An Educational Game for Teaching Search Algorithms

Foteini Grivokostopoulou, Isidoros Perikos and Ioannis Hatzilygeroudis

Department of Computer Engineering and Informatics, University of Patras, Patras, Greece

Keywords: Educational Game, Gabe-based Learning, Search Algorithms, Algorithm Visualization, Pacman,

Game Learning Efficiency.

Abstract: Search algorithms constitute an important topic in the Artificial Intelligence curriculum and are

acknowledged by most tutors to be a hard and complex domain for teachers to teach and students to deeply

understand. In this paper, we present an educational computer game, designed to teach search algorithms,

based on the popular Pacman game. The purpose of the educational Pacman game is to assist students to

understand the artificial intelligence topic of search algorithms in an entertaining, interactive and motivating

way. During their experience with the game, students can examine the behaviour of various search

algorithms and a graphical annotated depiction of them through suitable visualizations. Visualizations can

demonstrate the operational functionality of algorithms and are designed in line with the principles of

student’s active learning. Various learning activities were designed and request students to apply specific

search algorithms in various example cases with or without the assistance and feedback of the game. An

evaluation study was conducted in real classroom conditions and revealed quite satisfactory results. The

results indicate that the educational Pacman game is an effective way to enhance students’ engagement and

help them to deeper understand the AI search algorithms.

1 INTRODUCTION

Over the past few years, special attention has been

focused on the integration of digital technologies

and games in education and there is an increased

interest in the utilization of games as educational

instruments to assist students’ learning and teachers’

teaching procedures (Wu et al., 2012). More and

more educational systems and serious games are

developed and utilized by teachers in the context of

their courses. The rapid advancement of web and the

technological spread of devices like tablets,

smartphones and laptops have greatly facilitated the

integration of digital games in educational

procedures.

Recently, the development and integration of

computer games and game based learning

approaches in educational procedures has become a

significant focus of attention and has attracted the

interest of tutors, educational institutes and

researchers. In an effort to attract and engage

students and enhance the overall efficiency of

learning procedures, digital technologies and

computer educational games are examined to add

fun factors and make teaching more attractive and

appealing (Mihail et al., 2013). In our days, a great

part of the students, like most individuals, spend a

large part of their free time playing computer games

and in this line, the integration of games into courses

curriculum could increase students’ interest and

stimulation and provide opportunities for learning in

an entertaining way.

Computer games can be used to teach almost

every area of computer science and researchers point

out that they could constitute an effective way to

provide more interesting learning environments for

knowledge acquisition and construction (Sung and

Hwang, 2013). Computer games have been used

successfully in both introductory computer science

courses (Parberry et al. 2005, Bayliss 2007) and

general artificial intelligence classes (Wong et al.,

2010; Taylor, 2011; Sosnowski et al., 2013; DeNero

and Klein, 2010) to scaffold learning and bring

excitement and enthusiasm among students. Indeed,

through games students are given motives to

increase their interest and teachers to implement the

learning by doing or by participating principle

(Papastergiou, 2009). So, the students have a

framework for better grasping or understanding

computer science and artificial intelligence concepts.

In an Artificial Intelligence (AI) curriculum, a

fundamental topic is the domain of search

Grivokostopoulou, F., Perikos, I. and Hatzilygeroudis, I.

An Educational Game for Teaching Search Algorithms.

In Proceedings of the 8th International Conference on Computer Supported Education (CSEDU 2016) - Volume 2, pages 129-136

ISBN: 978-989-758-179-3

129

algorithms and it is a part of almost any introductory

artificial intelligence and computer science course

and textbook (Russell and Norvig, 2003). It is vital

for students to get a strong understanding of the way

search algorithms work and also of their application

to various problems. However search algorithms,

including blind and heuristic search algorithms, are

considered to be challenging for the tutors to

effectively teach and students to deeply understand.

Specifically, from a tutors’ perspective, teaching and

explaining how search algorithms operate is

challenging and in general requires a lot of

explanations, illustrations and teaching aids other

than blackboard to assist students in understanding

algorithms better (Baecker, 1998). In addition, from

a learners’ perspective, the algorithms constitute a

very challenging task for students to deeply

comprehend as they usually model complicated

concepts and also refer to abstract mathematical

notions (Shabanah et al., 2010). When students learn

new abstract concepts such as algorithm heuristics, it

can be hard for them without appropriate connection

to concrete examples (Ma et al., 2014). Visualization

of their functionality and the interactive application

in various exercises and learning activities can help

students connect abstract concepts and procedures to

concrete experiences and examples. Furthermore,

learners can recognize and comprehend virtual

graphical representations faster and deeper than

textual instructions and static representations

(Shabanah et al., 2010).

