Music lessons: effects/correlations on/between general and musical intelligence according to age and gender

Abstract – To investigate how music lessons, affects one’s improvement in general and musical Intelligence Quotient (IQ) by looking at IQ scores using tests. The review discusses the past research conducted identifying experimental flaws and limitations. The study intends to replicate previous research correcting their flaws and limitations. 54 participants (27 males/27 females) between the ages of 6 and 40 years old were recruited to complete 3 tests conducted at the University of Otago.

Participants will be matched into groups according to age periods, gender, general intelligence and musical intelligence. The study hypothesises that musical lessons (Standard Keyboard, Singing, Drama) will cause an increase of general and musical IQ in a significant way.

Across generations, the topic of music and it’s implications on one’s personality, moods, cognition, intelligence, and behaviour has been given much focus and attention. The concept of music is by no means universal; in other words, music is always the result of some kind of human intervention, intention or organization through production practices such as composition, arrangement, performance or presentation (Davies, J. B., 1978, p. 25-46). In the following review, previous research conducted on music and intelligence will be discussed and analyzed to provide a basis for current study.
First, the hypothesis that music lessons improves the Intelligence Quotient (IQ) of children has led a psychologist Schellenberg, E. G. (2004) to conduct a research on six-year-old children. His research concluded that music lessons do in fact increase IQ in a relatively small way (Schellenberg, E. G., 2004). The participants, randomly assigned into four lessons groups (Standard Keyboard, Kodaly voice, Drama, No lessons) were recruited by means of advertisement through the local community newspaper. Schellenberg’s study measured the IQ levels [Wechsler Intelligence Scale for Children-¬Third Edition (WISC-III)] of children before and after undergoing a year of musical (Standard Keyboard, Kodaly voice), non-musical (Drama) and no lessons (Schellenberg, E. G., 2004). Although the results obtained were significant, there were several flaws within the study. First, Schellenberg’s sample of participants was significantly limited to the method of recruitment in which focused only on people or families that read or bought newspapers. Geographically, families who lived in close proximity to the lessons centre (Royal Conservatory of Music, Toronto) would be most assumed to participate in the study. Further, participants whose families owned a keyboard were only chosen for the experiment, thus recruiting participants who were of a certain socioeconomic status (SES). According to (Sigelman, C. K. & Rider, E. A., 2003), families of a particular SES would differ from another family of a different SES in ways of parenting style. In turn, they denoted that parenting styles would influence how parents encourage growth and learning, middle and upper class families being the most conducive (Sigelman, C. K. & Rider, E. A., 2003, p. 408). Clearly, the sample of participants cannot be generalized to the whole world or even Canada’s population. Second, the manner of which the musical/non-musical lessons were taught was not specified or standardised. In fact, the same two instructors taught the three experimental groups (Standard Keyboard, Kodaly voice/singing, Drama) which could open possibilities of extraneous variables (EV) such as different teaching methods between the two instructors. EV’s such as facial expression, physical states, emotional states, and mental states could compromise or impair the process of learning. Third, the periods of testing (IQ, educational achievement, social functioning) were only limited to before and after a year of lessons. This leaves Schellenberg unaware of the children’s development during the lessons. Factors such as absenteeism, family activities, etc could influence the participants IQ. Fourth, the short duration of the entire study was not enough to see long term improvement/deterioration in IQ performance as children develop. This sets a basis for the current study that desires to research about music lessons and general intelligence.

