How can we expand the computational capabilities of online cognitive tasks to realize dynamic task behavior?
This repository implements a containerized 'template' of an architecture that defers advanced computations from an online cognitive task to a cloud server configured to perform these computations.
Comprehensive documentation describing how to customize and deploy this template can be found in the wiki.
Beyond manipulation of task stimuli and static web components, dynamic behavior is constrained within the limitations of JavaScript itself. Dynamic task behavior can be implemented by deferring advanced computations from the participant’s browser to a cloud server. We propose a task-server architecture that receives participant task input, uses an R or MATLAB script to perform advanced computations, and returns an output to inform dynamic task behavior.
This concept is still in the early stages of piloting, but has been successfully setup and utilized on both AWS (API Gateway, Lightsail) and Microsoft Azure to communicate with the Gorilla platform.
Requirements for integration and deployment of this template:
- Cloud computing resource that supports API path configurations or a similar concept.
- A domain name with an SSL certificate to ensure secure communications between the tasks and the computing resources.
The wiki contains further detail of specific requirements for each prerequisite.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.