spell. In particular, the automatization of arbitrary associations warrants further investigation. Associative learning is a time-consuming process for children with dyslexia in learning musical notation. It is imaginable that the lack of contextual support (top-down processing) is an extra handicap for them. The absence of compensatory procedures probably elicits a more structured weakness in this type of learning. For the musical learning process to succeed, it is important that music teachers take the effects of this laborious automatization into account. With music subjects in school, or with other types of musical education, dyslexic children will need more time not only to achieve automatization in music reading but also possibly for the automatization of the link between notation and playing techniques on an instrument.

Alternative musical notation systems such as color or picture codes (Hubicki 1994; Oglethorpe 1996) do not seem to be satisfactory aids for dyslexics, as the process of automating a system of arbitrary conventions is not the same as learning all the codes separately. When learning to read, dyslexic children often manage to learn the separate connections between graphemes and phonemes, but cannot automatically apply this knowledge in the reading process (Gough 1996). They do not progress beyond reading slowly, sound-by-sound. Put differently, explicit instruction in subskills does not automatically result in an integrated and partly unconscious (automated) final process.

A solution for this problem may, therefore, be found in putting less emphasis on explicit subskills. Instead, instruction should be structured in such a way that subskills are also acquired implicitly. At present, this approach is used in reading instruction by employing series of orthographically overlapping words. By the process of covariant learning these overlaps become, as it were, implicitly reinforced (cf. Van Orden and Goldinger 1994; Van den Broeck 1997).

Applying this carefully structured process of implicit learning in the musical instruction of dyslexic children would, for instance, mean that less attention would be given to the explicit naming of notes. Learning to read musical notation would then be integrated in its application to the playing of an instrument. Particular emphasis would have to be placed on becoming familiar with the constantly shifting patterns of musical notation. Knowledge of subpatterns and of separate notes would, in this way, be implicitly acquired and reinforced. Only when a reasonable proficiency is attained at this level should we implement an explicit phase to make the pupil aware of the more specific body of knowledge.

Whether or not such an approach is desirable, of course, depends on the purpose of the music instruction. This explicit phase would be a standard part of a professional course, but not of regular musical education in schools.

A follow-up study with a larger sample of children may afford more insight into specific automatization problems, and allow for more generalized conclusions about typical error patterns of dyslexic children.

Address correspondence to: Birgit Jaarsma, Leiden University/Faculty of Social and Behavioral Sciences, Pieter de la Court Building, Wassenaarseweg 52, P.O. Box 9555, 2300 RB Leiden, the Netherlands. Tel: +31-71-5274072, Fax: +31-71-527-3619,E-mail: jaarsma@rulfsw.fsw.leidenuniv.nl

1 This `implied information’ falls into two categories of symbols: those which represent time (time symbols follow one another across the page from left to right) and pitch (pitch symbols are placed one above or below another) (Hubicki 1994).

2 The Western system of musical notation comprises 8 octaves. One octave contains,in this connection, the string c d e f g a b c. The numbers following the note names are a form of official musical notation and refer to the octave in question. A ‘1’ following a note indicates the first string. A ‘2’ following the note refers to a second string of notes which have the same names, but are in a different octave and hence are graphically represented in a different way, and sound at a different pitch.

3 A third is a distance of two whole tones between notes. In the investigation we mainly used notes from the scale of c: cl dl el fl gl al bl c2 (and d2 e2 f2 g2). A third transposition occurs when two notes separated by two whole tones are interchanged (for instance el-gl).

References

American Psychiatric Association. Diagnostic and statistical manual of mental disorders, Fifth Edition. Washington, DC. American Psychiatric Association 1996. Atterbury, B. W. 1983. A comparison of rhythm pattern perception and performance in normal and learning-disabled readers, age seven and eight. Journal of Research in Music Education 31(4):259-70.

Atterbury, B. W. 1984. Music teachers need your help. Journal of Learning Disabilities 17(2): 7977.

