The Relationship of Pitch Sight-Singing Skills With Tonal Discrimination,Language Reading Skills,and Academic Ability in Children

Abstract

This study investigated correlates that might explain variance in beginning sight-singing achievement, including tonal discrimination, reading fluency, reading comprehension, and academic ability. Both curriculum-based and standardized tests were used, including the Intermediate Measures of Music Audiation, Otis-Lennon School Ability Test, and Dynamic Indicators of Basic Early Literacy Skills. Sight-singing ability of second-grade students (N = 170) was individually assessed for pitch accuracy only using four-note tonal patterns following a 16-week instructional period and again 8 weeks later following a period of no practice. A factor analysis explained 62% of the variance across 13 variables, revealing correlated factors of Music Ability, Reading Ability, and Academic Ability. Regression analyses with individual variables as predictors indicated that significant variance in sight-singing achievement beyond that explained by pitch matching ability could be explained by reading comprehension ability. Similar results were found with both sight-singing tests. Findings are discussed in relation to Patel’s shared syntactic integration resource hypothesis and the need to advocate for music education programs.

Keywords

sight-singing reading comprehension elementary children music reading advocacy for music education

The ability to sight-sing pitches is a complex skill involving many subskills. The performer must perceive and decode the relevant information in a visual stimulus, formulate aural images of the specific sound frequencies represented, and then execute motor skills required to accurately produce those frequencies. Developing sight-singing skill to some level of proficiency eludes most of the general population and is challenging even among many of those actively engaged in music training. Henry and Demorest (1994) found that high school students in choirs with outstanding contest ratings for group sight-singing achieved a mean accuracy of only 66% when individually sight-singing a very simple melody. According to Demorest (2004), “sight-singing is one of the most important, and perhaps most difficult, skills for the choral singer to master” (p. 3). The failure to acquire music-reading fluency can hinder a music student’s progress and has been suggested as a major reason for discontinuing music study (Mills & McPherson, 2006). Observing that an increasing number of students auditioning to be collegiate music majors were unable to read music, Asmus (2004) reaffirmed the importance of prioritizing music-reading skill acquisition in music education. Gudmundsdottir (2010) pointed out that the study of staff notation reading is an important but neglected area of music education research, particularly related to the acquisition in childhood, even though that is the time when music-reading skills are most often developed. She suggested that research studies need to examine the cognitive components involved in music-reading tasks to improve instructional strategies.

Studies that investigate music-reading skills frequently employ sight-reading tasks. Sight-reading refers to the act of performing an unfamiliar musical passage on an instrument while reading the music notation for the first time. The term sight-singing is often thought of similarly except the sounds represented by the notation are produced by the voice. Singing notated music, however, requires additional skill, particularly as it relates to pitch production, because the performer must mentally construct aural images of notated pitches and then produce them, all without the aid of a mechanical pitch source. In the absence of absolute pitch, a singer relates to internal representations of structures such as intervals, scale–degree relationships, and harmonic function to formulate the auditory representation of the next note before singing it. Because of this additional and formidable complexity, the term sight-singing is broadened in certain circumstances to include singing while reading previously seen notation. If a singer is using previously seen notation and is not simply recalling an aural memory of the passage but instead is dependent on the notation to guide the mental construction of auditory imagery and vocal production, he or she is usually considered to be sight-singing. To identify possible constructs related to the development of beginning sight-singing skills in young students, I examined previous research that demonstrated music experiences or skills are related to abilities in general intelligence, beginning language reading, aural discrimination, and processing of language syntax.

Music learning in general has been associated with gains in overall intellectual ability and academic achievement (for a review, see Rauscher, 2009). Studies with 5- and 6-year-olds, for example, found that the students who participated in a series of music lessons experienced significantly greater gains in IQ scores than those who did not receive music lessons (Kaviani, Mirbaha, Pournaseh, & Sagan, 2014; Schellenberg, 2004, 2006), though a tenable explanation for an association between music learning and IQ gains remained unclear. While Schellenberg (2004) reasoned that music lessons may improve IQ because they provide additional educational experiences, gains might be attributed to improved performance on a particular shared underlying cognitive process or construct.

Theorists have suggested that music and language may share some cognitive mechanisms because the two domains seem to employ similar executional demands, such as the perception and interpretation of acoustic sequences and the interpretation of a system for mapping mental images to sounds (Forgeard et al., 2008; Jackendoff, 2009; McMullen & Saffran, 2004; Patel, 2008; Slevc & Okada, 2015). For example, a common pitch processing mechanism is believed to underlie aural processing of both music and speech (Asaridou & McQueen, 2013; Besson, Chobert, & Marie, 2011; Moreno et al., 2009). With a meta-analysis of 24 studies, Butzlaff (2000) concluded that there is a strong and reliable relationship between music study and performance on standardized reading tests. Searching for a potential basis for this association, studies focusing on auditory processing skills of young children have identified a relationship between aural discrimination of music and phonological awareness (Degé & Schwarzer, 2011; Forgeard et al., 2008; Gromko, 2005; Lamb & Gregory, 1993, Lucas & Gromko, 2007; Tsang & Conrad, 2011). Going a step further, Anvari, Trainor, Woodside, and Levy (2002) found music ability and pitch discrimination ability to be significant predictors of 4- and 5-year-olds’ reading test scores, even after removing the variance they shared with phonemic awareness. In the decade that followed, the annual number of neural studies investigating the relationship of music and language increased linearly (Peretz, Vuvan, Lagrois, & Armony, 2015). Peretz et al. (2015) pointed out that increasing our understanding in this area is important for educational purposes.

A theoretical framework to explain how the distinct domains of music and language may be related has been proposed with Patel’s (2003, 2008, 2012) shared syntactic integration resource hypothesis (SSIRH). According to the SSIRH, two types of networks are used in both language and music cognition. Representation networks are used to provide long-term storage of domain-specific knowledge. Therefore, mental representations of music and linguistic information would be stored in separate and disassociated representation networks. Resource networks are used to access and process information in working memory, and they are believed to be shared across domains. Therefore, the same cognitive processing resources would be used for the syntactic comprehension of both language and music, though each domain is very different in its structural organization. When cognitively processing a sequence of events in language or music, incoming elements (i.e. words or notes) must be related to and integrated with the cognitively maintained structure that has been established by previous elements. Successful sight-singers have been reported to use syntax processing strategies such as maintaining a sense of key (Henry, 2001), tonicizing to activate a tonal framework (Killian & Henry, 2005), and using harmonic predictions (Fine, Berry, & Rosner, 2006).