In this paper, we present an educational

computer game that is based on the famous Pacman

game. The purpose of the educational Pacman game

is to assist students to understand the artificial

intelligence search algorithms in an entertaining,

interactive and motivating way. During their

interaction with the game, students can see the

behavior of search algorithms and graphical

annotated depictions of them through algorithm

visualizations. Visualizations can demonstrate the

operational functionality of algorithms and are

designed in line with the principles of student’s

active learning. Various learning activities were

designed that instruct students to apply specific

search algorithms in various example cases with or

without the assistance and feedback of the game. An

extended evaluation study was conducted in real

classroom conditions and revealed quite satisfactory

results. The results indicate that the educational

Pacman game is an effective way to enhance

students’ motivation and help them to deeper

understand the AI search algorithms. In addition, the

game can benefit students' learning motivation and

also assist them get a deeper understanding of search

algorithm functionality.

The structure of the rest of this paper is as

follows: In Section 2, related work on the utilization

of games in teaching concepts of Artificial

Intelligence and Computer Science are presented. In

Section 3, we present a game to assist students in

learning search algorithms. In Section 4, the

evaluation study conducted and the results collected

are presented. Finally, Section 5 concludes the paper

and provides directions for future work.

2 GAMES FOR TEACHING AI

In recent years, the design and integration of games

in education and courses curriculum have attracted

the attention of researchers. A detailed and complete

overview of approaches can be found in (Michael

and Chen, 2005; Connolly et al., 2012; De Gloria et

al., 2014; Gibson and Bell, 2013).

In literature, there is great research interest and

many works study the design of educational

procedures and the development of games for

teaching the domain of computer science. In (Levitin

and Papalaskari, 2002), the authors present the using

of the puzzles in teaching design and analysis of

algorithms. In (Markov et al., 2006), authors

presents a work that uses machine learning as a

theme to unify core AI topics typically covered in

the AI course using the N-puzzle game and

provides several pedagogical possibilities for the

game. In (Sosnowski et al., 2013), authors present

SEPIA which stand for Strategy Engine for

Programming Intelligent Agents and is a game

environment for AI teaching. SEPIA is based on a

real-time strategy game, modified extensively to

preferentially support the development of artificial

agents. Another effort is presented in (Chang et al.,

2008) where a game-based learning approach is used

to help students learn graph theory topics and more

specifically Kruskal’s, Prim’s and Dijkstra’s

algorithms. The game is called Ticket to Ride and

the students through the missions that they choose

about connecting one city to another, come across

the implementation of the above algorithms.

In (Hatzilygeroudis et al., 2012), authors present

an educational game to assist students in

understanding the Constraint Satisfaction

algorithms. The game aims to offer an entertaining,

interactive and most of all motivating way to

students to experience with and learn about aspects

of constraint satisfaction problems, constraint

propagation and algorithms for constant consistency.

CSEDU 2016 - 8th International Conference on Computer Supported Education

130

Robocode (Hartness, 2004) is a programming Java

game where the goal is to develop a robot battle tank

to battle against other tanks programmed by other

players. It is designed to help students to learn Java

programming and used into an artificial intelligence

class to provide students tools for developing

practical versions of algorithms. Moreover, the robot

battles are running in real-time and are suitable to all

kind of programmers from beginners to experts. In

the work presented in (Eagle and Barnes, 2008),

authors introduce the Wu’s Castle game that is a two

dimensional role playing game teaching loops and

arrays in an interactive, visual way. The game

provides to the students immediate feedback and

helps them visualize the execution of their code in a

safe environment.

3 TEACHING AI ALGORITHMS

VIA A GAME

3.1 Learning Objectives

The learning objective of the game has an

educational goal and it aims to assist and to motivate

students to learn and deeper understand AI search

Algorithm. In our game, several learning activities

were designed to offer students various opportunities

to study and examine the way that search algorithms

operate and the learning activities are based on the

revision of bloom taxonomy.

