Rhythmic and melodic sight reading interventions: Two meta-analyses

Abstract

The purpose of these meta-analyses was twofold: 1) to determine whether experimentally - tested sight reading interventions positively influenced rhythmic or melodic sight reading performance and if so; 2) to explore whether the interventions differentially affected rhythmic and melodic sight reading. Two meta-analyses were conducted; one comprised experimental research measuring rhythmic sight reading (46 studies; 68 individual analyses) and another comprised research measuring melodic sight reading (21 studies; 35 individual analyses). Analyses revealed small overall effect sizes (Rhythmic: d = −0.34, 95% CI [−0.50, –0.19]; Melodic: d = −0.35, 95% CI [−0.59, –0.12]). Moderator variables were used to examine how effect size was influenced by treatment type and other study-level design elements. Treatment type significantly influenced both rhythmic and melodic sight reading. For rhythm, treatments that focused on counting systems and included movement or rhythmic drill positively affected sight reading. Melodic results are preliminary due to the limited number of studies, but based on the information available, treatments using collaboration activities or instrumental training positively affected sight reading. Varying notation did not improve sight reading. The type of sight reading (rhythmic, melodic) significantly affected the effectiveness of some treatments.

Keywords

meta-analysis rhythm reading sight reading solfege

To effectively sight read music, both melodic and rhythmic information must be processed and performed simultaneously, but melodic and rhythmic information is presented very differently on the page. Melodic information is presented spatially on the staff while decoding rhythmic information requires recognizing features of the symbols that denote duration (e.g. flags, beams, stems). Differences between the physical features of the written systems suggest that pitch and rhythm may be read differently. There is growing evidence that this is the case.

In music education, improving sight reading performance has been of great concern, with hundreds of research studies and pedagogical texts and articles written on the topic. Recently, I attempted to determine which types of treatments were most effective at improving sight reading (Mishra, 2014). Using a meta-analysis, I found that treatments focusing on aural training, creative activities, controlled reading, or solfege were most effective. One limitation of the study (Mishra, 2014) was that only primary research studies measuring both melodic and rhythmic sight reading were included; research studies focusing on rhythmic-only or melodic-only sight reading were excluded. Reading rhythm separately from melody is a common practice strategy. Many researchers have investigated ways to improve either rhythmic or melodic sight reading with a view to improving overall sight reading.

Historical context

Modern musical notation is rooted in the early eleventh-century system of staff notation developed by Guido d’Arezzo. The focus of this early notation was the spatial representation of exact pitch height on a staff. By the thirteenth century, the Guidonian system had evolved into the five-line staff still in use today (Bent et al., n.d.). However, problems with decoding notation were already evident to d’Arezzo. Along with developing staff notation, he developed a system of solmization to assist his choir in translating the notated pitch into sound. This system is the basis of the solfege system used today.

Rhythmic notation developed later. By the thirteenth century, the proportional division of long notes (breves) was represented in notation, though the symbol system continued to evolve. It was not until the eighteenth century that the rhythmic and metric symbols used today became standardized (London, n.d.). In the twentieth century, a number of pedagogical counting systems, analogous to d’Arezzo’s solmization system, developed to assist in the reading of rhythms, including speech rhythms as used in the Orff-Schulwerk method and those developed by Kodály (ta, ti-ti) and Gordon (du, du-de).

Research supporting differences between rhythmic and melodic sight reading

The idea that musical sight reading functions differently for rhythm and melody (a succession of pitches) was investigated by researchers as early as the 1930s. In 1932, Sanderson tested elementary school children on their ability to read rhythmic material, read melodic material, and read both together. He found a low correlation (r = .33) between rhythmic and melodic sight reading. Byo (1992) found a similarly low correlation (r = .24) between melodic and rhythmic reading. These low correlations suggest that melodic and rhythmic reading proficiency may involve unrelated skills.

Researchers investigating eye movement during sight reading also found evidence of a disassociation between rhythmic and melodic sight reading. Van Nuys and Weaver (1943) documented eye movements and eye-hand span of pianists in three reading conditions: rhythm-only, pitch-only, and combined. Eye-hand span is the distance between where the eyes are reading in the notation and the notation being performed (by the hands). The larger the eye-hand span, the further ahead the eyes are reading. Van Nuys and Weaver found eye movement differences across the conditions, though eye-hand span was only slightly larger for rhythm-only or pitch-only conditions than it was for the combined condition. More recently, Adachi, Takiuchi, and Shoda (2012) found that eye-hand span was affected by melodic structure and meter during sight reading. Complexity of melodic structure reduced eye-hand span for the beginners while metric difficulty reduced eye-hand span for the advanced musicians. Investigating the problem in a slightly different way, Henry (2011) asked singers to sight read stimuli of varying melodic and rhythmic difficulty. She found that in general, singers focused on melodic over rhythmic accuracy, with success in rhythmic accuracy predicting success in melodic accuracy, but not vice versa. These findings indicate the possibility of an asymmetrical relationship between melodic and rhythmic sight reading.

