Research on the Influence of Learning Based on Game Design on the Deep Learning of Middle School Students

Abstract

The knowledge of deep learning experts, the cultivation of high-order thinking and the input of emotions are the goals pursued by current education. Game-based learning refers to the use of appropriate tools to build games to support student learning. This approach provides learners with powerful contextual and real-world questions that effectively stimulate students' interest in learning and motivation to learn. Develop high-level thinking and creativity in the process of design, development and debugging to achieve deep learning. The study collected the test scores of the students, the final works and the programming notes. Through qualitative and quantitative analysis, the results are as follows: The game design-based learning is enough to promote students' deep understanding and application of programming knowledge; in terms of thinking, the experimental group performance There are more complex task interactions and problem solving; in terms of emotions, the experimental group has invested more time energy and expressed more interest. At the same time, the study found that collaboration can make up for the lack of learning methods to a certain extent and improve students' thinking level. This method provides an idea for improving the current programming teaching.

Full text

1
Research on the Influence of Learning Based on Game Design on the Deep Learning
of Middle School Students
Xinyi Li1, Xiaomei Yan1, Yue Xiao1 and Jingying Wang1
1. Beijing Normal University
ABSTRACT
The knowledge of deep learning experts, the cultivation of high-order thinking
and the input of emotions are the goals pursued by current education. Game-based
learning refers to the use of appropriate tools to build games to support student
learning. This approach provides learners with powerful contextual and real-world
questions that effectively stimulate students' interest in learning and motivation to
learn. Develop high-level thinking and creativity in the process of design,
development and debugging to achieve deep learning. The study collected the test
scores of the students, the final works and the programming notes. Through
qualitative and quantitative analysis, the results are as follows: The game design-
based learning is enough to promote students' deep understanding and application
of programming knowledge; in terms of thinking, the experimental group
performance There are more complex task interactions and problem solving; in
terms of emotions, the experimental group has invested more time energy and
expressed more interest. At the same time, the study found that collaboration can
make up for the lack of learning methods to a certain extent and improve students'
thinking level. This method provides an idea for improving the current
programming teaching.
Keywords: Game-based learning, deep learning, open source hardware programming
INTRODUCTION
Deep learning is a new requirement of teaching in the new era and an important indicator of
the quality of learning. In the Horizon Report, deep learning is one of the key trends driving the
application of school education technology(Wang & Zhang,2016)
i
. Deep learning refers to actively
and critically learning new ideas and knowledge, using diverse learning strategies to deeply process
knowledge information, establishing new and old knowledge information, building personal
knowledge systems and effectively migrating them to real situations. Solving complex problems(Bu
& Feng,2016). Deep learning emphasizes students' meaningful construction of knowledge, the
cultivation of high-order thinking and the input of emotions, which are the goals pursued by current
education.
With the rapid development of computers and networks, programming skills have gradually
become the basic skills of the 21st century. More and more schools are teaching programming,
cultivating students' mastery of programming knowledge and the ability to use programming
thinking to solve practical problems. However, as programming teaching tends to be younger, there
is a general difficulty in understanding programming concepts in the actual programming process(Di
Lieto, M.C., et al.,2017), students' learning stays in skill imitation, difficulty in knowledge transfer,
lack of creativity and learning motivation, and inability to achieve deep learning(Yan& Zhong,2018).
Therefore, how to find a suitable learning method for students and improve the deep learning of
students' programming knowledge is a problem that needs to be solved in the current programming
teaching of primary and secondary schools.

