Property verification

This cognitive science research project intends to determine the extent to which our knowledge of what things look like is represented in long-term memory in a visual format. If visual knowledge is represented using visual mechanisms, then visual interference might disrupt one's ability to answer even simple yes or no questions about what things look like.

This repo includes the materials, methods, and results of our experiments designed to address these questions. The experiments are run using python, and the analyses are done in R. The results of the experiments can be downloaded as an R package from this repo.

To run the experiment using python, you need to clone the repo and install the required python packages. It is recommended that you set up a virtualenv to keep the python packages for this experiment isolated from the rest of your system:

$ virtualenv --python=python2 ~/.venvs/property
$ source ~/.venvs/property/bin/activate
(property)$ git clone https://github.com/lupyanlab/property-verification.git
(property)$ cd property-verification/experiment
(property)$ pip install -r requirements.txt
(property)$ python run.py  # Run the question first experiment

To download the data in this repo as an R package, you need to have the devtools package installed, and then you can run this command in R to download the data:

> devtools::install_github("lupyanlab", "property-verification",
                           subdir = "propertyverificationdata")
> library(propertyverificationdata)
> data("question_first") # Load the data for the question first experiments

For our interpretation of the results, see the reports directory, or the OSF page for this research at <osf.io/gctbx>. The manuscript reporting our initial findings is under review.