Brady, S. A., and Shankweiler, D. P. (eds.) 1991. Phonological Processes in Literacy. Hillsdale, NJ: Lawrence Erlbaum Associates.

Edgington, E. S. 1987. Randomization Tests (2nd Ed.). New York: Marcel Dekker. Fawcett, A. J., Nicolson, R. I., and Dean, P. 1996. Impaired Performance of Children with Dyslexia on a Range of Cerebellar Tasks. Annals of Dyslexia 46:259-83. Ganschow, L., Lloyd-Jones, J., and Miles, T. R. 1994. Dyslexia and Musical Notation. Annals of Dyslexia 44:185-202.

Gough, P. 1996. How children learn to read and why they fail. Annals of Dyslexia 46:>20. Haywood, H. C., and Tzuriel, D. (eds.) 1992. Interactive Assessment. New York: Springer Verlag.

Hubicki, M. 1994. Musical problems? Reflections and suggestions. In G. Hales, ed., Dyslexia Matters. London: Whurr Publishers.

Lidz, C. S. 1991. Practitioner’s Guide to Dynamic Assessment. New York: The Guilford Press.

Manis, F. R., Savage, P. L., Morrison, F. J., Horn, C. C., Howell, M. J., Szeszulski, P. A., and Holt, L. J. 1987. Paired associate learning in reading-disabled children: Evidence for a rule-learning deficiency. Journal of Experimental Child Psychology 43:25-43.

Morrison, F. J. 1993. Phonological processes in reading acquisition: Toward a unified conceptualization. Developmental Review 13:279-85.

Nicolson, R. I., and Fawcett, A. J. 1990. Automaticity: A new framework for dyslexia re

search? Cognition 35:159-82.

Nicolson, R. I., and Fawcett, A. J. 1994a. Comparisons of deficits in cognitive and motor skills among children with dyslexia. Annals of Dyslexia 44:147-4. Nicolson, R. I., and Fawcett, A. J. 1994b. Reaction times and dyslexia. Quarterly Journal of

Experimental Psychology 47A:29-48.

Nicolson, R. I., and Fawcett, A. J. 1995. Dyslexia is more than a phonological disability. Dyslexia: An International Journal of Research and Practice 1:19-37. Oglethorpe, S. 1996. Instrumental Music for Dyslexics: A Teaching Handbook. London: Whurr Publishers.

Ruijssenaars, A. J. J. M. and Hamers, J. H. M. 1989. Assessment of Learning Ability. Learning Ability Tests and Analysis of Learning Processes. In H. Mandl, H. Bennet, E. De Corte, and A. F. Friedrich, eds., Learning and Instruction, Vol II. Oxford: Pergamon Press.

Scarborough, H. S. 1990. Very early language deficits in dyslexic children. Child

Development 61:1728-43.

Stanovich, K. E. 1988. The right and wrong places to look for the cognitive locus of reading disability. Annals of Dyslexia 38:154-77.

Van den Broeck, W. 1997. Phonological Processes in Reading Development (Ph.D. diss.) University of Leiden.

Van Orden, G. C. and Goldinger, S. D. 1994. Interdependence of form and function in cognitive systems explains perception of printed words. Journal of Experimental Psychology: Human Perception and Performance 20:1269-91. Visser, J., Laarhoven, A., and van Beek, A. 1996. AVI-toetspakket. Handleiding. ‘sHertogenbosch: KPC.

Yap, R., and Van der Leij. 1994. Automaticity deficits in word reading. In R. I. Nicolson and A. J. Fawcett, eds., Dyslexia in Children: Multidisciplinary Perspectives. New York: Harvester Wheatsheaf.

B. S. Jaarsma

A. J. J. M. Ruijssenaars

W. Van den Broeck

Leiden University, the Netherlands

Copyright International Dyslexia Association 1998

Provided by ProQuest Information and Learning Company. All rights Reserved