2
In order to promote the realization of deep learning in programming, the existing research
mainly intervenes in programming teaching from two aspects: teaching strategy and teaching mode.
The teaching strategy mainly explores three aspects: programming tool selection, tool allocation
strategy and collaborative form. However, this kind of research pays more attention to the influence
of strategy on programming performance, and does not deeply explore its role in understanding the
specific programming concept. Fundamentally improve the imitation of this form of teaching. The
exploration of teaching mode provides ideas for improving this problem as a whole. For example,
Zhong Bochang cooperates with first-line teachers to extract four kinds of new programming
teachings: invention and creation, scientific inquiry, interesting interaction and experimental
simulation. Mode. These four types of teaching have all improved the shortcomings of imitation and
lack of creativity in the original teaching mode to a certain extent, and also provided a good idea for
the source of teaching cases. However, this part of the research stays at the theoretical stage, and
there is a lack of empirical research to quantify the data to present its impact on students' learning
outcomes .
Learning based on game design refers to a learning method in which the learner actively
designs and develops the game in the real problem situation, and uses the feedback to improve the
function to realize meaningful knowledge construction. The method provides students with an
effective learning situation through “game design”, which includes real tasks and complex
backgrounds, which can promote students to learn independently and achieve meaningful
knowledge construction (Robertson, J. 2012).
LITERATURE REVIEW
Definition and characteristics of deep learning
The concept of deep learning was first proposed by Marton and Saljo. When they studied the
strategies adopted by students in reading academic papers, they found that learning can be divided
into shallow learning and deep learning(Marton F, Saljo R.1976). Scholars have different
understandings of deep learning and summarize relevant literature. Deep learning is mainly
reflected in three aspects of cognition, thinking and emotion. (1) Cognitive aspect, deep learning
means that the cognitive structure of learners changes and achieves high Hierarchical cognitive
goals (Duan & Yu,2013). (2) In terms of thinking, students improve their metacognitive level and
other high-order thinking skills through reflection, conceptual interaction and other behaviors
during the learning process (De Corte, E. 2003). (3) In terms of emotions, deep learning indicates that
students have high behavior and high emotional input. Although scholars hold different opinions on
definitions, the researchers' views are more consistent on the results of deep learning, that is, deep
learning shows that students apply what they have learned to new scenarios and solve practical
problems (Robertson, J., & Howells, C. 2008) .
Learning definition and application based on game design
Learning based on game design, that is, the student is the designer of the game, the process of
game design is the process of learning. In the Chinese literature, no scholars explicitly put forward
the concept of "learning based on game design", but in the foreign literature, the author found many
similar expressions, such as computer game making, game-base construction learning, computer -
assisted game making, learning through game-making, etc., although there are differences in the
name, in terms of connotation, the game design is the content of teaching, allowing students to
actively learn and construct knowledge in the process of designing the game. Develop high-level
thinking skills.
Game design is a comprehensive task that provides students with an effective learning context
that encourages students to use a variety of skills to create complex artifacts such as game rule
design, character creation, visual design, and Computer Programming(Robertson, J. . 2012) .

3
Compared with the single identity of the “player” that the learner only has in the game, the learner
will have different identities such as user, designer, copy plan and programmer in the game design
process. It promotes the initiative and creativity of students, and there are constant trial and error
and modification iterations in the game production process. This process cultivates students'
problem-solving ability and deep understanding of knowledge (Vos, N., van Der Meijden, H., &
Denessen, E. 2011) .
Many researchers have conducted empirical research to verify the effectiveness of learning
based on game design. Some researchers apply game design to programming and mathematics to
improve students' learning performance. For example, Wilson uses Scratch for programming
teaching, and evaluates the game designed by students from the perspective of procedural
knowledge, programming structure and practicality. It is finally found that students use interactive,
loop and conditional statements in the game. The programming concept is well mastered (Wilson,
A. , Hainey, T. , & Connolly, TM . 2013). Denner asked students to create games with programming
software after class, and found that programming activities can promote students' understanding of
programming concepts (Denner, J., Werner, L., & Ortiz, E.,2011). Jiang Fengguang and others used
Scratch to write math games to promote students' mastery of mathematical equations (Chiang, F. K. ,
& Qin, L. .,2018). In addition to promoting the mastery of specific subject knowledge, more
researchers have explored the impact of this approach on student abilities, such as problem-solving
skills, critical thinking, and creativity. Gary built the game design pattern (GMP, full name Game
Making Pedagogy) to solve the learning problems students encounter when using interactive media.
The model sets up seven stages to guide students to design, develop and debug their own games.
Can promote students' problem-solving skills, learning motivation, and creativity in interactive
media (Cheng, G. ,2009). Robertson explored the impact of students using Neverwinter Nights 2
software to make their own successful learning in the UK's excellence program. The study found
that students' motivation to learn, knowledge transfer, and determination to complete tasks were
achieved through the production of computer games. Upgrade . Thomas explored the impact of this
approach on 21st century skills, and found that students have improved problem-solving skills and
creativity in the production of game projects (Thomas, MK , Ge, X. , & Greene, BA ,2011).
In summary, in the programming teaching of primary and secondary schools, there are
problems such as difficulty in understanding the concept of students' knowledge, lack of motivation,
and low creativity, which cannot achieve the goal of deep learning. The game design-based learning
effectively improves students' learning motivation, knowledge construction level and problem
solving ability by providing real problem situations, rich tasks and multiple roles, which promotes
students' deep learning. Therefore, this study will apply the game design-based learning to the
programming teaching of open source hardware, and explore its impact on deep learning of
programming. The specific research questions are as follows:
Can game-based learning promote deep learning in middle school programming?
At the same time, according to the evaluation dimension of deep learning, the problem is
divided into the following sub-problems:
(1) Can learning based on game design improve the cognitive level of learner programming
knowledge?
(2) Can learning based on game design promote the development of students' high-level
thinking?
(3) Can learning based on game design improve the emotional level of students' learning
programming?

