Method and system for providing a brain-computer interface - Patents.com

Abstract

To provide a novel useful system and method for providing a brain computer interface.SOLUTION: There is provided a method for providing a brain computer interface, the method including the steps of: detecting a neural signal of a user responding to a calibration session having a time-locked component and a spontaneous component; generating a user-specific calibration model based upon the neural signal; prompting the user to receive a verification session having the time-locked component and spontaneous component; detecting the neural signal simultaneously with delivery of the verification session; generating output of the user-specific calibration model from the neural signal; determining an authentication status of the user based upon a comparison operation between processed outputs; and performing the authenticated action.SELECTED DRAWING: Figure 1A

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority to U.S. Provisional Patent Application No. 62 / 382,445, filed September 1, 2016, U.S. Provisional Patent Application No. 62 / 382,433, filed September 1, 2016, and U.S. Provisional Patent Application No. 62 / 360,640, filed July 11, 2016, each of which is incorporated herein by reference in its entirety.

[0002] Technical Field The present invention relates generally to the field of neural signal processing, and more particularly to improved systems and methods for authentication, authorization, and ongoing evaluation of user profiles in the field of neural signal processing. [Background technology]

[0003] background Brain-computer interface (BCI) systems and methods can be used to seamlessly connect a user with the user's environment and enhance the user's experience in the digital world. Such BCI systems can be used to connect one or more users to the digital universe, provide mechanisms for security (e.g., authentication, authorization) related to accessing sensitive information, and / or for any other suitable purpose. In particular, cybersecurity systems are essential for protecting sensitive information, user authentication, and user authorization, and must be robust against external and internal attacks. Current cybersecurity systems focus on multi-factor authentication and / or the use of specific biometric signature types. However, such systems are becoming increasingly susceptible to hacking, and new systems and methods need to be developed that provide enhanced security solutions that are impenetrable. Summary of the Invention

[0004] Thus, there is a need to create new and useful systems and methods for providing a brain-computer interface. The present invention provides such new and useful systems and methods. [The present invention 1001] detecting, within a first time window, a first neural signal of the user in response to a calibration session having at least one of a first time-locked component and a first spontaneous component, the first time-locked component including a first randomization factor; generating a user-specific calibration model based on the first neural signal; prompting the user to undergo a verification session having at least one of a second time-locked component and a second voluntary component within a second time window, the second time-locked component including a second randomization factor; detecting a second neural signal concurrently with delivery of the validation session; generating an output of the user-specific calibration model from the second neural signal; determining an authentication status of the user based on a comparison operation between the processed output of the second neural signal and an output of the user-specific calibration model; and performing an authorized action based on said authentication status. 23. A method for providing a brain-computer interface, comprising: [The present invention 1002] The method of claim 1001, wherein the first time-locked component includes a first sound stimulus and the second time-locked component includes a second sound stimulus, and the first and second sound stimuli are substantially identical. [The present invention 1003] The method of claim 1001, further comprising the steps of detecting an ancillary biometric signal within the first time window, and generating the user-specific calibration model based on the first neural signal and the ancillary biometric signal. [The present invention 1004] The method of claim 1001, wherein the first voluntary component includes prompting the user to imagine a stimulus, and the second voluntary component includes prompting the user to imagine the stimulus. [The present invention 1005] The method of the present invention 1001 further includes a step of retrieving user preferences from a database based on the authentication status of the user, and the step of performing the authenticated action includes a step of adjusting environmental parameters of a physical environment associated with the user in accordance with the retrieved user preferences. [The present invention 1006] The method of claim 1001, wherein performing the authenticated action comprises authorizing the user to access a computing system. [The present invention 1007] The method of the present invention 1001, wherein the step of performing the authenticated action includes a step of searching for digital content associated with a physical object with which the user interacts by performing a user-specific action in communication with a brain-computer interface device. [The present invention 1008] The method of any one of claims 10 to 12, wherein performing the authenticated action comprises retrieving a model of the user's neurological signals characterizing an interaction between the user and a digital input device. [The present invention 1009] detecting a first neural signal of the user responsive to the calibration session having at least one of a time-locked component and a spontaneous component within a first time window; generating a user-specific calibration model based on the first neural signal; prompting the user to undergo a verification session within a second time window; detecting a second neural signal concurrently with delivery of the validation session; determining an authentication status of the user based on a comparison operation between the processed output of the second neural signal and the processed output of the user-specific calibration model; and performing an authorized action based on said authentication status. 23. A method for providing a brain-computer interface, comprising: [The present invention 1010] The method of claim 1009, wherein the step of detecting the first neural signal includes the step of obtaining a signal from a brain activity sensor of a brain-computer interface device proximate to the user. [The present invention 1011] The method of the present invention 1009, wherein the time-locked component of the calibration session includes stimuli provided to the user at a series of time points within the first time window, and the step of detecting the first neural signal is synchronized in time with the series of time points. [The present invention 1012] The method of claim 1011, wherein the stimulus includes at least one of an acoustic stimulus, a visual stimulus, an olfactory stimulus, and a tactile stimulus. [The present invention 1013] The method of any one of claims 10 to 11, wherein the time-locking component includes a randomization factor that accounts for a predetermined percentage of the stimuli. [The present invention 1014] The method of claim 1009, further comprising the step of prompting the user to imagine a stimulus, and wherein the spontaneous component of the session includes the user imagining the stimulus. [The present invention 1015] The method of the present invention 1014, wherein the step of prompting the user to undergo the validation session includes the step of prompting the user to imagine the stimulus, and the comparison operation includes comparing a first output of the user-specific calibration model generated based on the first neural signal with a second output of the user-specific calibration model generated based on the second neural signal. [The present invention 1016] The method of claim 1009, further comprising the steps of detecting an ancillary biometric signal within the first time window, and generating the user-specific calibration model based on the first neural signal and the ancillary biometric signal. [The present invention 1017] The method of claim 1009, wherein performing the authenticated action includes reading a set of characters into an input field rendered on a display of a computing device associated with the user. [The present invention 1018] The method of the present invention 1009, wherein the step of determining the authentication status of the user includes a step of determining a state of the user including at least one of the user's alertness level, the user's drinking state, the user's mood, and the user's discomfort level, and the step of performing the authenticated action includes a step of enabling unlocking of a steering column of a vehicle associated with the user based on the state of the user. [The present invention 1019] The method of the present invention 1009 further includes retrieving user preferences from a database based on the authentication status of the user, and performing the authenticated action includes adjusting environmental parameters of at least one of a physical environment and a virtual environment associated with the user in accordance with the retrieved user preferences. [The present invention 1020] The method of claim 1009, wherein the authentication status includes a cognitive state of the user, and performing the authenticated action includes alerting an entity associated with the user of the cognitive state of the user. [Brief description of the drawings]

