To answer your question about the social interactions, I want to first clarify that MHS is a single player game; so traditional social interactions would all take place outside of the game (class discussions, student coaches, extension activities, etc.), and we do have some evidence of those interactions. However, more specifically to what we mean, one of our goals is for players not to see science just as a theoretical subject to study, but also as a social practice, critical for a successful community as well. To do that, we had each of our NPCs represent a specific social faction (environmentalist, capitalists, etc.), and tried to use that as a lens into how others may see or benefit from the success the player is having in game.

As far as just in time instruction, in MHS that is primarily communicated through gameplay interactions. So for example if the student believes a certain branch of the river is polluted; they can toss a sensor in the water, and immediately see it pour out red if the water is contaminated or green if it is clean. Then to help illuminate some of the areas we've found to be less intuitive to players, both in terms of curriculum or gameplay, we have dialogue delivered through an AI dog NPC named ARF, which the player receives early in the game. Acting as a Navi, or Cortana type pedagogical agent, ARF is both always there and always able to provide a correction or hint when needed. Determining what is needed and when "just in time" is, are the more difficult challenges. In order to do that, we test with new players as often as we can, and also collect a lot of log data which our Learning Analytics team then crunches to help us answer both curricular and gameplay design questions.

There have certainly been a ton of challenges along the way. Currently we're focused on distributing the game; so optimizing the game to determine things like minimum system requirements, and looking towards porting the game to tablet platforms are some of the big technical challenges we're tackling. Then as I mentioned above, getting the NPCs to be really engaging for the players is important to us. For this we're currently trying to improve their facial animations, and add voice overs for all of the dialogue. We feel like these improvements will add a lot to the player experience and hopefully contribute to their immersion and interest in the game.

Thanks for the positive feedback :)

Scott,

I'll add on a bit to what Joe described in challenges. I'd say a major challenge is to examine our initial set of data and identify areas for improving the learning outcomes. As a team we did NOT want to build a learning game that was not fun so we made some choices about interactions, the story line, how to handle failures, etc.....that I think we as a team are very happy with. However, when we look at the data we see areas where we think kids are guessing rather than problem solving, where kids may be just trying to get through a task rather than mastering it.....of course there are lots of students doing lots of different things, but to some extent we can see fairly large numbers making poor choices (from our vantage point). We can now go back and make adjustments, maybe do some renovation, etc....  As a designer that is very exciting and one of the really attractive aspects of building games for learning....improvement now and into the future as the analytics tell more stories.

In our last Field Test we scheduled 10 science class periods for the game and the pre/post testing (about 8 hours total). The game is designed as a curricular replacement for an earth science: water systems unit, which also teaches scientific argumentation. 

The argumentation interface can be seen at the 0:43 and the 1:30 times in the video. Players are limited to sentences which they can select by dragging them into their argument. However, the driving questions, which are shown at the top of the interface, are always critical narrative decisions with major outcomes for the newly established colony. The Blue circles represent claim statements, which will be one of the other NPCs answer to the driving questions. The layer then goes out into the environment and collects evidence by interacting with the world, which populate the green circles in the video. The red circles represent reasoning statements, which are provided for the player. All of this is summarized under the driving question at the top of the interface in a paragraph form. 

The video mentions one student in particular, but we see a student like that in almost every class :) 

I've heard a few other serious game designers discuss the "Game of School", although I'm unsure if it's been researched or published. Basically some students get used to doing well in school, some get used to doing mediocre, and some get used to struggling. Once we change up the rules or modes of interaction for that game, which happens in different class settings all the time (e.g. different subjects, labs vs. tests, etc.); then we see different students become the experts and find solutions fastest, in turn becoming leaders, sometimes for the first time, for their peers. If I had to guess; I'd say it has something to do with, self-efficacy, interest, and experience, but it would be very interesting to study this phenomena in particular to find out. 

Yep, I agree...and I've used that idea...that some students understand the 'game of school' and so when we change the game, they become lost...

cool idea to think about finding a way to study that!

wow, Jim! That is definitely something cool...i wonder if there is a way to quantify that pre/post...it'd make a very cool pub

I'd love to hear the team hypothesize about what factors in the game might be producing these social changes. Is is just using a game in the first place, or is there something specific about this game particularly? If so, what do you think it might be?