Systems, methods, and computer program products for cognitive state assessment

Abstract

A system for assessing a cognitive state of a user, comprising: a hardware output user interface providing user progression information at performing a task; an input user interface controllable by the user for performing the task; wherein response dynamics of the input user interface to user instructions are changed by the system during the performance of the task, thereby resulting in highly fluctuational motion of a body portion; a processor and a memory module including computer program code, collectively configured to cause the system to: receive over at least one hardware communication channel sensor based kinematic data indicative of kinematic parameters of the body portion at different times during the performance of the task; process the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and determine a cognitive state of the user based on the motion patterns.

Description

SYSTEMS, METHODS, AND COMPUTER PROGRAMPRODUCTS FOR COGNITIVE STATE ASSESSMENTFIEED

[0001] The invention related to systems, methods, and computer program products for assessing a cognitive state of a person, and especially to systems, methods, and computer program products for assessing a cognitive state of a person based on analysis of kinematic data of the person during the performance of a task.BACKGROUND

[0002] When a person executes a relatively larger movement, the execution of that movement is accompanied by highly fluctuational motion (such as but not limited to corrective submovements). Such highly fluctuational movements are thought to play a crucial role in motor learning and adaptation. For example, the relationship between corrective submovements and cognition has been studied extensively in the context of motor control and learning. One line of research has investigated the role of cognitive processes such as attention, working memory, and decision-making in the generation and control of corrective submovements. For example, studies have shown that attentional load can affect the size and timing of corrective submovements during a reaching task (Plamondon & Alimi, 1997), and that working memory demands can influence the planning and execution of corrective submovements during a visuomotor adaptation task (Galea & Darian-Smith, 1994). Another line of research has focused on the relationship between corrective submovements and cognitive control processes such as inhibition and cognitive flexibility. For example, studies have shown that inhibition of an ongoing movement can lead to the generation of corrective submovements (Stinear & Byblow, 2003), and that cognitive flexibility can facilitate the adaptation of corrective submovements to changing task demands (Noble, Smout, & Byrne, 2018).

[0003] Beyond the scientific literature there are several patents dealing with the connection between corrective submovements and cognition. In these applications, Hochman (WO2022123568A1; US 11,141,113 B2) presented a methods for analyzing corrective submovements as a means for monitoring a person's cognitive state during daily activities.

[0004] The following publications discuss interrelationship perturbation tasks and resulting corrective submovements, organized by year of publication: Jones and Hunter (1983) investigated the effects of perturbations on reaching movements. They found that sudden perturbations to the reaching hand resulted in corrective submovements that helped to bring the hand back to the target. Soechting and Lacquaniti (1983) conducted a study on the effects of sudden perturbations on the trajectory of reaching movements. They found that the corrective submovements produced were consistent with the predictions of the optimal feedback control model. Ghez and colleagues (1995) examined the role of visual feedback in the production of corrective submovements. They found that when visual feedback was removed, the magnitude and timing of corrective submovements were reduced. Scheidt and colleagues (2000) investigated the effects of perturbations on the adaptation of arm movements to novel force fields. They found that sudden perturbations could disrupt the adaptation process, but that corrective submovements helped to compensate for the perturbations and improve performance. Franklin and Wolpert (2008) proposed a computational model of corrective submovements that incorporated both sensory feedback and feedforward control. The model was able to reproduce many of the key features of corrective submovements observed in experimental studies. Shemmell and colleagues (2010) investigated the effects of perturbations on the coordination of muscle activity during reaching movements. They found that perturbations resulted in changes to the timing and magnitude of muscle activity, consistent with the production of corrective submovements. Kagerer and colleagues (2013) conducted a study on the effects of perturbations on the coordination of hand and eye movements. They found that perturbations resulted in changes to the timing and direction of eye movements, as well as the production of corrective submovements in the hand. Mawase and colleagues (2014) investigated the effects of perturbations on the adaptation of grasping movements to novel object properties. They found that perturbations coulddisrupt the adaptation process, but that corrective submovements helped to compensate for the perturbations and improve performance.

[0005] In order to better appreciate the background of the present invention and therefor its novelty, reference is made to the following collection of scientific articles:

[0006] Desmurget, M., & Grafton, S. (2000). Forward modeling allows feedback control for fast reaching movements. Trends in cognitive sciences, 4(11), 423-431.

[0007] Fishbach, A., Roy, S. A., Bastianen, C., Miller, L. E., & Houk, J. C. (2005). Kinematic properties of on-line error corrections in the monkey. Experimental Brain Research, 164, 442-457.