Current neurological research reinforces the idea that rhythmic and melodic reading may be separate processes. Midorikawa, Kawamura, and Kezuka (2003) presented a case study of a subject with musical alexia and agraphia (the inability to read or write musical notation respectively). In this case, the reading of pitch notation was preserved while the reading of rhythmic notation was impaired. Other researchers have reported cases where only the ability to read pitch was impaired (e.g. Brust, 1980; Cappelletti, Waley-Cohen, Butterworth, & Kopelman, 2000). Neuroimaging research confirms unique processing for melodic and rhythmic reading. In an ERP study, Schon and Besson (2002) found distinct activation patterns for reading key and meter. Bengtsson and Ullen (2006) measured brain activation (fMRI) of pianists who sight read rhythms and melodies in isolated conditions. The researchers reported distinct patterns of activation for rhythm and melody. Melodic information moved through the dorsal visual stream, which processes spatial information, and rhythm through the ventral visual stream, which focuses on feature recognition.

There is growing evidence that rhythmic and melodic sight reading are two distinct neurological processes. It would, therefore, follow that an executive neurological process would be needed to integrate rhythmic and melodic information. Paraskevopoulos, Kuchenbuch, Herholz, and Pantev (2012) found a unique multi-sensory integration region in the right superior temporal gyrus. They trained non-musicians to read and play simple melodic patterns (aural-visual-motor condition) or read and listen to recordings of the patterns without playing them (aural-visual condition). The multi-sensory integration region was only activated in the aural-visual-motor group. Paraskevopoulos and colleagues concluded that the individual processes (aural, visual, motor) did not improve with training, but the ability to integrate the processes did. This finding may direct future research towards a specialized multi-modal system required for sight reading.

Research using multiple types of methodologies (e.g. correlational, eye-movement, neurological) strongly suggests that rhythmic sight reading is distinct from melodic sight reading. Further, results from the Paraskevopoulos et al. (2012) study support the idea of a unique integration process during sight reading. However, research in music education has been primarily based on the hypothesis that improving abilities in one mode (i.e. rhythm) will improve overall sight reading.

Effects of rhythmic training on sight reading

One conclusion from the Van Nuys and Weaver (1943) study was that “rhythmic factors constitute the limiting conditions for rate of reading … an increase in rate of reading depends upon improvement in ability to grasp rhythmic figures” (p. 50) and several researchers have confirmed that a large number of sight reading errors are rhythmic (Elliott, 1982; McPherson, 1994; Thomson, 1953). One interpretation from this line of research has been that focusing instruction on rhythmic sight reading improves sight reading in general. This finding has been particularly influential in music education.

In a seminal study, Boyle (1970) experimentally tested whether rhythmic training that included body-movement would improve sight reading. Boyle’s research took place in a naturalistic setting, using 200 students from 24 junior high school bands. Teachers in the control group were asked to discourage body movements, while teachers in the treatment group were asked specifically to include the following activities: marking time, clapping rhythms while tapping beat, playing rhythmic patterns while tapping beat. The groups were tested using the Watkins-Farnum Performance Scale (WFPS), and a rhythmic-only version of the WFPS. While both control and experimental groups improved, the experimental group showed greater improvement in both overall sight reading and rhythmic sight reading. However, subsequent researchers have not been able to replicate Boyle’s findings (Allen, 1987; Barlar, 2010; Earney, 2008; Egbert, 1990; Laing, 2007; Massingale, 1979; Parker, 1979; Pike & Carter, 2010; Searle, 1985; Stegall, 1992; Tucker, 1969). The meta-analysis I conducted (Mishra, 2014) determined that rhythmic training was not an effective treatment for improving overall sight reading (d = 0.03).

The current meta-analyses are needed to expand on my previous findings (Mishra, 2014); however, now investigating rhythmic and melodic sight reading as unique processes. The purpose of these meta-analyses was twofold: 1) to determine whether experimentally tested sight reading interventions positively influenced rhythmic or melodic sight reading performance, and if so; 2) to explore whether the interventions differentially affected rhythmic and melodic sight reading, as might be expected from previous research.