The Bloom Taxonomy (Bloom, 1956) is a

classification of different levels of cognitive learning

objectives that tutors can set for students. It is an

important instrument in designing teaching

procedures that can provide a detailed understanding

of the learning objectives and can also help to design

activities based on the learning goals. For the

learning objectives of the game, the Bloom's

Revised Taxonomy (Anderson et al., 2001) was

utilized. It describes six progressive levels of

learning, which are starting from the foundation

towards the pinnacle and are the following:

Remembering, Understanding, Applying, Analyzing,

Evaluating, Creating.

Initially, regarding the theoretical aspects of the

search algorithms domain, the game can present

basic background topics and the description of the

algorithms. Specifically, it presents the basic textual

description of algorithms and their corresponding

graphical flowchart along with their pseudocode.

The aim is to assist students in studying and

constructing their knowledge. The playing process

mainly requires of students to apply their knowledge

of the algorithms in specific scenarios in our

educational Pacman game. During the game playing,

students need to analyze a specific algorithm

selected to study into its main sub-steps and specify

correctly the next moves. Also, the students can

select a specific algorithm to study and the

theoretical topics are presented to illustrate the way

the algorithm operates. Finally, evaluating the

students engage in checking and critiquing the

incorrect selection and it helps the student think

about why they have made an incorrect choice.

3.2 Design of the Game

The design of the game was based on the popular

Pacman game and was developed by our university

team using Java programming language. The game

is a one player game that the students can study,

examine and implement the blind and heuristic

search algorithms in various maze scenarios. The

game consists of two main modes that are the

"Educational mode" and the “Playing mode” and the

starting menu of the game is illustrated in Figure 1.

Figure 1: The main game modes of the game.

In the educational mode, the student can select

the type of the algorithm to study and the game can

present textual description of the algorithm and the

graphical flowchart along with its pseudocode. The

game offers the student the opportunity to study the

algorithm via visualizations and in this approach the

game illustrates the functionality of the algorithm in

example mazes of Pacman. Thus, students can study

the theoretical aspects of an algorithm in line with

appropriate explanations and algorithm

visualizations on various Pacman mazes.

Algorithm visualizations and animations are well

pointed to assist students in learning algorithms

(Hundhausen et al., 2002). Indeed, the

visualizations, when used properly in a learning

process, can help a student deeper understand the

way that an algorithm operates, by demonstrating

how it works and how it makes proper decisions

An Educational Game for Teaching Search Algorithms

131

based on parameters, such as heuristic and cost

functions (Hansen et al., 2002; Naps et al., 2002). In

the game, during the visualization of an algorithm,

every decision that the algorithm makes, such as for

example which node(s) to expand/visit, is properly

presented and explained to the student. The Pacman

game explains how a decision was made by the

algorithm and how the values of the parameters,

such as the heuristic and the cost functions (if any),

were calculated for each algorithm’s step. Although

visualizations are utilized in various systems and

games, they are in most cases integrated without

aspects and opportunities to engage and interact with

students during the animation process (Shabanah et

al., 2010). In the game, a noticeable aspect of

algorithm visualizations is that they have been

developed according to the essence of student active

learning. More specifically, the visualizations have

been designed based on the principle of engaging the

student as much as possible in the demonstration

process and making student to think hard at every

step of the algorithm’s animation. The principles of

active learning maintain that the more the users

directly manipulate and act upon the learning

material, the higher the mental efforts and

psychological involvement and therefore the better

the learning outcome ( Lee and Rößling, 2010).

In this spirit, during an animation demonstrating

the implementation of an algorithm in a maze, the

game and the Pacman can stop at a random step of

the algorithm and ask the student to specify some

aspects regarding the operation of the algorithm. The

animation may engage the student and request from

him/her to specify the next grid position on the maze

to be visited or ask him/her to justify why a

movement was made. In general, such justifications

mainly concern either the last action (or actions)

conducted by the Pacman simulating the algorithm

or the specification and proper justification of the

next action to be conducted. The interaction with the

student and the questions asked are either interactive

questions or multiple choice questions. The

interactive exercises may require of the student to

interact with Pacman in the maze and specify the

next movement based on the algorithm’s step. For

example, during visualization the Pacman can pause

and ask the student to specify the next algorithm’s

step by selecting the proper next grid position. In

case of a correct student’s answer, it can also request

from student to justify the reason, by offering

additional multiple choice question(s). In case of an

erroneous answer, knowledge of correct response

and proper explanations are immediately offered to

the student. After an interaction with the learner, the

animation process continues. In this line, during an

algorithm’s visualization in an example exercise

scenario, multiple interactions with the learner can

be made.

