Prerequesites for symbiotic brain-machine interfaces

Abstract

Recent advancements in the neuroscience and engineering of Brain-Machine Interfaces are providing a blueprint for how new co-adaptive designs based on reinforcement learning change the nature of a user's ability to accomplish tasks that were not possible using static methodologies. By designing adaptive controls and artificial intelligence into the neural interface, computers can become active assistants in goal-directed behavior and further enhance human performance. This paper presents a set of minimal prerequisites that enable a cooperative symbiosis and dialogue between biological and artificial systems.

Conceptualization of the Human-Machine Symbiosis – A Literature Review

“…A third main objective of HMS is to create humanlike communication and interaction, which is crucial to enable actors to accept each other as equal partners [39]. Humanlike communication includes both verbal (spoken and written language) and non-verbal communication (gestures, facial expressions, emotions) [40,44].…”

“…Even if not made explicit in the papers, in most cases we are talking about one human and one machine. Three papers specifically highlight this 1:1 relationship [11,39,42]. Xu et al, however, postulate that a limit of only two actors is not suitable for grid computing [48].…”

“…Twelve papers emphasize that the system in HMS must be dynamic [2,13]. In concrete terms, this means that both humans and machines must be able to adapt to new situations in real time [39]. All actors must play several clearly defined roles, and must also be able to learn in any context [21,38].…”

“…The second requirement is representation. In addition to awareness of the environment, the machine must also have an understanding of the actors and the system [39]. This requires that the machine possess a model or profile of the human, the machine itself, and the system in which data can be collected [23,44].…”

“…This goes hand in hand with the idea that data are collected constantly or at least periodically [17]. This information retrieval process takes place through information and data input by both actors (r1) [23,38,39]. The models of the human (c11) and the machine (c12), which each represent the state of the actors, are extended by a context and a task model [39].…”