Method

Study inclusion

Studies included in the current meta-analyses measured only rhythmic or only melodic sight reading. Theses and published studies were discovered by searching for keywords related to sight reading and sight singing, along with word variants (e.g. “sightread,” “sight-sing”) and phrases such as “rhythmic reading.” The following indices were searched: Dissertations and Theses Full Text, Google Scholar, International Index of Music Periodicals, JSTOR, Education Full Text, ERIC, PsycINFO, Web of Science, Pubmed, and WorldCat. Reference lists from the identified studies were used to discover additional relevant research.

Only quasi-experimental research studies were included. In each study, the researcher tested the effectiveness of an intervention (i.e. treatment) hypothesized to improve accuracy of either rhythmic or melodic sight reading. In situations where a dissertation was later published, the two reports were considered as a single analysis. Studies that analyzed rhythmic-only or melodic-only sight reading in addition to analyzing a combined rhythm-melodic sight reading were included (e.g. Boyle, 1970). Excluded were studies that tested sight reading using a written measurement (e.g. recognition task) or those that used training stimuli as testing material, as previously encountered stimuli could not be literally “sight read.”

Based on this search, 68 potentially relevant rhythmic and 29 potentially relevant melodic experimental studies were identified and could be obtained. Twelve were excluded because basic statistical information necessary to calculate effect sizes (e.g. descriptive statistics, t, or F values) was not reported or could not be computed from available raw data. Nine were excluded because the design lacked a comparison group or did not compute comparisons, six did not establish group equivalency, and the treatment was unclear in one study and was excluded.

Ultimately, 46 rhythmic and 21 melodic experimental studies conducted between 1954 and 2012 were included in the analysis (see Tables 1 and 2 in Supplemental Material Online). The majority of studies were conducted in the 1990s (25%) or after 2000 (36%). Most were unpublished theses (n = 56; 84%). Forty-seven (70%) used a pretest–posttest, control-group design, 15 (22%) used an equivalent-group, posttest-only design, and five (7%) were repeated measures. The number of participants in the included studies ranged from 8 to 193 (M = 67.23; SD = 45.06; Mdn = 58). The large number of unpublished studies in the analysis is a source of concern as the research quality may be variable. On the other hand, publication bias, often considered a problem when conducting meta-analyses, would not appear to be an issue here. Publication type was coded as a moderator variable to determine whether the results were different between published and unpublished research.

Coding

Individual studies were used as the unit of analysis. In cases where the researcher tested multiple populations (e.g. elementary school students and college music majors) or multiple treatments, sub-group was used as the unit of analysis. This procedure resulted in 68 rhythmic and 35 melodic analyses. All studies were dual coded to reduce the possibility of coding errors.

Moderator variables included: type of publication (unpublished thesis or published paper), participants’ ages and experience levels (elementary, secondary, college non-music majors, college music majors), response type (body percussion, counting aloud, instrumental, vocal), and the type of sight reading test (researcher-constructed, rhythmic portions of WFPS, other published sight reading test, repertoire). Also coded when applicable was information about the control groups’ task (equivalent sight reading training, no sight reading training) and whether the posttest measure was identical to the pretest or was an equivalent test (see Tables 3 & 4).

Studies were grouped into treatment categories similar to those used previously (Mishra, 2014). Rhythmic studies were grouped as follows: Aural Training (n = 7; 10%), Controlled Reading (n = 8; 12%), Counting System (n = 10; 15%), Creative Activities (n = 1; 1%), Movement (n = 11; 16%), Notation (n = 12; 18%), and Rhythmic Drill (n = 8; 12%). The treatments in 11 cases seemed unique and were labeled “Non-Categorized.” Aural Training studies were those that used rote methods or aural dictation, Controlled Reading studies used the tachistoscope or some other way of controlling how quickly notation was presented, and treatments labeled as Creative Activities included composition activities. Studies in the Notation category in some way varied standard musical notation (e.g. proportional rhythmic notation, color-coding). Studies using various counting systems (e.g. Gordon, speech-cue) or movement activities (e.g. foot tapping, Breath Impulse Technique) were grouped accordingly. Rhythmic Drill treatments most often tested a developed sequence of programmed instruction. Non-Categorized treatments included instruction in sight reading, two different types of beat delivery systems (aural via metronome and tactile via a vibrational pulse), various feedback systems (visual versus aural), instruction focusing on patterns, or inclusion of mental rehearsal prior to sight reading.