The theory of shared syntactic processing has been supported by experiments in which participants were asked to process aurally presented music containing harmonic or melodic violations simultaneously with lyrics or visually presented sentences containing syntactic violations (Fedorenko, Patel, Casasanto, Winawer, & Gibson, 2009; Slevc, Rosenberg, & Patel, 2009). For example, listening to syntactically complex music was found to impair recall of visually presented sentences but did not impair recall of a word list, which did not impose syntactical processing demands (Fiveash & Pammer, 2014).

In an educational setting, processing of language syntax might be more effectively assessed by a reading comprehension measure as opposed to a simple oral reading assessment that would not require the same level of cognitive processing. Similarly, processing of music syntax might be effectively assessed by a sight-singing task in which success is dependent on comprehension and ongoing processing of an overall tonal context and harmonic structure. Corrigall and Trainor (2011) found that reading comprehension could be predicted by length of music training for 6- to 9-year-olds who were taking instrumental music lessons. Further investigation is warranted given the absence of an authentic musical skill variable and the fact that children who elect to take music lessons are likely to be similar to each other on a number of variables.

In addition to the aural and syntactical processing associations of linguistic and music reading, both domains subsume visual decoding demands, requiring the dynamic transformation of visual symbols into the phonemes or musical sounds represented, which may or may not be rendered orally. Moreno et al. (2011) suggested that learning grapheme–phoneme mapping of word decoding is similar to learning notation–sound mapping of music reading. In their study with children ages 4 to 6, significantly greater improvement in ability to associate unfamiliar visual symbols with familiar words occurred after participating in music theory training versus art training. Reading words and sight-reading music notation were suggested to be neurally and functionally separable abilities in previous cognitive neuroscience studies (for a review, see Peretz & Zatorre, 2005); however, a word reading task may not demand the same higher level cognitive processing that music sight-reading requires.

While researchers and teachers have affirmed the importance of learning sight-singing skills and the need to examine possible cognitive components as a way to possibly improve teaching and learning of music reading skills, very little research has investigated development and components of sight-singing skill. Much research has been conducted to show that skills and experiences in music are related to general intelligence and early reading development, but sight-singing skill has not been considered specifically. Furthermore, Patel’s recent hypothesis of SSIRH that theorizes a common underlying syntax processing mechanism across both language and music has not been considered. Because syntax processing of music has been shown to be integral to successful sight-singing, SSIRH would suggest that sight-singing skills may be related to language processing skills.

This study examined the interrelationship of pitch sight-singing achievement with other variables that have been shown by previous research to be related to music achievement in general, including tonal discrimination, general academic ability, and language reading measures of reading fluency and reading comprehension. Each of these variables, described in more detail in the following, are assessments that might typically be administered in an educational setting. Similar to how linguistic reading skills are taught and assessed in schools, sight-singing skills were taught to all students, not just those who self-elected to pursue additional music training. These skills were taught to students at the elementary grade level, perhaps the most ideal time to render the opportunity for the acquisition of music-reading skills across the general population, similar to that of the acquisition of linguistic reading skills. The investigation was designed to answer the following research questions: (1) After examining the strength of the interrelationships of the variables being considered, what larger underlying constructs can be defined by the commonalities among the variables? and (2) After controlling for vocal pitch matching accuracy, do any of the specific variables contribute significantly to explaining variance in the development of pitch sight-singing skills? To examine the first research question, factor analysis was used to objectively and parsimoniously observe the underlying variable structure, providing a means of examining the construct validity of individual assessments. The identification of underlying constructs also informed the second research question because two variables measuring the same construct would have much in common; therefore, the explanatory capacity of one variable would be obfuscated in the presence of the other. To examine the second research question, regression analyses were conducted using sight-singing achievement scores as a dependent variable to identify explanatory variables that accounted for significant variance in pitch sight-singing achievement with beginners.

Participants and Instructional Method

Participants in this study were second-grade students (N = 170), 87 girls and 83 boys, ranging in age from 6.25 years to 8.75 years (M_age = 7.5 years) at the midpoint of the study, from three public schools in an urban northeastern United States school district. With second-graders, singing has been found to be the most frequently employed instructional activity in general music classes, and the percentage of class time focused on reading and notating music has been found to increase substantially from Grade 1 to Grade 2 (Orman, 2002). The students participated in a prescribed program of sight-singing instruction during a 25-minute portion of once-weekly general music classes for 16 sessions beginning around mid-October. Because the students were at the very beginning stages of learning to sing from notation, sight-singing instruction and assessment was focused exclusively on accurate vocal pitch responses to visual pitch stimuli, excluding any rhythm component. The instruction, provided by the students’ regular music teacher during regularly scheduled music time, involved reading pitch notation while singing tonal patterns and singing rote songs while playing singing games.

The tonal patterns, notated as large flash cards, consisted of four noteheads placed on a five-line staff. Only five different pitches were included across all patterns with the range extending from F⁴ (do) to D⁵ (la) on a treble staff, though no clef sign was included. The pitches were within the range that second-grade children should be singing (Goetze, Cooper, & Brown, 1990). The key of F made the reading task as easy as possible; all the notes appeared within the staff lines, and a flat sign was not needed because the fourth scale degree (fa) was not used. Also, a consistent key could be mentally referenced with all patterns when processing visual and aural imagery. Patterns began on F⁴ (do), A⁴ (mi), or C⁵ (so), as was typical for beginnings of songs the students sang. Some patterns were extracted from the beginning of songs, and some were researcher constructed.

During each class, the students read notation to sing one new pattern and all previously introduced patterns for three trials both at the beginning and near the end of the 25-minute portion of the class. On the first trial, the students sight-read the notation to sing the pattern. To promote accurate learning, on trials two and three, the students first heard the teacher sing the pattern and then echoed while reading the notation. Between the pattern practice portions of the class, the students sang rote songs. In week 16, the 15 previously introduced patterns were reviewed. Instructional procedures and duration were consistent across all classes and ended around mid-March.

Variables

Sight-singing performance was assessed with two separate tests that were administered following the instructional period. Both previously practiced and unfamiliar tonal patterns were used to assess sight-singing skills. Though the students were believed to be sight-singing as previously defined, the task will be referred to as a notation-singing test. Data were also collected from additional tests administered during the year assessing the students’ abilities in tonal discrimination, linguistic reading fluency and comprehension, and general academic ability. All variables were metric measures and were believed to be related to the acquisition of sight-singing skills in children based on previous research findings. Table 1 presents a list of the variables along with information about range and reliability. Reliabilities ≥.80 are generally considered adequate, although reliabilities ≥.70 may be considered sufficient in the early stages of research (Nunnally & Bernstein, 1994).

