Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders

Babak Mahmoudi, Jose C. Principe, Justin C. Sanchez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

The design of Brain-Machine Interface (BMI) neural decoders that have robust performance in changing environments encountered in daily life activity is a challenging problem. One solution to this problem is the design of neural decoders that are able to assist and adapt to the user by participating in their perception-action-reward cycle (PARC). Using inspiration both from artificial intelligence and neurobiology reinforcement learning theories, we have designed a novel decoding architecture that enables a symbiotic relationship between the user and an Intelligent Assistant (IA). By tapping into the motor and reward centers in the brain, the IA adapts the process of decoding neural motor commands into prosthetic actions based on the user's goals. The focus of this paper is on extraction of goal information directly from the brain and making it accessible to the IA as an evaluative feedback for adaptation. We have recorded the neural activity of the Nucleus Accumbens in behaving rats during a reaching task. The peri-event time histograms demonstrate a rich representation of the reward prediction in this subcortical structure that can be modeled on a single trial basis as a scalar evaluative feedback with high precision.

Original languageEnglish
Title of host publication2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10
Pages1682-1685
Number of pages4
DOIs
StatePublished - Dec 1 2010
Externally publishedYes
Event2010 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10 - Buenos Aires, Argentina
Duration: Aug 31 2010Sep 4 2010

Other

Other2010 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10
CountryArgentina
CityBuenos Aires
Period8/31/109/4/10

Fingerprint

Brain
Feedback
Decoding
Reinforcement learning
Prosthetics
Artificial intelligence
Rats
Neurobiology

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Mahmoudi, B., Principe, J. C., & Sanchez, J. C. (2010). Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders. In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10 (pp. 1682-1685). [5626827] https://doi.org/10.1109/IEMBS.2010.5626827

Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders. / Mahmoudi, Babak; Principe, Jose C.; Sanchez, Justin C.

2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10. 2010. p. 1682-1685 5626827.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mahmoudi, B, Principe, JC & Sanchez, JC 2010, Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders. in 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10., 5626827, pp. 1682-1685, 2010 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10, Buenos Aires, Argentina, 8/31/10. https://doi.org/10.1109/IEMBS.2010.5626827
Mahmoudi B, Principe JC, Sanchez JC. Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders. In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10. 2010. p. 1682-1685. 5626827 https://doi.org/10.1109/IEMBS.2010.5626827
Mahmoudi, Babak ; Principe, Jose C. ; Sanchez, Justin C. / Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders. 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10. 2010. pp. 1682-1685
@inproceedings{d8501d793d2f4c27acba097a22442163,
title = "Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders",
abstract = "The design of Brain-Machine Interface (BMI) neural decoders that have robust performance in changing environments encountered in daily life activity is a challenging problem. One solution to this problem is the design of neural decoders that are able to assist and adapt to the user by participating in their perception-action-reward cycle (PARC). Using inspiration both from artificial intelligence and neurobiology reinforcement learning theories, we have designed a novel decoding architecture that enables a symbiotic relationship between the user and an Intelligent Assistant (IA). By tapping into the motor and reward centers in the brain, the IA adapts the process of decoding neural motor commands into prosthetic actions based on the user's goals. The focus of this paper is on extraction of goal information directly from the brain and making it accessible to the IA as an evaluative feedback for adaptation. We have recorded the neural activity of the Nucleus Accumbens in behaving rats during a reaching task. The peri-event time histograms demonstrate a rich representation of the reward prediction in this subcortical structure that can be modeled on a single trial basis as a scalar evaluative feedback with high precision.",
author = "Babak Mahmoudi and Principe, {Jose C.} and Sanchez, {Justin C.}",
year = "2010",
month = "12",
day = "1",
doi = "10.1109/IEMBS.2010.5626827",
language = "English",
isbn = "9781424441235",
pages = "1682--1685",
booktitle = "2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10",

}

TY - GEN

T1 - Extracting an evaluative feedback from the brain for adaptation of motor neuroprosthetic decoders

AU - Mahmoudi, Babak

AU - Principe, Jose C.

AU - Sanchez, Justin C.

PY - 2010/12/1

Y1 - 2010/12/1

N2 - The design of Brain-Machine Interface (BMI) neural decoders that have robust performance in changing environments encountered in daily life activity is a challenging problem. One solution to this problem is the design of neural decoders that are able to assist and adapt to the user by participating in their perception-action-reward cycle (PARC). Using inspiration both from artificial intelligence and neurobiology reinforcement learning theories, we have designed a novel decoding architecture that enables a symbiotic relationship between the user and an Intelligent Assistant (IA). By tapping into the motor and reward centers in the brain, the IA adapts the process of decoding neural motor commands into prosthetic actions based on the user's goals. The focus of this paper is on extraction of goal information directly from the brain and making it accessible to the IA as an evaluative feedback for adaptation. We have recorded the neural activity of the Nucleus Accumbens in behaving rats during a reaching task. The peri-event time histograms demonstrate a rich representation of the reward prediction in this subcortical structure that can be modeled on a single trial basis as a scalar evaluative feedback with high precision.

AB - The design of Brain-Machine Interface (BMI) neural decoders that have robust performance in changing environments encountered in daily life activity is a challenging problem. One solution to this problem is the design of neural decoders that are able to assist and adapt to the user by participating in their perception-action-reward cycle (PARC). Using inspiration both from artificial intelligence and neurobiology reinforcement learning theories, we have designed a novel decoding architecture that enables a symbiotic relationship between the user and an Intelligent Assistant (IA). By tapping into the motor and reward centers in the brain, the IA adapts the process of decoding neural motor commands into prosthetic actions based on the user's goals. The focus of this paper is on extraction of goal information directly from the brain and making it accessible to the IA as an evaluative feedback for adaptation. We have recorded the neural activity of the Nucleus Accumbens in behaving rats during a reaching task. The peri-event time histograms demonstrate a rich representation of the reward prediction in this subcortical structure that can be modeled on a single trial basis as a scalar evaluative feedback with high precision.

UR - http://www.scopus.com/inward/record.url?scp=78650835050&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650835050&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2010.5626827

DO - 10.1109/IEMBS.2010.5626827

M3 - Conference contribution

SN - 9781424441235

SP - 1682

EP - 1685

BT - 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10

ER -