4
THEORETICAL FRAMEWORK
Learning framework based on game design
The task-driven approach believes that the learners, learners, and learning activities are the
core elements of the teaching process. The richer the learning activities, the richer the teacher's
behavior will be, the better the corresponding student behaviors and learning effects will be. The
relationship between them can be Expressed as: Q=F(AT, AS, T) (Du & Fan,2006). When setting up
tasks, the driving problem should have practical directivity. The problem of designing mosaic
knowledge must have certain problem space and the characteristics of complexity and multi-path,
and cultivate students to use knowledge to solve practical problems (Lindh, J., & Holgersson, T.,2007).
The completion of deep learning requires three stages: knowledge acquisition, skill development,
and application of knowledge to solve problems. The transition between each stage does not occur
naturally, and it depends on certain strategies (Kucuk, S., & Sisman, B.,2017) . Based on this, this
study adopts the framework and process of task-driven method, and uses game design as a strategy
to support the promotion of each teaching link, in order to promote the deep learning of student
programming. The framework built is shown below:
In the pre-class preparation stage, the teacher refers to the reverse engineering DAA activity
model and transforms the actual game into a teaching case through the three steps of
“deconstruction-analysis-reorganization”. The task setting of the experimental group follows three
stages of deep learning, namely, the acquisition of knowledge, the improvement of skills, and
finally the use of knowledge to solve problems. Specifically, in the task setting, the first stage
"interactive task" corresponds to the knowledge acquisition, that is, the basic use method of the
master component, and the second phase "feedback task" corresponds to the skill improvement, and
the phase is a shallow to deep transition link, so the task Set to semi-open, pointing to problem
solving, while also retaining a certain problem space.
MATERIAL & METHOD
Context
The study subjects were 50 high school students from the affiliated middle school of Peking
University, including 25 in the experimental group and the control group. All the participants were
enrolled in the open source hardware course through the school elective system. The students
mainly came from the first year of high school. Most of the students had taken the python course
and had a certain programming foundation. However, most students did not have access to open
Figure 1 Experimental group "learning based on game design" teaching framework

5
source hardware. Therefore, the operational skills are relatively lacking. At the same time, the
development level of high school students' thinking is relatively perfect, with high abstraction and
theory.
Due to the randomness of the course selection, this study analyzed the initial level of the
subjects. The source grades, elective courses, and instructors of the subjects were all different, and
they were not comparable. Therefore, the whole study was selected. Uniform evaluation criteria:
GPA, as a pre-test data. There was no significant difference in the grade points between the
experimental group and the control group, which met the requirements of quasi-experimental
research and could be followed by experimental intervention.
Procedure
The study began in July 2018 and ended in July, with a total of 36 weeks of 9 weeks, including two lessons
per week, each with a duration of two hours. In the first 16 weeks, the experimental group and the control
group completed the basic knowledge of software and hardware. From 17 to 28 hours, the experimental
group used game design-based learning, and the control group used skill-based learning. In this process,
students' programming was collected. Notes; After all the courses have been completed, the students'
knowledge mastery will be tested by paper and pencil, and the students' final works will be collected. Finally,
the collected materials and data are analyzed to draw conclusions.
Design method for evaluation of deep learning
This study uses a game design-based learning approach to design the curriculum, and in the
teaching to verify its promotion of student programming deep learning. The evaluation of deep
learning is carried out in three dimensions: cognition, thinking and emotion. The study collects the
test scores of the students, the final works of the students and the programming notes. The
corresponding measurement dimensions and methods are shown below:
Table 1 Deep learning evaluation dimensions and tools
Data Collection
Assessment Dimension
Research Tool
Knowledge test score
(subjective, objective)
Cognitive
Test
Product
（program、text、picture、video etc.）
Cognitive、
Thinking、Emotion
Assessment Scale
Programming Notes
Cognitive、
Thinking、Emotion
Qualitative coding
DATA ANALYSIS & DISCUSSION
Game-based learning is less effective in applying and analyzing cognitive goals
Since the experimental and control samples in this study were all 25 samples and belonged to
small samples, the Kolmogorow-Smirnov Test and the Levene Test were used to test the normality
and the homogeneity of the variance, respectively. The sample was found to satisfy a normal
distribution with no difference in variance. An independent sample t-test was performed on the data
of the experimental group and the control group, and the data were as follows:
Table 2 Experimental group and control group test score independent sample t test
Item
Category
Num
Avg
SD
SE
t
Sig.