Table 1.

Ranges and Reliabilities of Music and Academic Assessments.

Task Type	Variable	Maximum	Reliability
Sight-singing	Notation-singing (pitch accuracy) test 1	75	.88
Sight-singing	Notation-singing (pitch accuracy) test 2	75	.88
Singing accuracy	Starting pitch matching test 1	25	.89
Singing accuracy	Starting pitch matching test 2	25	.89
Tonal discrimination	Intermediate Measures of Music Audiation Tonal Subtest	40	.78^a
	Tonal subtest pretest		.77
	Tonal subtest posttest		.77
Language reading	Dynamic Indicators of Basic Early Literacy Skills	150	.89–.94^a
	Reading guide—pretest and posttest	20	.90
	Reading comprehension—pretest and posttest	12	.73
Academic ability	Otis-Lennon School Ability Test
	Verbal subtest	30	.79–.80^a
	Nonverbal subtest	30	.80–.82^a

Denotes reliability reported by publisher.

Singing and Notation Reading

Around mid-March, each student’s sight-singing skill was assessed with notation-singing test 1, which involved reading and singing 25 randomly ordered tonal patterns, only 15 of which had been practiced in class. Around mid-May, skill retention was assessed on the same patterns with notation-singing test 2. Throughout the 8 weeks between the two notation-singing tests, no pattern reading practice occurred. The tests were individually administered in a quiet place. During the test, the student was shown a pattern and heard the first note of the pattern played on a resonator bell, as had been practiced during the instructional period. A preliminary exercise of echoing an adult soprano while reading three practice patterns familiarized the student with the procedure and established the tonality of the patterns. The first note of each pattern sung by the student yielded a score for pitch-matching accuracy (25 points possible). The three remaining notes sung by the student yielded a score for sight-singing pitch accuracy (75 points possible). The students’ singing tests were recorded for later scoring.

To score the notation-singing tests, the recordings were randomly ordered and given to a retired elementary music teacher who listened to the recordings and wrote down the announced pattern number and the letter names of the four pitches the student sang for each pattern. Due to limitations in vocal control for children at this age (Cooper, 1992), precise intonation was not required as long as the teacher, who had extensive experience with listening to singers at this age, was able to make a determination of which step of the scale the student was attempting to sing. The judge had no knowledge of what notes were actually used in each pattern. To assess the reliability of the scoring, a second retired elementary music teacher independently judged approximately 50% of the tests. Reliability of the singing variables, as determined by the number of agreements divided by the number of agreements plus disagreements, ranged from r = .88 to .89 (see Table 1).

Tonal Discrimination

The students’ tonal discrimination ability was assessed with the tonal subtest of E. E. Gordon’s (1982) Intermediate Measures of Music Audiation (IMMA-T) prior to the beginning and after the end of the instructional program (approximately October and March). The IMMA is a standardized measure of musical aptitude designed for children in Grades 1 through 4. While the test includes tonal and rhythm subtests, only the tonal subtest was used. It consists of a recording of three-note pattern pairs for which the listener must identify whether the two patterns are the same or different for 40 trials. Isochronous sinewave tones are used, requiring the listener to discriminate exclusively by pitch. In a longitudinal study of the predicative validity of the IMMA involving boys in Grade 4, IMMA tonal subtest scores correlated with achievement in singing, learning to play violin, and learning to play recorder in the range of r = .54 to .90 (E. E. Gordon, 1986). The reported split-halves reliability of the test with second-grade students is r = .78. In the present study, split-halves reliability assessed with the Spearman-Brown formula was r = .77 for both administrations.

Academic Ability

The students’ academic ability was assessed with the Otis-Lennon School Ability Test (7th ed.; OLSAT; Otis & Lennon, 1997), administered in October. The OLSAT is designed to measure abstract thinking and reasoning ability required for learning and success in school. There are two subsections of the test, with the distinction made based on whether language is requisite to answering the item. The verbal component includes 30 items related to following directions, aural reasoning, and arithmetic reasoning. The nonverbal component includes 30 items related to picture classification, picture analogies, figural classification, figural analogies, pattern matrix, and figural series. Scores from the verbal component (OLSAT-V) and the nonverbal component (OLSAT-N) provided a standardized measure of the students’ general academic ability. According to the publishers, the seventh edition changed the phrase mental ability used in the previous edition to school ability as a way to discourage overinterpretation of abilities that the test purports to measure (Psychological Corporation, 2002). On Form C (used in this study) with Grade 2 students, publisher reported reliability coefficients were r = .79 to .80 for the verbal component and r = .80 to .82 for the nonverbal component.

Linguistic Reading

Reading fluency

To assess linguistic reading ability, students were individually administered a subtest of the Dynamic Indicators of Basic Early Literacy Skills (DIBELS), a widely adopted standardized test of early reading ability (Good & Kaminski, 2002). In the oral reading fluency subtest (DORF), the student is required to orally read the provided passage for 1 minute, and the total number of words read within that time becomes the score representing the student’s oral reading fluency rate. The benchmark goal for the end of second grade is 90. In the Administration and Scoring Guide, Good and Kaminski (2002) report reliabilities for curriculum-based measurements (CBM) of reading upon which this instrument is modeled. Using data reported in 1983 by Tindal, Marston, and Dino (as cited in Good & Kaminski, 2002), reliabilities for CBM in reading fluency with elementary students have been obtained in the r = .89 to .94 range for alternate forms. These coefficients are based on older studies, however, not on the passages used with the DIBELS assessments. The DIBELS assessment has come under criticism by some reading experts because of the lack of research evidence establishing a firm relationship between fluency and comprehension or overall reading proficiency and a lack of independent peer-reviewed validity studies (Shelton, Altwerger, Jordan, 2009). The present study included additional assessments of language reading ability.

Reading comprehension

Reading ability was also assessed with curriculum-based assessments constructed by the publisher of the students’ reading textbook. Two types of curriculum-based tests were administered: Reading Guide and Reading Comprehension. The Reading Guide test required the student to read a passage and then retell the passage details listed on the test administration guide. The Reading Comprehension test required the student to answer questions requiring a more general comprehension of the passage that was read. Both tests were individually administered by a reading specialist near the beginning of the year and the end of the year. Because the pre- and posttests were not equivalent forms but should be measuring the same underlying construct, reliability was determined by comparing the performance of all students across the two administrations of each assessment type with Cronbach’s alpha (Eisinga, Grotenhuis, & Pelzer, 2013). Reliability was α = .90 for the Reading Guide tests and α = .73 for the Reading Comprehension tests, indicating that the two administrations of each test type were reliably assessing the same construct.