[0005] [Figure 1A] 1A-1D illustrate embodiments, variations, and examples of methods for providing a brain-computer interface. [Figure 1B] 1A-1D illustrate embodiments, variations, and examples of methods for providing a brain-computer interface. [Figure 1C] 1A-1D illustrate embodiments, variations, and examples of methods for providing a brain-computer interface. [Diagram 2] 2A-2B are diagrams illustrating an embodiment of a system for providing a brain-computer interface. [Diagram 3] 3A-3B show an embodiment of an ear-coupled system for providing a brain-computer interface. [Figure 4] 4A-4B are diagrams illustrating components of a variation of a system for providing a brain-computer interface. [Diagram 5] FIG. 1 illustrates a specific use case of the system for providing a brain-computer interface. [Figure 6] FIG. 1 illustrates variations of components of a system for providing a brain-computer interface. [Figure 7A] FIG. 1 illustrates a variation of an ear-worn system for providing a brain-computer interface. [Figure 7B] FIG. 1 illustrates a variation of an ear-worn system for providing a brain-computer interface. [Figure 7C] FIG. 1 illustrates a variation of an ear-worn system for providing a brain-computer interface. [Figure 7D] FIG. 1 illustrates a variation of an ear-worn system for providing a brain-computer interface. [Figure 8] FIG. 1 illustrates an application example of the use of the method and system for providing a brain-computer interface. [Figure 9] 9A-9B show variations of eyewear systems for providing a brain-computer interface. [Figure 10] 10A-10C are flow diagrams illustrating variations of a method for providing a brain-computer interface. [Figure 11] 11A-11C illustrate example neural signal patterns associated with generating a user-specific calibration model and authenticating a user based on the neural signals and the user-specific calibration model. [Figure 12]FIG. 1 illustrates an example of data flow in a method for providing a brain-computer interface. [Figure 13] 13A-13B are diagrams illustrating exemplary authenticated operations of a method for providing a brain-computer interface. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0006] Description of the Preferred Embodiments The following description of preferred embodiments of the invention is not intended to limit the invention to those preferred embodiments, but rather to enable any person skilled in the art to make and use the invention.

[0007] 1. Overview As shown in FIGS. 1A-1C, the method 100 of providing a brain-computer interface includes the following steps: collecting a first neural signal of a user in response to a calibration session having at least one of a first time-locked component and a first spontaneous component within a first time window S110; generating a user-specific calibration model based on the first neural signal S115; prompting the user to undergo a validation session within a second time window S120; collecting a second neural signal S130 concurrently with delivery of the validation session; generating an output of the user-specific calibration model from the second neural signal S135; determining an authentication status of the user based on a comparison operation between a processed output of the second neural signal and an output of the user-specific calibration model S140; and performing an authenticated operation based on the authentication status S150.

[0008] The method 100 functions to provide data transfer, control, and general interface between a user's brain and a computing system, as shown by way of example in FIG. 10A. The method 100 can also function to implement a neural signal analysis system for cybersecurity and other solutions (e.g., accessing a universe of electronic / digital content, receiving assistance from a digital assistant, etc.), and in certain embodiments, the invention described herein provides systems, methods, and protocols for evaluating and inferring user identity and brain-derived behavior information from brain activity measurements and analysis, as shown by way of example in FIG. 10B and FIG. 10C. The present invention relates to both a system for calibrating neural signal parameters that characterize and differentiate individuals and groups of individuals, and a software system for granting authorization and permission based on neural signal evaluation. As defined and combined in the present invention, these novel systems and methods may have utility within a wide range of applications and environments.

[0009] More specifically, the method 100 can implement an algorithm configured to receive signals from a user and classify brain activity, such that neural signals can be used as a single authentication factor and / or integrated into a broader multi-factor authentication platform. As an example, as shown in FIG. 11, signals of different users can have variable and invariant features, and the variable and / or invariant features of a given user can vary between users. Thus, the methods described herein can be used to understand specific and highly diverse features of brain activity (e.g., as shown in FIG. 12), and these features can be used as inputs to a classification and sorting system that verifies user identity with a higher degree of confidence than existing solutions. In the context of method 100, other elements of the multi-factor authentication platform may include one or more of the following: alphanumeric password input (e.g., characters, etc.), tokens, fingerprint input (e.g., fingerprint shape and characteristics), manual input (e.g., handprint shape and characteristics), peripheral perfusion measurement related input, iris pattern input (e.g., iris characteristics, etc.), facial characteristics, voice characteristics, gait characteristics, other biometric characteristics, keystrokes, computer usage logs, input into a digital object, and / or other possible current or future identifying observables.

[0010] The method 100 can be implemented, at least in part, using a brain-computer interface (BCI) device 200 to provide a brain-computer interface-based identity verification technique that helps authenticate a user by evaluating the user's brain activity. As shown in Figures 2A-2B, 3A-3B, 7A-7D, and 9A-9B, the BCI can have at least one of the following form factors: a headset, a head-mounted wearable device, an earbud, an ear-mounted wearable device, a face-mounted wearable device, a local environment device (such as a room with a magnetic coil that can detect electromagnetic activity from the user's head), and any other suitable device capable of generating a brain-computer interface. In a variation, the BCI can have an ear-mounted form factor, as shown in Figures 3A-3B and 7A-7D, and in another variation, the BCI can have an eyewear form factor, as shown in Figures 9A-9B. However, in further variations, the BCI may have a combined ear-worn and eye-worn aspect, and / or any other suitable form factor.

[0011] The BCI can interface with peripheral devices 201 (e.g., USB-connected devices, Bluetooth-connected devices, loaded with user-specific calibration model(s) and configured to output commands) to act on device states of devices associated with a user that interacts with method 100, as shown in Figures 2A-2B, 4A-4B, 5, and 6.

[0012] Additionally, the BCI / peripheral device may include or otherwise be in communication with a computing system including at least one of a computer or processor non-transitory readable medium, a computer or processor non-transitory storage medium (e.g., memory, disk drive, flash memory) configured to encode and / or store instructions (i.e., computer-executable instructions associated with a portion or portions of the methods described above).

[0013] Additionally or alternatively, each part of the system may include or cooperate with one or more of sensors 210, sensory stimulus delivery systems 220 (e.g., an audio stimulus delivery system as shown in Figures 3A-3B and 7A-7D, a visual stimulus delivery system as shown in Figures 9A-9B, a tactile stimulus delivery system, an olfactory stimulus delivery system, a taste stimulus delivery system), and associated embedded software (or firmware) modules, algorithms, and / or procedures 230.