6
Total
Experimental
22
76.09
14.395
3.069
0.917
0.365
Control
21
72.00
14.873
3.246
Deep-learning Score
Experimental
22
43.59
10.276
2.191
0.446
0.658
Control
21
42.14
11.001
2.401
Surface learning
Experimental
22
25.50
5.125
1.093
1.587
0.120
Control
21
22.86
5.790
1.264
As can be seen from Table 11, the scores of the total score, deep knowledge and shallow
knowledge of the experimental group were higher than those of the control group, but they did not
reach significant differences. Therefore, there was no difference in the scores of the knowledge test
between the experimental group and the control group.
Learning based on game design has a positive impact on students' deep knowledge application and
emotional input
The students' work was scored from the three dimensions of cognition, thinking and emotion,
and the scores of the experimental group and the control group were tested differently. Since the
sample size is small, all the data are tested for homogeneity and normality of the variance. In the
end, only the two scores of the total score and the depth of knowledge application are satisfied. An
independent sample t test can be used, and other items are used. Mann-Whitney U test of
parameters.
It can be seen from the table that the works of the experimental group are superior to the
control group at the overall level (P=0.031<0.05), and the experimental group is also more obvious
in the depth of knowledge application (P=0.035<0.05). It shows that the experimental group has a
deeper understanding of knowledge when making works. At the same time, in terms of emotion, the
experimental group showed more time input (P=0.01<0.05) and resource call (P=0.01<0.05) than
the control group, indicating that the use of game design-based learning promoted students'
emotional aspects. Level of commitment.
Table 3 Experimental results of different dimensions of the experimental group in the experimental
group
Dimention
Test
Group
Avg
SD
t
Sig.
Score
Experimental
32.09
7.661
2.322*
.031
Control
24.64
7.393
Cognition
Experience
Experimental
3.18
.874
.300
Control
2.82
.603
Width of application
Experimental
3.27
.905
.116
Control
2.64
.809
Depth of application
Experimental
3.55
1.073
2.268*
.035
Control
2.45
1.214
Emotion
Explore
Experimental
3.45
.934
.797
Control
3.18
1.328
Resource
Experimental
3.91
.701
.010*
Control
2.55
1.293