Results

The means and standard deviations for the variables and the correlations among the variables can be seen in Table 2. All bivariate correlations were significant at p < .05, with all but three significant at p < .01. Subsequent analyses were conducted to (a) synthesize the complex interrelationships among these correlated variables and (b) identify variables that explained a significant portion of the variance in sight-singing achievement. A preliminary analysis assessed the appropriateness of conducting an exploratory factor analysis (Tabachnick & Fidell, 2012). The sample size was deemed adequate with a participant to variable ratio of 13:1. The Bartlett test of sphericity was significant (χ² = 1,385.9, p < .001), indicating the variables were sufficiently intercorrelated to produce representative factors. The Kaiser-Meyer-Olkin measure of sampling adequacy was .84, well above the threshold of .50, indicating that the variables as a set were appropriate for factor analysis. Also, the measure of sampling adequacy values for each of the variables were all well above .5; therefore, all 13 variables were retained for the analysis.

Table 2.

Means, Standard Deviations, and Intercorrelations of Music and Academic Assessments With Second-Graders (N = 170).

Assessment	M	SD	2	3	4	5	6	7	8	9	10	11	12	13
1. Notation-singing test 1	20.19	13.59	.78	.80	.64	.33	.33	.24	.32	.32	.32	.25	.36	.22
2. Notation-singing test 2	20.04	13.61		.66	.79	.34	.31	.25	.21	.34	.29	.22	.30	.19
3. Starting pitch matching test 1	10.04	5.91			.64	.33	.33	.20	.23	.27	.35	.22	.33	.21
4. Starting pitch matching test 2	10.21	6.32				.39	.38	.25	.21	.24	.30	.20	.28	.22
5. IMMA-tonal pretest	29.75	5.43					.40	.18	.24	.24	.20	.16	.34	.29
6. IMMA-tonal posttest	31.26	4.36						.25	.35	.24	.24	.24	.24	.34
7. DIBELS (reading fluency)	82.42	34.51							.59	.69	.77	.77	.38	.27
8. Reading comprehension pretest	5.10	2.48								.57	.58	.58	.49	.43
9. Reading comprehension posttest	7.24	3.04									.59	.63	.54	.34
10. Reading guide pretest	10.06	2.76										.81	.42	.36
11. Reading guide posttest	14.27	3.02											.40	.36
12. OLSAT-verbal	14.47	5.25												.63
13. OLSAT-nonverbal	15.09	5.28

Note: Values of r ≥ .16 and r ≥ .20 are significant at p > .05 and p > .01, respectively. IMMA = Intermediate Measures of Music Audiation; DIBELS = Dynamic Indicators of Basic Early Literacy Skills; OLSAT = Otis-Lennon School Ability Test.

Principal axis factoring was employed to consider the shared variability among the variables while disregarding the unique and error variability. Using the latent root of ≥1.0 as a stopping criterion, three factors were retained. Both the eigenvalue of .85 for a potential fourth factor and the inflection point on the scree plot supported the appropriateness of a three-factor solution. The three factors explained 62% of the shared variance, indicating that no variables with low communalities needed to be removed from the analysis. Because the factors were believed to be correlated, the factor solution incorporated direct oblimin rotation to obtain simple and interpretable factors.

The loadings for each of the variables on the three factors can be seen in Table 3. Given the sample size of 170, factor loadings of ±.30 are considered minimally acceptable, and loadings of ±.45 or greater are considered significant for interpretative purposes (Hair, Black, Babin, & Anderson, 2010). Using this threshold, all variables except the IMMA-T loaded significantly on one and only one factor. One cross-loading >.30 occurred with the Reading Comprehension pretest variable, but it loaded significantly on only one factor. Many variables have loadings in excess of .70, indicating that the factor is accounting for at least 50% of the variance that the variable shares with all other variables in the analysis.

Table 3.

Oblimin-Rotated Principal Axis Factor Loadings, Communalities, and Cronbach’s Alpha of Music and Academic Assessments With Second-Graders (N = 170).

Variable	Factor			Communality	Cronbach’s Alpha
Variable	1	2	3	Communality	Cronbach’s Alpha
Reading Achievement					.42
DIBELS (reading fluency)	.947	.003	−.106	.732
Reading guide posttest	.901	−.040	−.015	.729
Reading guide pretest	.858	.084	−.035	.755
Reading comprehension posttest	.645	.054	.161	.613
Reading comprehension pretest	.533	−.023	.321	.525
Music Achievement					.86
Notation-singing test 2	.036	.905	−.085	.764
Notation-singing test 1	.035	.856	.005	.770
Starting pitch matching test 2	.003	.823	−.005	.684
Starting pitch matching test 1	.019	.804	.002	.695
General Academic Ability					.70
OLSAT-nonverbal	.032	−.090	.803	.473
OLSAT-verbal	.164	.034	.668	.560
IMMA-tonal pretest	−.053	.299	.339	.270
IMMA-tonal posttest	.046	.271	.300	.313
Rotated sums of squared loadings	4.21	3.87	3.08
Variance accounted for	40.4%	15.5%	6.2%	Total: 62.2%

Note: Factor loadings ≥ ±.45 in bold are considered significant for interpretive purposes. DIBELS = Dynamic Indicators of Basic Early Literacy Skills; OLSAT = Otis-Lennon School Ability Test; IMMA = Intermediate Measures of Music Audiation.

The factor structure that emerged represents three distinct groups of variables. Factor 1 included five variables and accounted for 40% of the variance shared across all 13 variables. Each of the variables assessed the students’ skill in oral reading or comprehension of written language, suggesting that Factor 1 represents a general component of reading achievement. Factor 2 included four variables and accounted for 16% of the variance shared across all variables. All the variables loading on Factor 2 assessed the students’ music performance skills in pitch matching and pitch accuracy when sight-singing. These variables represent a general component of music achievement, albeit specific to singing and sight-singing. Factor 3 accounted for an additional 6% of the shared variance but is more difficult to define due to cross-loadings and a lack of some significant loadings. The factor included the school ability measures and the music aptitude measures. These disparate variables suggest that the factor may represent a component of general academic ability. The communalities, also shown in Table 3, indicate how well each variable is explained by the three factors. Communalities for the IMMA-T indicate that about 70% of the variance in their scores was not accounted for, suggesting that the IMMA-T is measuring some other construct not defined by this factor solution. The factors are correlated, supporting the appropriateness of oblique rotation. Music Achievement has a moderate positive association with General Academic Ability (r = .42) and a weak positive association with Reading Achievement (r = .33). Reading Achievement also had a moderate association with General Academic Ability (r = .48).

