changThe interview with Dr. Maiga Chang

December, 2015

Interviewers: Soumya Narayana & Kavya Alse

Interviewer: Welcome to this IEEE TCLT interview at seventh international conference on technology for education at Warangal, India. I am Kavya Alse, currently a PhD Student at IIT Bombay. With me I have my colleague Soumya, who is also a PhD student at IIT Bombay and we have Dr. Maiga Chang with us today for the interview.

Dr. Maiga Chang is an Associate Professor in the School of Computing Information and Systems at Athabasca University, in Canada. His researches mainly focus on mobile learning and ubiquitous learning, museum e-learning, game-based learning, educational robots, learning behavior analysis, data mining, intelligent agent technology, computational intelligence in e-learning, and mobile healthcare.

He is a prolific writer and has coauthored more than 185 edited books, book chapters, journal and international conference papers. He has given 66 talks and lectures in different conferences, universities, and events. He has participated in more than 260 international conferences and workshops as a Program Committee Member. He is a member of many societies like ICM, IEEE, ACM, AAAI, INNS, and Phi Tau Phi Scholastic Honor Society.

Thank you Sir, for having accepted our invitation. During this privileged time with you, we would like to hear your thoughts about intelligence and gaming in educational research.

Dr. Maiga Chang: Thank you so much for having me here and I hope the questions will not be too difficult. Sure, you can ask anything you want.

Question: Do games inspire your research work? Could you elaborate more on that?

Interviewer: So, we have seen a lot of your games, you are an avid gamer. Do games inspire your research? How do you connect your research and your gaming?

Dr. Maiga Chang: Yeah, basically the point is if you want to do game-based learning research, on my personal point of view, you need to know what a game is. So, of course, playing games will help you to know what kind of features a game has and what kind of features can attract people to playing the game, to keep them playing the game. So that is very important, especially when you have an educational game, you always want the students to play the game and you want them to play a lot, continuously for many months, not just one hour or two hours in a week. So, yes, games definitely can help me to think about what kind of educational game we can have.

Interviewer: Features in games, non-educational games, do you think they contribute to learning in educational games?

Dr. Maiga Chang: Yes, I must say that, because in most of educational games, you will find that they miss something, some elements which commercial games or what we call video games have because probably we are too emphasized on the learning/educational part and then we sacrifice the playing part. So if the educational game is not fun, then it is not an educational game, it is what we call a serious game. Yeah, that is the idea I would like to emphasize – most old people think a serious game is, you know, serious. No, if you want your game to be playable, fun, and attractive to the students, you need to consider this game. So, almost all elements in commercial games can be used in educational game. But the problem is how you adopt that, because you need to try to change the contest. For example, we have fighting, we have fighting in the commercial game. How to transform, how to convert that fighting part to an educational game so you don’t have, you know, violence in your educational game but you still keep the competitive element for your educational game? So that would be what we need to emphasize.

Interviewer: What would you say makes an application qualify as a game?

Dr. Maiga Chang: You mean simulation?

Interviewer: Any application, like if you have a program, how would you say that this program is not a game, or this program is a game?

Dr. Maiga Chang: Okay, so, to answer this question we first need to know that previous researchers have some kind of definitions saying why a game can attract people to play. So basically they summarize many elements or we say many features or characteristics of the game which can reach that goal. So the most important part will be like a control so players can have control in the game and also we mention curiosity, so again, we need to be very special and make students have a curiosity to try to do something, try to explore the world, something like that. And, of course, we need to have fancy, you know, user interface, graphics art. That is also what our research likes of that. But that is also important. So in that case, I think the most important part will be control. You can see that when you play a game, you can always decide what you want to do. I don’t want to do it today; I can do that tomorrow. I want to play a lot of games today because this is Sunday/Saturday, so I want to play a lot of games I can do that. Or, I want to play that game but actually I don’t want to do anything; I just want to check with my friend. That’s okay. So, everything is under the user’s control, the players’ control. So, most of the time, our educational games or simulations are fixed. They have a predefined procedure and they don’t allow students to do extra things or to do different things because you need it to do this. Otherwise, I cannot respond.

