Developing Minds in the Digital Age

Chapter 1. Introduction

Soo-Siang Lim

Sonia Guerriero

Dirk Van Damme

Abstract

This chapter presents an overview of the book's organisation and the events that led to its publication. It introduces an interdisciplinary Science of Learning that integrates across levels of analysis and disciplinary perspectives to provide new advances in the study of learning. It argues for renewed efforts to make use of scientific findings of how people learn, in order to advance education goals, human development and 21st century workforce preparation. It also calls for the science of learning to inform the design of technology that more effectively support and enrich learning in this digital age. The chapter ends with implications for research, education policy, and practice and presents the potential of international networking for collaborative action to make transformative advances in learning and education.

Why science of learning? Why now?

Learning is a sine qua non of effective education and human accomplishment. It encompasses complex changes in the learner from his/her dynamic engagement with an equally ever-changing environment. At best, the interplay of the biological, physiological, cognitive and behavioural processes supporting the learner in interactions with the environment remain robust and resilient over a life-time of positive and negative experiences – at home, in school, at the workplace and in old-age. The key in the journey through this increasingly knowledge-intensive and fast-changing society is life-long learning.

Learning is an important research topic in many disciplines, but much knowledge is confined within disciplinary boundaries and practices. To address these challenges, the US National Science Foundation established the Science of Learning Centers (SLC) programme in 2003. Significantly larger than usual awards, the SLCs afforded large-scale, convergent and interdisciplinary efforts that integrate across levels of analysis and disciplinary perspectives – from molecular/cellular mechanisms of circuits and brain systems that underlie cognitive and behavioural processes, to social/cultural influences that affect learning – in individuals and in groups. The launch of this programme was timely, taking advantage of technological advances, particularly in neuroscience, engineering, and computer and information sciences, which enabled a re-examination of longstanding problems in learning, raised new questions and offered new approaches to the study of learning previously not possible. It fostered interdisciplinary team science and expanded the range of expertise traditionally identified with the study of learning in the context of education goals.

The premise was that new insights would be found through convergent, integrative knowledge at the interfaces of many disciplines, and that advances in our understanding of learning would have wide ranging societal impacts. The new insights were expected to not only transform how we learn and teach, but also lead to innovations in diverse technologies (e.g. in health and national security) that permeate 21st Century work and work-force preparation.

Overview of book organisation:

The idea for this book stemmed from a 2012 Conference, “Connecting How We Learn to Educational Practice and Policy; Research Evidence and Implications”. Focused on the science of learning, this conference was jointly organised by awardees of the US Science of Learning Centers programme and the Centre for Educational Research and Innovation at the OECD. In response to interest from researchers and educators in Australia, China and South America, a series of additional conferences followed in Brisbane, Shanghai and Rio de Janeiro, respectively.

This book highlights some of the increasing bounty of scientific findings about how people learn from participants in these conferences, who represented multiple disciplinary perspectives including neuroscience, social, cognitive and behavioural sciences, education, computer and information sciences, artificial intelligence/machine learning, and engineering. It seeks to catalyse discussions that examine the implications of these research findings for education practice and policy, and in turn, examine how knowledge and experience from real-world education practice and policy could challenge and inform research agendas and theory building. This book is not intended to be a comprehensive treatise of all topics central to learning, and does not presume to prescribe solutions to the myriad of complex educational dilemmas.

Nonetheless, taken in total, the Key Themes and Findings illustrate the diverse disciplinary and expanding scientific groundwork in the study of learning, explain the implications of research findings for educational practice and policy and present some cases of successful models for bridging basic research and education practice through interventions that improve learning and education.

Chapters 2 to 7 underscore the dynamic and interactive elements of biological, behavioural and cognitive development across the life-span, and the interplay of the learner with his/her environment. Just as technology increasingly pervades every human endeavour, the confluence of technology affordances in service of learning and education has been inevitable. Chapters 8 to 18 provide examples of technologies that promote spatial learning and others that create new learning opportunities; and how learning can be tracked, assessed and shaped at scale.

