VOLUME 10July 2019

A Science of Learning SymposiumFrom Laboratories to Learning

by Azilawati Jamaludin, Brian Butterworth, Daniel Ansari, David Hung, Michael Chee, Paul Dux & Suzy Styles

RESEARCH EMPOWERS

OER KNOWLEDGE BITESOFFICE OF EDUCATION RESEARCHNATIONAL INSTITUTE OF EDUCATIONSINGAPORE

OER Know l edge B i t e s

Launched in May 2016 by the Office of Education Research at the National Institute of Education, Singapore, OER Knowledge Bites aims to share education research discussions and issues as seen in the Singapore context. It also serves as a platform for researchers to share thoughts and concepts of education research with policymakers, educators and the public.

For enquiries pertaining to this publication, e-mail Nur Haryanti Sazali: haryanti.sazali@nie.edu.sg

About this VolumeThe Office of Education Research (OER), with support from the National Research Foundation, organised the inaugural Science of Learning (SoL) symposium on 2 November 2018. SoL is a field that has gained attention in recent years and proponents extoll its potential to impact educational practice and improve learning outcomes. This 10th volume of OER Knowledge Bites is based on individual talks given at this symposium, which is themed From Laboratories to Learning.

Acknowledgements

We would like to thank the following individuals from NIE, OER for contributing to this volume of OER Knowledge Bites:

» Tan Min Ying (Knowledge Mobilisation Unit) » Nur Haryanti Sazali (Research Engagement &

Publications Unit)

CONTENTS

04

05

06

08

10

12

14

Poor Numeracy and the Science of Learning: The Case of Dyscalculia by Brian Butterworth

On the Capacity Limits of Cognitive Control and its Enhancement: Neural Mechanisms & Transfer

by Paul E. Dux

Translational Specifications of Neural-Informed Game-Based Interventions for Mathematical Cognitive Development of Low-Progress Learners:

A Science of Learning Approach by Azilawati Jamaludin & David Hung

Building Blocks of Early Math: The Importance of Numerical Symbols by Daniel Ansari

Improving Adolescent Sleep: A Compelling Target for Improving Educational Outcomes

by Michael Chee

How Language Mixes Contribute to Effective Bilingualism and Effective Biliteracy in Singapore: Surveying the Language Landscape

by Suzy Styles

Highlights from the Panel Discussion

4 | OER Knowledge Bites

There are many causes of poor numeracy, including inadequate

or inappropriate teaching, few numerical activities in the home, going to school hungry, illness and missing crucial lessons. There are also cognitive factors, including problems with memory, concentration and reasoning. These factors affect all kinds of learning and not just numeracy.

There is also a cognitive difference that affects just learning arithmetic and has distinct characteristics from just being “bad at maths”. Standardised arithmetic tests or curriculum-based tests cannot identify the distinct causes of an individual’s poor numeracy because of the many factors that can affect test performance. To help an individual learner who is falling behind classmates to improve, the cause must be identified.

There is now evidence from developmental psychology, brain imaging and genetics that a learner can have a very specific disability that seriously impacts acquiring arithmetic competence. For example, recent studies suggest that small regions in the parietal lobes of the brain are abnormal in dyscalculics. There may also be abnormal connections from and to these regions. These regions are responsible for extracting numerical information from the environment—how many objects are experienced. My claim is that this mechanism is working inefficiently in about 4–5% of learners and this creates real difficulty in learning arithmetic the typical way.

Following the pioneering study of Ladislav Kosc 46 years ago, this is

P o o r N u m e r a c y a n d t h e S c i e n c e o f L e a r n i n g : T h e C a s e o f D y s c a l c u l i a

nowadays called “dyscalculia”. There is a parallel with dyslexia: the mechanism for analyzing syllables into phonemes is impaired, which makes learning an alphabetic script difficult. Dyslexics are sometimes referred to as “word blind”. In actuality, this isn’t strictly true—most can see the words perfectly well, but they have difficulty understanding or pronouncing them. Similarly, dyscalculics are sometimes termed “number blind”. Again, this is misleading. They can see the numbers perfectly well, but they do not understand them in the usual way. They also have trouble with spoken numbers and spoken calculations.

Dyscalculia, unlike other forms of poor numeracy, is congenital—you are born with it—and, as far as we currently know, it persists into adulthood and old age. It may be that with the right intervention, the condition can be ameliorated or even “cured”, but there have not been long-term prospective studies to see whether this is the case. We do know, however, that without the appropriate intervention, young dyscalculics become old dyscalculics.