But one kind of simulation could be enhanced to become a game. For example, in the industry we use simulation to teach pilots how to pilot their airplanes. Similarly, we have a simulator, a flight simulator in Microsoft XBox that is a game. So, what is the difference? The difference is, basically with XBox, I can make a flight crash, right? I can control it. Of course, in the industry, I cannot do that; if I do that, probably I won’t get my license, something like that. But they can still make a flight crash, right? It’s just that they cannot do in the training mode, in other states they can do what they want. So those kinds of simulations, they can become a game and sometimes, for example, I can fly them a little bit fast. I can see the city, that is fun, that is a fun part. So, they use simulation to teach; at the same time they enjoy the simulation. So I think control is very important.

So, as long as an application can provide students enough control ability or make students capable to control what they want to do, then it can still become a game. Of course that is just my definition. You can still say that most of educational games right now we see are games because they still provide fun, they still have very good graphic arts, so basically you don’t have to have all the elements, all the features, all the characteristic realized in your educational game but from my point of view I will say that the real educational game should also consider the control part, but of course that is very difficult to realize because you always want students to learn.

Question: What are the key decisions that you take while introducing intelligence in your interventions?

Interviewer: That leads me to this question. You say that the main element a game should have is control. So the game should know what the user is doing; the game should have some form of intelligence to it. And even if it is not a game, I have seen that all your systems will have some sort of intelligence added to it. So, how do you take decisions about what type of intelligence to add? Where to add it and why to all?

Dr. Maiga Chang: Yeah, best games usually have two parts; we apply that for intelligence. The first one is what we call NPC or the monster’s behavior. So, we gave intelligence to the monsters so they can react to the environment, they can even react to your user’s behavior. So that is the first thing. The second thing in what we can apply the intelligence is quest. In the game, there are many, many quests. For example, if you are playing a shooting game, you’ll see a lot of different quests and you need to protect some kind of case you’re transferring and/or you need to find someone and kill him, Okay, actually, I shouldn’t say that, but it is a game. So in that case, how the quest is showing to students or to the users that part can also be applied to intelligence or we say you can apply the data mining, data analytics or artificial intelligence, computational intelligence to that part. So there are two parts you can apply the AI or data mining, data analytics. The fun thing is usually that we won’t consider to apply the artificial intelligence or data mining techniques to the NPC’s behavior or monster’s behavior. Why? Because, for example, if I’m playing a game, if the monster has intelligence, what will happen? I will never beat him, right? Because it is very intelligent and can beat, you know, adapt to what you have done earlier. So, I will never win. That is not a game, that is just I’m looking for someone to beat me. I don’t need that, I need to have some situation which I can win, that perhaps has a challenge but in which I can still win. So, that part of artificial intelligence or intelligent system is a little bit more difficult because usually you cannot make up computer too smart to beat.

The second one, for the quest, that can be done because all the users, all the students, all the players, they have their own personality, they have their own preference, experience and even prior knowledge. So, in that case, we can give educational games or any kind of educational system intelligence. We can put artificial intelligence or data analytics module to the game or to the application to provide users personalized quests. Yea, so that’s it; I’m not sure if I answered your questions but that is what we want to do because if you don’t put something like intelligence or data analytics into your system, then your system is just an object filled with all the predefined rules, which means everyone, when they use the system, will find that it looks like the same one, and nowadays, we have very good computers, mobile devices and computing powers, so it is time for us to consider to have a smarter system rather that rule-based systems like before.

Question: How much of subject knowledge is required to design interventions?

Interviewer: You have developed a lot of interesting systems for education; they have spread over a wide variety of domains, like finance in the Pecunia application and then you have this landmark based system, then the museum walk system and you have one for autism as well. All of them are so different domains; so how much of domain expertise do you require for developing such systems?

Dr. Maiga Chang: Yeah, in order to develop such system, we needed to have a very good understanding about the domain knowledge, the concepts and other things. So, when you develop such a kind of system, you may need to have full knowledge about it. But, one thing we can consider is that you can develop a system which is just a kind of interface between the students and the background knowledge, the knowledge behind the thing. If we can do that, then we can always change the knowledge behind the thing and then our system can automatically provide learning activities for students for different subjects.