Melodic studies were grouped as follows: Aural Training (n = 2; 6%), Collaborative (n = 3; 9%), Controlled Reading (n = 4; 11%), Counting System (n = 1; 3%), Instrumental Training (n = 1; 3%), Movement (n = 1; 3%), Notation (n = 2; 6%), Reading Drill (n = 3; 9%) and Solfege (n = 7; 20%). The treatments in 11 cases were labeled “Non-Categorized.” Studies in the Aural Training, Controlled Reading, Counting Systems, Movement, and Notation used treatments similar to those used in the rhythm-only sight reading studies described above. Reading Drill was similar to Rhythmic Drill treatments described above and consisted of a developed sequence of programmed instruction. Collaborative treatments included sight reading with accompaniment or ensemble and Instrumental Training included learning instruments such as keyboard to enhance sight reading. The studies that used Solfege as part of the training method were grouped together. Non-Categorized treatments included varying mental rehearsal time and compositional structure of the material (e.g. atonal v. tonal).

Cohen’s d was used to compute a standardized effect size between posttest means and standard deviations for the control and experimental groups for each study. Cohen’s d standardizes mean differences across studies and weights the results based on population size. Effect sizes were computed from descriptive statistics or reported t or F values. Effect sizes (ES) ranged from −4.27 to 1.84.¹ All effect sizes and statistics were computed using the Comprehensive Meta-Analysis software package (2010).

Results

Rhythmic sight reading

A meta-analysis using a random effects model was conducted. The results of the meta-analysis including rhythmic sight reading studies showed a significant (p ⩽ .01) overall effect for treatment (d = −0.34, 95% CI [−0.50, –0.19]). The influence of the moderator variables was determined through the sub-group analyses and Q tests (which test for statistically significant differences) (see Table 1).

Table 1.

The overall significance of tested moderator variables (p) on rhythmic sight reading between treatment and control groups and the resultant effect size for each level of the moderator variables (d).

Moderator variable	Effect size		N	p
	d	SE
Publication type				.49
Unpublished thesis	−0.30*	0.09	53
Conference proceeding	0.00	0.71	1
Published paper	−0.50*	0.16	14
Age & experience level				.68
Elementary	−0.35*	0.17	22
Secondary	−0.30*	0.10	24
College non-musicians	−0.74	0.42	10
College musicians	−0.38*	0.12	9
Multiple	−0.03	0.29	3
Type of sight reading test^a				.26
Constructed for study	−0.45*	0.09	39
Published SR test	−0.05	0.19	9
Repertoire	0.26	0.93	3
WFPS-Rhythms	−0.36*	0.12	12
Response type^b				.14
Body percussion	−0.67*	0.19	14
Counting aloud/vocal	−0.41	0.25	5
Instrumental	−0.21*	0.09	44
Control task				.16
Equivalent task	−0.22*	0.10	41
No task	−0.63*	0.18	18
Within design	−0.28*	0.13	9
Pre- to posttest equivalency^c				.64
Equivalent	−0.37*	0.08	23
Same	−0.44*	0.15	30
Treatment				.03*
Aural Training	−0.10	0.11	7
Controlled Reading	0.12	0.13	8
Counting System	−0.82*	0.25	10
Creative Activities	−0.38	0.36	1
Movement	−0.33*	0.16	11
Notation	−0.06	0.20	12
Rhythmic Drill	−0.73*	0.19	8
Non-Categorized	−0.29	0.23	11

Note. Only analyses that provided information for the listed comparisons were included. Posttest-only studies were also excluded from this analysis.

Five studies did not report this information and were not included in analysis.^bFive studies did not report this information and were not included in analysis. ^cThree studies did not report this information were not included in analysis.

p ⩽ .05.

Of the coded moderator variables, only the treatment category was significant (p = .03). The treatments that significantly affected rhythmic sight reading were those focused on counting systems, movement, and rhythmic drill. Additional analyses were computed to determine whether coded moderator variables affected these treatments. For studies investigating counting systems, the published studies had a significantly higher effect size than theses (d = −1.65 versus −0.43) and those using a true control group had a significantly higher effect size than studies where the control group was given an equivalent sight reading task (d = −1.18 versus 0.03). Counting systems were also significantly affected by the type of sight reading test, and rhythmic drill studies were affected by the equivalency of the pre- and posttest. However in these last two instances, only one study was represented in one of the comparison groups, making the comparison unbalanced.

Melodic sight reading

The results of the analysis showed a significant (p ⩽ .01) overall effect for treatment (d = −0.35, 95% CI [−0.59, –0.12]) (see Table 2). Melodic sight reading was significantly influenced by most of the coded moderator variables; only type of sight reading test and response type were insignificant. The results of unpublished theses were more likely to favor treatments. Also, sight reading treatments influenced older students (secondary and college) more than younger students (elementary). Sight reading treatments were more influential in studies with a control group design rather than a repeated measures design as well as studies using the same test pre- to posttest. Type of treatment also had a significant effect on melodic sight reading. Collaborative and instrumental treatments were effective, but these results are based on only a few primary studies. The unique treatments included in the Non-Categorized category seemed to have the greatest effect on melodic sight reading.