As an example, consider the simple case depicted

in Figure 2 where the visualization demonstrates the

operation of the breadth search algorithm in a case

where the aim is for Pacman to reach the cherries in

the maze starting from the default position. The

visualization has paused just after the first

movement of the algorithm and asks student to

specify the grid to be visited by Pacman based on

the algorithm. The student can either click on the

grid or move the Pacman with the keyboard.

Figure 2: An example visualization.

Additionally, the game provides students the

capability to select the type of algorithm and then to

see the application of it on the maze with the

additional explanations and information at the steps

of the algorithms such as heuristic functions, cost

calculated and other. For example, Figure 3 presents

the application of Depth-First algorithm as

implemented by the Pacman in order to get the

power-up in the bottom-left corner starting from the

default starting position. The corresponding tree

representation is presented in an expletory window.

Figure 3: Depth-First Algorithm in PacMan.

CSEDU 2016 - 8th International Conference on Computer Supported Education

132

Figure 4: Experimental procedure of the Pacman Game.

The graph represents the maze and the way that

the Pacman moved in it. The nodes of the graph are

the tiles of the maze and the additional information

like heuristic and cost are illustrated on the graph to

help students to understand how movement

decisions were made.

Furthermore, the game provides various

demonstrations of how a ghost agent can move in a

move according to a specific algorithm. Specifically,

the game can highlight aspects of how a ghost

‘operates’ according to a specific algorithm and

moves in the maze. For example, in Figure 4, the

way that a ghost operates and is moving is

illustrated. The ghosts are generally aiming to reach

the Pacman and then the game ends. The student can

move the Pacman in the maze to collect the dots and

evade the ghosts and can observe, while Pacman is

moving, the way that the ghost applies a specific

algorithm and how is examining the maze tiles and it

decides to move. In this spirit, in every state the

maze’s tiles that are expanded and are in the open

list of the algorithm, are highlighted with purple

background. The green line between the Pacman and

the ghost highlights the shortest path between them.

The game illustrates in every state the distance in

tiles between the ghost and the Pacman and also the

number of tiles that the ghost and the algorithm has

explored and has expanded. In the example state

presentend in Figure 4, the ghost has explored 188

tiles, has expanded 88 tiles and the distance between

the Pacman and the ghost is 13 tiles.

The second game mode of the game, as

mentioned above, is the “playing mode”. It is

designed to be more challenging and fun oriented. In

this mode, the student has to solve predetermined

maze levels of Pacman under different conditions

and in a specific amount of time, something that can

make the playing mode more challenging and

motivating. Also, in this mode students can complete

maze levels and proceed to next ones that are more

complex and challenging in a similar manner like in

the original version of the Pacman game. The

various levels are designed in the spirit to necessitate

students to apply a specific search algorithm and

properly move the Pacman in the maze in order to

accomplish the level requirements. The level

requirements in general concern Pacman to reach in

the maze a specific fruit, a power-up and also eat

dots. In this approach, the student is requested in a

maze level, starting from a random position, to reach

the goal (e.g. a cheery or a power-up) by moving

Pacman based on the specific algorithm that the

level specifies. The student using the keyboard can

move the Pacman in the maze and specify the

direction to follow. In case of an incorrect

movement, the ghosts that are in the maze can move

faster towards to Pacman. The only case for the

student to complete the level is to correctly apply the

algorithm and properly move the Pacman in the

maze towards the goal(s). As the student proceeds,

next levels are getting more complex in terms of

maze characteristics, number of ghosts in the maze,

goals to achieve and most of all, the complexity of

the algorithm and its parameters that are requested

from student to apply.

4 EVALUATION

The purpose of the study is to examine the

efficiency, the motivation and the effectiveness of

the educational Pacman game in learning of AI

search algorithms. We conducted an evaluation

study for the educational Pacman game during the

An Educational Game for Teaching Search Algorithms

133

Artificial Intelligence course at our department. The

participants of this study were 38 students (male and

female) from those enrolled in the Artificial

Intelligence course. Initially, all the students were

randomly divided into two groups of 19 students

each, namely experimental group (EG) and control

group (CG). The two groups consist of almost the

same number of girls and boys. Also, we used a pre-

test, a post-test study and a questionnaire survey for

measuring the learning effectiveness, the learning

attitudes and the motivation from using the Pacman

game in the context of Artificial intelligence course

in our department. The experimental procedure used

to evaluate the game is illustrated in Figure 5.

Figure 5: Experimental procedure of the Pacman Game.

All the students took a pre-test to evaluate the

prior knowledge on AI search algorithms. The pre-

test aimed to ensure that the groups had equivalent

prior knowledge on AI search Algorithms. The

pretest consisted of twelve questions on the AI

algorithms and the duration of the pretest was 45

minutes. Then, all the pretests were marked by an

expert-tutor and the score of the test ranged from 0

to 10 points.

After that, the experimental group interacted

with the Pacman game for two weeks and the control

group was selected to study the lectures and discuss

with the teacher. Then, all the students took a post-

test. The purpose of the post-test was to evaluate

the learning performance and achievement of the

students after the participation in the learning

activities. The post-test consisted of twelve exercises

of the same difficulty levels with those in pre-test

and the students were given 45 minutes to complete

the test and submit their answers. After the learning

activities the students were asked to fill in a

questionnaire and express their feelings and opinions

towards the game and assess its learning assistance.

4.1 Evaluation Results

Initially, a one-way Analysis of Variance (ANOVA)

was performed on the pretest. The means of pre-test

for Control Group (CG) and Experimental group

(EG) were 3.63 and 3.72 respectively. The results

showed no significant difference among the students

of the groups (p=0.648> 0.05, F=0.212) so it is

concluded the two groups had equivalent prior

knowledge before using the game. Then, we

conducted an Analysis of Covariance (ANCOVA) to

extract the difference between the two groups using

the pre-test scores as the covariate and the post-test

scores as dependent variables. Table 1 summarizes

the descriptive statistics for the post-test conducted.

The ANCOVA results indicate the differences in

post-test scores are statistically and significantly

different between the two groups (F=83.143

p=0.00<0.05). Finally, the results showed that the

performance of the students of experimental group,

who used the Pacman game, was better than that of

control group.

Table 1: Post test Results.

Groups N Mean SD

Control Group (CG) 19 4.605 0.698

Experimental Group (EG) 19 6.861 0.782

Then, the students of the experimental group

were asked to fill in a questionnaire including

questions for evaluating usability of the game,

stating their experience and their opinions about the

learning impact of the game in teaching search

algorithms. The questionnaire included 15 questions.

The questions Q1-Q12 were based on the Likert

scale (1: not at all, 5: very much). Questions 13-15

were open type questions and concerned strong and

weak points of the game or problems faced and also

improvements that can be made to the game.

After analyzing the students’ responses to the

questionnaire, the reliability of the questionnaire was

checked using the Cronbach’s alpha (Cronbach

1951) metric. The reliability of the scale was good

and the internal consistency coefficient was 0.87.

The questionnaire results indicate that the

students’ feeling about the game was very positive,

as summarized in Table 2. Results point out that the

All students

Control group

(CG)

Experimental group

(EG)

Traditional

Learnin

Using Educational

Pacman Game

Post-Test

Evaluating learning achievements

(Duration 45 Minutes)

uestionnaire

Pre-Test

Evaluating prior knowledge

(

Duration 45 Minutes

)

CSEDU 2016 - 8th International Conference on Computer Supported Education

134

Table 2: Questionnaire Results.

Q QUESTIONS

ANSWERS (%)

1 2 3 4 5

1 I enjoyed playing the game 0 0 0 5.3 94.7

The interface of the game is easy

to use.

0 0 0 10.5 89.5

The game made me more active

in the course

0 0 0 21.0 78.9

The game can increase my

motivation

0 0

10.

10.5 78.9

The game can enhance my

engagement in the course.

0 0 5.3 15.8 78.9

The game can enhance my

learning interest.

0 0 5.3 15.8 78.9

The using of the game for

learning is more interesting than

other ways of learning.

0 0 0 10.5 89.5

The game assisted me in learning

more effectively the search

algorithms.

0 0 0 5.3 94.7

The game assisted me in getting

a deeper understanding of the

functionality of the algorithms

after playing.

0 0 5.3 10.5 84.2

The interaction with the

visualizations of the algorithms

assisted me in understanding the

algorithm way of function.

0 0 0 10.5 89.5

Using the game provides me

with new way of thinking about

AI search algorithms

0 0 0 10.5 89.5

Will you recommend the

educational game to other

classmates and be integrated in

the course curriculum?