So, yes, no matter what kind of approach we take, we need to be well-informed about the knowledge. However, one suggestion or one comment I should make is we don’t need to create a system with complete knowledge of a particular subject. We can do it one by one, stage by stage, just like when we learn for the day 1, for the first week we learn this kind of knowledge and for the second week we learn another unit. So you can create a system step by step, by feeding more content, more knowledge into the system. So, at the end, your system can be used for the whole course but at the beginning probably your system can only be used for the first two or three units or the first two or three weeks. But that is okay. You can add more and more later on.

Question: Can you give some insights into development of systems for developing thinking skills in students.

Interviewer: Coming to the matter of different subjects, students require certain types of thinking skills to easily apply or jump from one subject to another and apply that skill in that domain. So, in order to teach such skills, how can a system be developed?

Dr. Maiga Chang: Yeah, I think that is a very difficult question because the thinking skill itself is difficult or we can say complex, complicated. So, if in that case, we cannot have a very straightforward solution for a particular thinking skill because the first thing we need to do is to analyze what sub-skills that thinking skill has and then try to transform the matter that we are going to teach in the real situation, into the system. So in order to teach students this kind of sub-skill in the real world, in the real classroom, we try to convert it to a kind of quest or a kind of learning activity in the virtual world or virtual system. And then, even if we do that, we only finish one sub-skill. If we want to teach students the whole thinking skill, we need to realize, we need to implement all the sub-skills one by one. As you know, there are so many thinking skills, which means we need to have a lot of time to develop such systems which can teach students all the thinking skills. So I will say that it is a huge and very difficult work, but is worth to do it. But of course, we also need to have a lot of help from different experts because no one is the expert of all thinking skills and the thinking skills can be applied, used, and learned in different large domains. So in that case, if you really want to do such kinds of systems which teach students or train students thinking skills, then you need to have a lot of help from different experts, different subject domain experts.

Question: What skills do researchers need to do research in Education Technology?

Interviewer: Sir, What kind of skills would researchers need to do research in educational technology?

Dr. Maiga Chang: Okay, so first of all, researchers need to have skills of doing in depth and comprehensive survey. That is the first thing. Because you don’t want to do a research and after one year find that, oh, someone has already done this.

That is not good for anyone, especially for students. See, a master’s student will only have two or three years and two years later, if someone has done this in 2010, then he or she needs to redo his or her master’s thesis. So, that is the first thing. The second thing I want to emphasize is researchers should have the ability to do cross domain or we say interdisciplinary survey because the backbone or the theory of most researchers and most of research we see in the educational technology domain, is actually coming from another domain. Not educational technology but probably psychology or even business, like teamwork or grouping strategy. Actually, the business area has a lot of theory about how to form a group, what kind of person you need to have in your group so your group can have good productivity. So, a good researcher should be able to know that the problem you are currently working on or the methodology you want to adopt, is probably not in your domain, in your area, but is probably in another area. So you need to make sure that you have good vision, not just like me having these glasses, but you need to see far away, see wide enough. And the third one is that researchers in educational technology domain need to have at least an ability to use technology. So, for example, if you want to use web 2.0 for your course or you want to use web 2.0 to create a research to create a platform or to create a framework for teaching something, of course you need to try to use web 2.0 first. There are so many different web 2.0 applications. After this, you need to try to use it, and once you use it, you have the idea about how it works and whether or not a student or a teacher who sees this kind of technology will use it or like it? If not, then you shouldn’t give up and that leads to another ability or skill. That is you need to identify and know what the equivalent technology of a particular things is. For example, say you want to use Wikipedia. But if you find Wikipedia is not good enough, what kind of equivalent technology can you also use and that is like Wikipedia. So those are the four things and the fifth… (sorry, there are too many things a researcher in educational technology should have) The fifth thing I think is they need to have at least an idea of how an educational technology or how an educational system was implemented. For example, when we use second life as virtual world, at least you need to know that second life is actually created by some kind of things and it can import some kind of 3D object, and this kind of 3D object can be created by what kind of application. You don’t need to be able to implement the system but at least you need to know the foundation, how this system is built. Okay, last. You see, the sixth and last thing is researchers in educational technology need to be able to collaborate with others because educational technology, I think in most of research nowadays, cannot be done by one person. They need to be conducted by many different people who have different abilities or skills. So, educational technology researchers are probably very good in programming, in doing surveys, in modelling, so they need to collaborate with others. Don’t just lock your room and do research in your own room. That is not good. There are so many things in educational technology researchers should have.