If I am right, and many scientists and special needs teachers working in the science of learning agree with me, then dyscalculia is due to a “core deficit” in the ability to represent and manipulate the number of objects in a set. Sets and their number—often called their “numerosity” or “cardinality”—are foundational for learning arithmetic in school. If a learner can’t link a number name (e.g., five) or a number symbol (5) to a set with five members, then the foundations will be very shaky. And if a learner doesn’t understand what happens to the number when sets are combined, or subsets removed, then again the foundations of arithmetic will be shaky.

But knowing this is not sufficient to identify dyscalculics or to help them. First, we need reliable and valid tests to pick out those learners who are having trouble with sets and their number, and distinguish them from other learners

making slow progress. Such tests do exist. My Dyscalculia Screener is one of them. For interventions, we know that “more of the same” doesn’t work for dyscalculics. Five year olds who are equally bad on a standardized test (bottom 10%), but who are not dyscalculic, make progress with a standard intervention, but the dyscalculics’ progress is much slower, if it happens at all.

Interventions should be targeted at counting, combining and splitting sets and the results of these operations on number, and only when learners can do this reliably, then move on to written arithmetic. We have developed a digital game designed to move the learner, especially the young learner in need of learning support in mathematics, from sets to symbolic arithmetic. It’s called NumberBeads, and it is being trialed here in Singapore in a project funded by National Research Foundation, Singapore, and is led by Professor David Hung and Assistant Professor Azilawati Jamaludin from NIE’s Office of Education Research.

Useful ResourcesButterworth, B. (2019) Dyscalculia: From

science to education. Routledge. Butterworth, B., Varma, S., & Laurillard,

D. (2011). Dyscalculia: From brain to education. Science, 332, 1049–1053.

By Brian Butterworth

Brian Butterworth is Emeritus Professor of Cognitive Neuropsychology at the Institute of Cognitive Neuroscience, University College London (UK). His research ranges from speech errors and pauses, short-term memory deficits, dyslexia, reading both in alphabetic scripts and Chinese, and mathematics and dyscalculia.

Butterworth, B. (2019). Poor numeracy and the science of learning: The case of dyscalculia. OER Knowledge Bites Volume 10 (p. 4). Singapore: National Institute of Education.

How to Cite

5Vo lume 10 | 2019 |

How can an individual’s psychological processing be

enhanced? Can the benefits from such interventions transfer across tasks to facilitate learning? What are the cognitive and neural underpinnings of such benefits? And how can studying these help us to tailor learning protocols?

In recent years, there has been much excitement, both in the scientific literature and popular media, around claims that short periods of practice/training can enhance learning. For example, it has been suggested that cognitive training (e.g., prolonged exposure to a multitasking paradigm) can increase fluid intelligence, enhance working memory, improve the rate of numerical learning, and help treat neuropsychological and psychiatric impairments. Although it is clear that repeated exposure to certain tasks can enhance performance on those same tasks, evidence for transferable and generalised learning, and enduring cognitive benefits are yet to be established.

Similarly, “do-it-yourself” brain stimulation devices have recently been introduced, which have garnered significant media

O n t h e C a p a c i t y L i m i t s o f C o g n i t i v e C o n t r o l a n d i t s E n h a n c e m e n t : N e u r a l M e c h a n i s m s & Tr a n s f e r

attention in outlets such as The Guardian, The New York Times and The Wall Street Journal. These battery-operated systems deliver transcranial direct current stimulation (tDCS) to the brain via electrodes placed on the wearer’s scalp. When applied during training, tDCS can improve performance on motor, language and numeracy tasks. Again, however, the extent to which such effects generalise across a range of tasks and cognitive processes, or endure beyond the immediate training regime, is yet to be established.

The “Information Age” frequently requires individuals to perform multiple decisions concurrently. Under such multitasking conditions, one must use cognitive control processes in order to manage capacity limited attentional resources. Of import, attention plays a crucial role in learning and its impairment is associated with poor educational outcomes.

In this seminar. Dux presented a range of cognitive neuroscience methods to argue that such cognitive control limitations occur because frontoparietal and subcortical (FP-SC) brain regions both serve a broad range of mental functions and are limited information processors. His research showed that training/learning improves multitasking ability by segregating task representations in FP-SC regions.

Specifically, as individuals improved in multitasking with learning, the brain processed these tasks more distinctly which reduced interference. This outcome

provides a potential mechanism for how the brain solves the problem of dealing with competing sources of information during learning.

In addition, Dux presented evidence that brain stimulation (tDCS), when paired with training, can lead to generalised benefits for attention, decision-making and multitasking by increasing the rate with which evidence about the stimulus is accumulated. This work illustrates that learning, when paired with brain stimulation, can transfer across psychological processes.