[0008] Grafton, S. T., & Tunik, E. (2011). Human basal ganglia and the dynamic control of force during on-line corrections. Journal of Neuroscience, 31(5), 1600- 1605.

[0009] Kawato, M. (1999). Internal models for motor control and trajectory planning. Current opinion in neurobiology, 9(6), 718-727.

[0010] Koshland, G. F., & Hasan, Z. (2000). Electromyographic responses to a mechanical perturbation applied during impending arm movements in different directions: one-joint and two-joint conditions. Experimental brain research, 132(4).

[0011] Novak, K. E., Miller, L. E., & Houk, J. C. (2000). Kinematic properties of rapid hand movements in a knob turning task. Experimental Brain Research, 132, 419- 433.

[0012] Novak, K. E., Miller, L. E., & Houk, J. C. (2002). The use of overlapping submovements in the control of rapid hand movements. Experimental brain research, 144, 351-364.

[0013] Prodoehl, J., Corcos, D. M., & Vaillancourt, D. E. (2009). Basal ganglia mechanisms underlying precision grip force control. Neuroscience & Biobehavioral Reviews, 33(6), 900-908.

[0014] Prodoehl, J., Yu, H., Little, D. M., Abraham, I., & Vaillancourt, D. E. (2008). Region of interest template for the human basal ganglia: comparing EPI and standardized space approaches. Neuroimage, 39(3), 956-965.

[0015] Spraker, M. B., Yu, H., Corcos, D. M., & Vaillancourt, D. E. (2007). Role of individual basal ganglia nuclei in force amplitude generation. Journal of neurophysiology , 9S(2), 821-834.

[0016] Tunik, E., Houk, J. C., & Grafton, S. T. (2009). Basal ganglia contribution to the initiation of corrective submovements. Neuroimage, 47(4), 1757-1766.

[0017] Turner, R. S., & Anderson, M. E. (2005). Context-dependent modulation of movement-related discharge in the primate globus pallidus. Journal of Neuroscience, 25(11), 2965-2976.

[0018] Turner, R. S., Desmurget, M., Grethe, J., Crutcher, M. D., & Grafton, S. T. (2003). Motor subcircuits mediating the control of movement.

[0019] Turner, R. S., Grafton, S. T., Votaw, J. R., Delong, M. R., & Hoffman, J. M. (1998). Motor subcircuits mediating the control of movement velocity: a PET study. Journal of Neurophysiology, 80(4), 2162-2176.

[0020] Plamondon, R., & Alimi, A. M. (1997). Speed / accuracy trade-offs in target- directed movements. Behavioral and Brain Sciences, 20(2), 279-303.

[0021] Stinear, C. M., & Byblow, W. D. (2003). Impaired inhibition of a pre-planned response in patients with focal hand dystonia. Neuroreport, 14(2), 151-155.

[0022] Soechting, J. F., & Lacquaniti, F. (1983). Invariant characteristics of a pointing movement in man. Journal of Neuroscience, 3(8), 1574-1588.

[0023] Scheidt, R. A., Dingwell, J. B., & Mussa-Ivaldi, F. A. (2000). Learning to move amid uncertainty. Journal of Neurophysiology, 84(2), 971-985.

[0024] Franklin, D. W., & Wolpert, D. M. (2008). Computational mechanisms

[0025] Flanagan, J. R., & Rao, A. K. (2005). Trajectory adaptation to a nonlinear visuomotor transformation: evidence of motion planning in visually perceived space. Journal of neurophysiology, 94(1), 460-468.

[0026] Pruszynski, J. A., Kurtzer, I., & Scott, S. H. (2011). Rapid motor responses are appropriately tuned to the metrics of a visuospatial task. Journal of Neuroscience, 31(47), 16620-16629.

[0027] In light of the cited prior art, there remains a need for novel systems, methods, and computer program products for cognitive state assessment.GENERAL DESCRIPTION

[0028] This disclosure describes various systems, methods, and computer program products which can analyze motion patterns during highly fluctuational motion by a person while performing a task during which task the response dynamics of a user interface used by the user are changed. The results of such an analysis of motion patterns during the highly fluctuational motion of the person are an assessment of a cognitive state of the user, which in turn can be further utilized to select which followup action should be taken, considering the detected cognitive state. Tasks in which the response dynamics of the user interface are changed during the performance of the task are also referred below to as “perturbation tasks” or “viscosity tasks”. For example, perturbation may be induced during the task by frequent changes in the amount of force the person is required to exert in order to follow a moving target on a monitor. The proposed analysis enables assessment of the cognitive state of the person (e.g., detecting fatigue, intoxication, or motion sickness) based on very slight motion (e.g., submovements, corrective submovement, highly fluctuational motion) of one or more body parts of that person.