To estimate the reliability of each factor, the internal consistency of the scale created with all variables having significant loadings on the factor was calculated with Cronbach’s alpha. For this analysis, the IMMA-T was included with the General Academic Ability factor. Results indicated α = .86 for Music Achievement, α = .70 for General Academic Ability, and α = .42 for Reading Achievement. It should be noted that an estimated reliability of the Reading Achievement scale would be α = .87 if the DIBELS variable was omitted. The lower obtained reliability appears to be an artifact of the greater variability of the DIBELS scores occurring as the result of the greater range in scores on that assessment as compared to the other reading assessments.

The goal of the second research question was to determine if a significant portion of variance in learning to sing pitches from notation could be explained by academic ability, language reading ability, or tonal discrimination ability after controlling for pitch matching ability. These variables were examined with multiple regression using a combination of hierarchical and stepwise methods; performance on notation-singing test 1 and notation-singing test 2 were criterion measures in separate analyses. The single variable of starting pitch matching on the respective test was entered on Step 1 to control for the students’ pitch matching ability before considering the contribution of the other variables in explaining notation-singing achievement. Step 2 statistically evaluated the following nine additional independent variables for inclusion in the model: academic ability (OLSAT verbal and nonverbal), reading fluency (DIBELS), and the pretest and posttest scores for the assessments in reading guide, reading comprehension, and tonal discrimination (IMMA-T). Scores from the specific assessments were used rather than composite factor scores to facilitate direct comparisons of these results to variables in past and future studies. The two separate notation-singing tests allowed for a comparison of results across different analyses as a form of cross-validation in the absence of a completely new sample of students.

Initial screening for outliers that might adversely affect the interpretation of results was conducted. Three cases were identified as multivariate outliers using a p < .001 criterion for Mahalanobis distance, but none of these cases were identified as bivariate outliers using each dependent variable in combination with each of the independent variables; therefore, all cases were retained. The data were also examined for conformity with the assumptions of normality, homoscedasticity, and linearity. Univariate normality assessed with Kolmogorov-Smirnov tests indicated all but one variable had nonnormal distributions; however, multivariate normality examined with residuals plots indicated normally distributed errors. Levene tests indicated no significant departure from homogeneity of variance for any of the variables. A scatter plot of the standardized residuals against the standardized estimates showed a random dispersion of the data, indicating the assumption of linearity was met.

The regression analysis using notation-singing test 1 as the criterion indicated that starting pitch matching accounted for 63.5% of the variance in achievement scores on Step 1 (see Table 4). On Step 2, significant additional variance beyond that explained by pitch-matching ability was explained by achievement on the reading comprehension pretest, ΔR² = .019, total R²_adj = .650, F(2, 167) = 157.8, p < .001; no other variables entered the model. Because variables within the same factor exhibit a high degree of multicollinearity and are therefore unlikely to enter the model, the unique contribution of the other variables in explaining variance after accounting for pitch-matching ability was considered by examining partial correlations. The squared partial correlation of each variable with notation-singing indicates the proportion of variance in notation-singing achievement that is associated with that variable but is not associated with pitch matching. After partialing out pitch matching, significant variance in achievement on notation-singing test 1 was explained by the reading comprehension pretest (p = .003, partial r² = .05), OLSAT verbal subtest (p = .02, partial r² = .03), and reading comprehension posttest (p = .02, partial r² = .03).

Table 4.

Summary of Hierarchical-Stepwise Regression for Variables Explaining Variance in Second-Graders’ Sight-Singing Achievement (N = 170).

Step	Predictor	Beta		R ²	F	ΔR²
Step	Predictor	Step 1	Step 2	R ²	F	ΔR²
Criterion measure = notation-singing test 1
1	Starting pitch matching test 1	.797**	.759**	.635	291.9**
2	Reading comprehension pretest		.142*	.654	157.8**	.019
Criterion measure= notation-singing test 2
1	Starting pitch matching test 2	.787**	.749**	.619	272.8**
2	Reading comprehension posttest		.155*	.641	149.4**	.023

Note: Variables entered on Step 2: Intermediate Measures of Music Audiation tonal pre- and posttest, reading guide pre- and posttest, reading comprehension pre- and posttest, reading fluency, and Otis-Lennon School Ability Test verbal and nonverbal.

*p < .01. **p < .001.

The regression analysis using notation-singing test 2 as the criterion indicated that a significant portion of additional variance beyond that explained by pitch-matching ability was explained by achievement on the reading comprehension posttest, ΔR² = .023, total R²_adj = .637, F(2, 167) = 149.4, p < .001; no other variables entered the model. Partial correlations of the variables with notation-singing test 2 after partialing out pitch matching indicated that significant variance in achievement on notation-singing test 2 was explained only by the reading comprehension posttest (p = .001, partial r² = .06).

Discussion

Skill in reading music notation has been cited as an important but neglected area of music education in need of improvement (Asmus, 2004; Demorest, 2004; Gudmundsdottir, 2010; Mills & McPherson, 2006). The goal of this study was to identify general constructs and specific skills that are related to and may help explain differences in achievement when learning to read and sing pitch notation. Young students at the beginning stage of sight-singing skill acquisition were assessed with a notation-singing task limited to pitch only. Student performance data were also collected from other assessments that are typically administered in an educational setting and have been found to be related to music achievement. A common factor analysis objectively summarized the data, revealing three underlying factors that parsimoniously described the observed variance among the variables. Regression analyses then identified and determined the relative importance of variables that were significant in explaining variance in the beginning sight-singing achievement.

A Music Achievement factor revealed with the factor analysis showed that the combined measures of music achievement had strong internal consistency, indicating that they were measuring the same general construct. Likewise, all the reading assessments loaded on a Reading Achievement factor, providing an indication of the construct validity across the reading assessments. Reading Achievement correlated positively with Music Achievement (r = .31) as hypothesized. The moderate relationship of the two factors for second-graders in this study is in line with McMullen and Saffran’s (2004) theory that similar mechanisms are likely to be involved in the acquisition of musical and linguistic knowledge. The Music Achievement factor also had a strong positive relationship with the factor that seems to represent General Academic Ability (r = .42), with nearly 18% overlap in variance between the two factors. With 6- to 11-year-olds, Schellenberg (2006) reported that length of music study accounted for 12% of the variance in full-scale IQ (Wechsler Intelligence Scale for Children) and 12% of the variance in academic achievement (Kaufman Test of Educational Achievement). Similarly, with fifth- and sixth-graders, Hedden (1982) found that 25% and 40% of the variance in achievement on the Colwell’s Music Achievement Test was explained by the Iowa Test of Basic Skills. While participation in music has been associated with better performance in academic areas, the point has been made that students who excel academically are more likely to choose and persist in music participation (Costa-Giomi, 2012; Schellenberg, 2011). The present study ruled out self-selection as a potential confounding variable because it included all students, not just those choosing to pursue music study. Also, an important extension of these previous findings with the present study is that the Music Achievement factor was comprised of actual music performance tasks.

A surprising finding in the factor analysis was that the tonal discrimination measure, which is purportedly a measure of music aptitude, had stronger loadings on the General Academic Ability factor than the Music Achievement factor. Furthermore, the tonal discrimination measures’ communalities of .27 and .31 across all factors are indicative that the majority of variance in the students’ IMMA-T scores would be explained by one or more other constructs not disclosed by this analysis. From a practical standpoint, a communality below .50 indicates less than half of variable’s variance is taken into account, making it a candidate for deletion from the analysis. Rutkowski (2015) reported similar findings with first-graders; the relationship of tonal aptitude measured by the IMMA-T with singing accuracy was weak and nonsignificant (r = .19). Results from the present analysis corroborate Hornbach and Taggart’s (2005) suggestion that singing voice use and tonal aptitude are apparently separate constructs. Given that the IMMA is used in both research and practice in music education, its validity needs to be investigated further.

Regression analyses in this study showed that measures of reading comprehension explained significant variance in students’ sight-singing achievement after controlling for the students’ pitch matching ability. A previous study examining nearly the same variables used in this study as possible predictors of fifth-graders’ performance achievement after first- and second-year instrumental music instruction found similar results (Klinedinst, 1991, 1992). Among music aptitude (IMMA), school ability (OLSAT), and reading and math achievement (Stanford Achievement Test), only reading achievement was found to be significant in explaining adjudicators’ ratings of performance achievement, accounting for 18% and 28% of achievement variance after 1 and 2 years of instruction, respectively.

In the present study, the area of reading achievement was considered in greater detail by including several different achievement measures, among which only the reading comprehension variable explained significant variance in second-graders’ notation–singing skill acquisition. Similar results were found by Gromko (2004), who examined predictors of sight-reading performance with high school instrumental students; reading comprehension was the strongest predictor, explaining 24% of the variance in performance. Partial correlations in the present study, controlling for students’ pitch matching ability, indicated that reading comprehension explained 5% and 6% of the variance in sight-singing achievement on test 1 and test 2, respectively. It should be noted that the percentage of variance explained in this study may have been attenuated to some degree because of several factors, including the following: the 16-week instructional treatment period, which is relatively short, especially considering the young age of the students and the difficulty of the task; the inclusion of all students, not just those motivated to elect extra participation in music and who are more likely to practice outside of the instruction time; the limited range of possible scores on the reading comprehension test, which reduced statistical variability, thereby resulting in lower correlation coefficients; and the reliability of the reading comprehension assessment used in this study, which is adequate but may also be attenuated by the limited number of items on the test. Nevertheless, a robust association between reading comprehension and notation–singing achievement is suggested by the observation of similar results obtained with different administrations of the assessments at different points in time. These findings concur with those reported by Corrigall and Trainor (2011), in which length of music training was associated with reading comprehension even after controlling for other variables, including full-scale IQ. The present study extends this finding by reversing the perspective and using an actual music performance measure as the criterion.

Reading achievement as measured by the reading fluency variable in the present study was found to be nonsignificant in explaining variance in notation-singing achievement. This also concurs with findings from other studies; a meta-analysis of experimental studies examining the effect of music training on reading fluency concluded that music training did not have a significant impact on reading fluency (R. L. Gordon, Fehd, & McCandliss, 2015; see also Slater et al., 2014) or decoding skills (Corrigall & Trainor, 2011). The present study extends this finding by reversing the perspective and using an actual music performance measure as the criterion.

This study proffers evidence that aligns with Patel’s (2003, 2008, 2012) shared syntactic integration resource hypothesis, which proposes that the cognitive resource networks used when processing information in working memory are shared across the domains of language and music. While the present study does not use a simultaneous presentation paradigm in which both music and language syntax need to be processed at the same time, it does suggest that the two domains are related as it was found that ability in reading comprehension explains variance in sight-singing achievement. Greater demands on resource networks are believed to occur when a task requires rapid and selective activation of items in representation networks to process and connect with incoming elements. Similarly, music performance requires a foundational knowledge of tonal and metric relationships and the ability to maintain that tonal and metric reference in working memory while also mentally processing and executing incoming performance directions.

Future research is needed to further examine the cognitive processes underlying association of reading comprehension and music reading achievement. Corrigall and Trainor (2011) observed, “It appears that the association between length of music training and reading comprehension skills extends beyond auditory sensitivity and even symbol-to-sound decoding, which implies that higher-level processes might better explain this link” (p. 153). It would seem that greater ability to regulate cognitive resources, such as working memory, might be fundamental to greater achievement on syntactically challenging tasks. Other researchers examining the connection between music and language have suggested that the underlying construct affecting performance in both domains may be related to cognitive control (Slevc & Okada, 2015), allocation of attentional resources (Perruchet & Poulin-Charronnat, 2013), or specific components of executive functions (Slevc, Davey, Buschkuehl, & Jaeggi, 2016).

While the focus of the present study was on skill development in sight-singing pitches only, both tonal and rhythmic aspects of music training have been shown to be related to language reading development in childhood (for a review, see Strait & Kraus, 2011). Studies have found that when reading music notation, the decoding of pitch information and timing information are separate processes, even when the two must be integrated in motor output (for a review, see Gudmundsdottir, 2010). Future research might examine components related to the development of rhythm reading and performance skills or the development of sight-singing skills with materials that incorporate both rhythm and pitch but with older participants.