Finally, Dux presented work examining neural predictors of the response to brain stimuli. To wit, in a large-scale individual differences study that employed Magnetic Resonance Spectroscopy, Dux and his colleagues demonstrated that concentrations of neurotransmitters in Prefrontal Cortex—the balance of Gaba and Glutimate concentration—predict an individual’s response to tDCS. Collectively, the results shed light on the neuro-cognitive mechanisms of cognitive control limitations and those that underlie the enhancement of associated operations via training and stimulation.

By Paul E. Dux

Paul Dux is Professor in the School of Psychology at University of Queensland (Australia). He is a psychologist and neuroscientist, and his area of research interests lies in the mechanisms of attention and the efficacy of cognitive training and brain stimulation.

Dux, P. E. (2019). On the capacity limits of cognitive control and its enhancement: Neural mechanisms & Transfer. OER Knowledge Bites Volume 10 (p. 5). Singapore: National Institute of Education.

How to Cite

6 | OER Knowledge Bites

Tr ans l a t i ona l Spec i f i c a t i ons o f Neu r a l - I n f o rmed Game-Based I n t e r ven t i ons f o r Ma thema t i c a l Cogn i t i v e Deve l opmen t o f Low-P rog ress Lea r ne r s : A Sc i ence o f Lea r n i ng App roachBy Azilawati Jamaludin1 & David Hung

1Azilawati Jamaludin is Assistant Professor at the Office of Education Research, National Institute of Education (Singapore). Her research interests are in applying the science of learning to education, with current projects on theories of learning, neuroscientific games and education innovation impact.

The challenge of levelling up low progress (LP) learners in

Mathematics, particularly those who continue to struggle despite multi-pronged behavioral intervention approaches, remains a prevalent phenomenon in schools. A 20-year analysis of the PSLE Mathematics trend shows at least 13% of every annual population cohort who do not attain a minimum of fifty percentile points during the annual PSLE assessment (DataGovSg, 2019). The study of children who consistently attain low achievement scores in mathematics is challenging because there could be many underlying reasons for LP Math, both cognitive and affective, which include language difficulties, sensory processing deficits, inappropriate instructions, poor attitude towards mathematics, high anxiety or lack of home support. These factors may be exacerbated if the child has other specific learning differences such as attention difficulties, sensorial disorders or other underlying deficits that affect his/her cognitive and affective development (Steedly, Dragoo, Arefeh, & Luke, 2008).

In recognising that LP learners in Math may constitute a heterogeneous group with diverging profiles and potentially dissociable characteristics (Bartelet, Ansari, Vaessen, & Blomert, 2014; Wong, Ho & Tang, 2014), this study adopts a Science of Learning (SoL)1 approach to

harness novel research methods of neural, psychobehavioral, and social measures to facilitate early identification and remediation for young children struggling with math. Specifically, in enacting an ecological framework that recognises the interrelations between factors at different analytical levels as equally important to the influence of factors within a single level (Hung, Lee, Toh, Jamaludin, & Wu, 2019), the study aims to provide a more precise identification and remediation of low progress trajectories in Math (see Figure 1 above).

Our SoL ecological framing is aimed towards recognising the interrelations between i) micro individual level, through investigating cognitive characteristics, affective dispositions and imaging neural activations and functional connectivity; ii) meso–social level, through analyzing home, school and social factors; and

iii) macro level, through foregrounding evidence for research to practice translation.

The objectives of this study are:

1. Characterise and validate deficits and mechanisms of low progress Math learner population

2. Investigate growth and performance change (neural and behavioral) over time

3. Develop an account of the role interventions play for learners’ remediation

As we work towards more systematic approach(es) to understanding the problems of LP students, this research aims to establish a replicable method for identifying and tracking LP learners across crucial schooling transitions that is timely and necessary, through ecologically valid methods. In combining the use of behavioural, social and fNIRS neurological measures, this study can

Figure 1. A Science of Learning ecological framework.

1 The OECD defines SoL as a term that describes the type of scientific research possible when neuroscience, psychology, cognitive physiology and/or other relevant disciplines joins with educational research and practice to study how people learn.

7Vo lume 10 | 2019 |

lend new depth to the literature of early discrepancies in academic performance. It will have implications on not only the way in which precursors or struggles of math learning may be identified earlier but so too for early interventions that may be tailored and personalised to bridge achievement gaps and in predicting mathematics dysfunction before it compounds into lifelong struggles. Such interventions are optimally personalised, engaging, adaptive and predictive (PEAP), and the long-term objective is to bridge research into the cascading effects of PEAP interventions and remediation for LP learners.