0 0 5.3 10.5 84.2

majority of the students greatly enjoyed studying

and playing with the game (94.7%) and a

considerable portion of them (89.5%) found the

interface of the game easy to use. Also, (78.9%) of

the students indicated that the game is more

interesting that other educational approaches and

ways of learning (89.5%). In addition, a great

portion of the students stated the game increased

their motivation (78.9%), engagement (78.9%) and

interest (78.9%) and made them more active in the

course (78.9%). Regarding the learning efficiency of

the game, students stated that the game helped them

in learning more effectively (94.7%) and in getting a

deeper understanding of the algorithms (84.2%) after

playing. In addition, the interactions with the game

during the algorithm visualization assisted students

to understand the way an algorithm function (

89.5%).

In general, the results showed that the game assisted

the students to get a deeper understanding of the AI

search algorithms and the way they operate in an

entertaining way. Moreover, approximately (89.5%)

of the students stated that the game can provide them

a new way of thinking about AI search algorithms.

Finally, the majority of students (84.2%) suggested

the game to be integrated in the course curriculum

and be used by the next year students.

5 CONCLUSIONS

The educational games can offer various

possibilities for learning. In this paper we present an

educational computer game based on Pacman that

aims to assist the students to learn the artificial

intelligence search algorithms in an entertaining,

interactive and motivating way. During the game,

students can observe the behaviour of the search

algorithms and graphical annotated depictions of

them, through algorithm visualizations.

Visualizations can demonstrate the operational

functionality of algorithms and are designed in line

with the principles of student’s active learning.

Various learning activities require of students to

apply specific search algorithms in various example

cases with or without the assistance and the

feedback of the game. An extended evaluation study

was conducted in real classroom conditions and

revealed quite promising results. The results indicate

that the educational Pacman game is an effective

way to enhance students’ motivation and help them

to deeper understand AI search algorithms.

As future work, a bigger scale evaluation will be

designed to provide a more complete insight of the

learning efficiency of the game and also evaluate

specific educational capabilities such as the

feedback and assistance offered to learners.

Furthermore, a learning analytics module will be

developed to record students learning actions while

playing and analyse them with the aim to extract

knowledge from them. Exploring this direction is a

key aspect of our future work.

REFERENCES

Anderson, L. W., Krathwohl, D. R., & Bloom, B. S.

(2001). A taxonomy for learning, teaching, and

assessing: A revision of Bloom's taxonomy of

educational objectives. Allyn & Bacon.

Baecker, R. (1998). Sorting out sorting: A case study of

software visualization for teaching computer science.

Software visualization: Programming as a multimedia

experience, 1, 369-381.

An Educational Game for Teaching Search Algorithms

135

Bayliss, J. D. (2007). The effects of games in CS1-3. In

Microsoft Academic Days Conference on Game

Development in Computer Science Education,. 59-63.

Bloom, B.S. (1956). Taxonomy of educational objectives

(Vol. 1). New York: David McKay Company.

Chang, W. C., Chiu, Y. D., & Li, M. F. (2008). Learning

Kruskal’s Algorithm, Prim’s Algorithm and Dijkstra’s

Algorithm by board game. In Advances in Web Based

Learning-ICWL 2008 (pp. 275-284). Springer Berlin

Heidelberg.

Connolly, T. M., Boyle, E. A., MacArthur, E., Hainey, T.,

& Boyle, J. M. (2012). A systematic literature review

of empirical evidence on computer games and serious

games. Computers & Education, 59(2), 661-686.

Cronbach, L. J. (1951). Coefficient alpha and the internal

structure of tests. psychometrika, 16(3), 297-334.

De Gloria, A., Bellotti, F., & Berta, R. (2014). Serious

Games for education and training. International

Journal of Serious Games, 1(1).

DeNero, J., & Klein, D. (2010). Teaching introductory

artificial intelligence with pacman, In Proc. of the

EAAI Symposium.

Eagle, M., & Barnes, T. (2008). Wu's castle: teaching

arrays and loops in a game. In ACM SIGCSE Bulletin

(Vol. 40, No. 3, pp. 245-249). ACM.

Gibson, B., Bell, T. (2013, November). Evaluation of

games for teaching computer science. In Proceedings

of the 8th Workshop in Primary and Secondary

Computing Education (pp. 51-60). ACM.

Hansen, S., Narayanan, N. H., & Hegarty, M. (2002).

Designing educationally effective algorithm

visualizations. Journal of Visual Languages &

Computing, 13(3), 291-317.