Question: What has been the most surprising/unexpected result that you have observed so far in your research?

Interviewer: As a researcher, what was the most surprising or unexpected result that you got in your experiments?

Dr. Maiga Chang: Actually, not so much because when you do research, you usually will have your research model created and based on the research model you will have a hypothesis. So, of course, when we do research, we always want to see that our research result is in line with our hypothesis or supports our hypothesis. So most of the time, we do have that happened, but still there will be some surprises. But usually when you see the surprising results or unexpected results, you need to know that there are so many factors that may make this thing happen. So, that is not bad thing, you need to know why – what is the possible reason to see this kind of result. So, in our case there are two findings that are really surprising. The first one is when we do the context-aware mobile game, and of course our game is context-aware so we want to make sure that this context-aware feature that is part of the game does make students feel this game is good for them to use or good for them to learn. Surprisingly, they don’t think so. On the other hand, the context-aware mobile educational game we have also has another feature called a story, so we want to use a story to make them feel like they are playing a kind of role in the virtual world. And yes, they think a story is good. So, we are wondering why context-aware is not good? And after that, we found that when we do our context-aware experiment, the context-aware is not real context-aware. What does that mean? When we do the experiment, we try to simulate the whole world in the school campus, which means actually the school campus is not the real world. And then we are trying to say, okay this building is North America, this building is Europe. Students say, no, this is not North America, this is building 1, and this is building 2. So what we think is context-aware, students don’t think it to be context-aware, although our application, because we have some kind of coordinated settings, thinks you are in North America. However, students say: oh no, this is building 2. Okay, so that is why it is surprising. The second unexpected result is when we do the game-based learning. Usually we think of female students as being not very good in playing game and that they don’t like games. They have their attitude and it will be different from the male students. And we found that is not 100% true. Female students just aren’t so comfortable in playing games or in winning games. They don’t have confidence to win the game, something like that. But, on the other hand, when we say, “Do you want to use educational game for learning?” we found that, actually, the difference between female and male students is in the nature of their attitude, which means do they really think a game is useful for learning? If not, if a person says no, then of course they will not want to use the game. Okay, that is a very important thing. So, although this is not our expected answers, however, because of this finding, we found that it is not a problem. So you can create educational games for the students as long as students think playing the games may help them for learning. Here, I need to emphasize one thing. Never treat or never have that kind of thinking saying okay I want to create a game for everyone or I want to create a system for everyone. Because there are some students that don’t like games, or some students whose learning styles don’t allow them to have good feelings for using games. For example, some students want to read the hard copy of the book; they don’t want to use any kind of learning. Or they say watching a display will cause them to have eye problem. So I won’t say that. So in that case, of course they don’t like games. So I always say game-based learning is just an option for students who don’t like other ways of learning; they can have a choice, that’s all. So those kinds of unexpected findings make us say yes, that’s true, that is correct because, if you’re naturally born not liking games, then of course nothing will help. So, that is unexpected. We have two unexpected finding, but of course we all find the answer or reason why they are unexpected.

Interviewer: What was the most challenging project that you have handled so far?

Dr. Maiga Chang: Okay, the most challenging project at this moment is mega world. It is multi-player educational game for all platforms, because we want to create a game for everyone, for all subjects as you can see, or for everyone. That is a contradiction to what I just said. So we want to build such a kind of platform which means the platform needs to be very complete, very detailed, everything needs to be considered. So, that is one challenge. Of course, we have progressed, and we are trying to move on from time to time but that one is very, very challenging.

Interviewer: Thank you for your time and for your valuable solutions and for all the answers that you gave.

Dr. Maiga Chang: No problem, it’s my pleasure.