Predictive Neural Information for Proactive Actions:
From Monkey Brain to Smart House Control.
The Plan4Act project will provide new emerging technologies that address “How neural activity, representing high-level cognitive processes of planning and mental simulation of action sequences can be extracted and used to proactively control smart home environments”.
The basis for this are recent experimental results that show that complex planning and sequencing information is represented by neural activity in the (monkey) brain . These investigations shall here be extended and transferred to a BMI-setup for controlling devices with more foresight than present in the currently existing systems. The here planned research and innovation action will, thus, open up a future path for people with disabilities to interact with their (smart) environment in a more robust way by – for the first time – including predictive neural information towards improving the quality of their life.
The above descriptions can be captured by four precise objectives of this project:
1. Identify sequence-predicting activity in the monkey in a non-restricted environment, i.e., a smart cage.
2. Develop mathematical models based on the interaction of activity and plasticity mechanisms to understand this sequence-predicting neural activity and to provide the algorithmic basis of adaptive neural control (Objective 3).
3. Develop a generic hardware controller that uses the models from Objective 2 to process the recorded sequence-predicting neural activity, and generate complex action sequences to manipulate devices in a Smart House.
4. Employ the results from Objectives 1-3 to develop an advanced BMI system that can robustly extract
sequence-predicting neural activity for proactive Smart House controls.
Project results are
This deliverable describes the proof-of-concept for wireless neural recordings in freely moving rhesus monkeys. We established and developed the hard- and software for registering behavioral and neural data while the animals participate in a controlled experiment.
This deliverable describes the basic neural network model developed and used to encode sequences in a biological plausible, cortical microcircuit-like neural network, and the results of sequence encoding and underlying processes.
This deliverable describes the status of the Plan4Act-designed and -developed devices and interfaces that will allow the connection between the Smart House Living Lab and the Smart Cage environment (WP1) via the neural network controllers from WP2 and WP3.
The Data Management Plan will address how data will be collected, generated and processed during the project, following what methodology and standards will be used, what data will be shared and/or made openly available and how will it be curate and preserved.
This document aims at the presentation of a suitable dissemination plan for making the project known over the scientific community as well as for end users