[0014] Thus, the developed system elements and / or methods may work together as a processing pipeline configured to detect brain activity, extract unique and informative activity features and characteristics for each user, define a model of each user (e.g., in a population-wide manner, in a user-specific manner), refine the model(s) such that each unique user has a separate and highly identifiable profile, and then compare future identity claims to said models. The system elements and / or methods may then be operable to enable users to interact with their environment and / or digital content in a more advanced way. Thus, the system components, in combination with novel algorithms implemented in software, can operate as one or more of the following: neural signal scanner, neurobiometric signature module, neural ink decoder (e.g., for watermarking to encrypt and / or decrypt brain activity upon command), electrophysiological wave evaluation engine, electrophysiological classifier, biomagnetic signature analyzer, thought-print scanner, personal identification physiological number generator, and archetypal topology analyzer.

[0015] 2. Profit The present methods and / or related systems, and variations thereof, may provide several benefits and / or advantages.

[0016] First, variations of the method may enable metadata extraction from the real world by monitoring neural signals and performing actions based on the extracted metadata. For example, a user may imagine walking through a garden and recalling information about each plant. The information provided to the user may increase in detail until the user provides neural signal-based input (e.g., imagine a "pause", imagine a spreadsheet), whereupon data associated with the information is moved to an electronic document (e.g., a spreadsheet) that the user can later view. In another example, a user may gaze at a physical item in a grocery store and imagine purchasing the physical item, and in response, a digital indicator of the physical item may be added to a virtual shopping cart associated with the user. In another example, a user may interact with physical objects in the real world and, in response to monitoring of neural signals associated with the user's interaction with the physical objects, manipulate digital objects in the virtual world (e.g., with a head-mounted wearable display).

[0017] Second, variations of the method may allow for iterative improvement of the mind-physical data link. For example, a user may interact with physical objects in the real world, and in response to monitoring of neural signals associated with the user's interactions with the physical objects, digital objects may be manipulated in the virtual world, and the user's mental responses to the manipulation of the digital objects may be monitored to assess the quality of the digital representation (e.g., the user's mental state may be assessed as pleased or dissatisfied with the virtual representation). A feedback loop between the user's physical behavior, the virtual representation of the user's physical behavior, and the user's mental responses to the virtual representation may be repeated multiple times to improve the virtual representation of the user's physical behavior (e.g., until the user's state of pleasure and / or satisfaction is consistently determined). In one example, as shown in FIG. 13A, a user wearing an augmented reality headset may automatically initiate a video call with a desired entity by imagining the need to speak to the entity according to a user-specific model of neural signatures associated with speaking to the entity.

[0018] Third, variations of the method allow for monitoring a user's desires and addressing such desires without relying on externally observable user actions (gestures, eye movements, etc.) since the method involves monitoring the user's internal mental state via neural signals. The user's brain itself can provide commands and / or indications of the user's desires (e.g., via motor cortex, visual cortex, language centers, etc.) even when the action is not explicitly and / or physically performed. For example, a user can select a selection box via a brain-computer interface as shown in FIG. 13B.

[0019] Fourth, variations of the method may improve the performance of dedicated and non-dedicated hardware. For example, variations of the method may enable a user to unlock their mobile device via a brain-computer interface by authenticating the user identity and unlocking the device in response. In this example, a user may utilize their mobile device more quickly and efficiently while maintaining secure safeguards against unauthorized access. In another example, the performance of a brain-computer interface device may be improved by generating user-specific calibration models that account for differences between users that may otherwise reduce the effectiveness of the brain-computer interface device.

[0020] Fifth, the variant of the method can solve problems rooted in computer and machine technology. For example, the method can enable deconvolution of variable and invariant features present in neural signals. Invariant features can be used to determine the authentication status of a user, and variable features can be deciphered through some implementations of the method to determine the user's emotional state(s), intention, desire, etc. Thus, the variant of the method can solve the problem of simultaneous user authentication and neural signal evaluation, which is a problem rooted in computer and machine technology.

[0021] However, the methods and / or systems herein may provide any other suitable benefits and / or advantages.

[0022] 3.1 Method - Calibration Block S110 includes performing a calibration operation using a neural calibration signal data set from a user received in response to a session having a time-locked component and a spontaneous component within a first time window. Block S110 functions to record and process data in a single environment or in various environments to calibrate a user-specific model of brain activity that can be used for user identity verification. The user-specific model can be used to make inferences about new brain activity signal samples (from the user or from other users). Additionally or alternatively, the user-specific model can be used as an information source for grouping individuals, characterizing personality traits, and / or any other suitable purpose.

[0023] Block S110 preferably includes receiving bioelectrical signals indicative of the user(s)' brain activity, including electroencephalography (EEG) signals collected during a calibration session using a suitable BCI coupled to the user. However, the bioelectrical signals (or other signals received in blocks S110 and S130) may additionally or alternatively include any one or more of the following: magnetoencephalography (MEG), impedance signals, galvanic skin response (GSR) signals, electrocardiogram (ECG) signals, heart rate variability (HRV) signals, electronystagmogram (EOG) signals, electromyogram (EMG) signals, and / or any signals that may be indicative of brain activity or that may supplement a protocol for user verification using signals derived from brain activity. Furthermore, block S110 may additionally or alternatively include collecting other biosignal data obtained from or associated with the user's biological tissue or biological processes and the user's environment.

[0024] Block S110 may be implemented in a real physical environment and / or a simulated physical environment (e.g., using a virtual reality device). In one variation, block S110 may include receiving a signal in a physical or simulated environment (e.g., an office, a room, a parking lot, a server room, a war zone, a daycare, a machine shop, a test facility, etc.) associated with a use case of the authentication / authorization system. The real and / or simulated environments may have any suitable noise figure associated therewith, related to electrical noise, acoustic noise, and / or any other suitable type of noise associated with signal acquisition, such that the calibration model(s) generated and validated in block S110 are substantially adaptive / invariant, robust, and functional in a variety of environments having different noise or other profiles.

[0025] Similarly, block S110 may be performed in association with environments associated with various mental states. In a variation, such mental states of the user(s) may be associated with one or more of the following: mental states associated with concentration (e.g., a focused mental state, a distracted mental state), emotional / affective states (e.g., a happy state, a sad state, a frightened state, a relaxed state, an angry state, etc.), cognitive states (e.g., a high cognitive load state, a low cognitive load state, etc.), and / or any other suitable mental states. Thus, in generating calibration models in various mental environments, the calibration model(s) generated and validated in block S110 may be substantially adaptive / invariant, robust, and functional in association with various user mental states.

[0026] However, calibration can be performed in a manner that takes into account a wide variety of other variables to generate a substantially adaptive / invariant, robust, and functional user model. For example, the signal acquisition in block S110 can be performed in association with various physiological states of the user (e.g., the user's metabolic state, the user's sleep state, the user's state associated with cardiovascular stress, the user's state associated with another physiological state, etc.). Additionally or alternatively, the signal acquisition in block S110 can be performed in association with various environmental conditions (e.g., associated with the sensors of the BCI used). Such environmental conditions can include one or more of the following: various environmental temperature conditions, various environmental pressure conditions, various environmental humidity conditions, and / or any other suitable environmental conditions that may affect sensor performance and / or performance of any other suitable hardware components associated with the system(s) implementing the method(s) described above.

[0027] Further, in connection with signal acquisition in various environments / conditions, such environments may be specifically designed / devised and provided during the calibration session of block S130, or responses may be made to such environments autonomously during calibration, with the various environmental / condition conditions being recorded during signal collection.

[0028] The calibration operation of block S110 may be performed once or may be repeated. In variations in which the calibration operation is repeated, the repetitions may be performed any suitable number of times and with any suitable frequency in association with a particular triggering event (e.g., time of day, triggered by a user, triggered by a system administrator, triggered by a system operator, triggered in response to an environmental event, etc.). In a specific example, the calibration operation may be performed whenever a user desires the system to learn a model of the user's particular mental phenomenon, such that the calibration operation is performed whenever the user's particular mental phenomenon occurs.

[0029] As mentioned above, the calibration session of block S110 may have a time-lock component, and a time-lock signal is collected in response to a provided stimulus. Preferably, the time-lock component of the calibration session includes a stimulus provided to the user within a time window (e.g., a series of time points within the time window, a single time point within the time window, continuously within the time window, etc.), and the signal acquired in association with the time-lock component is synchronized in time (e.g., locked to the neural signal). Alternatively, the time-lock component may have any suitable temporal characteristics in association with the signal acquisition (e.g., asynchronous, delayed by a predetermined time shift, etc.). In a variation, the stimulus of the time-lock component may include one or more of the following: an auditory stimulus (such as a sound stimulus) provided by an audio output device (e.g., an audio output device integrated with or in communication with an EEG signal acquisition system and / or other sensor system), a visual stimulus provided by a display (e.g., a display integrated with or in communication with an EEG signal acquisition system and / or other sensor system), a tactile stimulus provided by a tactile object / tactile system, an olfactory stimulus, a taste stimulus, and any other suitable stimulus. In specific examples, the stimuli for the time-locked components can include audio stimuli (e.g., music samples, speech samples, harsh sound samples, etc.) Additionally or alternatively, in other examples, the stimuli for the time-locked components can include visual stimuli (e.g., a series of images, a video, etc.).

[0030] The time-locked component of the calibration session may have a randomization factor that occupies any desired portion (e.g., a predetermined percentage) of the time-locked stimuli. More specifically, the randomization factor may contribute to decoupling environmental and psychological factors associated with generating a user profile. In a specific example, the randomization factor may occupy 1%-50% of the stimuli. However, in variations of the example, the randomization factor may occupy any other suitable portion (e.g., 0%-100%) of the stimuli. In one example, the sound stimuli may be sound samples that include a randomization factor of 5%-30% (compared to an unedited version of the sound sample). In another example, the readable stimuli may be written samples that include a randomization factor of 5%-30% (compared to an unedited version of the written sample). For example, the unedited written sample may state "The man unlocked the door in front of him" and the randomized version may state "The man unlocked his elephant." The responses to the randomized versions / portions of the stimuli and the responses to the unmodified stimuli can then be processed to extract features, which can be analyzed for brain activity signatures used in other blocks of method 100.

[0031] As previously mentioned, the calibration session of block S110 can have a spontaneous component, where signals are collected in response to free-form events. In specific examples, the free-form events can include imagined sounds (e.g., sound stimuli imagined by the user, whereby the spontaneous component includes a session in which the user imagines listening to a particular sound sample), imagined odors (e.g., olfactory stimuli imagined by the user, whereby the spontaneous component includes a session in which the user imagines smelling a particular odor), imagined tastes (e.g., taste stimuli imagined by the user, whereby the spontaneous component includes a session in which the user imagines tasting a particular consumable), user memories (e.g., whereby the spontaneous component includes prompting the user to recall a memory), and / or any other suitable free-form events.

[0032] Furthermore, the calibration session can be performed with the user aware that a calibration is occurring, or can be performed without the user's awareness. For example, visual stimuli can be provided subconsciously (such as images spliced ​​between video frames), and / or sound stimuli can be provided subconsciously (such as at frequencies outside the perceptible range), and / or data can be analyzed retrospectively (such as after the fact), so that new models are calibrated based on data obtained outside of the calibration session in which the user was aware. However, prompting and signal acquisition can be performed in a calibration session in any other suitable manner.

[0033] Although a variation of block S110 is described above that includes both a time-locked component and a spontaneous component, a variation of block S110 may alternatively omit the session having the time-locked component or omit the session having the spontaneous component.

[0034] In a specific example of block S110, the spontaneous component includes a session in which the user imagines a stimulus (e.g., a needle stimulus, shaking hands with a particular person, the sensation of falling, etc.) and collects signals in response to the user imagining the stimulus. In a related example, the spontaneous component can include a session in which the user freely imagines a needle stimulus (or other stimulus) and then later logs that the user imagined the needle stimulus (or other stimulus) at a prior time. However, variations of the spontaneous component can be configured in any other suitable manner.

[0035] Furthermore, with regard to calibration, the calibration process can be governed by the desired analysis range and any other appropriate boundary conditions so that unintended use of the system can be prevented. For example, in certain applications for government use, an assessment of an individual's identity can be made without going too deep into the assessment of brain health (i.e., in a manner that would compromise health-related privacy laws). Furthermore, such calibration models can be encrypted according to industry accepted protocols and / or made anonymous with private and secure storage means.

[0036] 3.1.1 Calibration - Model Generation The method preferably includes block S115, which includes generating a user-specific calibration model based on the first neural signal. Block S115 functions to generate a calibration model of the evaluated user(s) with respect to received signals (e.g., EEG signals, signals reflecting brain activity or other activity) in response to the time-locked stimuli and / or spontaneous or free-form events.

[0037] Generating the calibration model may implement a machine learning (ML) algorithm configured to generate user-specific and / or population-wide models from the training data received in block S110, more specifically, to fuse data features and trained classifiers by coding for specific fusion and using automated combination techniques. In one variation, generating the calibration model in block S110 may implement a support vector machine (SVM) classifier trained on specific data features (invariant features, variable features, etc.) derived from the neural signals (and other data) received in the calibration operation in block S110. In another variation, the received data may be processed to be understood and applied in a particular setting by using the features as inputs to one or more artificial neural networks, natural neural networks, and / or other deep learning architectures / learning systems (e.g., nonlinear learning systems). Metaheuristic algorithms may then be applied to the output to generate even more accurate, reliable, and robust models for these particular settings. However, the calibration model may be generated and / or improved in any other suitable manner.

[0038] In variants, the metaheuristic algorithms applied at this stage can be classified according to one or more of the following: local search strategies, global search strategies, single solution approaches, population-based approaches, hybridization algorithm approaches, memetic algorithm approaches, parallel metaheuristic methods, nature-inspired metaheuristic methods, and any other suitable approaches. In specific examples, the metaheuristic algorithms applied to the received calibration data can include genetic algorithms (e.g., genetic adaptive algorithms), although any other suitable approaches can be used. Additionally or alternatively, non-metaheuristic algorithms (e.g., optimization algorithms, iterative methods) can be used.