The aspirations of interdisciplinary research to make transformative advances in educational practice and policy are shared by other countries. Chapters 19 to 22 are examples of like-minded efforts in Australia, Hong Kong and Brazil that grew out of mutual support and collaborations with the US Science of Learning community. Each is a story about local motivators and grass roots organisation to create a national/regional core of experts, inspired by the use of scientific understanding of how people learn to make a difference in education outcomes. These and many other groups around the world represent the potential building blocks to realise the full power of networked knowledge and resources needed for transformations in learning and education.

Key themes and findings

Interplay between the learner and the environment

• Learning is life-long, starting with the important early years and the interplay of learner and environment.

Chapter 2 (Kuhl and Ferjan Ramírez) links neuroscience to behaviour to demonstrate the importance of social interactions and sensitive periods in early childhood that impact early language learning. The quantity and quality of early input also matters in how children master basic understanding of numbers. In Chapter 6 (Gibson, Berkowitz and Levine), cognitive science-based interventions based on better understanding of sources of variability in children’s understanding of number, factors that make learning about numbers challenging, and characteristics of effective number input were used successfully in promoting early number knowledge and reducing achievement gaps. Spatial skills are important foundational skills for science, technology, engineering and mathematics (STEM) learning. Chapter 16 (Toub, Verdine, Golinkoff and Hirsh-Pasek) demonstrates how everyday activities can create spatial learning opportunities through play with shapes, blocks, puzzles and origami. Utilizing these ways to develop spatial skills early in childhood will help scaffold formal learning later on. Across age, methods for training broad cognitive, social and communicative skills are critical. Chapter 7 (Congdon, Novack, Wakefield and Goldin-Meadow) shows that modifying everyday actions produced as gestures can causally improve learning outcomes. Gestures facilitate learning through its capacity to engage the motor system, be integrated with speech, and can be used as a spatial tool to reflect and to communicate thinking not apparent in spoken explanations. Chapter 15 (Khalil, Minces, Iversen, Musacchia, Zhao and Chiba) describes other enriching experiences that support learning, including music as a stimulant for learning that transfers to attentional behaviour, language, literacy and the ability to integrate easily into a social-cultural environment. Chapter 17 (Suarez, Samano, Yu, Snyder and Chukoskie) explores STEM learning opportunities that are facilitated by engagement of researchers with community educators and the general public

• Social identity and socio-emotional interactions matter in learning and STEM engagement

Underrepresentation of women and minority populations in STEM are existent inequalities in education and opportunity that are of a global concern. Chapter 3 (Meltzoff and Cvencek) combines psychology and education research to demonstrate how the gender stereotypes in girls develop from early social experiences that shape identity and self-concepts to negatively impact later academic performance and choices. Science-based interventions designed to strengthen children’s resistance to STEM stereotypes are described that help spark children’s continued engagement in and success in STEM disciplines. In Chapter 4 (Nasir), the intertwining of racial identity and learning is explored in the construct of Racialized Learning Pathways, and points of intervention that break the negative cycle between racialization and learning are highlighted. Chapter 5 (Rozek, Levine and Beilock) reveals that young children are sensitive to the maths anxiety experienced by adult role models, such as teachers and parents. Children with maths-anxious parents or teachers tend to show less growth in their maths knowledge than those without a maths-anxious role model. Home and classroom interventions that were effective in mitigating harmful learning consequences of negative affect are highlighted.

Science of learning in design and use of technology for learning

Technological affordances create significant possibilities for supporting learners and teachers. Aligned and combined with principles of how people learn, they provide novel avenues for enhanced learning through new ways of presenting the curriculum, adaptivity in how information is presented, and the provision of feedback contingent on learner needs, knowledge and behaviour. Technology presents new avenues and scales of access for the learner, as well as new ways for learners to express themselves through natural language and sketching to communicate, interact and collaborate with other. These new technologies promote learning through argumentation and develop learners’ identities as reasoners and problem solvers.

Chapter 8 (Forbus and Uttal) describes how modern digital technology provides new options for enriching spatial learning with a focus on technologies for: 1) improving communication via sketching where the software incorporates cognitive models of human visual, spatial, and conceptual representations and processes; and 2) Geographic Information Systems (GIS), computer-based mapping systems that facilitate spatial data analysis and visualization, leading to improved spatial reasoning in high school students. Recognizing that digital technologies have become an integral part of children’s lives, Chapter 9 (Barron and Levinson) urges us to consider not only the content children use, but how families can engage collaboratively in technologies through joint attention to support development in and out of school. This brings to the fore the need for programmes and policies to support equitable learning opportunities for parents and educators, in order to avoid exacerbating the digital gap. Chapter 10 (Llorente, Moorthy and Dominguez) shows how Designed Joint Engagement with Media (DIEM) can promote co-reviewing and joint attention which in turn improves children’s development of early science literacy.

In Chapter 11 (Okita), the use of programmable robotic systems, virtual avatars and computer agents have revealed new insights into the role of social relationships in learning. For example, the mere “belief” that an avatar was human (vs. a computer agent) promoted significant learning gains and higher arousal measures. Incorporation of well-grounded theories of learning, such as Learning-by-Teaching and Recursive Feedback into technology design optimizes the learning relationship between learner and technology. Chapter 12 (Klahr and Siler) introduces the TED tutor, an intelligent computer-based tutor that adapts instruction to individual students based on its assessment of their knowledge and ability. The tutor is used to address children’s difficulty in acquiring an understanding of basic experimental design known as Control of Variables Strategy (CVS). This essential component of STEM education all too often receives inadequate instruction; the TED tutor can be used as an online instruction to augment classroom instruction.

Chapters 13 (Koedinger) and 14 (Rosé, Clarke and Resnick) address technology-mediated possibilities of scale – and address issues for scaling iterative course improvements, learning research in classrooms that are practical and aligned with real-world contexts, and professional development in instructional practice. Chapter 13 demonstrates how systematic and iterative use of intelligent tutoring systems in large-scale classroom experiments has led to the development of the Knowledge Learning Instruction (KLI) framework. This new learning theory is driven and fuelled by Big Data, AI algorithms, education data mining and learning analytics, and has provided important guidance and a “roadmap” in choosing among the trillions of instructional strategies to improve learning and education. Chapter 14 provides another example of AI-enabled scaffolding of learning through collaborative and discussion-based strategies that benefit student learning and teacher professional training. It draws on classroom facilitation practices by teachers referred to as Accountable Talk (AT), and students’ articulation of reasoning and transactive exchange. Findings derived from teacher-led classroom discussions and computer-supported collaborative learning (CSCL) were positive. For example, a district-wide professional development effort in promoting transactive exchange through automated AT facilitation in small groups led to enhanced learning, and greater uptake of AT facilitation practices among teachers; a second investigation leveraging the same concept of Transactivity in a crowdsourcing environment and then tested in a real MOOC deployment, yielded promising results as well. Mandated tests exert strong influence on what is taught in schools and how it is taught. Chapter 18 (Means, Cheng and Harris) argues that rich technology-based environments will be necessary to assess the science proficiencies described in national standards, to capture students’ ideas, concepts and practices simultaneously. Innovations in science and technology-based assessments have the potential to resolve the challenge that many countries face in trying to align and reconcile classroom-based and national learning assessments.

Implications for research, education practice and policy: Opportunities and challenges

• The need for more effective communication and shared knowledge across research, education practice and policy communities

Large disconnects exist among the diverse basic science disciplines and education research, and among researchers, education practitioners, policy makers and other stakeholder communities. Inadequate communication and sharing of knowledge have been among the varied reasons why longstanding, complex problems about learning continue to elude our understanding, and why research findings have not found traction and adoption in education practice and policy.

The good news is that for the most part, the need for and the importance of interdisciplinary approaches to study learning is recognized, and the availability of research funding opportunities have increased collaboration among the basic science disciplines. The gaps between basic science researchers and education researchers continue to be challenging, complicated in part by the inherently different contexts in which research are conducted: the “messy” classroom with difficult to control variables, and in contrast, the use of abstracted stimuli and highly controlled experimental conditions in the laboratory.

The application of laboratory-derived scientific principles of learning to classroom practice remains challenging. To date, there are varying models and outcomes, usually confined to a few classrooms, except in cases of technology-mediated interventions that are more scalable. While there is implicit acceptance that knowledge about how people learn should be the foundation for how we teach and educate, the reality is that researchers, education policymakers and education practitioners rarely have opportunities to examine and discuss the issues surrounding application of evidence-based learning practices. More accessible and active collaboration between researchers and practitioners are needed to affect the following: 1) ensure that the research conducted is relevant to practitioners; 2) engage practitioners in the research process, thereby building capacity for educators to become more knowledgeable about the science of learning; and 3) contribute to the building of a knowledge base and conceptual framework for translation of research to practice.

• Renewed investment in preparing researchers to work in a data-driven society

The continued development of new technologies will advance acquisition, sensing and processing capabilities of data collection in real-time during learning and in real-world contexts. This promises exciting possibilities to better understand the complex dynamics of the many factors that impinge upon the learner simultaneously, including the biological and physiological influences as well as those from the external, physical and social environment. Current capabilities have already raised concerns about workforce preparedness to harness the big data promise in ways that properly acquire, curate, analyse and share these data resources. There is urgent need to train researchers to effectively generate and use sensitive, public and private big data that is actionable and of value.

• Renewed investment in teacher education and research about teaching

Teacher education has traditionally focused primarily on subject content, classroom management, teaching methods and a rudimentary understanding of learning, child development and the use of assessment for diagnostic purposes. Teacher education has not kept pace with developments in the science of learning and more in-depth understanding of learning processes. This leads to a serious weak link as the teacher is a critical partner in the implementation of science-based interventions to improve learning. There is need for policy change to create incentives and new requirements in the reform of teacher education curricula and the qualifications of teacher educators to enable the incorporation of the new knowledge and practices. This is a complex issue because teacher education takes place in universities and universities have autonomy over their programmes and hiring practices.

There is also need to better understand the process of teaching and the underlying cognitive, behavioural and social processes. Teaching is a complex cognitive skill that requires real-time learning (e.g. of cues from the learner, class) for real-time decision-making involving many factors (the student, the task, the classroom environment, the teacher’s expertise with the material, the teacher’s cognitive processes). Currently there are few studies on how teacher decision-making affects student learning; such studies would also benefit from collaborations among basic science researchers, teachers and teacher educators.

New technologies with learning analytic functions can be used to support teaching and learning in the classroom by supporting teachers to adapt lessons for students of different abilities or as a teacher’s aide with large or multi-grade classrooms. Teacher training and technical support of teachers are needed for timely incorporation of learning technologies for pedagogical purposes.

• Investment in socio-technical infrastructure to facilitate knowledge convergence and collaboration among research, educator and policymaker communities

Effective collaboration among disparate disciplines spurs innovation and is critical for solving society’s complex problems. Investment in cyberinfrastructure for the Science of Learning community will take advantage of the affordances of technology to bridge disciplinary and geographical barriers so that researchers and stakeholders can better collaborate and converge their expertise and resources towards shared goals and problem-solving. Such investment in infrastructure will capitalise on the growing international expectations that the science of learning can play a vital role in training the research community to master the skills necessary to benefit from an increasingly data-intensive environment; to build a foundation of shared values and standards that facilitate transparent, high-quality science; and to foster a culture of ethical, responsible innovation that anticipates and addresses the threats and vulnerabilities accompanying the digital age.

The views expressed in this article are those of the authors. They do not necessarily represent the views of the United States National Science Foundation, UNESCO, nor the United States Government.