Table 2.

The overall significance of tested moderator variables (p) on melodic sight reading between treatment and control groups and the resultant effect size for each level of the moderator variables (d).

Moderator variable	Effect size		N	p
	d	SE
Publication type				.03*
Published	0.39	0.37	5
Unpublished thesis	−0.47*	0.12	30
Age & experience level				< .01*
Elementary	−0.22	0.20	8
Secondary	−0.34*	0.09	9
College non-musicians	−2.74*	0.50	5
College musicians	−0.43*	0.17	8
Type of sight reading test^a				.43
Constructed for study	−0.08	0.15	19
Published SR test	−0.06	0.04	4
Repertoire	−0.12	0.18	2
WFPS	−0.40	0.21	2
Response type				.11
Instrumental	−0.59*	0.21	17
Vocal	−0.17	0.15	18
Control task				.02*
Equivalent task	−0.52*	0.17	24
No task	−0.48*	0.21	5
Within design	−0.19	0.20	6
Pre- to posttest equivalency^b				.01*
Equivalent	−0.07	0.16	17
Same	−0.99*	0.30	11
Treatment				.01*
Aural Training	−0.24	0.04	2
Collaborative	−0.20*	0.07	3
Controlled Reading	0.29	0.18	4
Counting System	0.17	0.34	1
Interval Drill	−0.64	0.38	3
Instrumental	−0.84*	0.35	1
Movement	−0.43	0.27	1
Notation	−0.19	0.17	2
Solfege	−0.35	0.33	7
Non-Categorized	−0.89*	0.34	11

Note. Only analyses that provided information for the listed comparisons were included. Posttest-only studies were also excluded from this analysis.

Eight studies did not report this information and are not included in analysis. ^bOne study did not report this information and is not included in analysis.

p ⩽ .05.

Sight reading comparisons

To determine whether there was a significant difference in how effective treatments were in enhancing rhythmic or melodic sight reading versus a combined sight reading task, a third meta-analysis was conducted which included studies measuring combined sight reading (previously published in Mishra, 2014) and the studies measuring only rhythmic sight reading or only melodic sight reading. Not all treatments that were included in the previous study (Mishra, 2014) had investigated rhythmic-only or melodic-only sight reading, but comparisons were made for treatments which had. For instance, the effect of collaboration on rhythmic sight reading has not been tested nor has the effect of creative activities on melodic sight reading (see Table 3). Treatments were coded to reflect whether the measurement reflected combined sight reading, rhythmic-only sight reading, or melodic-only sight reading.

Table 3.

Effect sizes (ES) by treatment using random effects model for combined sight reading, rhythmiconly sight reading, and meldoic-only sight reading.

Treatment	Combined		Rhythmic		Melodic		Overall ES	p
	ES	n	ES	n	ES	n
Aural Training	−0.33*	15	−0.10	7	0.24	2	−0.10	.07
Collaboration	−0.41	14			−0.20*	3	−0.22*	.37
Controlled Reading	−0.62*	11	0.12	8	0.29	4	−0.10	.01*
Counting Syllables			−0.72*	9	0.17	1	−0.42*	.03*
Creative Activities	−0.28*	9	−0.38	1			−0.29*	.78
Instrumental	−0.31	3			−0.84*	1	−0.50*	.24
Movement	0.27	2	−0.53*	6	−0.43	1	−0.08	.01*
Notation	0.16	5	−0.03	10	−0.19	2	−0.04	.41
Sight Reading Drill	0.08	18	−0.73*	8	−0.64	3	−0.07	< .01*
Solfege	−0.33*	26			−0.35	7	−0.33*	.95
Not Categorized	−0.15	21	−0.25	10	−0.89*	11	−0.28*	.18

p ⩽ .05.

The type of sight reading (rhythmic, melodic, combined) significantly affected the effectiveness of some treatments. Controlled Reading treatments significantly enhanced combined sight reading, but not rhythmic or melodic sight reading in isolation. Rhythmic sight reading was enhanced through the use of Counting Syllables, Movement, and Drill though these treatments were less effective when used with either melodic and/or combined sight reading. Changes in Notation was the only treatment that did not have an effect on sight reading in any of the conditions. Solfege was equally effective in enhancing combined and melodic sight reading.

