Learn a new language by singing

Hello Dear Reader

Today I am thinking about foreign languages. People who come from monolingual families and who are raised in the UK school system are notorious across the world for being poor at speaking foreign languages.

My Kenyan friend Kim once told me a joke that I have repeated to knowing laughter on many occasions since: ‘ People who speak more than two languages are called multi-lingual, people who speak two language are called bilingual….people who speak one language are called British’

This may be an unfair generalisation but it applies to me, no argument. I learned French for 3 years at school and German for 1 year. I lived in Switzerland for the best part of a year. My fiancé is Spanish.  And yet, I speak only English. For shame.

photo mamaThis week my mother-in-law-to-be has been visiting from Zaragoza, Spain. We have known each other for 7 years and yet we can hardly speak. ‘Frustration’ is hardly the word. We can share the odd joke and passing comment about the weather, a meal or my outfit. We even managed to co-ordinate a wedding dress trip between us this visit. But communication remains minimal.

You can imagine my excitement therefore when I came across a new article in Psychology of Music journal that tested the efficacy of singing for foreign-language learning. The article, by Arla Good and colleagues, tested the theory that song is a more effective way of teaching a new language than spoken passages.

Might there be hope for my terrible Spanish, if only I sing along?


Singing is used the world over as a means to facilitate learning. Surprisingly however, there is little empirical evidence that singing boosts learning of new words anymore than learning by listening to speech or speaking normally.

Earlier this year Hi Jee Kang and I published an article where we found that background music boosted learning of Mandarin Chinese as compared to exactly the same material that lacked a musical backing track. We proposed several mechanisms for this effect, including:

1) Music grabs more attention, which is then directed to the sound of the new words

2) The structure of the music facilitates processing of the structures in the new words

3) Song creates a more elaborate memory trace, made from multiple sources and stronger as a result.

Interestingly we found that the music did not help with pronunciation, only basic memory for the new words and phrases in Chinese.


The new study took place over a 2-week learning period. 38 Spanish-speaking children from a junior school (age 9-13) in Ecuador were taught a novel lyrical passage, one group (16 students) heard the lyrics as a song and one group (22 students) heard them as a poem spoken rhythmically.

None of the children had been given any formal English instruction, though there is no doubt they had been differentially exposed to English through the media.

Wooden_Table_-_SketchUpThe children were tested after the 2 week testing session and also, uniquely, 6 months later. Each child received 20 exposures to the lyrics before testing. The lyrics contained 29 words in total and the children were tested on their ability to translate  10 of the words (like ‘table’, ‘feet’, and ‘daddy’). The children were also tested on their ability to recall the whole lyrics and their pronunciation was scored.

After 6 months 13 of the children (7 from the spoken condition) were available for re-testing.


Pronunciation – overall consonants were pronounced better than vowels. There were no differences between the groups in their pronunciation of consonants but the children who had heard the lyrics on song form were better at pronouncing the vowels.

Recall -  Children in the sung condition recalled more words in sequence than the children in the poem condition. The children from the song condition also showed fewer syllable errors, suggesting that the song had facilitated the rhythmic structure in the lyrics.

Translation – Of the 10 new English words, the singing children were able to recall on average 4 whereas the children in the poem group recalled 2.6 – a significant difference.

The long-term recall test had low participant numbers but the data suggested that the children in the singing group had maintained their advantage in better recall of the lyrics. However, the group difference in translation success had disappeared.


These findings support the use of sung lyrics to teach children words in a new language, suggesting that they may be able to learn twice as many in the short-term.

In the long-term the children may lose their overt knowledge of how to translate new foreign words correctly but they seem to retain a new foreign song more than a poem, which may form the basis for quicker learning and better pronunciation down the line.

rhythmThe findings support  one of Hi Jee’s suggestions: music helps to facilitate the processing of language structure through the medium of rhythm. Sadly there was no evidence to support or rule out other reasons why music facilitated the childrens’ learning, such as grabbing attention (through enjoyment) and building a stronger memory trace. Testing these theories, and differentiating them, will have to be the remit of future studies.

Right – I am going to practice what they preach and listed to a few Spanish tracks to see if it improves my communication with my future in-laws. There is still time before the wedding!

Paper: Good, A.J., Russo, F.A., & Sullivan, J. (2014) The efficacy of singing in foreign language learning. Psychology of Music, DOI: 10.1177/0305735614528833


Are musicians better at multitasking?

Hello Dear Reader,

coldIt got chilly all of a sudden. We were enjoying a mild late summer in the UK until this week. Now people have broken into their supplies of heavy winter scarves and gloves.

I have had the pleasure of bringing out my Grandmother’s legacy of winter hats. Today I am debuting her white 1920s bonnet.

My Grandma’s hats remind me of her, a lovely feeling.  They also fit my head, which is a miracle. My head measures 60cm circumference. Why not measure your head Dear Reader? You will likely find yours is a neat and well-shaped cranium, not like the bulging mass that sits atop my neck. Large heads run in my family; my brother, sister, my father, and my grandma. Wearing her hat reminds me I am her granddaughter – big head and all.

musicianThe idea of transfer from one to another leads me into my reading for this week. Not genetics per se, but rather the transfer that may occur when people train as musicians. Do musicians gain skills/ abilities that become generalized beyond musical ability? This long-standing debate in the literature regarding so-called ‘far transfer effects’ is one I covered in my book (You Are The Music) and in my blog (musicians and their wider skills).