Many questions remain about the link between music and language, though a considerable amount of research is currently being conducted to investigate this relationship. Extending previous findings, the present study further delineates the nature of the relationship between music and language, finding reading comprehension to be a significant predictor of young children’s achievement in reading and singing pitch notation. This finding is another indication that music learning may have an effect on language learning not only because of an association in perceptual skills across domains but also because of an association of cognitive skills. The overlapping of cognitive processes in both language and music has important implications for educational and therapeutic purposes. If certain cognitive processes affect learning in both domains, developing skill with those processes in one domain may bolster the ability to use those processes in the related domain. Because students learn in a variety of ways, greater opportunities to learn occur when multiple ways of learning are provided. Music-related activities also have additional potential to benefit the educational process because most students find music highly enjoyable and engaging. Further research is needed to investigate the mechanism underlying the relationship between achievement in music reading and linguistic reading comprehension. The identification of possible mediating mechanisms would provide additional guidance for refining instruction or designing innovative instructional approaches that might improve educational outcomes.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

Author Biography

James L. Reifinger, Jr. is an assistant professor and a coordinator of music education at Southern Illinois University, Carbondale. His research interests include music cognition, music literacy, and assessment.

References

Anvari

S. H.

Trainor

L. J.

Woodside

Levy

B. A.

(2002). Relations among musical skills, phonological processing, and early reading ability in preschool children. Journal of Experimental Child Psychology, 83, 111–130. doi:10.1016/S0022-0965(02)00124-8

Asaridou

S. S.

McQueen

J. M.

(2013). Speech and music shape the listening brain: Evidence for shared domain-general mechanisms. Frontiers in Psychology, 4, 321. doi:10.3389/fpsyg.2013.00321

Asmus

E. P.

Jr. (2004). Music teaching and music literacy. Journal of Music Teacher Education, 13, 6–8. doi:10.1177/10570837040130020102

Besson

Chobert

Marie

(2011). Language and music in the musician brain. Language and Linguistics Compass, 5, 617–634. doi:10.1111/j.1749-818X.2011.00302.x

Butzlaff

(2000). Can music be used to teach reading? Journal of Aesthetic Education, 34, 167–178.

Cooper

N. A.

(1992). Selected factors related to children’s singing accuracy (Doctoral dissertation). Available from Dissertations & Theses: Full Text. (Publication No. 9307489)

Corrigall

K. A.

Trainor

L. J.

(2011). Associations between length of music training and reading skills in children. Music Perception, 29, 147–155. doi:10.1525/mp.2011.29.2.147

Costa-Giomi

(2012). Music instruction and children’s intellectual development: The educational context of music participation. In MacDonald

R. A. R.

Kreutz

Mitchell

(Eds.), Music, health, and wellbeing (pp. 339–355). New York, NY: Oxford University Press.

Degé

Schwarzer

(2011). The effect of a music program on phonological awareness in preschoolers. Frontiers in Psychology, 2, 124. doi:10.3389/fpsyg.2011.00124

10.

Demorest

S. M.

(2004). Choral sight-singing practices: Revisiting a web-based survey. International Journal of Research in Choral Singing, 2, 3–10.

11.

Eisinga

Grotenhuis

Pelzer

(2013). The reliability of a two-item scale: Pearson, Cronbach or Spearman-Brown? International Journal of Public Health, 58(4), 637–642. doi:10.1007/s00038-012-0416-3

12.

Fedorenko

Patel

Casasanto

Winawer

Gibson

(2009). Structural integration in language and music: Evidence for a shared system. Memory & Cognition, 37, 1–9. doi:10.3758/MC.37.1.1

13.

Fine

Berry

Rosner

(2006). The effect of pattern recognition and tonal predictability on sight-singing ability. Psychology of Music, 34, 431–447. doi:10.1177/0305735606067152

14.

Fiveash

Pammer

(2014). Music and language: Do they draw on similar syntactic working memory resources? Psychology of Music, 42, 190–209. doi:10.1177/0305735612463949

15.

Forgeard

Schlaug

Norton

Rosam

Iyengar

Winner

(2008). The relation between music and phonological processing in normal-reading children and children with dyslexia. Music Perception, 25, 383–390. doi:10.1525/mp.2008.25.4.383

16.

Goetze

Cooper

Brown

D. J.

(1990). Recent research on singing in the general music classroom. Bulletin of the Council for Research in Music Education, 104, 16–37.

17.

Good

R. H.

Kaminski

R. A.

(Eds.). (2002). Dynamic indicators of basic early literacy skill (6th ed.). Eugene, OR: Institute for the Development of Educational Achievement.

18.

Gordon

E. E.

(1982). Intermediate measures of music audiation. Chicago, IL: GIA Publications, Inc.

19.

Gordon

E. E.

(1986). Manual for the primary measure of music audiation and the intermediate measures of music audiation music aptitude tests for kindergarten and first, second, third, and fourth grade children. Chicago, IL: GIA Publications, Inc.

20.

Gordon

R. L.

Fehd

H. M.

McCandliss

B. D.

(2015). Does music training enhance literacy skills? A meta-analysis. Frontiers in Psychology, 6, 1777. doi:10.3389/fpsyg.2015.01777

21.

Gromko

J. E.

(2004). Predictors of music sight-reading ability in high school wind players. Journal of Research in Music Education, 52, 6–15. doi:10.2307/3345521

22.

Gromko

J. E.

(2005). The effect of music instruction on phonemic awareness in beginning readers. Journal of Research in Music Education, 53, 199–209. doi:10.1177/002242940505300302

23.

Gudmundsdottir

H. R.

(2010). Advances in music-reading research. Music Education Research, 12(4), 331–338. doi:10.1080/14613808.2010.504809

24.

Hair

J. F.

Jr. Black

W. C.

Babin

B. J.

Anderson

R. E.

(2010). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Pearson Education, Inc.

25.

Hedden

S. K.

(1982). Prediction of music achievement in the elementary school. Journal of Research in Music Education, 30, 61–68. doi:10.2307/3344867

26.

Henry

(2001). The development of a vocal sight-reading inventory. Bulletin of the Council for Research in Music Education, 150, 21–35.

27.

Henry

Demorest

S. M.

(1994). Individual sight-singing achievement in successful choral ensembles: A preliminary study. Update: Applications of Research in Music Education, 13(1), 4–8.

28.

Hornbach

C. M.

Taggart

C. C.

(2005). The relationship between developmental tonal aptitude and singing achievement among kindergarten, first-, second-, and third-grade students. Journal of Research in Music Education, 53, 322–331. doi:10.1177/002242940505300404

29.

Jackendoff

(2009). Parallels and nonparallels between language and music. Music Perception, 26, 195–204. doi:10.1525/mp.2009.26.3.195

30.