[0029] The system uses kinematic data collected sensors located either on the person's body, in electronic devices that come in contact with that person, or which can otherwise detect movement of the person (e.g., camera). This kinematic data may be collected in real-time during the perturbation task and may be stored for later analysis. Optionally, the processing of the kinematic data may take place during the perturbation task, in which case the perturbation itself may be induced based on results of such processing.

[0030] The collected kinematic data (possibly after preprocessing procedures) is then processed by a suitable model. Such models may be based, for example, on machine learning, on heuristic or rule -based algorithms, on combination of both, or on any other suitable algorithm. The systems and methods infer the cognitive state of the person by analyzing (a) the kinematic data corresponding to the highly fluctuational motion of the respective body part that occur when the person is required to perform a task (e.g.,follow a moving target) and (b) the dynamics of how the person should operate the user interface (e.g., the amount of force they are required to exert) changes during the performance. For example, such highly fluctuational motion may include subtle changes in body angles, angular velocities, and angular accelerations that are not visible to the eye.

[0031] Generally, “cognitive assessment” refers to the process of evaluating an individual's cognitive abilities, such as memory, attention, language, perception, and problem-solving skills. It has been an important aspect of neuropsychology, especially in the diagnosis and treatment of neurocognitive disorders such as dementia, traumatic brain injury, and attention deficit hyperactivity disorder (ADHD). In recent years, with the advancement of technology and the widespread use of smartphones, there has been a surge in the development of computerized cognitive assessment applications (e.g., smartphone apps). Such apps aim to provide an efficient and accessible way to assess a person's cognitive state.

[0032] There are two main categories of smartphone applications designed to assess a person's cognitive state: clinical and self-administered. Clinical apps are typically used by healthcare professionals to administer standardized cognitive assessments, while self-administered apps are designed for individuals to assess their cognitive state on their own. Clinical apps are commonly used by neuropsychologists, rehabilitation specialists, and other healthcare professionals to diagnose and monitor cognitive impairments. These apps typically include well-established and validated cognitive assessments, such as the Montreal Cognitive Assessment (MoCA) and the MiniMental State Examination (MMSE). The results obtained from these assessments can provide valuable information about an individual's cognitive state and support the diagnosis of neurocognitive disorders. Self-administered apps, on the other hand, are designed for individuals to assess their cognitive state on their own. These apps typically include a variety of cognitive tests that assess various aspects of cognitive function, such as memory, attention, and reaction time. Some popular selfadministered cognitive assessment apps include Brain Test, Brain Fitness Pro, and Luminosity.

[0033] While smartphone applications for cognitive assessment have many advantages, there are also some limitations that need to be considered. One of the mainlimitations is the lack of standardization in terms of the tests included in these apps and the validity of the results obtained. Many self-administered apps may not have undergone the same level of rigorous scientific testing and validation as clinical apps, and the results obtained may not be as reliable. Another limitation is the potential for user bias. Self-administered cognitive assessment apps may not accurately reflect an individual's true cognitive state if the user is motivated to manipulate the results or if the user is not paying attention while completing the assessments.

[0034] In conclusion, computerized cognitive assessment applications, including smartphone apps, have the potential to provide a convenient and accessible way to assess a person's cognitive state. However, it is important to consider the limitations and limitations of these apps, especially when using self-administered apps. Healthcare professionals should continue to play a vital role in the assessment and diagnosis of neurocognitive disorders, and individuals should be cautious about relying solely on self-administered apps for a comprehensive evaluation of their cognitive state.

[0035] Some of the most commonly used indices in computerized tests for assessing cognitive state include: a. Reaction Time (RT): Reaction time measures the amount of time it takes a person to respond to a stimulus. It is often used to evaluate a person's attention, processing speed, and decision-making abilities. RT is often expressed in milliseconds and is calculated as the time elapsed between the presentation of the stimulus and the response. b. Error Rate: Error rate is the number of incorrect responses made by a person during a test. It is often used to evaluate a person's attention, memory, and decision-making abilities. c. N-Back Test: The N-Back test measures working memory and attention. In this test, a person is presented with a series of stimuli and must indicate whether the current stimulus matches the one presented "n" stimuli ago. d. Digit Span Test: The Digit Span test measures a person's short-term memory. In this test, a person is presented with a sequence of numbers and must repeat the sequence in the same order.e. Stroop Test: The Stroop test measures a person's attention and processing speed. In this test, a person is presented with words and must indicate the color of the ink in which the word is written. f. Continuous Performance Test (CPT): The Continuous Performance Test (CPT) measures a person's attention and reaction time. In this test, a person is presented with a series of stimuli and must respond to specific targets while ignoring others.