Under the theory that “there is no smoke without fire” it seems likely that musical skills result in some kind of transfer to other domains. But where and, more importantly, how might this occur?

A new paper by Linda Moradzadeh and colleagues (York University, Canada) has looked at task switching and dual-task performance in musicians.

OLYMPUS DIGITAL CAMERATheir background comes from research into bilingualism. Learning more than one language is associated with general advantages to linguistic processing and task switching skills.

Musical training by contrast is frequently associated with verbal advantages (in memory, verbal IQ, linguistic processing).

Few studies have put these two strands of evidence together to ask the question, is musical training associated with task switching abilities in normal adults, as it is in bilinguals? Furthermore, can you double up – do bilingual musicians have the biggest advantage in task switching?

Task switching and dual task performance

Task switching is the ability to swap operations between two different mental challenges. It is a skill that is thought to reflect our cognitive flexibility. There are two types of cost associated with task switching:

1) Local cost occurs when we temporarily slow down following a switch;

2) Global cost occurs when we slow down in general because we know that we might need to switch soon.

Local cost reflects the cognitive effort needed to switch our mindset at any one moment and Global cost is associated with the overall demand of keeping two tasks in mind.

Bilinguals often need to switch between their languages, resulting in switch costs. Studies have demonstrated that bilinguals perform better on general task switching tasks as early as preschool age.  However, many of these studies lack control over confounds such as socio-economic status (SES) and IQ.

multitaskingAnother skill in bilinguals is dealing with dual task demands. Dual task performance, doing two things at once (multitasking), relies on executive function skills. These overlap with those skills used in task switching (working memory) but they are unique in one important way – dual tasks require the simultaneous processing of two things.

In comes our new paper. Moradzadeh et al. (2014) predicted that musicians would outperform non-musicians on task switching (lower local and global costs) and dual task performance (more accurate under high demands). They made the same predictions for bilinguals vs. monolinguals. Finally, they looked at the possibility of an additive effect of musical training and bilingualism on task switching and dual task ability.

The new experiment

153 participants (age 18 – 31) were divided into four groups: 1) monolingual musicians, 2) bilingual musicians, 3) monolingual non-musicians, and 4) bilingual non-musicians. All the musicians had at least 12 years training and the bilinguals were fluent except for 8% who were conversational level.

Groups were matched on SES (socio economic status) and non-verbal IQ. Receptive vocabulary was higher in musicians compared to non-musicians, and higher in monolinguals compared to bilinguals. These differences were expected and dealt with statistically at the analysis phase.

Computer_keyboardTask switching was tested with the Quantity/Identity task. Participants saw various digits on a screen and had to indicate either the total number they could see (Quantity: 1 or 3) or their type (Identity: 1s or 3s).

For dual task the participants completed the Krantz paradigm, a visual paradigm that required people to track a white dot while also attending to letters. They also completed a dual n-back task where they had to track and report a target letter in a series.


  • Musicians performed better on task switching and dual task challenges compared to non-musicians.
  • Bilinguals showed no task advantages compared to monolinguals.
  • There were no additive effects of musicianship and bilingualism.


Musicians’ advantage in task switching related to their reduced Global costs. On average they were 150ms faster in processing trials when they may have needed to switch at any moment. The authors conclude that their result reflects musicians’ superior ability to maintain and manipulate competing information in their working memory.

A lack of musician effect in Local switch costs by contrast may reflect the fact that while musicians can better target their working memory ability, they are not superior at immediate inhibitory control.

The lack of any effects in bilinguals is not in line with the literature – the authors speculate that their non-finding may be a result of their task, which was verbal in nature. Previous positive findings may also have been driven by group differences in SES or non-verbal IQ, which were controlled in this study.

Given that the authors found no bilingualism effects it is no surprise that they failed to find any additive effects of musicianship and bilingualism. This does not mean that no effect exists – just that no evidence was found in this study.

Final points

Blue_question_markOnce again we see evidence for a musician advantage in complex cognition. It is tempting to draw conclusions but we must remember that my original questions remains unanswered by this kind of study – how might this effect occur? Why does musicianship render advantages in task switching and dual task challenges? Following results of this kind we can only speculate.

Future studies in the area of musician transfer effects need to address the limitations of cross sectional designs (comparing two groups). Ideally studies would investigate skills on a continuum (musicians of a different number of years in training, or bilinguals on a scale of fluency) to show linear impacts of training rather than presence or absence.

The ultimate design is random assignment to groups and pre/post testing. This is possible though it requires risk as the studies could take many months if not years while the training goes on. However, these kinds of studies are necessary to establish if musical training or bilingualism positively affects our mental abilities in a general way.

Right. I will leave you with that Dear Reader. Time to put away my tea cup, don my Grandma’s hat, and break out once again into the cold November morning.