Experiment 1: direct replication of Berdahl et al (2013) reported in cogsci paper (note that large pre-rendered background file is needed to run this expt -- contact authors for details)
Experiment 2: single-player micro-experiments to directly test model with simulated bots
Experiment 3: multi-player experiment with single, smooth spotlight & binary reward
Note: This is an unofficial fork of MWERT, adapted to allow arbitrary numbers of players to join a single room and participate in a collective behavior task. This task was originally used by Ian Couzin's lab to investigate collective behavior in fish.
Local demo (from scratch)
Git is a popular version control and source code management system. If you're new to git, you'll need to install the latest version by following the link for Mac or Windows and downloading the first option in the list. On Mac, this will give you a set of command-line tools (restart the terminal if the git command is still not found after installation). On Windows it will give you a shell to type commands into. For Linux users, more information can be found here.
On Mac or Linux, use the Terminal to navigate to the location where you want to create your project, and enter
git clone https://github.com/hawkrobe/couzin_replication.git
at the command line to create a local copy of this repository. On Windows, run this command in the shell you installed in the previous step.
Navigate into the repository you created. You should see a file called package.json, which contains the dependencies. To install these dependencies, enter
npm installat the command line. This may take a few minutes.
Finally, to run the experiment, enter
node app.jsat the command line. You should expect to see the following message:
info - socket.io started :: Express :: Listening on port 8888
This means that you've successfully created a 'server' that can be accessed by copying and pasting
in one tab of your browser. You should see an avatar in a waiting room. To connect the other player in another tab for test purposes, open a new tab and use the same URL. Repeat as many times as you'd like!