Kaviani

Mirbaha

Pournaseh

Sagan

(2014). Can music lessons increase the performance of preschool children in IQ tests? Cognitive Processing, 15, 7–84. doi:10.1007/s10339-013-0574-0

31.

Killian

J. N.

Henry

M. L.

(2005). A comparison of successful and unsuccessful strategies in individual sight-singing preparation and performance. Journal of Research in Music Education, 53, 51–65.

32.

Klinedinst

R. E.

(1991). Predicting performance achievement and retention of fifth-grade instrumental students. Journal of Research in Music Education, 39, 225–238. doi:10.2307/3344722

33.

Klinedinst

R. E.

(1992). Predicting performance achievement of beginning band students—second-year results. Contributions to Music Education, 19, 46–59.

34.

Lamb

S. J.

Gregory

A. H.

(1993). The relationship between music and reading in beginning readers. Educational Psychology: An International Journal of Experimental Educational Psychology, 13, 19–27. doi:10.1080/0144341930130103

35.

Lucas

Gromko

(2007). The relationship of music pattern discrimination skill and phonemic awareness in beginning readers. Contributions to Music Education, 34, 9–17. doi:10.1080/0144341930130103

36.

McMullen

Saffran

J. R.

(2004). Music and language: A developmental comparison. Music Perception, 21, 289–311. doi:10.1525/mp.2004.21.3.289

37.

Mills

McPherson

G. E.

(2006). Musical literacy. In McPherson

G. E.

(Ed.), The child as musician: A handbook of musical development (pp. 155–172). New York, NY: Oxford University Press.

38.

Moreno

Friesen

Bialystok

(2011). Effect of music training on promoting preliteracy skills: Preliminary causal evidence. Music Perception, 29, 165–172. doi:10.1525/mp.2011.29.2.165

39.

Moreno

Marques

Santos

Castro

S. L.

Besson

(2009). Musical training influences linguistic abilities in 8-year-old children: More evidence for brain plasticity. Cerebral Cortex, 19, 712–723. doi:10.1093/cercor/bhn120

40.

Nunnally

J. C.

Bernstein

I. H.

(1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill, Inc.

41.

Orman

E. K.

(2002). Comparison of the National Standards for music education and elementary music specialists’ use of class time. Journal of Research in Music Education, 50, 155–164. doi:10.2307/3345819

42.

Otis

A. S.

Lennon

R. T.

(1997). Otis-Lennon school ability test (7th ed.). San Antonio, TX: Harcourt Brace Education Measurement.

43.

Patel

A. D.

(2003). Language, music, syntax and the brain. Nature Neuroscience, 6, 674–681. doi:10.1038/nn1082

44.

Patel

A. D.

(2008). Music, language, and the brain. New York, NY: Oxford University Press.

45.

Patel

A. D.

(2012). Language, music, and the brain: A resource-sharing framework. In Rebuschat

Rohrmeier

Hawkins

Cross

(Eds.), Language and music as cognitive systems (pp. 204–223). New York, NY: Oxford University Press.

46.

Peretz

Vuvan

Lagrois

M-É.

Armony

J. L.

(2015). Neural overlap in processing music and speech. Philosophical Transactions of the Royal Society B, 370, 20140090. doi:10.1098/rstb.2014.0090

47.

Peretz

Zatorre

R. J.

(2005). Brain organization for music processing. Annual Review of Psychology, 56, 4–26. doi:10.1146/annurev.psych.56.091103.070225

48.

Perruchet

Poulin-Charronnat

(2013). Challenging prior evidence for a shared syntactic processor for language and music. Psychonomic Bulletin and Review, 20, 310–317. doi:10.3758/s13423-012-0344-5

49.

Psychological Corporation. (2002). Otis-Lennon school ability test (7th ed.) multilevel fall norms form 3 & 4. San Antonio, TX: Harcourt Brace Education Measurement.

50.

Rauscher

R. H.

(2009). The impact of music instruction on other skills. In Hallam

Cross

Thaut (Eds.), The Oxford handbook of music psychology (pp. 244–252). New York, NY: Oxford University Press.

51.

Rutkowski

(2015). The relationship between children’s use of singing voice and singing accuracy. Music Perception: An Interdisciplinary Journal, 32, 283–292. doi:10.1525/mp.2015.32.3.283

52.

Schellenberg

E. G.

(2004). Music lessons enhance IQ. Psychological Science, 15, 511–514. doi:10.1111/j.0956-7976.2004.00711.x

53.

Schellenberg

E. G.

(2006). Long-term positive associations between music lessons and IQ. Journal of Educational Psychology, 98, 457–468. doi:10.1037/0022-0663.98.2.457

54.

Schellenberg

E. G.

(2011). Examining the association between music lessons and intelligence. British Journal of Psychology, 102, 283–302. doi:10.1111/j.2044-8295.2010.02000.x

55.

Shelton

N. R.

Altwerger

Jordan

(2009). Does DIBELS put reading first? Literacy Research and Instruction, 48, 137–148. doi:10.1080/19388070802226311

56.

Slater

Strait

D. L.

Skoe

O’Connell

Thompson

Kraus

(2014). Longitudinal effects of group music instruction on literacy skills in low-income children. PLoS One, 9(11), e113383.

57.

Slevc

L. R.

Davey

N. S.

Buschkuehl

Jaeggi

S. M.

(2016). Tuning the mind: Exploring the connections between musical ability and executive functions. Cognition, 152, 199–211. doi:10.1016/j.cognition.2016.03.017

58.

Slevc

L. R.

Okada

B. M.

(2015). Processing structure in language and music: A case for shared reliance on cognitive control. Psychonomic Bulletin and Review, 22, 637–652. doi:10.3758/s13423-014-0712-4

59.

Slevc

L. R.

Rosenberg

J. C.

Patel

A. D.

(2009). Making psycholinguistics musical: Self-paced reading time evidence for shared processing of linguistic and musical syntax. Psychonomic Bulletin & Review, 16, 374–381. doi:10.3758/16.2.374

60.

Strait

Kraus

(2011). Playing music for a smarter ear: Cognitive, perceptual and neurobiological evidence. Music Perception, 29, 133–146.

61.

Tabachnick

B. G.

Fidell

L. S.

(2012). Using multivariate statistics (6th ed.). New York, NY: Pearson Education, Inc.

62.

Tsang

C. D.

Conrad

N. J.

(2011). Music training and reading readiness. Music Perception, 29, 157–163. doi:10.1525/mp.2011.29.2.157