Conclusion

The first purpose of this study was to determine whether experimentally tested sight reading interventions positively influenced rhythmic or melodic sight reading performance. Interventions that included a counting system, movement, or rhythmic drill effectively enhanced rhythmic sight reading while melodic sight reading was positively affected by collaboration and instrumental training. However, the benefits of collaboration and instrumental training on melodic sight reading should be taken as preliminary as there were fewer melodic sight reading studies overall and very few studies representing each type of treatment. All three analyses in the Collaborative category were based on one primary research study (Antholz, 1993) and the large effect of instrumental training was also based on one primary research study (Holland, 1985). Treatments that were not categorized also improved melodic sight reading, but it should be noted that four analyses with extremely large effect sizes in this category were part of one primary study (Brucksch, 1991). When the Brucksch study was removed from the analysis, the effect size for the Non-Categorized treatment category falls to 0.08. At this point, it is premature to base conclusions on the amount of data currently available. Much more research into treatments that may improve melodic sight reading are needed.

The study of rhythmic sight reading has been much more active and the effect of treatment type is clearer than with melodic sight reading. The addition of a counting system enhanced rhythmic sight reading. Rhythmic counting systems included the Gordon system (du, du-de), the Kodály system (ta, ti-ti), speech rhythms (e.g. “Mississippi”), the Sueta (too, ta-ta), and the traditional numerical counting system (1e&a). In a few of the studies, a syllabic counting system was compared with the numerical system, but most were compared with a control group. Though there was some fluctuation in the effectiveness of the various counting systems, over half (n = 6) of the primary studies resulted in large effect sizes (exceeding −0.70) and no one system was obviously better than the others. For instance, Palmer (1976) tested fourth grade students’ rhythmic performance before and after receiving instruction using either the Gordon counting system or Richard’s version of the Kodály counting system. She compared the results to a control group who had received no special rhythmic instruction. Both the Gordon and Richards systems resulted in significantly better rhythmic sight reading than the control group, though there was no difference between the two experimental groups. Only one study, comparing the Sueta to the numerical counting system, resulted in a positive effect size favoring the control group (numerical counting system) (Littlehale, 1991). The addition of pitch syllables (i.e. solfege) has been found to be effective in enhancing combined sight reading (Mishra, 2014) and the addition of solfege to melodic sight reading was also effective, though not significantly so, in the current meta-analysis (see Table 3). The addition of syllables (counting syllables or solfege) to sight reading appears to positively affect rhythmic, melodic (though additional research is needed), and combined sight reading indicating that this treatment is particularly robust in enhancing sight reading.

Rhythmic sight reading also was significantly enhanced by movement activities. Movement activities were broadly defined for this study and included clapping, foot tapping, conducting, as well as the Breath Impulse Technique (BRIM). BRIM uses breath pulsing through a wind instrument to help performers properly subdivide rhythmic patterns. There was no obvious advantage to the various movement activities used. Movement as a treatment to improve sight reading in general has not been widely investigated. The few studies that have investigated this treatment with melodic or overall sight reading reported mixed findings. It is possible that movement activities are only effective when attention can be focused on rhythmic production. When attention must be divided between melodic and rhythmic production, the positive effects of movement training dissipate. However, because of the moderately large (though insignificant) effect size found in the study (see Table 3) investigating the effects of movement on melodic sight reading, future research is warranted.

The third treatment to significantly enhance rhythmic sight reading was Rhythmic Drill, which commonly manifested in the primary research as some type of programmed instruction. It is unclear at this time what elements of the drill were particularly effective. The control group in most of the rhythmic drill studies (n = 5) received an equivalent training program, but the rhythmic drill treatment generally resulted in better results. However, rhythmic drill was not found to be an effective treatment in enhancing sight reading in general (Mishra, 2014). This treatment may only be effective when attention can be focused on rhythmic production.

Previous researchers suggest that rhythmic and melodic sight reading are different processes. This suggests that the interventions used to improve rhythmic and melodic sight reading might vary. The results of this study support this hypothesis. Some treatments were effective in enhancing either combined, rhythmic, or melodic sight reading, but not all three. Movement enhanced only rhythmic sight reading as did rhythmic drill. There are preliminary indications that collaboration and instrumental training enhance melodic sight reading, though these results are limited by the lack of available research. In a previous meta-analysis, aural training, controlled reading, and creative activities were found to enhance only combined sight reading (Mishra, 2014). In some cases, these results require much more research to verify, as few studies have investigated the use of these treatments (e.g. creative activities, instrumental experiences). However, the dissociated effects of treatment on the three forms of sight reading coupled with previous research reporting low correlations between rhythmic and melodic sight reading and neurological evidence suggests that sight reading consists of at least two unique skills (rhythmic and melodic) that are coordinated by an executive process.