Second, similar to the study by (Schellenberg, E. G., 2004), a study by (Thompson, W. F., Schellenberg, E. G., & Hussain, G., 2004) conducted a study on the effects of music lessons on the ability to decode speech prosody or emotions. This ability is explained as an enhanced sensitivity to emotions (Thompson et. al., 2004). They conducted three separate experiments within the study, one of which is the current focus, using the exact hypothesis statement of Schellenberg’s study however, instead of testing on intelligence they tested on the ability to decode speech prosody. The results of their study showed that children with musical training performed better at prosody decoding as compared to children without musical training. Although the results obtained were significant, there were several flaws within the study. First, similar this study is to Schellenberg’s, the procedure of testing the children encompassed flaws in the process of testing before and after the music lessons during summer breaks. Why is it flawed? This is because the period of testing was a holiday in which participants could have/not practiced harder/not at all due to free time or scheduled activities such as outings, camps or even extra academic classes. Second, in addition to the first, participants were encouraged to practice at home without a specified or encouraged time or period of hours for practice. Without proper guidelines, the participants’ motivation to practice depended on parental or personal bases (extraneous variable). This further strengthens the basis for the current study that desires to research about music lessons and general intelligence.
Third, the research by (Ho, Y. -C., Cheung, M, -C., & Chan, A. S., 2003) studied the effects of music training on verbal memory in children. Their study focused on verbal memory instead of intelligence. Similar to Schellenberg’s study, their experiment consisted of participants between the ages of 6 to 15 that were assigned to two experimental groups [Musical training (MT), No musical training (NMT)]. The participants were matched to the experimental groups according to IQ, health, age, educational level, SES, parental educational level, and family income. The results of their study showed that children with musical training performed better in verbal memory as compared to those who those with no musical training. Although results obtained were significant, there were several flaws within the study. First, by recruiting participants aged 6 to 15 only, the study limits the results to represent only the people whose age is similar. Subsequently, according to (Sigelman, C. K. & Rider, E. A., 2003, p. 4) the participants’ age periods were of the middle childhood (6-12 years) and adolescence (12-15) periods, thus suggesting a difference of cognitive abilities [e.g. Piaget’s Preoperational and Concrete stages (middle childhood); and Piaget’s Formal operations stage (adolescence)] between the two age periods. The flaw within this design was the usage of the same tests for the different age period groups. For example, the verbal memory test (which tested on memory recollection of words) would definitely be biased towards the older children because the words used within the test would be more easily understood by them. This sets a basis for the current study that desires to research about gender and age period differences when measuring IQ performance with/without musical training.

In conclusion, all the research as mentioned above had similar results, however all three were testing on different measures in which allowed multiple experimental flaws to surface. The question the present study intends to investigate is the effects of musical training on general intelligence (IQ), to determine how general intelligence correlates with musical intelligence, to investigate the gender and age period differences when measuring IQ performance with/without musical training, and to answer how past research is relevant to New Zealand’s sample of people; eliminating any flaws or limitations within previous research.

To begin, (Rauscher, F.H., Shaw, G.L. and Ky, K.N., 1995, as cited in Jausovec, N. & Habe, K., 1996) developed research on the Mozart effect. They argued that if a similar music piece to that of Mozart’s in tempo, melody, organization, and predictability would be presented to a group of participants it would increase IQ. From their research, the current study is interested in whether music lessons (instead of listening to music) would increase IQ. Paul R. Farnsworth in his book, The Social Psychology of Music, wrote about a relationship between general intelligence and musical abilities. He denotes that there is a correlation between one’s general intelligence and musical abilities (Farnsworth, P. R., 1969, p. 156-157). However it is unknown whether an increase in general intelligence causes or facilitates the increase in musical abilities. Subsequently, a psychologist, Howard Gardner, developed a theory on intelligence called the Theory of Multiple Intelligence (Sigelman, C. K. & Rider, E. A., 2003, p. 220), denoting eight distinct intellectual abilities one of which being musical intelligence. Musical intelligence can be defined as “what you would find in a composer or performer, someone who can think musically” (Gardner, H., 2001), and one’s “acute sensitivity to sound patterns” (Sigelman, C. K. & Rider, E. A., 2003, p. 220). Interestingly, the definition of musical intelligence is very similar to the definition of musical abilities according to (Farnsworth, P. R., 1969, p. 151) which is defined it as the capability for musical performance and appreciation. This sets a basis for the current study that desires to research about how general intelligence correlates with musical intelligence.

As argued in the review about the flaws and limitations of previous research, the present study intends to eliminate and correct any flaws and limitations identified replicating previous research in a local setting. The hypotheses of my present study is to answer whether participants under musical training would perform better than participants under non-musical training or no training, general intelligence will correlate positively/negatively with musical intelligence, and whether there will be gender and age period differences when measuring IQ performance with/without musical training.

54 participants between the ages of 6 and 40 years, consisting of equal number of males and females, will be recruited to participate in an experiment conducted across three years in the University of Otago. The participants will be randomly matched into four experimental groups (Standard Keyboard, Singing, Drama, No Training) according to gender, age periods [middle childhood (6-12 years old), adolescence (12-20 years old), early adulthood (20-40 years old)], general intelligence and musical intelligence in which, each age period group has 18 participants. Two groups, standard keyboard and singing are the musical training groups however; drama (Non-musical) and no training are the control groups. Middle childhood and adolescence participants will be recruited from different schools, colleges and universities randomly selected from cities within New Zealand. However early adulthood participants will be recruited via newspaper advertisements. Only those participants, who/whose families (for children) voluntarily agree to participate in the experiment will participate.

