Are BCI Devices Cognitively Demanding?

This blog is going to discuss whether BCI devices are cognitively demanding.It is built on the paper: Andéol Evain, Ferran Argelaguet, Nicolas Roussel, Géry Casiez, and Anatole Lécuyer. 2017. Can I Think of Something Else when Using a BCI? Cognitive Demand of an SSVEP-based BCI. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (CHI ’17). Association for Computing Machinery, New York, NY, USA, 5120–5125
The main idea behind the experiments conducted by the authors is to test whether we can perform parallel tasks while using BCI devices.The hypotheses is based on the 4D Model of multiple resources which in a nutshell says that if different tasks utilize different resources then it should be possible to do them simultaneously.Steady State Visually Evoked Potentials (SSVEP) are common BCI devices that try to stimulate a certain frequency in the brains signals by making participants visually observe the same frequency.The experiment for proving that BCI devices are in fact not cognitively demanding involved using these along with a demanding task which in this case was an N back task.N back tasks are memory based tasks where the participant has to recall the nth previous word that was show visually or heard.EEG data form the participants brain was recorded from 6 electrodes placed on the scalp.The study has two main factors:the difficulty of the n back tasks and the presentation of information which in this case was auditory and visual.Since an SSVEP is modulated by the localization of visual attention and not by its target it was observed that the visual n back task was not affecting the SSVEP although SSVEP’s are thought to be very sensitive to visual attention.Let us discuss the multiple resource theory which was the motivation behind this experiment. The Multiple Resources Theory is an approach to describe the extent to which dual-task performance will lead to decreases in time-sharing ability. This model considers four dimensions of resources. It classifies the cognitive resources by stages of processing (perception, cognition, and responding), by codes of proceeding (spatial and verbal), and by modality (auditory and visual). The model is generally refined with a fourth dimension, differentiating between focal and peripheral attention. The general idea between these distinctions is that if two tasks use different levels along each of the four dimensions, time-sharing will be better.The results of the experiments were statistically analyzed and it was proven that the hypotheses that it is possible to use BCI devices in tandem with secondary tasks.Further studies and experiments can be conducted such as this one to see how far can we extend this concept?Different kinds of BCI devices have been emerging and continue to emerge with recent trends on functional near infrared spectroscopy caps being used to track brain activity.Industry trends also show that people love to multitask operations on computing devices more and more which makes multitasking the core focus of most recent processors such as the latest generation intel processors.Would it be possible to have devices that passively automate certain tasks based on brain signals without active participation from the user?For example by passively looking at something of a particular frequency,users could indicate a certain command to be performed such as to turn a switch on or even send a message.The possibilities are endless with recent advancements in AI.For example Google Assistants can passively conduct conversations to reserve a dinner table as soon as the command is received.The automation of tasks can be of considerable interest so that users can focus most of their cognitive workload on tasks at hand instead of focusing on these devices to work.For example a person could be sending a command to a car while focusing on the cognitively demanding task of navigating the road.BCI devices are yet to hit the consumer market but that space can quickly evolve soon once industries start accepting them as consumer devices for day to day use.There are still a few hurdles to be crossed for such systems such the cost of the device,the feasibility of the devices in the given situation as well as accuracy of the device in question.There is still huge interest in devices that can be of use to people.For example Neurosky is one such company that manufactures BCI devices.The human brain being one of the most complex structures with workings we are yet to understand makes research on this topic a challenging field,however simple SSVEP devices for simple operations such as making a selection can be used by people as a part of daily life such as wearing a smartwatch.This field of research has huge potential and it is up to us to direct it in the right direction.