The finding that movement activities enhanced rhythmic sight reading is important as it highlights sight reading as a multi-modal activity. Sight reading seems to include visual, aural, and motor processes. During sight reading the individual processes are combined into a multi-modal activity. Neurological research supports the idea of a multi-sensory executive process that can be enhanced by training (Paraskevopoulos et al., 2012). This finding is particularly relevant in the current investigation of interventions thought to improve sight reading. If rhythmic sight reading is distinct from melodic sight reading, it would follow that pedagogical interventions would differentially assist these skills. Some treatments would enhance rhythmic sight reading while others would enhance melodic sight reading. However, an executive process would be needed to unify the components during combined sight reading and this may be influenced by yet a different set of treatments. Sight reading, then, may not be simply summative; an addition of melodic and rhythmic sight reading. Each component of sight reading may function uniquely and then must be combined using an integration process. Training rhythmic and melodic sight reading in isolation would not necessarily result in an enhanced overall sight reading performance. This would explain the observation noted earlier that researchers attempting to improve sight reading by focusing on rhythmic drill were, on the whole, unsuccessful. Rhythmic drill does appear to enhance rhythmic sight reading, but the treatment is not sufficiently robust to enhance sight reading when melodic material is also produced.

Viewing sight reading as having distinct components changes how teachers approach students with sight reading difficulties. Musicians may be impaired only in rhythmic or melodic sight reading or musicians may only have difficulty combining the components. The most effective pedagogical intervention would then depend on the nature of the difficulty. Musicians with rhythmic sight reading difficulties may utilize movement or rhythmic drill activities to address the problem while musicians who have difficulty integrating rhythm and melody may focus on aural skills training or forcing eyes forward in the music (Mishra, 2014).

Results of these meta-analyses indicate that treatment type significantly influenced rhythmic and melodic sight reading. The addition of verbal syllables, whether counting or solfege, seems to be a robust treatment that may benefit all aspects of sight reading. On the other hand, varying notation did not improve any type of sight reading (rhythmic-only, melodic-only, or combined). Further results from melodic sight reading are preliminary due to the limited number of primary research studies that have investigated treatments specifically thought to improve melodic sight reading. Much more research is needed in this area. However, based on the information available, sight reading with accompaniment or in a group (Collaborative) or studying instruments such as keyboard (Instrumental Training) positively affected melodic sight reading. More information is available for interventions that influence rhythmic-only sight reading. For rhythm, treatments focused on counting systems, included movement, or rhythmic drill positively affected sight reading.

Footnotes

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Adachi

Takiuchi

Shoda

(2012). Effects of melodic structure and meter on the sight-reading performances of beginners and advanced pianists. In Cambouropoulos

Tsougras

Mavromatis

Pastiadis

(Eds.), Proceedings of the 12th International Conference on Music Perception and Cognition. Retrieved from http://icmpc-escom2012.web.auth.gr/sites/default/files/papers/5_Proc.pdf

Allen

D. H.

(1987). The effect of mastery of selected music theory and ear training skills presented in a computer-assisted format on the sight playing performance of second-year band students (Master’s thesis). University of Missouri, Kansas City.

Antholz

J. R.

(1993). The effect of accompaniment types on sightsinging note accuracy (Unpublished master’s thesis). University of Missouri, Kansas City.

Barlar

N. E.

(2010). The effect of sight-reading instruction on the language reading fluency and music sight-reading ability of middle school band students (Doctoral dissertation). Boston University, MA.

Bent

I. D.

Hughes

D. W.

Provine

R. C.

Rastall

Kilmer

Hiley

Chew

(n.d.). Notation. In Grove Music Online. [Online subscription access]. Oxford, UK: Oxford Music Online.

Bengtsson

S. L.

Ullen

(2006). Dissociation between melodic and rhythmic processing during piano performance from musical scores. Neuroimage, 30, 272–284.

*Boyle

J. D.

(1970). The effect of prescribed rhythmical movements on the ability to read music at sight. Journal of Research in Music Education, 18(4), 307–318.

*Brucksch

A. E.

(1991). The effect of mental rehearsal on sight reading by beginning college guitarists (Unpublished master’s thesis). Bowling Green State University, OH.

Brust

J. C. M.

(1980). Music and language: Musical alexia and agraphia. Brain, 103, 367–392.

10.

*Byo

J. L.

(1992). Effects of barlines, pitch, and meter on musicians’ rhythm reading performance. Journal of Band Research, 27(2), 34–44.

11.

Cappelletti

Waley-Cohen

Butterworth

Kopelman

(2000). A selective loss of the ability to read and to write music. Neurocase, 6, 321–332.

12.

