Item

Abstract

Historically, neurocognitive experimental tasks have been administered to participants within a controlled laboratory environment. In recent years it has become popular also to administer such tasks remotely to reach a broader segment of the population and to increase sample size substantially. This movement has been accelerated by the ongoing COVID-19 pandemic. Due to current uncertainty in how changing the testing environment, i.e., in-person vs online, may potentially affect participant performance, there is a pressing need to test participants across multiple environments in order to conduct a comparative analysis on at least a subsample of a larger cohort to identify any online biases. Operating across multiple environments presents an engineering challenge to guarantee consistent and replicable experimental task behavior. Each environment presents a set of unique operational requirements that must be addressed to ensure complete data collection and task completion. There is a currently unfulfilled need for tools that enable neurocognitive tasks to be easily administered across multiple environments. We present Neurocog.js, a JavaScript library that enables jsPsych experiments to operate across three unique experimental environments: over the internet via the Gorilla behavioral experiment builder platform, in-lab within an MRI spectrometer, and in-lab in a standard psychophysics/decision-making environment. Additional features are introduced including support for button-boxes and alternate input configurations, seeded random number generation, global experiment state and state management, and error handling with data preservation. Neurocog.js is an open-source project designed to easily integrate with existing jsPsych experiments while remaining extensible; the project is freely available online via GitHub at https://github.com/Brain-Development-and-Disorders-Lab/Neurocog.js.