Apparatus and method for supporting brain function enhancement

1. An apparatus for supporting brain function enhancement of a prescribed brain function by enabling effective learning, comprising:

a brain activity detecting device for detecting a signal indicating brain activity at a prescribed area within a brain of a subject;

a storage device storing information of a target class corresponding to a target activity pattern obtained beforehand with respect to brain function enhancement; and

processing circuitry configured to, during a brain activity decoder configuration stage, perform the following steps

present, for a predefined time period, the subject with a plurality of stimulus events, where a stimulus event is a perceptual stimulus which activates portions of the subject's brain;

receive brain activity signals from the brain activity detecting device while the subject is exposed to the plurality of stimulus events, each of the plurality of stimulus events corresponding to one of a plurality of classes including the target class; and

train, using a machine learning algorithm, the processing circuitry to decode cranial nerve activity patterns from brain activity signals and classify the cranial nerve activity patterns to one of the plurality of classes of activity patterns classified in advance;

the processing circuitry further configured to, during a neurofeedback stage for training the subject, repeatedly perform the following steps until a predetermined condition is satisfied,

exposing the subject to a reward stimulus so that the subject tries to increase the reward while maintaining the subject in the absence of the awareness of the relation between the reward stimulus and the target class;

receiving a brain activity signal inducted by the subject from the brain activity detecting device while the subject is exposed to the reward stimulus;

deriving, from the received brain activity signal, a current cranial nerve activity pattern;

calculating a value representing a degree of possibility of said current cranial nerve activity pattern belonging to the target class,

computing, based on said calculated value, in accordance with degree of similarity of said current cranial nerve activity pattern to said target class, a reward value corresponding to said degree of similarity; and

altering the reward stimulus to represent a magnitude of said reward value without revealing to the subject the stimulus event corresponding to the target class; and

outputting said altered reward stimulus to said subject, wherein

said stimulus event corresponding to the target class is an object of perception leading to an identification problem of which class it is classified to in the brain; and

said processing circuitry is further configured to calculate likelihood of which class said activation pattern of cranial nerve activity corresponds to.

2. The apparatus for supporting brain function enhancement according to claim 1 , wherein the prescribed area within the brain of the subject corresponds to specific portions of the brain.

3. The apparatus for supporting brain function enhancement according to claim 1 , wherein said brain activity detecting device includes a functional Magnetic Resonance Imaging device.

4. The apparatus for supporting brain function enhancement according to claim 1 , wherein said brain activity detecting device includes a device for measuring electroencephalogram and near infrared light from outside of a skull.

5. A neurofeedback method of supporting brain function enhancement, using a brain activity detecting device for detecting a signal indicating a brain activity at a specific area within the brain of a subject and processing circuitry trained to decode cranial nerve activity patterns from brain activity signals and classify the cranial nerve activity pattern to one of a plurality of classes of activity patterns classified in advance respectively corresponding to stimulus events in association with a possibility, said method comprising the steps of:

training, during a brain activity decoder configuration stage, said processing circuitry through machine learning such that said plurality of classes includes a target class corresponding to a target activity pattern obtained with respect to brain function enhancement;

exposing, during a neurofeedback stage for training the subject, the subject to a reward stimulus and asking the subject to increase the reward without the subject having any knowledge of how to alter it;

receiving, during the neurofeedback stage, a brain activity signal induced by the subject from the brain activity detecting device while the subject is exposed to the reward stimulus;

deriving, from the received brain activity signal, a current cranial nerve activity pattern;

classifying, using the trained processing circuitry, said current cranial nerve activity pattern to one of the plurality of classes respectively corresponding to stimulus events presented to the subject prior to said neurofeedback stage;

calculating, based on the current cranial nerve activity pattern, a value representing a degree of possibility of said current cranial nerve activity pattern belonging to said target class;

calculating, in accordance with a degree of similarity of said current cranial nerves activity pattern to said target class, a reward value corresponding to said degree of similarity based on said calculated value;

altering the reward stimulus the subject is exposed to during said neurofeedback stage to represent a magnitude of said reward value, without revealing to the subject the stimulus event corresponding to the target class; and

during said neurofeedback stage, presenting the altered reward stimulus to the subject without revealing to the subject the stimulus event corresponding to the target class, wherein

said stimulus event corresponding to the target class is an object of perception leading to an identification problem of which class it is classified to in the brain; and

said processing circuitry is further configured to calculate likelihood of which class said activation pattern of cranial nerve activity corresponds to.

6. The apparatus for supporting brain function enhancement according to claim 1 , wherein

said processing circuitry is trained based on a sparse logistic regression; and

said degree of possibility is calculated using the sparse logistic regression, said degree of possibility being the likelihood of said current cranial nerve activity pattern belonging to the target class.

7. The neurofeedback method for supporting brain function enhancement according to claim 5 , wherein

said processing circuitry is trained based on a sparse logistic regression; and

said degree of possibility is calculated using the sparse logistic regression, said degree of possibility being the likelihood of said current cranial nerve activity pattern belonging to the target class.