[0036] These indices provide a comprehensive evaluation of a person's cognitive function and can be useful in the diagnosis of various neurological conditions. The disclosed systems and methods, however, utilize kinematic data of the person, which is not used in these indices. Especially, motion patterns during this highly fluctuational motion duration (such as but not limited to corrective submovements) of the person may be analyzed to assess the cognitive state of the person.

[0037] An example of how perturbation tasks may be used by the systems and methods below for assessment of cognitive state of the person is to give the person a visually guided task of moving an object (real-life object or a computer-generated object such as a cursor) while the response dynamic of the control user interface are changed frequently during the task. Even in ideal circumstances, controlling visually guided behavior demands real-time adjustments to correct performance errors. These real-time adjustments can be seen at the beginning of a movement, indicating that they are based on an internal representation of expected movement dynamics (Kawato, 1999; Desmurget and Grafton, 2000). One way of detecting early error correction is by analyzing submovements in the kinematic trace during one-dimensional target capture, submovements may be identified, for example, as separate peaks in the velocity trace (Novak et al., 2000, 2002) and modeled as small bell-shaped velocity profiles that appear either on top of or after the main movement (Fishbach et al., 2005). Neural recordings from monkeys performing one-dimensional target capture tasks reveal a modulation of activity in the internal segment of the globus pallidus (GPi) during submovement generation (Roy et al., 2008). In the case of a one-dimensional target capture task in unpredictable viscous loads, human imaging studies have shown a correlation between putamen activity and the number of submovements per trial (Tunik et al., 2009).

[0038] Meanwhile, functional imaging provides strong evidence that the human basal ganglia (BG) play a critical role in controlling several kinematic properties, including movement velocity and amplitude in the putamen and GPi (Turner et al., 1998, 2003; Spraker et al., 2007), force duration in the putamen (Vaillancourt et al., 2004; Prodoehl et al., 2008), and force rate of change and amplitude in the GPi and subthalamic nucleus (STN) (Spraker et al., 2007; Prodoehl et al., 2008). It is worth mentioning that behavioral experiments reveal that both agonist and antagonist muscles can increase grip force in response to mechanical disturbances, also coined perturbations (Koshland and Hasan, 2000; Spraker et al., 2007), including those used in one-dimensional target capture.

[0039] A study by Grafton & Tunik (2010) presented evidence that the basal ganglia is only active during production of corrective submovements in tasks involving a viscous challenge, where subjects were required to move a cursor on a computer screen using a joystick and a torque motor applied resistive forces that varied unpredictably in severity from trial to trial. The basal ganglia were not active during production of corrective submovements in tasks involving a target challenge where the motor was turned off resulting in consistent reactive grip forces. The difficulty in these trials was increased by unpredictably varying the target width from trial to trial.

[0040] Indeed, the basal ganglia (BG) play a crucial role in error monitoring, which is the ability of the brain to detect and respond to performance errors. In particular, the error-related negativity (ERN) is a brain wave that is often measured in the event- related potential (ERP) to study error monitoring. The ERN is a negative shift in the EEG signal that occurs approximately 50 milliseconds after an error is committed. The basal ganglia are essential for the generation of the ERN and has been shown to reflect the processing of errors and the subsequent adjustment of behavior to correct for the error. In summary, the basal ganglia are an important part of the brain's error monitoring system, and the ERN is a key measurement of error processing and correction. Recent studies suggest that the basal ganglia are only active when submovements are produced in order to overcome a viscous challenge.

[0041] Various factors altering a person’s cognitive and emotional state can impact the ERN and influence error processing. Alcohol, fatigue, inattention, drugs, and diseases such as Parkinson’s disease and anxiety are examples of such factors that havebeen shown to affect the ERN. Alcohol consumption has been shown to impair error monitoring, as evidenced by a reduction in the ERN. Fatigue can also impact error processing, as it leads to a decrease in the ERN and a reduction in the ability to detect and respond to errors. Inattention can also negatively impact the ERN, leading to a reduction in the ability to detect and respond to errors. Drug use, such as the use of stimulants, can also impact the ERN. Stimulants have been shown to increase the ERN, indicating an enhancement in error processing. However, excessive use of stimulants can lead to an overexcitation of the brain and eventually impair error processing. Patients with Parkinson's disease exhibit alterations in their ERN response compared to healthy individuals, suggesting that the ERN may be a potential marker for cognitive and motor impairments in Parkinson's disease. Also, individuals with high anxiety tend to show increased ERN amplitudes in response to errors, indicating an abnormal neural signal for error detection and correction.

