Brain-Computer Interface Programs (BCI)

A non-exhaustive list

For BCI designs, there are multiple toolboxes and design software that we can use. Each has their own pros and cons.

OpenViBE

Pros

• A few NCIL members already have some familiarity with it

• Compatible with Lab Streaming Layer (LSL) and OpenBCI

• Many ready-to-use paradigms included

• Commonly used in research

Cons

• Uses Lua programming language, not Python

• Poor documentation—not detailed or updated often

• Only compatible with Windows or Linux

NeuroPype

Pros

• Uses Python

• Compatible with PsychoPy, LSL, OpenBCI, and BrainProducts that we use in the lab

• Compatible with Mac, Windows, and Linux

• Has built-in nodes for LDA, SVM, and many other classifiers, and you’re able to code your own

• Has a web API so you’re not limited to a desktop/laptop to present your stimulus presentation

• Increasingly prevalent in the literature

• Documentation seems to be very detailed

• Free for academic research or for events like hackathons

Cons

• Requires a computer with four cores

  • Most newer laptops have over four cores, plus in-lab computers do. Older laptops often have less than four, so it might limit who can use NeuroPype at hackathons

• Subscription for commercial use

Timeflux

Pros

• Uses nodes written in YAML to acquire, monitor, and record data

• Uses nodes written in Python to process data and implement commands in real-time

• Uses packages we already use—pandas, scipy, scikit-learn, and others

• Compatible with Psychopy, LSL, OpenBCI devices

• Completely free and open-source

Cons

• Relatively new—documentation is still in development

• Not currently widely used