What Do These Mean for Practice? Through an ecological SoL approach, the aim is to optimise learning and maximise the potential of every learner through drawing connections between biological structure and processes, contexts of learning, and impactful educational outcomes.

Characterising underlying neurocognitive mechanism of math struggles yield specifications for empirically-based intervention design principles. Yet over and above students’ learning outcomes, effective intervention designs also entail establishing collaborative partnerships with educators.

In recognising the crucial role teachers play in the implementation of effective classroom intervention strategies, the

development of both learners and, in parallel, extending teachers’ professional competencies are key for educational system impact. Translating our research findings into practice, an online MOOC platform for educators functions as both a dialogic and development space for extending pedagogical repertoires using SoL approaches.

As an example, teachers can gain access to video case studies of students working on mathematics tasks and analyse the nature of their difficulties and math struggles. Screencasts of brain images of different learners and their respective neural development explanations, as well as links to neuroscientific math games and lesson plans on how the game can be introduced in class will be available. Additionally, teachers using the platform can also share and compare their experiences through online discussion forums, which serve as an opportunity for them to reflect on their own Math practices and learn collaboratively from one another.

References Data Government Singapore (DataGovSg).

(2019). Percentage of PSLE students who scored A* to C in Mathematics. Retrieved January 13, 2019 from https://data.gov.sg/dataset/percentage-of-psle-students-who-scored-a-c-by-subject?resource_id=d08a23b5-772d-4b95-9d5d-c847cfdda685

Steedly, K., Dragoo, K., Arefeh, S., & Luke, S.D. (2008). Effective mathematics instruction. Evidence for Education, 3(1) 1–12. Retrieved Oct 20, 2018 from //education.msu.edu/projects/leaps/proceedings/Default.html

Bartelet D., Ansari D., Vaessen A., Blomert L. (2014). Cognitive subtypes of mathematics learning difficulties in primary education. Research in Developmental Disabilities, 35, 657–670. doi: 10.1016/j.ridd.2013.12.010

Hung, D., Lee, S. S., Toh, Y., Jamaludin, A., & Wu, L. K. (2019). Innovations in Educational Change: Cultivating Ecologies for Schools. Singapore: Springer

Wong, T. T-Y., Ho, C., & Tang, J. (2014). Identification of children with mathematics learning disabilities (MLDs) using latent class growth analysis. Research in Developmental Disabilities, 35, 2906–2920.

Jamaludin, A., & Hung, D. (2019). Translational specifications of neural-informed game-based interventions for mathematical cognitive development of low-progress learners: A science of learning approach. OER Knowledge Bites Volume 10 (pp. 6–7). Singapore: National Institute of Education.

How to Cite

8 | OER Knowledge Bites

Bu i l d i ng B l ocks o f Ea r l y Ma th : The Impo r t ance o f Nume r i ca l S ymbo l s

By Daniel Ansari

Daniel Ansari is Professor and Canada Research Chair in Developmental Cognitive Neuroscience in the Department of Psychology and the Brain & Mind Institute at the University of Western Ontario in London, Ontario, where he heads the Numerical Cognition Laboratory

There now exists a large body of research showing that early

math skills are critical predictors of later school and life success (Duncan et al., 2007; Ritchie & Bates, 2013). In view of these data, it is important to identify the building blocks that enable young children to become numerate. This raises the question: what are some of the key foundational skills that preschool children need to develop in order to be well-prepared for formal mathematics education?

One of the key challenges faced by young children is to learn the meanings of numerical symbols (e.g., number

words and Arabic numerals). Numerical symbols are a human invention and children are not born with a knowledge of numerical symbols (Ansari, 2008). They have to learn these ways of representing numerical quantity (the total number of items in a set).

Indeed, when children first hear number words or see numerical symbols, they do not convey any meaning to them. The first foray into symbolic numerical thinking that children make is when they learn the meaning of counting. Children as young as 2 years of age may be able to recite the count sequence such as reciting the count sequence as they walk up a set of stairs. However, this does not mean that they understand the purpose of counting, which is to enumerate the number of items in a set. Put differently, a young child may be able to count to 10 but cannot give a teacher exactly five marbles from a bowl of marbles when being asked to.

This is because the child does not yet understand that the words in the count sequence refer to specific quantities.

Young children gradually learn the meaning of number words. Research has demonstrated that children first become one-knowers, then two-knowers, then three-knowers and eventually, they learn all the meaning of all the numbers in the count sequence (for a review see: Sarnecka, Goldman & Slusser, 2015). At that point, they are said to have learnt the so-called cardinality principle. The cardinality principle states that the last word in a count sequence refers to the total number of items in the counted set (e.g., a set of fruit).

Research has shown that there exist differences between young children’s understanding of the cardinality principle with some acquiring this understanding earlier than others. Many factors can

9Vo lume 10 | 2019 |

explain such differences, including the amount of numerical activities that the child has been exposed to and their language development (Merkley & Ansari, 2016). Importantly, longitudinal studies, in which children were followed over time, have shown that individual differences in children’s early understanding of cardinality predict their subsequent development of numerical and mathematical skills (Geary & vanMarle, 2018). Put differently, when children become cardinality principle knowers (i.e., when they have understood the meaning of counting), their growth in numerical skills accelerates.

In view of these data, preschool educators should pay attention to individual differences in young children’s understanding of the cardinality principle and provide opportunities for children who have not acquired this knowledge to help them develop an understanding of the meaning of counting. Such activities can include matching games that help children to see the relationship between sets of objects and number words (and later Arabic numerals).

Children need opportunities to gradually understand symbols as abstract representations of quantities (e.g., understanding that the number word “five” refers to any set of five items). Parents can also help children by talking to them about quantities (e.g., how many ducks do you see on this pond?) and relationship between quantities (is the line longer at this food stall or the one next to it?).

References Ansari, D. (2008). Effects of development and

enculturation on number representation in the brain. Nature Reviews Neuroscience, 9(4). Retrieved from: https://doi.org/10.1038/nrn2334

Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson, K., Huston, A. C., Klebanov, P., … Japel, C. (2007). School readiness and later achievement. Developmental Psychology, 43(6), 1428–1446. Retrieved from https://doi.org/2007-16709-012 [pii]10.1037/0012-1649.43.6.1428

Geary, D. C., & vanMarle, K. (2018). Growth of symbolic number knowledge accelerates after children understand cardinality. Cognition, 177, 69–78. Retrieved from https://doi.org/10.1016/j.cognition.2018.04.002

Ansari, D. (2019). Building blocks of early math: The importance of numerical symbols. OER Knowledge Bites Volume 10 (pp. 8–9). Singapore: National Institute of Education.

How to Cite

Merkley, R., & Ansari, D. (2016). Why numerical symbols count in the development of mathematical skills: Evidence from brain and behavior. Current Opinion in Behavioral Sciences, 10. Retrieved from: https://doi.org/10.1016/j.cobeha.2016.04.006

Ritchie, S. J., & Bates, T. C. (2013). Enduring links from childhood mathematics and reading achievement to adult socioeconomic status. Psychological Science, 24(7), 1301–1308. Retrieved from: https://doi.org/10.1177/0956797612466268

Sarnecka, B.W, Goldman, M.C, Slusser, E. B. (2015). How counting leads to children’s first representations of exact, large numbers. In A. Cohen Kadosh, R., Dowker (Eds.), Oxford Handbook of Numerical Cognition. Oxford University Press.

10 | OER Knowledge Bites

Imp ro v i ng Ado l escen t S l eep : A Compe l l i n g Ta rge t f o r Imp ro v i ng Educa t i ona l Ou tcomesBy Michael W. L. Chee

Michael Chee is Professor and Director of the Centre for Cognitive Neuroscience, Neuroscience and Behavioural Disorders Program at Duke-NUS Medical School (Singappre).

There is increasingly new evidence about the importance of sleep for

health and well-being. As such, a shift in attitudes and policies to improve sleep is long overdue. Data collected from five local and three international schools in Singapore involving over 2,300 students aged 13–19 years old, found that short sleep on school nights was associated with poorer self-rated health. Lack of sleep increases the odds of obesity, anxiety, feeling sad, irritable and worthlessness, and harboring self-harm/suicidal thoughts by 2–4 times compared to persons with appropriate sleep durations1.

Barriers to healthy sleep include later preferred sleep timing, lower parental supervision of bedtime, longer study time, early school start time, and longer travel time. Students at local schools were less likely to have a parent-set bedtime and spent more time doing homework or studying. Later bedtime in local schools attenuated the benefit of later school start time on nocturnal sleep duration (Yeo et al., 2019).

The US National Sleep Foundation recommends that teenagers obtain 8–10 hours of sleep a day. Our upper secondary and junior college students report averaging 6.5 hours of sleep on weekday nights. In our “Need for Sleep” series of studies on adolescent sleep, we found that students cannot stay vigilant with this amount of sleep if daytime naps are denied (Lo et al., 2019). Vigilance declines over successive nights of sleep restriction and does not recover

to baseline levels after two simulated weekend nights of 9 hours of recovery sleep opportunity (Lo, Ong, Leong, Gooley, & Chee, 2016).

