Research

Why should children learn computing?

Today, at every state-maintained primary school in England, children are learning concepts such as algorithms, abstraction, and decomposition (DfEd 2013). The national curriculum strongly emphasises computational thinking, encouraging the development of skills with application beyond the computing industry, as Jeanette Wing wrote: “computational thinking is taking an approach to solving problems, designing systems, and understanding human behaviour that draws on concepts fundamental to computing” (Wing 2006).

How do we learn concepts?

Recent neuroscientific studies have suggested that we store concepts in an amodal representational format, but that conceptual processing often involves strong interaction with sensory/motor systems (Mahon 2015). The context in which concepts are processed, including the thinker’s current task, determines which of the associated sensorimotor information is accessed. This implies that our ability to recall and process concepts may be enhanced by grounding them in rich sensory/motor experiences, via kinaesthetic and multi-media learning techniques.

Are Tangible Learner Interfaces effective?

Concrete manipulatives have been used in the formal / physical sciences for many years, however their required properties are not fully understood, and perhaps not intuitive (Clements 2000). An ineffective Tangible Learner Interface may enable the learner to complete tasks without developing an accurate conceptual understanding of their actions. “Although kinaesthetic experience can enhance perception and thinking, understanding does not travel through the fingertips and up the arm” (Ball 1992), and so the progression from perception to understanding requires instructional design consideration going beyond the physical attributes of an interface.

Can we scaffold the progression from concrete experience to abstract representation?

Educational psychologist Jerome Bruner theorized that knowledge could be understood using three distinct modes of representation: enactive (direct manipulation of physical objects), iconic (visual summary of conceptually relevant attributes), and symbolic (language based description), and that following this sequence would create an optimum learning experience (Bruner 1966). All three modes are commonly used in computing education, however research suggests the effectiveness of Bruner’s sequence may rely on the explicit linking and gradual transition between modes (Fyfe, McNeil, and Borjas 2015).

Is it possible to design a learning environment that fades between modes of representation?

Augmented reality (AR) enables embodied interaction with virtual objects, and the annotation of tangible objects with iconic and symbolic information (Santos et al. 2014). The designer of an AR learning environment has the power to manipulate modality, appearance, and quantity of virtual representations presented to the learner at any time, and therefore creating a gradual transition from concrete to abstract is possible. AR technology is sufficiently robust and affordable for implementation in primary classrooms, and has already been successfully applied to learning in a wide range of disciplines within this context.

Research Question #1:
Is Bruner’s modes of representation an effective method of learning computational concepts for children aged 7 to 11?

Empirical validation of Bruner’s theory has been mostly limited to within the field of mathematics education, therefore a study of its application to computational concepts will be undertaken across local primary schools. During an instructional phase enactive, iconic, and symbolic representations of novice-appropriate concepts will be presented in four distinct sequences: enactive only, symbolic only, enactive-iconic-symbolic, and symbolic-iconic-enactive, with the effectiveness of each sequence measured using a symbolic pre- and post-test.
The enactive-iconic-symbolic condition is expected to lead to significantly improved performance between the pre- and post-test, when compared to all other conditions.

Research Question #2:
Is a gradual transition between modes of representation more effective than an incremental approach?

Key computational concepts will be selected based upon their developmental and pedagogical appropriateness, and their compatibility with the modes of representation sequence. These will then be implemented in augmented reality learning environment prototypes, with each offering two distinct methods of presentation: an incremental sequence of representations, and a gradual transition between representations. The relative effectiveness of these two methods will be studied with children aged 7 to 11 in a local primary school, using symbolic pre- and post-tests.
Both conditions are expected to be beneficial, with the gradual transition showing significantly greater effect when compared to the incremental condition.

References

• Ball, D. L. (1992) Magical hopes: manipulatives and the reform of math education, American Educator, 16(2), pp. 14; 16-18; 46-47.
• Bruner, J. S. 1966. Toward a Theory of Instruction.Cambridge, MA: Harvard University Press.
• DfEd. 2013. “National Curriculum in England: Computing Programmes of Study.” https://www.gov.uk/government/publications/national-curriculum-in-england-computing-programmes-of-study/.
• Clements, D. 2000. “‘Concrete’manipulatives, concrete ideas.” Contemporary Issues in Early Childhood 45-60.
• Fyfe, E. R., N. M. McNeil, and S. Borjas. 2015. “Benefits of ‘concreteness Fading’ for Children’s Mathematics Understanding.” Learning and Instruction 35 (February). Elsevier Ltd: 104–20.
• Mahon, B. Z. 2015. “What Is Embodied about Cognition?” Language, Cognition and Neuroscience 30 (4). Routledge: 420–29.
• Santos, M. E. C., A. Chen, T. Taketomi, G. Yamamoto, J. Miyazaki, and H. Kato. 2014. “Augmented Reality Learning Experiences: Survey of Prototype Design and Evaluation.” IEEE Transactions on Learning Technologies 7 (1). IEEE: 38–56.
• Wing, Jeannette M. 2006. “Computational Thinking.” Communications of the ACM.