Comprehensive Meta-Analysis. (2010). Comprehensive Meta-Analysis (Version 2.2.057) [Computer software]. Englewood, NJ: Biostat. Retrieved from http://www.meta-analysis.com/

13.

Cooper

H. M.

(2010). Research synthesis and meta-analysis: A step-by-step approach. Thousand Oaks, CA: Sage.

14.

Earney

M. E.

(2008). The effects of aural rhythmic dictation on the sight-reading abilities of seventh and eighth grade band students (Master’s thesis, University of Kansas). Retrieved from Dissertations & Theses: Full Text. (Publication No. AAT 1459804)

15.

Egbert

J. B.

(1990). The effects of systematic rhythm reading instruction versus rote rhythm drill on the pitch and rhythm sight-singing performance of high school choral ensemble members (Doctoral dissertation). University of Missouri, Columbia.

16.

Elliott

C. A.

(1982). The identification and classification of instrumental performance sight-reading errors. Journal of Band Research, 18, 36–42.

17.

Henry

M. L.

(2011). The effect of pitch and rhythm difficulty on vocal sight-reading performance. Journal of Research in Music Education, 59(1), 72–84.

18.

*Holland

L. S.

(1985). A comparison of two variations of a behavioral model approach for teaching pitch reading at the junior high level: with keyboard training or without keyboard training (Unpublished doctoral dissertation). University of Missouri, Kansas City.

19.

Laing

D. R.

(2007). The effect of rhythm pattern instruction on the sight-reading achievement of wind instrumentalists (Unpublished doctoral dissertation). University of Missouri, Columbia.

20.

*Littlehale

(1991). Rhythm reading and the beginning band student: The effectiveness of two rhythm syllable systems for beginning instrumentalists (Unpublished master’s thesis). Indiana University, Bloomington.

21.

London

(n.d.). Rhythm. In Grove Music Online [Online subscription access]. Oxford Music Online.

22.

Massingale

G. W.

(1979). A study to determine the effect of a program of rhythmic training on the ability to perform music at sight (Doctoral dissertation). North Texas State University, Denton, TX.

23.

McPherson

G. E.

(1994). Factors and abilities influencing sight reading skill in music. Journal of Research in Music Education, 42(3), 217–231.

24.

Midorikawa

Kawamura

Kezuka

(2003). Musical alexia for rhythm notation: A discrepancy between pitch and rhythm. Neurocase, 9(3), 232–238.

25.

Mishra

(2014). Improving sightreading accuracy: A meta-analysis. Psychology of Music, 42(2), 131–156. doi: 10.1177/0305735612463770

26.

Parker

R. C.

(1979). The relative effectiveness of the TAP system in instruction in sight singing, an experimental study (Doctoral dissertation, University of Miami). Retrieved from Dissertations & Theses: Full Text. (Publication No. AAT 8014160)

27.

*Palmer

(1976). Relative effectiveness of two approaches to rhythm reading for fourth-grade students. Journal of Research in Music Education, 24(3), 110–118.

28.

Paraskevopoulos

Kuchenbuch

Herholz

S. C.

Pantev

(2012). Evidence for training-induced plasticity in multisensory brain structures: An MEG study. PLoS ONE 7(5), e36534. doi:10.1371/journal.pone.0036534

29.

Pike

P. D.

Carter

(2010). Employing cognitive chunking techniques to enhance sight-reading performance of undergraduate group-piano students. International Journal of Music Education, 28(3), 231–246.

30.

Sanderson

(1932). An objective study of reading musical notation (Unpublished master’s thesis). Northwestern University, Evanston, IL.

31.

Schon

Besson

(2002). Processing pitch and duration in music reading: A RT-ERP study. Neuropsychologia, 40(7), 868–878.

32.

Searle

J. W.

(1985). An investigation of movement to music and rhythmic pattern reading flash-slide training in fifth- and sixth-grade music classes (Unpublished doctoral dissertation). University of Michigan, Ann Arbor.

33.

Stegall

J. C.

(1992). The influence of isolated rhythmic training with a selected method of study on the ability to sing music at sight (Unpublished doctoral dissertation). University of Missouri, Kansas City.

34.

Thomson

(1953). An analysis of difficulties in sight reading music for violin and clarinet (Unpublished doctoral dissertation). University of Cincinnati, OH.

35.

Tucker

G. L.

(1969). The influence of isolated rhythmic drill on growth in sight singing (Unpublished doctoral dissertation, University of Oklahoma, Norman). Retrieved from Dissertations & Theses: Full Text. (Publication No. AAT 7002342)

36.

Van Nuys

Weaver

H. E.

(1943). Memory span and visual pauses in reading rhythms and melodies. Psychological monographs, 55(1), 1–50.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB