Granted Patent B2

US 12,461,595 · App. 18/386,907 · Granted Nov 4, 2025

Inventors: Tan Le (San Francisco, CA); Geoffrey Ross Mackellar (Sydney, AU)

Assignee: Emotiv 1nc.

G06F3/015G06F3/0487G06F16/24568G06F16/9535

Abstract

Claims (39)

1 . A method, comprising:

establishing contact between a biosignal detector and a user;

receiving a first set of cognitive state data parameters associated with the biosignal detector;

receiving a second set of cognitive state data parameters from a third-party, the second set of cognitive state data parameters comprising a data structure type for cognitive state data and a timing rule for cognitive state data generation;

receiving a third set of cognitive state data parameters associated with the user;

automatically collecting, at the biosignal detector, a bioelectrical signal dataset from a plurality of head regions of the user as the user is engaged with a stimulus provided by the third-party;

using a first model, generating the cognitive state data for the user based on the bioelectrical signal dataset and the third set of cognitive state data parameters associated with the user, according to the first set of cognitive state data parameters and the second set of cognitive state data parameters;

collecting a set of training data, wherein collecting the set of training data comprises: for each training user of a set of training users, collecting training data while the training user is engaged with a training stimulus and while a biosignal detector contacts the training user, wherein the training data comprises a response to the training stimulus and cognitive state data for the training user;

using a second model, classifying the user as a user group of a set of user groups based on the cognitive state data for the user, wherein the second model is a machine learning model trained to predict user groups using the set of training data, wherein the set of user groups is generated by segmenting the set of training users into the set of user groups based on the set of training data; and

predicting a preference for the user based on the user group.

2 . The method of claim 1 , wherein the plurality of head regions are associated with a plurality of brain lobes of the user.

3 . The method of claim 2 , wherein at least two brain lobes of the plurality of brain lobes are associated with a common brain hemisphere of the user.

4 . The method of claim 1 , further comprising, using a third model, classifying the user as a second user group of a second set of user groups based on the cognitive state data for the user.

5 . The method of claim 4 , wherein the third model is trained using a second set of training data comprising, for each of the set of training users, the cognitive state data for the training user and a second response to the training stimulus for the training user.

6 . The method of claim 1 , generating the set of user groups comprises, for each training user of the set of training users:

classifying a cognitive state of the training user based on the response to the training stimulus for the training user; and

segmenting the set of training users into the user groups based on the cognitive state classifications.

7 . The method of claim 1 , wherein, for each training user, the response to the training stimulus comprises a manual input from the training user associated with a cognitive state, wherein the set of user groups is generated based on the manual inputs.

8 . The method of claim 1 , wherein, for each training user, the training stimulus comprises a training stimulus provided by the third-party.

9 . The method of claim 1 , wherein generating cognitive state data for the user comprises processing the bioelectrical signal dataset using the first model, wherein the cognitive state data comprises the processed bioelectrical signal dataset.

10 . The method of claim 1 , further comprising: determining personalized content for the user based on the predicted preference; and presenting the personalized content to the user.

11 . A system, comprising:

a biosignal detector comprising a plurality of bioelectric signal sensors, the plurality of bioelectric signal sensors configured to automatically collect a bioelectrical signal dataset from a plurality of brain lobes of a user in response to a stimulus provided by a third-party, wherein the biosignal detector is associated with a first set of cognitive state data parameters; and

a processing system configured to:

receive a second set of cognitive state data parameters from the third-party, the second set of cognitive state data parameters comprising a data structure type and a timing rule for cognitive state prediction;

receive a third set of cognitive state data parameters associated with the user;

using a model, predict a cognitive state for the user based on the bioelectrical signal dataset for the user and the third set of cognitive state data parameters associated with the user, according to the first set of cognitive state data parameters and the second set of cognitive state data parameters, wherein the model is trained using a set of training data comprising, for each of a set of training users, a bioelectrical signal dataset for the training user and a cognitive state classification for the training user, wherein training the model comprises:

for each training user in the set of training users, selecting a user group from a set of user groups based on the set of training data; and

training the model to predict, for each training user in the set of training users, the user group for the training user based on the bioelectrical signal dataset for the training user.

12 . The system of claim 11 , further comprising a user interface configured to present personalized content to the user, wherein the personalized content is determined based on the predicted cognitive state for the user.

13 . The system of claim 11 , wherein the model comprises a machine learning model.

14 . The system of claim 13 , wherein the model is trained using a regularization method.

15 . The system of claim 11 , wherein predicting a cognitive state for the user comprises: using the model, classifying the user as a user group in the set of user groups; and determining the cognitive state for the user based on the user group classification.

16 . The system of claim 11 , wherein the processing system is further configured to, using a second model, predict a second cognitive state for the user based on the bioelectrical signal dataset for the user, wherein the second model is trained using a second set of training data comprising, for each of the set of training users, the bioelectrical signal dataset for the training user and a second cognitive state classification for the training user, wherein training the second model comprises:

for each training user in the set of training users, selecting a second user group from a second set of user groups based on the second set of training data; and

training the model to predict, for each training user in the set of training users, the second user group for the training user based on the bioelectrical signal dataset for the training user.

17 . The system of claim 11 , wherein, for each training user in the set of training users, the user group is selected based on the cognitive state classification for the training user.

18 . The system of claim 11 , wherein the cognitive state classification for each of the set of training users comprises a survey response.

19 . The system of claim 11 , wherein the biosignal detector does not obstruct a sense of smell of the user.

Assignments (1)

Continuity (5)

References Cited (177)