[0039] In connection with generating the calibration model in block S110, the fusion data may include data derived from user brain activity and data from other sources (e.g., non-brain derived sources, auxiliary biometric signals, auxiliary data, etc.).

[0040] In variants, the brain activity derived features used to generate the model can be derived from one or more of the following: event-related potential data (e.g. voltage / time values, etc.), resting state / spontaneous activity data (e.g. voltage / time values, etc.), response to different events (e.g. differences between event-related potentials obtained in response to events of different types, timing, categories, etc.), time domain signal subdivided (e.g. probabilistically, systematically) into many atomic units representing features at different time scales (e.g. sub-second, seconds, minutes, hours, days, months, years, more than a year, etc.), spectral aspects (e.g. minima, maxima, positive peaks, negative peaks, polarities or time points associated with other peaks, amplitudes of peaks, arithmetic relations calculated on peaks, etc.), similarity measures between defined "max-min spectra" and pre-computed templates (e.g. cosine similarity, various kernels, etc.), similarity measures between various equivalent atomic units of the raw time domain signal, similarity measures applied to combinations of features, from clustering of features (soft / fuzzy clustering, hard clustering, etc.). Derived measures, filtered signals (e.g. signals adaptively filtered based on quadratic parameters, signals filtered into discrete frequency bands with variance / autocorrelation in each band and cross-correlation / covariance across the bands, etc.), distance metrics between selection matrices (e.g. Riemannian distance), features derived from iterative matrix manipulation (e.g. using Riemannian geometry), entropy measures within features, between features and between the entire feature set of different users, subspace analysis (e.g. fixed subspace analysis separating the signal into fixed and non-fixed sources), recursive quantification analysis of different time segments, recursive plots processed by convolutional neural networks to extract additional features (e.g. shape, edges, corners, texture, etc.), features derived from principal component analysis, features derived from independent component analysis, features derived from factor analysis, features derived from empirical mode decomposition, features derived from principal geodesic analysis, features derived from sequence probabilities (e.g. by hidden Markov models, by application of the Viterbi algorithm, by comparison of estimated paths with templates, etc.), features derived from sequence probabilities (e.g. by application of the Viterbi algorithm, by comparison of estimated paths with templates, etc.), features derived from the principal component analysis (e.g.Features derived from single channel replicas (with or without application of different distortions to the replicas), invariant features, variable features, and any other suitable brain activity derived features.

[0041] In variations, the non-brain activity derived features may be derived from one or more of the following: personal information (e.g., username, user identification code, date of birth, location, relationships, family history, religion, etc.), physical attributes (e.g., height, weight, eye color, hair color, ethnicity, shoe size, etc.), biometric identifiers (e.g., fingerprints, iris scan data, hand scan data, palm print, voice print, epidermal features, odor, genomic information, blood type, endophenotypic information, phenotypic information, genotypic information, etc.), (e.g., electroencephalography data, electrocardiogram data, electrodermal response data, blood pulse or data derived from other sensors, near infrared spectroscopy data, electromyography data, pupillography data, hemoencephalography data, heart rate derived data, galvanic skin response data, galvanic nystagmus data, etc.); biometric data derived from other sensors, data derived from other sensors / emitters (e.g., gyroscope data, camera data, microphone data, Wi-Fi broadcast data, radio broadcast data, Bluetooth data, GPS data, RFID data, access control log data, elevator log data, etc.); data, etc.), user behavior interacting with the device (e.g., keystrokes, mouse activity, touchpad activity, joystick activity, reaction times, etc.), data derived from user surveys, environmental information (raw / processed data derived from computer vision based logs of the surrounding environment), data derived from video streamed in an augmented or virtual reality scene (e.g., timing, type, location, attributes, etc.), data derived from the user device (e.g., application derived information, digital service derived information, device state information, device sensor derived information), data from social networking applications, data from productivity applications (e.g., email, calendar, appointments, notes, etc.), data from other users, data from institutions (banks, hospitals, governments, schools, companies, etc.), network derived data (e.g., traffic properties, firewall logs, file metadata, internal server communications with other servers, mailroom activity logs, port opens, port closes, timing correlation with user IPs, user search terms, etc.), and any other suitable data.

[0042] While various algorithms are described above, the calibration model generated in block S110 may additionally or alternatively be based on an algorithm characterized by a learning style including any one or more of the following: supervised learning (e.g., using logistic regression, using back-propagation neural networks), unsupervised learning (e.g., using a priori algorithms, using K-means clustering), semi-supervised learning, reinforcement learning (e.g., using a Q-learning algorithm, using temporal difference learning), and any other suitable learning style.Additionally, the machine learning algorithm may implement any one or more of the following: regression algorithms (e.g., least squares, logistic regression, stepwise regression, multivariate adaptive regression splines, locally estimated scatterplot smoothing, etc.), instance-based methods (e.g., k-nearest neighbors, learning vector quantization, self-organizing maps, etc.), regularization methods (e.g., ridge regression, least absolute shrinkage and selection operator, elastic nets, etc.), decision tree learning methods (e.g., classification and regression trees, iterative dichotomiser 3 (ID3), C4.5, Chi-squared Automatic Interaction Detection (CHAID), decision strains, random forests, multivariate adaptive regression splines, gradient boosting machines, etc.), Bayesian methods (e.g., Naive Bayes, Averaged One-Dependence (AODE), etc.), and the like. We have a wide range of techniques for learning the following: Classifiers (e.g., Bayesian belief networks, etc.), Kernel methods (e.g., support vector machines, radial basis functions, linear discriminant analysis, etc.), Clustering methods (e.g., K-means clustering, expectation maximization, etc.), Association rule learning algorithms (e.g., Apriori algorithm, Eclat algorithm, etc.), Artificial neural network models (e.g., Perceptron, Backpropagation, Hopfield networks, Self-organizing maps, Learning Vector Quantization, etc.), Deep learning algorithms (e.g., Restricted Boltzmann Machines, Deep Belief Networks, Convolutional Networks, Stacked Autoencoders, etc.), Dimensionality reduction methods (e.g., Principal Component Analysis, Partial Least Squares Regression, Laplacian Eigenmap, Isomapping, Wavelet Thresholding, Sammon Map, Multidimensional Scaling, Projection Pursuit, etc.), Ensemble methods (e.g., Boosting, Bootstrapped Aggregation, etc.), aggregation, AdaBoost, stacked generalization, gradient boosting machines, random forests, etc.), and any suitable form of machine learning algorithm.

[0043] 3.2 Methodology - User Prompting and Signal Collection Block S120 includes prompting the user to undergo a verification session within a second time window, which serves to assess the user's identity at a next time point. The prompting preferably includes prompting the user for a short period of exercise (e.g., less than one minute long), but may alternatively include prompting the user for any exercise period. As with the previous block, the prompting may occur with or without the user perceiving the prompt.

[0044] The prompts in block S120 are preferably implemented in an electronic prompting device associated with system 100, the prompting device being further associated with a particular use application. In variations, the prompts may be provided within an application installed on a computing device, within an application running on a mobile computing device, within an application running on a wearable computing device (e.g., a head-mounted wearable computing device, a wrist-worn wearable computing device), within a website, and / or within any other suitable electronic environment. The prompts in block S120 may be provided in one or more of the following ways: visually, audibly, tactilely, and any other suitable manner. The prompts may be provided perceptually (e.g., via engagement with the user) or non-perceptually (e.g., a stimulus may be provided that is outside the range of consciously perceptible stimuli). Additionally, variations of the prompts in block S120 may be provided without device prompting and may instead rely on human direction to prompt the user.

[0045] As with block S110, the prompting in block S120 may occur at any suitable frequency and / or in response to any suitable triggering event. Thus, in a variation, the prompting may occur in block S110 whenever the user needs to be authenticated or authorized for a particular use application. Additionally or alternatively, in a variation, the prompting may occur multiple times periodically or at any frequency on an irregular basis (e.g., in connection with re-authorization, re-authentication). Additionally or alternatively, in a variation, the prompting may occur in connection with a triggering event (e.g., a period of inactivity, an attempt to access a different location, an attempt to access different content, etc.).

[0046] The exercises associated with the prompts in block S120 are preferably associated with the exercises of the calibration session of S110, but may include any combination or subset thereof. For example, in some variations, block S120 may include prompting with a combination of time-locked and / or spontaneous components. Alternatively, block S120 may include prompting with only a time-locked component. As a further alternative, block S120 may include prompting with only a spontaneous component. As a further alternative, block S120 may include prompting with unmodeled exercises, so that brain activity responses to these previously unmodeled exercises can be incorporated into the improved calibration model of block S110. In a specific example, the prompts in block S120 may include a spontaneous session that separates two time-locked sessions, each of short duration. Additionally, in connection with the randomization factors discussed above, any component of the prompts in block S120 may include an algorithmically selected randomization factor that promotes enhanced security and / or reliability of user identity.

[0047] Moreover, similar to the calibration session(s) of block S110, the prompting of block S120 can occur with the knowledge of the user or in a manner that is transparent to the user.

[0048] Block S130 includes collecting a neural signal dataset in coordination with the delivery of the validation session of block S120, and serves to collect the user's brain activity (and other) responses that can be tested according to the calibration model output from block S110.

[0049] Similar to block S110, block S130 preferably includes receiving bioelectrical signals indicative of the user(s)' brain activity, including brain signals (e.g., EEG or other signals) collected during a calibration session using a suitable BCI device coupled to the user. However, the bioelectrical signals (or other signals received in blocks S110 and S130) may additionally or alternatively include any one or more of the following: magnetoencephalography (MEG) impedance signals, galvanic skin response (GSR) signals, electrocardiogram (ECG) signals, heart rate variability (HRV) signals, electronystagmogram (EOG) signals, electrospinal electrogram (EMG) signals, electromyogram signals, and / or any signals that may be indicative of brain activity or that may supplement a protocol for user validation using signals derived from brain activity. However, in connection with the non-brain activity derived signals of the calibration model, block S130 may additionally or alternatively include collecting other biosignal data obtained from or related to the user's biological tissues or biological processes, as well as the user's environment, devices associated with the user, network aspects, and / or any other suitable source.

[0050] Any suitable signal processing operations may be performed in connection with the neural signal data sets received at blocks S110 and S130. For example, a noise reduction model may be applied to reduce or remove the effects of user-specific impedance, motion artifacts, environmental artifacts, and / or any other suitable noise. Additionally or alternatively, signal processing may include performing filtering, smoothing, clipping / winsorizing, deconvolution, standardization, detrending, resampling, and any other suitable data conditioning processes on the neural signal data (and / or other data) received at blocks S110 and S130.

[0051] 3.3 Method - User Authentication Block S140 includes determining an authentication status of the user based on a comparison operation between the processed output of the second neural signal and the processed output of the user-specific calibration model. Block S140 functions to authenticate the identity of the user with an authentication method that can be integrated with a multi-factor authentication protocol. The authentication in block S140 can be used to extract a user-specific model (e.g., of how the user interacts with digital content in relation to the neurological signals associated with the user behavior). The authentication in block S140 can additionally or alternatively be performed as a high-security service in an enterprise (e.g., a high-security enterprise) that can serve as a solution to key management, long passwords, and / or insufficient multi-factor authentication approaches (e.g., Yubikey). Thus, a portion or portions of the algorithms associated with method 100 can be integrated with third-party service provision software to enable time-specific authentication as described herein.

[0052] At block S140, the authentication status preferably includes a positive verification of the user's identity, a negative verification of the user's identity, or an inconclusive verification of the user's identity. However, the authentication status may additionally or alternatively include a confidence percentage in the analysis of the user's identity and / or any other suitable characteristic of the user's identity. The authentication preferably includes a comparison between a model of the signal received during the verification session and a calibration model. Additionally, the authentication preferably accounts for a randomization factor within the time lock component provided at the prompting of block S120. Upon executing the authentication process, block S140 may then include transmitting an authentication certificate to a target (e.g., a server, a device, etc.), where the quality of the authentication certificate triggers or prevents a subsequent action from occurring. However, the authentication may be performed in any other suitable manner at block S140.

[0053] In association with a positive identification of the user's identity, block S140 can drive various responses associated with the integrated platform / device. In a first variation, block S140 can include using a BCI device or other interface device to automatically extract a user model of how the user "clicks" a digital button associated with interacting with electronic content. In another variation, block S140 can include automatically loading a password (e.g., in an input box) in response to a positive identification of the user, so that the user does not need to remember the password or check another device for a token. In a specific example, the password can be loaded into an input box (e.g., an operating system login page, or a website, or an application, etc.) and / or into a clipboard of a computing device (e.g., a personal computer, a mobile device, a wearable computing device, etc.). In this embodiment, a peripheral device as shown in Fig. 6 (e.g., a USB connected device, a Bluetooth connected device loaded with user specific calibration model(s) and configured to mimic the key entry module of a computing device) may be implemented with pre-installed code and configured to communicate with at least one of the brain-computer interface identification system and / or computing system and execute instructions upon positive verification of the user. Such a peripheral device may further be portable between various computers / workstations that a user is attempting to access.