In the research conducted, three tests will be used in the experiment, first of which is the Wechsler Intelligence Scale for Children—Third Edition (WISC–III) (Wechsler, 1991, as cited in Thompson et. al., 2004) used to test for the general intelligence score for participants aged 6 to 16. Second, the Wechsler Adult Intelligence Scale-Revised (WAIS-R) (Wechsler, 1981, as cited in (Sigelman, C. K. & Rider, E. A., 2003, p. 223) used to test for the general intelligence score for participants aged 17 to 40, and third, the musical intelligence test (adapted from Douglas, N., 2004) used to test all three age periods. Kindly refer to appendix A for a sample of the Musical intelligence test.

The independent variables (IV) are manipulating music training, gender, and age periods. There are four levels of music training IV’s such as standard keyboard (SK), singing (S), drama (D), and no training (NT). Next, there are three levels of age IV’s such as middle childhood (MC), adolescence (A), early adulthood (EA); subsequently, males (M) and females (F) for the gender IV. The dependent variables (DV) measured is general intelligence and musical intelligence. Overall, the experiment is a between subjects multi-factorial design.
Participants will undergo musical and non-musical training weekly at Dunedin, New Zealand, University of Otago music department rooms. The lessons, taught to groups of 3 children, are approximately 45 minutes in length. Three professional female instructors (each for keyboard, singing, and drama) are the University’s lecturers or graduates who will teach 6 classes each. After every year of training, the instructors will swap classes for example, from keyboard to singing and drama the following year. The keyboard lessons designed by the lecturers at the University of Otago will consist of basic classical training. Participants in the singing group will receive training in the Kodály method (Schellenberg, E. G., 2004) an “intensive musical program that involves improvisation as well as singing, playing, and dancing to simple tunes and folk melodies” (Thompson et. al., 2004). However, participants in the drama group studied “simple scripts for plays, intonation, memorization of lines, staging, and acting” (Thompson et. al., 2004). All participants are expected to practice for 30 minutes a day. All will be were tested three times a year (January, May, September) on the 30th of each month on participants individually by a research confederate who is unaware of the experimental conditions. For children, the musical IQ test will be read out as questions rather than a written questionnaire.

Participants’ scores for the tests will be totalled and transformed for into means and standard deviations. For example, if the totalled general IQ score for early adulthood males after trained in keyboard is (369), the average score for the participants would be calculated by taking the totalled score (369) and dividing it by the number of early adulthood males after trained in keyboard (3) which would result in a mean score of 123. A larger mean score would indicate a greater level of general IQ as compared to the general average of 100. Likewise a lower mean score would indicate a lower level of general IQ. The mean score for the musical intelligence test will be calculated using the same method. Using the SPSS statistical analysis program, the results of the study will be analyzed. Kindly refer to appendix B for examples of mean tables.
As shown in table one and two, early adulthood males had an increase of general IQ after trained in keyboards from a mean score of 110 to 123 with a mean difference of 13. Both general and musical IQ will be calculated as shown above however, musical intelligence is based on a scale of 0 to 10 (10 signifying very high musical IQ and 0 signifying very low musical IQ). After three years of testing, the results will be calculated and presented in a graph form stating the increase/decrease in IQ levels for the three years.

Davidson, J. (2005). Multiple intelligence. Retrieved March 17, 2005 from the World Wide Web:
Davies, J. B. (1978). The psychology of music. United States: Stanford University Press.
Farnsworth, P. R. (1969). The social psychology of music (2nd ed.) Iowa: Iowa State University Press.
Gardner, H. (2001). Interview excerpt: Howard Gardner on Multiple Intelligences. Retrieved March 17, 2005 from the World Wide Web:
Ho, Y. -C., Cheung, M, -C., & Chan, A. S. (2003). Music Training Improves Verbal but Not Visual Memory. Neuropsychology, 17(3), 439. Retrieved March 17, 2005 from PsycARTICLES database from World Wide Web:
Rauscher, F.H., Shaw, G.L. and Ky, K.N. (1995). Listening to Mozart enhances spatial temporal reasoning: Towards a neurophysiological basis. Neurosci. Lett., 195, 44-47. In Jausovec, N. & Habe, K. (1996). The “Mozart Effect”: An Electroencephalographic Analysis Employing the Methods of Induced Event-Related Desynchronization/ Synchronization and Event-Related Coherence. Brain Topography, 16, 73. Retrieved March 17, 2005 from Proquest database from World Wide Web:
Schellenberg, E. G. (2004). Music lessons enhance IQ. Psychological Science, 15(8), 511-514.
Sigelman, C. K. & Rider, E. A. (2003). Life-span human development (4th ed.). United States of America: Wadsworth/Thomson Learning.
Thompson, W. F., Schellenberg, E. G., & Hussain, G. (2004). Decoding Speech Prosody: Do Music Lessons Help?, 4(1), 46-64. Retrieved March 29, 2005 from PsycARTICLES database from World Wide Web: