Chasing Detection Experiment

This is a evolutionary psychology experiment designed in a 2D video game form. I used pygame to develop the experiment because common tools for designing psychology experiment could not offer that much of flexibility for developing a videogame-like experiment. The experiment was developed for a scientific collaboration with Prof. Robert Biegler's lab at the department of psychology at Norwegian University of Science and Technology.

Among several moving circles, one circle chases the participant based on a variety. This vide of algorithms. The participant has to move its circle with the mouse to stay away from the chasing circle, if the it is detected. If the chasing circle arrives within a particular distance from the player's circle, the game (trial) is over.

There is a configuration file that allows you to specify the number of trials, algorithms, and all of the decisions needed for configuring the trials.

License

MIT

Installation

you need the following software packages:

• Python 2.7 32bit
• PyGame 1.9.1
• requests
• numpy

If you are a complete newbie, installing the packages required for running this experiment will keep you busy for a little while. I try to explain it "as basic as possible", assuming you are a biggener user.

First of all, you want to make sure that you install Python 32bit 2.7.12 or higher. Notice that I said 32bit, even if your machine is 64bit you still must install the 32bit Python 2.7.12 or higher. You must not install Python 3!

Next, you want to make sure that python is defined in your environment. This means that anytime you type python in your command line, python should be executed. However, if python is not in the Path, the command line will not recognize it. This is particularly a problem for Microsoft Windows users (as usual, whenever we talk about problems, Microsoft shines like a diamond!). There is two ways you can add python to the environment. The simplest way is to pay attention when you install Python 2.7. As shown in the image below, make sure add python.exe to path is selected. This option makes your life much easier but is not selected by default in the installation.

The second way is trickier. Microsoft users should open the System Properties and go to Advanced tab, and click on Environment Variables. A new windows will open, in the system variables click on Path and add the path to your Python directory which is something like:

C:\Python27

This tutorial might be helpful!. Then, also add the path to the Scripts Python directory which is something like:

C:\Python27\Scripts

Next, you will need to install a few required packages. The easiest way to install packages is using pip, which itself, needs an installation. See this tutorial about how to install pip script. Since you have previously added the Scripts directory to the path environment, now you can use the pip command to install other packages. If you have installed Python 2.7.12, it already includes pip but you should update it. To do so, run the cmd and paste the following code in it:

python -m pip install --upgrade pip

Now, you can install requests from the command line:

pip install requests

You also will need pygame. Visit http://www.pygame.org/download.shtml and download the correct version of python for your OS.

In addition, you also need numpy for doing some numeric manipulations on the trial numbers. The installation is just as before:

pip install numpy

Settings.cfg

The experiment's settings can be adjusted by updating the settings.cfg file. The file can change the following settings:

Configuration Argument	Description
screen_width	takes an integer which defines the width of the screen in pixels
screen_height	takes an integer which defines the height of the screen in pixels
FullScreen	can be True or False. If set to True it makes the game fullscreen. Otherwise, the screen will appear with the specified width and height.If you set the game to full screen without increasing the width and height, the display will stretch.
backgroundColor	takes a vector of 3 integers, separated by comma which represent an RGB values for defining the background color of the game. The default values are 100,100,100 which is light gray.
fontColor	takes a python color name which can be black, white, or any of the colors mentioned later in the documentation.
mouseCursor	can be True or False. If set to True the mouse curser appears during the trial. Otherwise it will be hidden during the trial.
showTimer	can be True or False. The experiment has a timer for tracking how long each trial lasts in terms of seconds. The timer can also be shown in the top-right corner of the screen if the value is set to True.
gameover_sound	The value can be True or False. When the trial ends, a buzz sound is played if the value is True.
playerSpeed	takes an integer which defines the movement speed of the player based on number of pixels. The default is 5 pixels
wolfSpeed	takes a real number and defines the movement speed of the wolf and sheep.
trialType	a vector of "0" and "1" showing the overall number of trials (counting 0 & 1) and whether a wolf exists or not (1 or 0 respectively). For example, if the experiment has 4 trials and only the first 2 trials have a wolf, the vector should be 1,1,0,0. This option only specifies whether Wolf actually exists or not. It doesn't provide any information about the wolf's strategy.
duration	a vector of number of seconds that each trial lasts. For example, 30,10, ... the integers should be separated by comma.
chaseRate	a vector of chase probability per turnRate. the integer should be between 0 to 100 and separated by comma.
chaseAngle	a vector of chase angles for each trial. the integer can be between 1 ro 180. the integers should be separated by comma. (1 is practically straightline chase, given the size of the game objects and screen)
scapeAngle	a vector of escape angles for each trial, only used when randomPlayer is True. the integer can be between 1 ro 180. the integers should be separated by comma. (1 is practically straightline chase, given the size of the game objects and screen)
killZone	the radius of the death circle, i.e. the distance between the center of the coordinates of the Wolf and the player which ends the game, where the wolf is considered to win. The minimum killZone should be 32 which is the diameter of the circles.
sheepNumber	a vector of comma separated integers that specifies the number of sheep including the wolf who appear in the game.
Radius	is an integer which defines the radius of the circle where the sheep and wolf should be located around it. The larger the radius, the more time the wolf will require to approach the player and also, the more time it takes other sheep to get close to the player. If set to anything below killZone there will be a chance of failure right at the start of the trial.
turnRate	is an integer which defines the number of seconds that it takes for sheep and the wolf to make a random turn. The default is 1 turn per second.
randomPlayer	can be True or False. If set to True the player will have a randomwalk similar to the other circles and the mouse will be disabled for moving the player during the trial.

Instructions

To run the experiment in Microsoft Windows click on RUN.bat file.

Available colors

For the fotColor you can pick any of the colors mentioned in the figure below. Write the name without any space in between, for example, redorange or midnightblue.