[0054] In another implementation of block S140, a peripheral may be implemented to protect copies of professional applications (e.g., video editing software, audio editing software, design software, etc.) and / or systems (e.g., automobiles, other vehicles, robotic agents, etc.). Such software and / or systems may be packaged with a peripheral that isolates access to a particular user or multiple users. In these implementations, the peripheral may communicate directly with the application (e.g., software application) or system being accessed. Additionally, in these implementations, the peripheral may facilitate installation of an agent on a host by implementing logic in a language(s) compatible with the system(s) of use.

[0055] In another example, block S140 may include performing an authentication operation within a web application platform (e.g., a banking system, a medical system, etc.). In this example, a positive identification may prompt direct communication with the provider's web service or communication via a web browser to allow the authenticated user to access content. As with the previous example, in this example, a peripheral may be implemented that facilitates installation of an agent on a host by implementing logic in a language(s) compatible with the system(s) of use. Additionally or alternatively, block S140 may include authorizing the user to access any suitable computing system (e.g., a mobile device associated with the user, a personal computer associated with the user, etc.).

[0056] In another example, block S140 may include comparing a first output of a user-specific calibration model to a second output of a user-specific calibration model, in this example, the first output is generated by a model based on a first neural signal (e.g., collected during a calibration session) and the second output is generated by a model based on a second neural signal (e.g., collected during a validation session).

[0057] However, block 140 may additionally or alternatively be performed in any other suitable manner.

[0058] 3.4 Methods - Authenticated Actions Block S150 includes performing an authorized action based on the authentication status. Block S150 functions to utilize the output of the authentication status determination to perform an action related to the authentication status of the user. Block S150 can also function to generate any suitable output based on the authentication status of the user. Some actions can require the user to be authenticated in order to perform them (e.g., authorized actions), and additionally or alternatively, the authorized actions can include actions personalized to the user or improved by basing the execution of the action on the authentication status (e.g., adjusting environmental actuators proximate to the user according to user preferences determined based on the user being authenticated). However, what is authorized can be any other suitable action and / or output.

[0059] The authenticated action can be performed by any suitable hardware and / or software module (e.g., communicatively coupled to the BCI and / or related components). In a first variant, performing the authenticated action can include rendering a graphic in a virtual head-up display worn by the user in response to determining the authentication status (e.g., rendering a graphic indicating that the option has been selected, as shown in FIG. 13B), and in this variant, determining the authentication status includes determining that the user has selected the option according to a user-specific model of how a user selects an object in a digital environment (e.g., with a button click, etc.). More specifically, in this example, the selection can be associated with communication via a telephone in the digital environment supported by a head-mounted display with input capabilities. In a second variant, performing the authenticated action can include manipulating a physical object (e.g., unlocking an electronic lock on a door the user wants to pass through). In a third variation, performing the authenticated action may include adjusting environmental parameters (e.g., room temperature, lighting levels, vehicle air conditioning fan speed, etc.) corresponding to the physical environment associated with the user via connected environmental parameter actuators (e.g., a thermostat, a connected light switch or lighting controller, a vehicle computer, etc.).

[0060] In a first particular embodiment, block S150 includes adjusting a set of vehicle setting parameters based on the authentication status of the user. The vehicle setting parameters may include radio station presets and other settings related to the user's sound preferences, cockpit setting parameters (e.g., seat position, steering wheel position, rearview mirror position, etc.), environmental setting parameters (e.g., air conditioning temperature, fan speed, window position, etc.), driving setting parameters (e.g., suspension settings, handling responsiveness settings, acceleration sensitivity settings, collision avoidance override sensitivity settings, etc.), and any other suitable vehicle-related settings. In this particular implementation, determining the authentication status of the user includes identifying the user (or verifying the user's claimed identity) and determining user preferences associated with the identified user. Determining the user preferences may include retrieving the user preferences from a database, dynamically monitoring the user's reactions to stimuli and / or other events (e.g., monitoring the user's pleasure and / or disappointment), or otherwise appropriately determining the user preferences.

[0061] In another particular implementation, block S150 includes performing an authorization action associated with the user's activity based on the authentication status. Thus, block S150 can function to grant permission to the user based on one or more criteria or rules defined by an application associated with the methods and systems described herein. In this particular implementation of block S150, the criteria for authorization can be based on an analysis of the user's brain activity such that the user is granted permission to perform the action only if the user's brain activity matches the criteria.

[0062] In connection with analyzing brain activity, block S150 may include determining activity (e.g., associated with spectral content, associated with neural oscillations, associated with evoked potentials, associated with event-related potentials, associated with various frequency bands of activity, associated with combinations of activity, etc.) from various electrode channels associated with various brain regions of the user to determine activity states of various regions associated with various brain states. In variations, the various brain states analyzed may include one or more of the following: vigilance state (e.g., sleep state, alertness level), concentration state (e.g., focused, distracted, etc.), emotional state (e.g., happy, angry, sad, scared, calm, surprised, etc.), mental health state (e.g., depression state, psychotic state, states characterized in a manual of mental health conditions, etc.), neurological health state (e.g., epileptic seizure, migraine, stroke, dementia, etc.), sobriety state, overt / covert attention state, response state to sensory stimuli, cognitive load state, imagery state (e.g., imagery of motor actions, scenes, sounds, procedures, etc.), memory function state, and / or any other suitable brain activity state.

[0063] Additionally or alternatively, the authorized action may be based on non-brain activity state (e.g., ancillary biometric signals, ancillary data, contextual data, etc.). For example, environmental factors (e.g., analysis of environmental threats) and / or equipment state (e.g., whether the user's device is wirelessly connected or otherwise connected to a network) may govern the execution of the authorized action in block S150.

[0064] However, additionally or alternatively, any of the actions in block S150 may be implemented and / or performed in any other suitable manner.

[0065] 3.5 Method - Example Variations of the method may include multiple time-locking components in the calibration session, the validation session, or both, or any other suitable neural signal analysis session. In some examples, the multiple time-locking components may be substantially identical (e.g., identical sound stimuli, identical visual stimuli, etc.). In other examples, the time-locking components may be related but not identical (e.g., sound stimuli with similar sounds in a different order, visual stimuli with similar images in different hue palettes, etc.). In yet another example, the time-locking components may be completely different (e.g., a sound stimulus followed by a video stimulus without sound). However, the method may include any other suitable time-locking components having any suitable relative relationship.

[0066] In an example, the method can include determining an emotional state of a user and alerting an entity associated with the user of the emotional state of the user. In a specific example, the method can include determining that a user is depressed and automatically alerting a mental health professional that the user is depressed.

[0067] In another embodiment, a method for providing a brain-computer interface includes the steps of: detecting a first neural signal of a user in response to a calibration session having a first time-locked component and a first spontaneous component within a first time window, the first time-locked component including a first randomization factor; generating a user-specific calibration model based on the first neural signal; prompting the user to undergo a validation session having a second time-locked component and a second spontaneous component within a second time window, the second time-locked component including a second randomization factor; detecting a second neural signal concurrently with delivery of the validation session; generating an output of the user-specific calibration model from the second neural signal; determining an authentication status of the user based on a comparison operation between a processed output of the second neural signal and an output of the user-specific calibration model; and performing an authenticated operation based on the authentication status.