7
Dimention
Test
Group
Avg
SD
t
Sig.
Time engagement
Experimental
3.73
1.218
.010*
Control
2.18
1.328
*p<0.05
Learning based on game design can promote students' deep application of knowledge, improvement
of program logic structure and emotional input
In terms of knowledge mastery, the control group stayed in the simple combing and review of
knowledge, while the experimental group was able to call more previous knowledge to solve the
problems encountered. In terms of thinking, the experimental group had higher complexity and
more functions than the control group. The control group mainly realized the basic functions. Some
students carried out the function development at the detail level on the original basis. The students
in the experimental group tend to propose new requirements and implement them on the basis of the
original. At the same time, in terms of debugging procedures, the experimental group showed more
exploration and deeper learning than the control group. The control group generally only
objectively lists the difficulties encountered in programming, and the experimental group basically
formed the behavior pattern of asking questions → analyzing the causes → solving the problems.
After asking questions, students in the experimental group tend to find out the cause of the bug and
debug it. In terms of emotions, the students in the experimental group recorded their sense of
accomplishment, interest in the task, time and energy invested in the programming notes, and some
students expressed their fears, while only one classmate in the control group expressed the task.
Interested.
CONCLUTION & FUTURE
In order to promote students' deep learning in programming, this study combines the elements
of deep learning with game design, proposes a learning style based on game design, and adopts a
paradigm of quasi-experimental research to explore the way in cognitive, emotional and thinking.
The impact of dimensions on students' deep learning, through data collection and analysis, is
concluded as follows:
Learning based on game design can promote students' deep application of knowledge
In this study, an independent sample t-test was conducted on the test scores of the students. It
was found that the experimental group was higher than the control group in terms of total score,
deep learning score and shallow learning score, but they did not reach significant difference.
Because of the limitations of the paper-and-pen test on the analysis and creation of the two
dimensions of the target evaluation, it shows that there is no difference between the experimental
group and the control group in the dimensions of memory, understanding, application and analysis.
The analysis of the student's work found that the students in the experimental group had more in-
depth application of knowledge in the production of the work. The functional design of the
experimental group was more complicated. When using the same knowledge point, the students in
the experimental group focused more on the analysis and evaluation. Application, not simple
operation. By analyzing the students' classroom coding, it is found that the experimental group can
link the newly learned knowledge with the existing knowledge, and realize the in-depth
understanding of knowledge through the interaction of concepts (Kucuk, S., & Sisman, B. ,2017).
Learning based on game design can promote the development of students' higher-order thinking
This study analyzes the student's programming notes and draws conclusions. The experimental
group showed more problem solving and improvement of the program logic structure, and higher-
order thinking was improved. This study conducted a differential test on the scores of the thinking

8
of students' works, and found that there was no difference in thinking between the experimental
group and the control group. The conclusions of the two data analyses are inconsistent. The reason
may be that the student works are completed by the peers. The way of peer collaboration makes up
for the lack of learning methods and promotes the development of students.
Learning based on game design has a positive impact on students' emotional level
This study analyzes the emotional scores of student works and draws conclusions. The
experimental group used more resources and invested more time in making the works, and had
more emotional input. The emotional input code of the student's programming notes was analyzed,
and it was concluded that the experimental group expressed more emotions and invested more
energy. This may be because the students in the experimental group have more immersion and more
emotional experiences when exploring and interacting. The study found that on the whole, girls are
more willing to express emotions than boys, and the proportion of statistical writing found that 90%
of the notes were written for girls. Other studies have similar findings, that boys do not like to
follow the teacher's instructions, while girls are more concerned with the task of writing (Thomas,
MK , Ge, X. , & Greene, BA ,2011) In the study of pupils' behavior patterns, Kucuk found that female
students expressed more feelings than boys, and that girls had more negative emotions than boys
(Cheng, G,2009). In terms of learning interest, students in the experimental group showed higher
interest in teaching cases. It may be because of the fun nature of the game itself that gives students
the interest and motivation to learn further.
The setting of game difficulty will affect students' cognitive level and emotional input.
Learning based on game design is a double-edged sword. This study found that in the ninth
lesson, students' programming notes have more positive feedback and functional design, but in the
notes of Lessons 10 and 11. There is a more extreme situation. Some students express the fun of the
game and the sense of accomplishment obtained through the design of the game. However, some
students express that the task is too difficult and the individual ability cannot be completed. The
main reason is the difficulty of game design. The game functions in the last two sections are more
complicated and difficult. Therefore, in future research, it is necessary to coordinate the balance
between the difficulty of the game and the new knowledge.
With the promotion of programming education in primary and secondary schools, how to
better promote the deep construction of students' knowledge and the development of higher-order
thinking is a common concern of researchers and front-line teachers. Compared with the existing
research, this study draws on the strategies of other disciplines to promote deep learning, and
integrates it with deep learning, and proposes a game-based learning method suitable for open
source hardware programming, which improves the current imitation of programming teaching. The
problem of light application has promoted the deep learning of student programming. At the same
time, in the game design-based learning, the difficulty setting of the game is the key factor for the
effect of the method, and future research can focus on the balance between the difficulty of the
game and the new knowledge. Finally, in many studies, collaboration is also an important way to
promote deep learning, and it is also an effective measure to reduce the difficulty of programming.
Although this study involves collaboration, it does not collect relevant data, nor does it have
programming collaboration and deep learning. The relationship is explored, and future research can
choose such concerns and conduct research.

9

10
REFERENCE
Wang Xiaochen,Zhang Jinbao,Du Jing,Zhang Dingwen,Gao Yuan.Exploration of the Development of
Educational Technology and the Application Model in the Context of Global Education
Informationization——Review of the First China-US Smart Education Conference[J].Journal of
Electrotechnical Education,2016,37 (03): 34-41.
Bu Caili, Feng Xiaoxiao, Zhang Baohui. The Concept, Strategy, Effect and Enlightenment of Deep
Learning——Interpretation and Analysis of American Deep Learning Project (SDL)[J]. Journal of
Distance Education, 2016, 34(05): 75-82.
Di Lieto, M.C., et al., Educational Robotics intervention on Executive Functions in preschool children: A pilot
study. Computers in Human Behavior, 2017. 71: p. 16-23.
Yan Ni, Zhong Bochang. Core Theory Research of Robot Education in Primary and Middle School——On
Invention and Creation Teaching Mode[J]. Research on Electron Education, 2018, 39(04): 66-72.
Zhong Baichang. Core Theory Research of Robot Education in Primary and Middle School——New
Classification of Robot Teaching Mode[J]. Journal of Electrotechnical Education, 2016, 37(12): 87-92.
Zhong Baichang, Han Lei. The Core Theory Research of Robot Education in Primary and Middle School——
On the Interesting Interactive Teaching Mode[J]. Research on Electron Education, 2018, 39(09): 88-95.
Robertson, J. . (2012). Making games in the classroom: benefits and gender concerns. Computers &
Education, 59(2), 0-398.
Marton F, Saljo R. On qualitative differences in learning: I-Outcome and process [J]. British Journal of
Educational Psychology, 1976, (46): 4-11.
Duan Jinju, Yu Shengquan. Research on e-Learning Deep Learning from the Perspective of Learning
Science[J]. Journal of Distance Education, 2013, 31(04): 43-51.
De Corte, E. (2003). Transfer as the productive use of ac- quired knowledge,skills,and motivations [J].
Current directions in psychological science,12( 4) : 142-146.
He Ling, Li thickening. Promoting students' deep learning [J]. Computer Teaching and Learning, 2005, (5):
29-30.
Robertson, J., & Howells, C. (2008). Computer game design: opportunities for successful learning.
Computers & Education, 50(2), 559-578.
Robertson, J. . (2012). Making games in the classroom: benefits and gender concerns. Computers &
Education, 59(2), 0-398.
Vos, N., van Der Meijden, H., & Denessen, E. (2011). Effects of constructing versus playing an educational
game on student motivation and deep learning strategy use. Computers & Education, 56(1), 127 –13.
doi:10.1016/j.compedu.2010.08.013.
Wilson, A. , Hainey, T. , & Connolly, TM . (2013). Using scratch with primary school children: an evaluation of
games constructed to gauge understanding of programming concepts. International Journal of Game-
Based Learning, 3, 93-109.
Denner, J., Werner, L., & Ortiz, E. (2011). Computer games created by middle school girls: Can they be used
to measure understanding of computer science concepts? Computers&Education,58(1), 240– 249.doi:
10.1016/j.compedu.2011.08.006.
Chiang, F. K. , & Qin, L. . (2018). A pilot study to assess the impacts of game-based construction learning,
using scratch, on students’ multi-step equation-solving performance.
Cheng, G. (2009). Using game making pedagogy to facilitate student learning of interactive multimedia.
Australasian Journal of Educational Technology, 25(2), 204-220.
Thomas, MK , Ge, X. , & Greene, BA . (2011). Fostering 21st century skill development by engaging
students in authentic game design projects in a high school computer programming class. Journal of
Educational Computing Research, 44(4 ), 391-408.
Qin Qin, Tan Liang, Zhou Xiongjun. Research on Task-Driven Teaching Model Based on Computational
Thinking[J]. Modern Educational Technology, 2011, 21(6): 44-49.
Du Jianxia, Fan Si Du Lin Tang, Anthony Olinzok. Dynamic Online Discussion: Deep Learning in an
Interactive Learning Environment [J]. Open Education Research, 2006, (8): 75-78.
J.S. Lanzuli, S.M. Rees. Rich teaching model: a book on quality education [M]. 2000:214-220
Lindh, J., & Holgersson, T. (2007). Does LEGO training stimulate pupils' ability to solve logical problems?
Computers & Education, 49(4), 1097–1111.
Kucuk, S., & Sisman, B. (2017). Behavioral patterns of elementary students and teachers in one-to-one
robotics instruction. Computers & Education, 111, 31–43.