Game thinking from Adam Clare

Tag: EEG

Emotion Sensing For Future Games

I’ve explored mindreading to enhance a game before and I like to stay up to date on what’s going on. This year we should be seeing more and more games using consumer EEG machines or other wearable technology that allows us to get a glimpse of what’s happening in people’s minds.

The Muse headband continues to be the best looking device, but on the experimental side there are some nifty new products.

This mindreading helmet records your stress level while you’re engaged in helmet-wearining activity like riding a bike. On it’s own, it’s not that novel since one can do the same measurements with existing head sensors. The useful part of MindRider is that it can be used to collect data on mass while not be an inconvenience to wear since you’re already wearing a helmet. FastCompany adds this informative bit of knowledge into the mix:

Cyclists use the feedback in different ways. “Most of our avid commuters are most interested in the mindfulness or relaxation aspect of MindRider,” says Ducao. “New cyclists are most interested how the high focus aspect–the red part of the spectrum–can help them know where to be more cautious.”

But the data may be most useful as it’s aggregated. Anyone with the helmet can opt to share it anonymously online, so everyone’s experience can be merged in an up-to-the-minute map showing exactly how a particular route will make you feel.

Sadly, their recent (and ambitious!) Kickstarter failed.

Don’t give up hope though!

While not a mindreader like the tools above, this experimental controller can sense emotions.

McCall added a 3-D printed plastic module packed with sensors to an Xbox 360 controller. Small metal pads on the controller’s surface measure the user’s heart rate, blood flow, the rate of breath, and how deeply the user is breathing. A light-operated sensor gives a second heart rate measurement, and accelerometers measure how frantically the person is shaking the controller.

Meanwhile, custom-built software gauges the intensity of the game, in this case, a simple but fast-paced racing game in which the player must drive over colored tiles in a particular sequence.

McCall can then compare all this data to generate an overall picture of the player’s level of mental engagement, which can be used to alter the pace of gameplay to better suit the player.

Obviously, these new tools can be used for play testing but it’ll be far more interesting to see how people can take these tools to create new meaningful gaming experiences.

When we combine the tools above with mind trickery we can really create some bizarre stuff!

Mind Reading, Horror, and TOJam

Last weekend myself and a great team of game designers set out to make a game which uses basic EEG readings as a mechanic. The main source of inspiration of the game idea came from Doctor Who’s weeping angels. The gist of the angels is that if you look at them, they will get you when you look away. Don’t even blink.

Here’s what I sent out to the team prior to the jam so that we’ll all be on the same page:

So in sum, here’s the basics of the game:

  • Single player (1st person perspective)
  • Escape a procedurally generated room. Over and over again; like Groundhog Day but meant to scare.
  • Gameplay: Player starts with the camera locked on the statue then they have to navigate to the doorway to escape the statue without physically blinking. If the player blinks then the statue teleports to right in front of them, they are granted only three blinks. Too much blinking = death.
  • Art: all we need is furniture and a statue. I think dim lighting will allow us to get away with lame textures and stuff.

After that email one of the team members suggested we make the game playable without monitoring physical blinks, so we’d have a version everyone can play and a “hardcore” version. The core mechanics then should be similar in concept to how Slender Man games function.

In short, we didn’t finish the game but we did make something playable. We blinked.

Here’s what happened:

First some context, TOJam is a fantastic annual game jam held in Toronto and it gets bigger and better every year. The event has hundreds of people all descend on George Brown College to make games starting on Friday and having something playable by Sunday evening.

Reading minds

We wanted to use an entry-level EEG device the MindWave because I have previously seen it used for other games built on Unity. It was more complicated than anticipated to use the MindWave with Unity. We spent a good chunk of the first day trying to get the device to work.

This is after some time spent on Thursday trying by myself to connect Unity and MindWave. The image below shows that the device is connected, the connection software (ThinkGear) knows it there, but the application claims the device is not there. I didn’t know if I should’ve directed my anger at MindWave or Bluetooth. Overall, this was frustrating and I was hoping that the talented programers would know where I went wrong.