[0068] In another embodiment, a method for providing a brain-computer interface includes the following steps: detecting a first neural signal of a user in response to a calibration session having at least one of a time-locked component and a spontaneous component within a first time window; generating a user-specific calibration model based on the first neural signal; prompting the user to undergo a validation session within a second time window; detecting a second neural signal concurrently with delivery of the validation session; determining an authentication status of the user based on a comparison operation between a processed output of the second neural signal and an output of the user-specific calibration model; and performing an authenticated action based on the authentication status.

[0069] The method 100 and / or system of the preferred embodiment, and variations thereof, can be embodied and / or implemented at least in part as a machine configured to receive a computer-readable medium storing computer-readable instructions. The instructions are preferably executed by computer-executable components that are preferably integrated with a portion or portions of the system and the processor and / or controller. The computer-readable medium can be stored in any suitable computer-readable medium, such as RAM, ROM, flash memory, EEPROM, optical device (CD or DVD), hard drive, floppy drive, or any suitable device in any suitable format. The computer-executable component is preferably a general or application-specific processor, but alternatively or in addition, any suitable dedicated hardware or hardware / firmware combination device can execute the instructions.

[0070] The figures show the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to preferred embodiments, configuration examples, and variations thereof. In this regard, each block of the flow diagrams or block diagrams may represent a module, segment, or code portion that includes one or more executable instructions for implementing a specified logical function(s). It should also be noted that in some alternative implementations, the functions shown in the blocks may occur in a different order than that shown in the figures. For example, two blocks shown in succession may in fact be executed substantially simultaneously, or the blocks may be executed in the reverse order depending on the functionality involved. It should also be noted that each block of the block diagrams and / or flow diagrams, and combinations of blocks in the block diagrams and / or flow diagrams, may be implemented by a dedicated hardware-based system that performs a specific function or operation, or a combination of dedicated hardware and computer instructions.

[0071] As those skilled in the art will appreciate from a reading of the above detailed description and the drawings and claims, modifications and variations can be made to the preferred embodiments of the invention without departing from the scope of the invention as defined in the appended claims.

Claims

1. 1. A method for authenticating a user using a brain-computer interface, comprising the steps of: receiving input from a user to perform an action using the brain-computer interface; Within the first time window, measuring a first set of neural signals of the user; and providing a first series of sensory stimuli from the brain-computer interface to the user; extracting features from the first set of neural signals that are temporally synchronized with the first series of the sensory stimuli; In response to the input, providing a second series of the sensory stimuli from the brain-computer interface to the user within a second time window; measuring a second set of neural signals of the user during the second time window; and applying a calibration model trained based on the first series of the sensory stimuli and the features to the second set of neural signals to authenticate the user. conducting a validation session comprising: wherein the first series of sensory stimuli includes at least one stimulus that is the same as but different from the second series of sensory stimuli; and In response to authenticating the user, performing the action of the user's input.

2. 10. The method of claim 1, wherein the brain-computer interface is one of a headset or a head-mounted wearable device.

3. 10. The method of claim 1, wherein the second series of sensory stimuli comprises auditory stimuli.

4. 4. The method of claim 3, wherein the auditory stimuli include one or more of a music sample, a speech sample, or a harsh sound sample.

5. 4. The method of claim 3, wherein the auditory stimulus is a sound sample that includes a randomization factor of between 5% and 30% compared to an unmodified version of the sound sample.

6. The method of claim 1 , wherein the step of taking an action comprises providing personalized digital content to the user.

7. 10. The method of claim 1, wherein the taking of an action comprises at least one of delivering digital content and coordinating operation of a user device in coordination with the detection of the user's brain state.

8. 2. The method of claim 1, wherein the step of performing an action includes reading a password into an input box.

9. 10. The method of claim 8, wherein the input box is associated with one of an operating system login page, a website, or an application.

10. A non-transitory computer readable medium comprising instructions that, when executed by a computing system, cause the computing system to perform processes including: receiving input from a user to perform an action using the brain-computer interface; Within the first time window, measuring a first set of neural signals of the user; and providing a first series of sensory stimuli from the brain-computer interface to the user; extracting features from the first set of neural signals that are temporally aligned with the first series of the sensory stimuli; In response to the input, providing a second series of the sensory stimuli from the brain-computer interface to the user within a second time window; measuring a second set of neural signals of the user during the second time window; and applying a calibration model trained based on the first series of the sensory stimuli and the features to the second set of neural signals to authenticate the user. performing a validation session, comprising: performing a calibration session, wherein the first series of the sensory stimuli includes at least one stimulus that is the same as but different from the second series of the sensory stimuli; and In response to authenticating the user, performing the action of the user's input.

11. 11. The non-transitory computer readable medium of claim 10, wherein the brain-computer interface is one of a headset or a head-mounted wearable device.

12. 11. The non-transitory computer-readable medium of claim 10, wherein the second series of sensory stimuli includes auditory stimuli.

13. 13. The non-transitory computer-readable medium of claim 12, wherein the auditory stimuli include one or more of a music sample, a speech sample, or a harsh sound sample.

14. 11. The non-transitory computer-readable medium of claim 10, wherein performing the action comprises reading a password into an input box.

15. 15. The non-transitory computer-readable medium of claim 14, wherein the input box is associated with one of an operating system login page, a website, or an application.

16. at least one processor; and Non-transitory computer-readable medium containing stored instructions A system comprising: The instructions, when executed by the at least one processor, cause the at least one processor to perform processes including: receiving input from a user to perform an action using the brain-computer interface; Within the first time window, measuring a first set of neural signals of the user; and providing a first series of sensory stimuli from the brain-computer interface to the user; extracting features from the first set of neural signals that are temporally aligned with the first series of the sensory stimuli; In response to the input, providing a second series of the sensory stimuli from the brain-computer interface to the user within a second time window; measuring a second set of neural signals of the user during the second time window; and applying a calibration model trained based on the first series of the sensory stimuli and the features to the second set of neural signals to authenticate the user. performing a validation session, comprising: performing a calibration session, wherein the first series of the sensory stimuli includes at least one stimulus that is the same as but different from the second series of the sensory stimuli; and In response to authenticating the user, performing the action of the user's input.

17. 17. The system of claim 16, wherein the brain-computer interface is one of a headset or a head-mounted wearable device.

18. 20. The system of claim 16, wherein the second series of sensory stimuli includes auditory stimuli.

19. 20. The system of claim 18, wherein the auditory stimuli include one or more of a music sample, a speech sample, or a harsh sound sample.

20. 20. The system of claim 16, wherein performing the action includes reading a password into an input box.

Images