Hartness, K. (2004). Robocode: using games to teach

artificial intelligence. Journal of Computing Sciences

in Colleges, 19(4), 287-291.

Hatzilygeroudis, I., Grivokostopoulou, F., & Perikos, I.

(2012). Using game-based learning in teaching CS

algorithms. In Teaching, Assessment and Learning for

Engineering (TALE), 2012 IEEE International

Conference on (pp. H2C-9). IEEE.

Hundhausen, C. D., Douglas, S. A., & Stasko, J. T. (2002).

A meta-study of algorithm visualization effectiveness.

Journal of Visual Languages & Computing, 13(3),

259-290.

Lee, M. H., & Rößling, G. (2010). Integrating categories

of algorithm learning objective into algorithm

visualization design: a proposal. In Proceedings of the

fifteenth annual conference on Innovation and

technology in computer science education (pp. 289-

293). ACM.

Levitin, A., Papalaskari, M. A. (2002). Using puzzles in

teaching algorithms. In ACM SIGCSE Bulletin (Vol.

34, No. 1, pp. 292-296). ACM.

Ma, T., Xiao, X., Wee, W., Han, C. Y., & Zhou, X.

(2014). A 3D Virtual Learning System for STEM

Education.In Virtual, Aug-mented and Mixed Reality.

Applications of Virtual and Augmented Reality (pp.

63-72). Springer International Publishing.

Markov, Z., Russell, I., Neller, T., & Zlatareva, N. (2006,

October). Pedagogical possibilities for the N-puzzle

problem. In Frontiers in Education Conference, 36th

Annual (pp. 1-6). IEEE.

Michael, D. R., & Chen, S. L. (2005). Serious games:

Games that educate, train, and inform. Muska &

Lipman/Premier-Trade.

Mihail, R. P., Goldsmith, J., Jacobs, N., & Jaromczyk, J.

W. (2013, July). Teaching graphics for games using

Microsoft XNA. In Computer Games: AI, Animation,

Mobile, Interactive Multimedia, Educational &

Serious Games (CGAMES), 2013 18th International

Conference on (pp. 36-40). IEEE.

Naps, T. L., Rößling, G., Almstrum, V., Dann, W.,

Fleischer, R., Hundhausen, C., ... & Velázquez-

Iturbide, J. Á. (2002, June). Exploring the role of

visualization and engagement in computer science

education. In ACM SIGCSE Bulletin (Vol. 35, No. 2,

pp. 131-152). ACM.

Papastergiou, M. (2009). Digital game-based learning in

high school computer science education: Impact on

educational effectiveness and student motivation.

Computers & Education, 52(1), 1-12.

Parberry, I., Roden, T., & Kazemzadeh, M. B. (2005).

Experience with an industry-driven capstone course on

game programming. ACM SIGCSE Bulletin, 37(1), 91-

95.

Russell, S. J., & Norvig P. (2003). Artificial Intelligence

A Modern Approach. Second Edition. Prentice Hall,

Englewood Cliffs, New Jersey.

Shabanah, S. S., Chen, J. X., Wechsler, H., Carr, D., &

Wegman, E. (2010). Designing computer games to

teach algorithms. In Information Technology: New

Generations (ITNG), 2010 Seventh International

Conference on (pp. 1119-1126). IEEE.

Sosnowski, S., Ernsberger, T., Cao, F., & Ray, S. (2013).

SEPIA: A Scalable Game Environment for Artificial

Intelligence Teaching and Research. In Fourth AAAI

Symposium on Educational Advances in Artificial

Intelligence.

Sung, H. Y., & Hwang, G. J. (2013). A collaborative

game-based learning approach to improving students'

learning performance in science courses. Computers &

Education, 63, 43-51.

Taylor, M. (2011). Teaching reinforcement learning with

mario: An argument and case study. In Proceedings of

the 2011 AAAI Symposium Educational Advances in

Artificial Intelligence.

Wong, D., Zink, R., and Koenig, S. (2010). Teaching

artificial intelligence and robotics via games. In

Proceedings of the 2010 AAAI Symposium

Educational Advances in Artificial Intelligence.

Wu, W. H., Hsiao, H. C., Wu, P. L., Lin, C. H., & Huang,

S. H. (2012). Investigating the learningtheory

foundations of gamebased learning: a meta

analysis. Journal of Computer Assisted Learning,

28(3), 265-279.

CSEDU 2016 - 8th International Conference on Computer Supported Education

136