MindWave, Bluetooth, sigh.

After some struggle we figured out how to connect the MindWave and rewarded with this screen (enlarge to see what the MindWave records):
A connection is made

Fantastic! Now we can really make this game interesting.:)

Then we discovered that the blink mechanic we wanted to use in the game was actually hard to read with the MindWave. They have a blink detection system but it’s not clear and requires some educated guessing. A lot of this effort could have been avoided if the company behind MindWave released some Unity project files and demos (they have one but the download link was broken).

Even though we successfully had Unity and the MindWave talking to one another we decided to put it aside because of Bluetooth chaos. It was time to make the rest of the game.

The room

For the atmospehre of the game we originally envisioned something that seemed OK, but then got creepy like a museum after opening hours. When looking for reference pictures, I chanced across a great series of photos from the decaying Prince Edward Hotel in Brandon Manitoba (of all places) and it looked perfect.


Saturday saw the game go from primarily creepy to something more typically found in the horror genre: the abandoned building. It wasn’t a big change and was decided upon before most of the modelling was started.

Getting the room to procedural generate wasn’t too complicated and we got that working. As the code was being written we had a statue constructed and then built all the furniture that will be placed around the room. Things were going smoothly. We had all the pieces of the game being built separately and by the end of Saturday it looked like all we had to do was put the pieces together.

It looked like we’d even have time to get the blink working. Even if couldn’t get blink functioning we had other mind-reading functions we could incorporate.

Here’s an early screenshot from the game:
BL_NK early screen

It’s so dark and gloomy to make the game hard without the MindWave.

It’s worth mentioning that we had off-site support for our audio and the fellow behind it made some insanely freighting sounds. The audio was done on time and, oddly, was the last thing we added to the game (just because it’s so easy).

The final day

After a late start, we got all the code from six separate computers on to one. We didn’t use Dropbox at the behest of the organizers, but in retrospect we should have broken that rule. Then we ensured that all the pieces would fit together.

I know it says 8:04, the Windows build was first.

I know it says 8:04, the Windows build was first.

This last days was the most hectic as it now relied mainly on the back of one of the programmers. We got the most important parts of the game working and some of the room elements working as well. We didn’t bother putting in all the assets we created as with what little time we had left we opted for function over form.

With that in mind, we decided not to incorporate the physical blink detection at all. It came down to a functioning game or a game that would probably break during the evening play sessions. One programmer went so far as to figure out some more of the MindWave but, due to too much going on, we didn’t incorporate it.

We figured we can easily add in the rest of the assets afterwards.

Around us other teams were play testing and celebrating their complete games. We, on the other hand, didn’t even have a chance to play test. TOJam game-building ends at 8pm and at 7:50 we tried our first complete build…and it worked!

This was particularly exciting for me as it was the first time at TOJam that I was part of a playable game by the end of the weekend. Of course, other games there were far better than ours, I was still pleased.

Next steps:

We will continue to work on the game to make it more playable and a better experience. By better I mean scarier. With luck, this will be playable for the TOJam Arcade for everyone to play.

Some stuff is obvious from what we ran out of time for:

  • More assets for the room and for the player to bump into.
  • Audio to accompany all the objects that move.
  • MindWave integration.
  • More early rooms to train the player before they get to the procedureally-generated rooms.

From play testing at the jam I noticed we need:

  • A way to end the madness as both “win” and “fail” states are the same – you end up in a new room. This was always intended but we never gave the player a stop button.
  • Better training levels to cue the player as to what’s going on.
  • The exit needs to be more visible.
  • We need a better way for the player to figure out where the statue initially is.

A BIG thank you to all of team Oh My Glob!

And thanks to all the wonderful people who made TOJam happen from the organizers to the sponsors to the volunteers! Without all their combined efforts we never would’ve had this game this far along.

In retrospect, I don’t know why I felt the need to write all this down. Thanks for reading!

Powered by WordPress & Theme by Anders Norén

%d bloggers like this: