Space Fortress
Aditya Prasad
I can't seem to find the game online though. Has anyone else come across it?
Curtis Warren
http://www.cogsci.rpi.edu/cogworks/?view=modules.research.spec&id=74
It doesn't seem like it's available online, though.
Aditya Prasad
Thanks, Curtis.
It's my understanding that the game has been around for many years, and that this team is just using it to test some new hypothesis. Maybe I'll check with them if the version they use is available more widely.
Aditya Prasad
Great news!
Dr. Destefano supplied me with this link:
http://linserver.cs.tamu.edu/~mmiller/RSF/
Anyone want to compile a list of references showing the benefits of this game? :)
Paul
Paul
The game is similar to the old Atari game Asteroids (navigate a ship on a frictionless zero-gravity 2d plane) with a few added twists and constraints.
I'm thinking of starting a new Python/pyglet project to create an easy-to-use cross-platform version of this game. :) Given the volume of research on Space Fortress, I'm surprised this hasn't been done already.
Martin Syk
It would be greatly appreicated I'm sure.
Martin
Tervitused
it was made to be pretty much platform independent. We'll see how it
goes.
Paul
that part is Windows specific (joystick.dll). I haven't found any way
of disabling the joystick and using keyboard controls instead but
there might be something hidden in the source.
Apparently Space Fortress was used to train navy air cadets for years
because they found that 10 hours of training improved actual flight
performance by 30%. They compared it with an ultra-realistic flight
simulator which was not successful at improving performance.
Check out this interview:
http://www.sharpbrains.com/blog/2007/02/08/brain-training-with-cognitive-simulations/
From the article:
- Cognitive performance can be substantially improved with proper training.
- It is not rigidly constrained by innate, fixed abilities.
- Cognitive task analysis enables us to extract major cognitive skills
involved in any task.
- Attention control and attention allocation strategies are a critical
determinants in performing at top level in complex, real-time
decision-making environments
- Those skills, and other associated, can be improved through training
- Research shows that stand-alone, inexpensive, PC-based training is
effective to transfer and generalize performance.
- The key for success is to ensure Cognitive fidelity, this is, that
the cognitive demands in training resemble those of the real life
task.
This opens up a whole new realm of possibility for cognitive training
software for any situation. The hard part is using Cognitive Task
Analysis to distill the optimal training strategy.
For any task requiring "constant shifts of attention, memory
retrievals, visual tracking, fine motor control, and dynamic decision
making", Space Fortress is the ideal training environment.
Other cognitive training software could be written for other
applications, for example IntelliGym (http://www.intelligym.com) for
basketball players. The possibilities are endless.
After I get my Python version of Space Fortress up and running, I'll
try making the cognitive tasks customizable, for example adding an
N-Back task concurrent with the Space Fortress game.
Paul
Tervitused
code.
I guess I'll stick to Asteroids. ;)
On Dec 2, 8:53 pm, Paul <plh...@gmail.com> wrote:
> Tervitused, you might have trouble with the joystick code - I think
> that part is Windows specific (joystick.dll). I haven't found any way
> of disabling the joystick and using keyboard controls instead but
> there might be something hidden in the source.
>
> Apparently Space Fortress was used to train navy air cadets for years
> because they found that 10 hours of training improved actual flight
> performance by 30%. They compared it with an ultra-realistic flight
> simulator which was not successful at improving performance.
>
karnautrahl
least the installer is! I can't get it to install on Vista.
Fingers crossed for a version that works well on Vista :).
>
> - Show quoted text -
dualnback
is your timeline for adding the other modes and changes discussed
before to BW? Whenever you do is fine, I just wanted to have an idea
of when.
On Dec 2, 9:53 pm, Paul <plh...@gmail.com> wrote:
> Tervitused, you might have trouble with the joystick code - I think
> that part is Windows specific (joystick.dll). I haven't found any way
> of disabling the joystick and using keyboard controls instead but
> there might be something hidden in the source.
>
> Apparently Space Fortress was used to train navy air cadets for years
> because they found that 10 hours of training improved actual flight
> performance by 30%. They compared it with an ultra-realistic flight
> simulator which was not successful at improving performance.
>
Paul
studying - I have 4 final exams in 4 days!
Paul
then froze up at the end, but by hitting ctrl-alt-del and ending the
installer tasks I was able to keep the files.
Here's a "quick" instruction guide:
Installation:
1. Go to the RSF directory
2. edit RSF\client\serverconf.txt and set "separate SimServer=no"
3. Open Start -> Control Panel -> Administrative Tools -> Data Sources (ODBC)
4. Add a user data source with driver 'Microsoft Access Driver
(*.mdb)' and data source name 'sf' .
5. Select as the database the sf.mdb file located in RSF\Data\sf.mdb
To launch the game:
1. Launch the server using RSF\Data\runccm.bat
2. Start Experiment -> AMEC (This one consists of the full game, I
haven't tried many of the others))
3. Start Experiment Condition -> AMEC111
4. Launch the client using RSF\client\runpvm.bat
5. Signup, choose a username & password (only necessary the first time)
6. go back to the Central Control Module window, select "Put test user in SFEC"
7. Select the experiment & experiment condition then look for your
name in the list and move it over to the right side (this only needs
to be done once per experimental condition)
8. Go back to the client Login screen and enter your name & password
9. Select "No" at the prompt "Login more users?"
10. Read the instructions, hit Space to continue
11. Play!
12. Read the file RSF\Documentation\Installation & User
Guides\RSFSubjectInstructions.doc for a guide to playing the game
Controls are with joystick & mouse. Since I don't have a joystick I
downloaded PPJoy and configured a virtual joystick, then ran PPJoyKey
to translate key presses to joystick movements.
Lots of trouble for a simple game! You can see why I want to code a
Python version...
dualnback
On Dec 3, 6:41 pm, Paul <plh...@gmail.com> wrote:
> I may have some time next week to work on BW. For now it's all-out
> studying - I have 4 final exams in 4 days!
>
Ben
Do you have an estimated time of completion for your more user-
friendly version of Space Fortress? I don't want to pressure you or
anything, especially considering the pressure you must already be
under with school and everything. So don't take this as a demand.
Rather, I am asking for purely informational purposes, so that I can
determine whether the trouble of installing the current convoluted
crappy version of Space Fortress would be worth it.
Either way, you deserve some props, as they say.
Paul
dualnback
forum and the original one, such Music, Image Rotation N Back, etc.
(if you can).
Further, can you make it possible for us to see where we went wrong in
Arithmetic N Back, so that we can improve that particular area of
mathematical operation. That is, if I consistently get my
multiplication wrong, feedback from BW would tell me where I need to
improve the most. I am not interested feedback for position, letter,
or color, etc. I am asking about arithmetic operations in
particular. I have not done Arithmetic N Back yet, so this feature
may already be there. If not, let me know what you think. Thanks.
On Dec 5, 12:49 am, Paul <plh...@gmail.com> wrote:
> Ben, I'm hoping to work on it over the break, so I might be able to get the
> basic game running by the end of December. Maybe earlier, depending on how
> much my programming skills have been improved by BW and dual n-back! The
> hard part will be making sure it plays as similarly to the existing version
> as possible, and making sure the basic code framework is scalable so that
> the game rules can be modified and new cognitive tasks added and subtracted
> at will.
>
Paul
I'm not sure about adding feedback for the different operations in Arithmetic mode - it might be hard to do considering the internal structure of the program. i'll take a look at it next week though.
dualnback
Dual N Back and/or BW help you on your tests. :)
On Dec 5, 5:10 pm, Paul <plh...@gmail.com> wrote:
> Don't worry, I haven't forgotten about BW :)
>
> I'm not sure about adding feedback for the different operations in
> Arithmetic mode - it might be hard to do considering the internal structure
> of the program. i'll take a look at it next week though.
>
dualnback
could you also add square roots and cube roots to Arithmetic N Back?
Just a thought.
Tervitused
Tervitused
training for basketball players developed by researchers who did pilot
training software for Israeli pilots. From what I have read, I am
pretty sure Space Fortress may have originated in that training
program. One of the unique pieces of the software is a pattern
recognition format. A series of dots representing players (I suppose)
move around the screen, and you have to indicate when a pattern that
you were shown earlier appears on the screen. The idea is that you
are replicating the task of recognizing correct teammate placement for
a play. It develops the ability to "see the floor".
I don't know what else happens in the task, that may be it. But
imagine playing Asteroids and having to indicate when Asteroids have
arranged themselves in a particular spatial pattern, and you have the
general idea. Pretty cool I think. Now, to increase the difficulty
multiplier, imagine that there is no up or down, left or right, you
just get points whenever you recognize whenever the little dots are
arranged in that pattern, regardless of the orientation. Or, if you
like, right and left handed versions of the pattern... Pretty
challenging, I would thin, and different than just showing a series of
images. Instead you have to maintain vigilance over several moving
pieces until just the right moment...
Tervitused
It is does not follow every rule of Space Fortress. For example,
there is no memory requirement for neutralizing mines. Also, points
are not deducted for veering outside of a perimeter outside the
fortress. However, you can replicate the second condition merely by
trying not to fly too far (as you define it) away from the fortress.
To make up for the lack of some of the difficulty, you have to shoot
through three layers of shield, making sure not to destroy anyone of
them completely, or a new shield grows to replace it.
Anyway, just Google, "Star Castle flash game" and you should be able
to find it. Its a actually an arcade game for the late 70's.
On Jan 1, 2:06 am, blank <peisistrat...@gmail.com> wrote:
> How is Space Fortress coming along? I'm a mac user so I have been
> waiting for the supposed release. Sorry for being impatient, hope
> everything is well...