The deficit in vigilance is compounded during an ensuing second cycle of sleep restriction and is accompanied by decrements in speed of processing and positive mood (Lo et al., 2017). Although sleepiness initially increases together with vigilance decline, change in its perceived magnitude plateaus and this dissociation distorts a person’s self-appraisal of the effects of cumulative sleep loss (Lo et al., 2016; 2017). Alongside, long term memory encoding falters (Cousins, Wong, Raghunath, Look. & Chee, 2019), an effect contributed by reduced hippocampal function with sustained wakefulness (Cousins, Van Rijn, Ong, & Chee, 2019). It is therefore highly likely that students fall asleep in the daytime during class to make up for inadequate sleep, consistent with self-report data showing that over a third of local teenagers feel tired “most of the time” (Yeo et al., 2019).

We found that a 60- (Lo et al., 2017) to 90-minute (Lo et al., 2019) mid-afternoon nap significantly boosts vigilance, memory

encoding and positive mood. The benefits of such naps extend into the mid-evening. Vigilance, on the other hand, can further extend into the following morning after 4–5 consecutive nights of restricted sleep (Lo et al., 2019). Time spent on a nap results in comparable or better delayed recall of learned material compared to cramming (Cousins, Sasmita, & Chee, 2018). The value of shorter naps, while shown in adults (Brooks & Lack, 2006), has not systematically been studied in adolescents. While tempting, one should not extrapolate benefits beyond what the data informs. For example, for reasons that are not clear, regular long naps (>1 hour) in older adults are associated with poorer cognitive outcomes and shortened lives (Cheng et al., 2018).

Repeatedly splitting restricted sleep opportunity clearly benefits cognition but it is associated with poorer handling of a glucose load (Lo et al., 2019). The latter may place a person at greater risk of diabetes mellitus with recurrent exposure to sleep restriction under this schedule. Thus, obtaining adequate nocturnal sleep is best for health, well-being and cognitive outcomes.

11Vo lume 10 | 2019 |

Chee, M. W. L.. (2019). Improving adolescent sleep: A compelling target for improving educational outcomes. OER Knowledge Bites Volume 10 (pp. 10–11). Singapore: National Institute of Education.

How to Cite

Starting school later is a strategy that has been shown to work in Singapore (Lo et al., 2018) and has been advocated by leading international professional bodies (Watson et al., 2017). Adolescents in their mid-teens have a biologically driven shift in preferred sleep time that is compounded by nighttime light exposure, electronic device use and homework related issues. Furthermore, changing prevailing indifference towards long-term consequences of poor sleep is critical for the national good, just as it is for the “war on diabetes”.

Sleep is a compelling target for intervention. Two decades ago, only dreamers would have thought that smoking in public places and transportation would be prohibited. Today, it is taken for granted. If for nothing else but plain economic benefit, the day will dawn when Singaporeans will value sleep like they do a smoke-free commute. Till then, educators need to be positive influencers by taking sleep to heart themselves and using the results of the class leading research from Singapore on adolescent sleep.

ReferencesBrooks, A., & Lack, L. (2006). A brief

afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep, 29(6), 831–840.

Cheng, G. H-L., Malhotra, R., Østbye, T., Chan, A., Ma, S., Lo, J. C. (2018). Changes in nocturnal sleep and daytime nap durations predict all-cause mortality among older adults: The panel on health and ageing of Singaporean elderly. Sleep, 41(7).

Cousins, J. N., Sasmita, K., & Chee, M. W. L. (2018). Memory encoding is impaired after multiple nights of partial sleep restriction. Journal of Sleep Research, 27(1), 138–145.

Cousins, J. N., Van Rijn, E., Ong, J. L., & Chee, M. W. L. (2019). A split sleep schedule rescues short-term topographical memory after multiple nights of sleep restriction. Sleep, 42(4).

Cousins, J. N., Wong, K. F., Raghunath, B.L., Look, C., & Chee, M. W. L. (2019). The long-term memory benefits of a daytime nap compared with cramming. Sleep, 42(1).

Lo, J. C., Lee, S. M., Lee, X. K., Sasmita, K., Chee, N. I. Y. N., Tandi, J., ... Chee, M. W. L. (2018). Sustained benefits of delaying school start time on adolescent sleep and well-being. Sleep, 41(6).

Lo, J. C., Lee, S. M., Teo, L. M., Lim, J., Gooley, J. J., & Chee, M. W. L. (2017). Neurobehavioral impact of successive cycles of sleep restriction with and wthout naps in adolescents. Sleep, 40(2).

Lo, J. C., Ong, J. L., Leong, R. L. F., Gooley, J. J., & Chee, M. W. L. (2016). Cognitive performance, sleepiness, and mood in partially sleep deprived adolescents: The need for sleep study. Sleep, 39(3), 687–698.

Lo, J. C., Twan, D. C. K., Karamchedu, S., Lee, X. K., Ong, J. L., Van Rijn, E., ... Chee, M. W. L. (2019). Differential effects of split and continuous sleep on neurobehavioral function and glucose tolerance in sleep-restricted adolescents. Sleep, 42(5).

Watson, N. F., Martin, J. L., Wise, M. S., Carden, K. A., Kirsch, D. B., Kristo, D. A., ... Chervin, R. D. (2017). Delaying middle school and high school start times promotes student health and performance: an American Academy of Sleep Medicine position statement. J Clin Sleep Med, 13(4), 623–625.

Yeo, S. C., Jos, A. M., Erwin, C., Lee, S. M., Lee, X. K., Lo, J. C., ... Gooley, J. J. (2019). Associations of sleep duration on school nights with self-rated health, overweight, and depression symptoms in adolescents: problems and possible solutions. Sleep Medicine, 60, 96–108.

12 | OER Knowledge Bites

How Language M i xes Con t r i bu t e t o E f f ec t i v e B i l i n gua l i smand E f f ec t i v e B i l i t e r ac y i n S i ngapo re : Su r ve y i ng t heLanguage Landscape

By Suzy Styles

Suzy Styles is Assistant Professor at the School of Social Sciences, Nanyang Technological University (Singapore). Her research interests include psycholinguistics, language develoment and lexicon growth, language comprehension in context, orthographic systems, and cross-modal perception of language and linguistic sound symbolism.

Language is the key to unlock learning throughout the life-span. However,

currently, most language research is conducted in monolingual communities, leaving a gap in our understanding of how language skills emerge for children in multilingual households, or how best to supporting their language development. The National Research Foundation (NRF) has funded a large-scale project allowing our interdisciplinary team to find out more about how different patterns of exposure in the home influence bilingual and biliterate language development.

What do We Know about Language Use in Singapore?The Singapore language landscape is changing rapidly, with notable increases

in the use of English. As can be seen in Figure 1 on the following page (adapted from Styles, 2019), in 2010, 30% of households reported that the main language spoken at home was English, and 10% of 15–24-year-olds reporting literacy only in English (DOS, 2011). 2010 was also the first time that English was reported as the main language at home for the majority of 5–9-year-olds (Wu, O’Brien, Styles, & Chen, in-press). However, the census data don’t tell us about how each family uses the combination of different languages when they are at home, and whether some home-language strategies are better for helping children develop strong language skills, and effective bilingualism. Our NRF Science of Learning project aims to find out.

What are We Discovering?We recently asked undergraduates from Nanyang Technology University, Singapore to complete a Language Fingerprint, informing us about the strengths and weaknesses across their different languages, and how much different caregivers spoke to them in each of their languages when they were

young. This allowed us to compute Composite Language Input Proportions (CLIPs) for each person (for example, the ratio of English to Chinese to dialect). We used a technique called Principle Components Analysis to find out how these different measurements are related to each other. Our preliminary findings suggest important links between the early years CLIPs, and adult proficiency in Mother Tongue (MT). In collaboration with Singapore’s GUSTO cohort study (Growing up in Singapore Towards Healthy Outcomes), we are investigating CLIPs reported by parents for several hundred infants at age 6- and 18-month-old. We evaluated infants CLIPs according to English, MT, and other regional varieties relevant to a particular MT (e.g., Chinese dialects; other varieties of Bahasa; other Indic languages) and a used cluster analysis to identify the dominant patterns of language exposure. Children’s’ CLIPs were best grouped into four clusters: English dominant; balanced English & MT; MT dominant; and Other Regional dominant. Between 6- and 18-month-olds, around 35% of infants change cluster; with most moving from a less standard cluster towards a more standard

13Vo lume 10 | 2019 |

cluster (e.g., MT dominant to balanced). These changes are most likely due to changing patterns of infant care over these ages, and are in line with general shifts in language use in Singapore. We are continuing our investigations into how these early CLIPs relate to the ongoing language development of the children, who are now 8 years old.

What are We Working On Now?Both of the studies described above rely on people estimating the amount of exposure children had to each of their languages, and language assessments that are taken some time after language exposure assessed in the CLIPs. We want a more detailed way to analyse what languages infants are really hearing at home (e.g., might be more mixed than some parents think), and we want to know more about how the developing brain starts to be shaped by the features of the languages that they hear. Our interdisciplinary team is developing a new kind of baby monitor—worn in a pocket on the front of the baby’s clothes. The monitor is able to detect and differentiate languages over a long stretch of recording, say, for a whole day. This will provide us with a more detailed picture

of the home language environments than ever before.

In order to create this device, we are recruiting 500 families in Singapore to take part in our Baby Talkathon. We will record all the sounds a baby hears during a day, and use this information to build a catalogue of how different languages sound like in Singapore, and what kinds of CLIPs different babies in Singapore hear. This Baby Talkathon collection will be used to train a new language detection system for the talk monitor.

We will also be working with some of these infants to find out how their different languages are developing (using vocabulary assessments), how quickly they recognise spoken words (using eye tracking) and how their brain is adapting to their unique language environment (using brain wave analysis).

A second group of children will be involved in investigations into the emergence of reading around the time they begin primary school. This project will allow us to understand more about the emergence of language skills across the diversity of Singapore’s language landscape, and to use that knowledge to support effective

bilingualism and biliteracy in Singapore’s littlest learners.

ReferencesDOS. (2011). Census of population 2010

statistical release 1: Demographic characteristics, education, language and religion. Singapore Retrieved from https://www.singstat.gov.sg/publications/cop2010/cop2010-sr1/.

Styles, S. J. (2019). Singapore’s languages at a glance. Retrieved from http://10.6084/m9.figshare.7781660.

Wu, C.-Y., O’Brien, B. A., Styles, S. J., & Chen, S. H. A. (in-press). The impact of bilingualism on skills development and education. In S. C. Tan & S. H. A. Chen (Eds.), Transforming teaching and learning in higher education: Chronicles of research and development in learning.

Styles, S. (2019). How language mixes contribute to effective bilingualism and effective biliteracy in Singapore: Surveying the language landscape. OER Knowledge Bites Volume 10 (pp. 12–13). Singapore: National Institute of Education.

How to Cite

Figure 1. Singapore Census Data for language and literacy.

14 | OER Knowledge Bites

HIGHLIGHTSF rom t he Pane l D i scuss i on

After the series of presentations, Professor David Hung chaired a panel discussion with speakers Professor Daniel Ansari, Professor Brian Butterworth, Associate Professor Paul Dux, Assistant Professor Azilawati Jamaludin and Assistant Professor Suzy Styles. Below are some highlights extracted from the discussion.

What has worked in order to build connections

between neuroscience and teachers, in which teachers

feel comfortable to approach neuroscientists

to share about their classroom issues?

COLLABORATIONS BETWEEN NEUROSCIENTISTS AND TEACHERS

SUSTAINING IMPACTS OF INTERVENTIONS

RESEARCH DIRECTIONS: THE WAY FORWARDHow can educational problems inform neuroscience or vice versa?

» Personalisation of learning that is scalable to typical classrooms; » How children with dyscalculia can be reliably identified; » Social component of learning; » Longitudinal research that links brain development to scholastic achievement; and » Learner’s own propensities that influence the environment they select for themselves and the

environments they elicit from others.

EXPOSURE TO MORE THAN ONE LANGUAGE IN THE EARLY YEARS

» Humility and respect are key; teachers don’t always respond positively to a researcher who criticises their practices with little understanding or experience of what goes on in the classroom.

» It is important to have avenues for researchers and educators to interact; to get people in the same room and speaking the same language. These avenues can include symposiums as well as teacher secondments at research centres.Associate Professor Paul Dux

» Neuroscientific measures may provide covert evidence of changes that may not be evident in behaviour. This data could be useful to paint a slightly different picture of behavioural outcomes and give a sense of why behavioural outcomes may not be sustaining.Associate Professor Paul Dux

» The fade-out of educational interventions is ubiquitous no matter how well-designed they are. If the impacts are to be sustained, then the intervention needs to be sustained, and at that point, policymakers play a critical role.Professor Daniel Ansari

» It is very normal to have more than one language spoken at home and most children still thrive in such linguistic environments.

» However, when children who are exposed to two languages since young are assessed on their vocabulary across the two languages they understand, their range of vocabulary is almost similar to that of one-language children. While their comprehension may be on track, their productivity is delayed.Assistant Professor Suzy Styles

How can the design of interventions among

disadvantaged populations be improved through

neuroscience research such that the impacts of the interventions can be

persistent and sustained?

Is it a good idea to learn a lot of different languages at a young age? Would it

confuse the child?

A publication of the Office of Education Research,NIE/NTU, Singapore © 2019