[0042] If so, it is possible to use ERN fluctuations in order to identify a person's cognitive and emotional state. Since the ERN is recorded using an electrode cap placed on the head, its use for cognitive assessment purposes is limited. However, since the basal ganglia are responsible for both the production of the ERN and the production of corrective submovements, the systems in methods discussed below utilize motion sensors (or, more generally, motion indicative sensors) in order to detect motion patterns within the highly fluctuational motion that correspond to fluctuations in the ERN. Such sensors may be located on the human body or in the devices that the person operates, or otherwise measure movements of one or more body parts of the person.

[0043] The systems and methods discussed below in greater detail are based on analysis of motion patterns during highly fluctuational movements of one or more body parts which are produced during tasks involving a viscous challenge. In a research conducted by the inventors, the inventors parametrically manipulated the amounts of alcohol, and the fatigue levels of the subjects. The subjects were then given tasks of two types: the first task involved a target challenge where the difficulty in trials was increased by unpredictably varying the target location, velocity or size; the second task involved a viscous challenge, where the amount of pressure that must be applied to the touch screen in order to move the cursor has been changed from time to time, or the ratio between the amplitude of the movement performed and the movementof the responding cursor has been changed. The inventors found that only in the second task, analysis motion patterns during such highly fluctuational motion allowed them to accurately detect the subject blood alcohol or fatigue.

[0044] It is important to note that there are a variety of situations in which viscosity can be created depending on the body part in focus and depending on the medium in which the task is conducted. Viscosity manipulation can be produced in any situation where control of the medium is achieved using one or more body parts of the individual. For example, when controlled of the medium is achieved one the person's hands, eyes, head, or even whole body are moved.

[0045] Few examples of ways in which a person (also referred to as “user”) may attempt to produce a change in the medium using a provided input user interface (input UI) include manually operating a joystick, a computer mouse, a game console, or a remote control. In such a case, the motion sensors which collect the kinematic data (which is used for detecting and analyzing motion patterns (such as during the aforementioned highly fluctuational motion) for determining cognitive state of the person) may be included in the device that the subject operates, on his body, or having the person within their field of detection (e.g., FOV of a camera). Some other ways in which the person may attempt to produce a change in the medium include moving the head or the eyes of the person, for example in a virtual reality game. In such cases, sensors can be placed, for example, in a helmet or dedicated glasses. In other cases, the movements can be tracked using radar or a camera. Another type of UI which may be used by the user is providing the user with feedback from a cursor, object, or target which the subject is required to control with his EEG or Autonomous activity.

[0046] It is noted that optionally, the person may create a change in the medium with no direct movement initiation of the person involved. This may be achieved, for example, using tasks in which the manipulation of the medium will be done by the subject through control of the breathing rate or heart rate or any other characteristic of the autonomic nervous system, as is often done in biofeedback processors. Some other tasks may involve manipulation of the medium by the person through control of the electrical activity of the person’s brain as is often done in neurofeedback processors. In these cases, the measurement of the highly fluctuational motion (especially following the change of the response dynamics of the input user interface) will be doneby analyzing the activity of the muscles affected by the autonomic or neurological reactions. For example, the movement of the muscles of the chest or the diaphragm or the movement of the eyes or the head.

[0047] As discussed below in greater detail, there are a variety of different ways to process the highly fluctuational motion (which may or may not include calculating submovements, e.g., following the change of the response dynamics of the input user interface), which may include at least two steps. One of these steps includes identifying a beginning and an end of a primary movement. A second one of these steps includes identifying and / or labeling various speed derivatives (e.g., acceleration, jerk) which appear within the primary movement (from beginning to end, and sometime slightly after the end). In many cases the primary movement is a sharp movement in one direction, while in other cases the primary movement may be a continuous and very long movement. In the latter case, the movement along its entire length can be defined as a primary movement. At the end of the process various calculations are made on the speed derivatives.

[0048] The processing of the kinematic data of the identified submovements (or other identified motion patterns within the measurement period) for identifying a cognitive state, may be achieved using various computational and algorithmic approaches, such as using heuristic algorithms, machine learning algorithms, or combination of both. It is noted that different methods of machine learning and / or different methods of heuristic algorithms may be used in synergy, in order to identify unique features in the highly fluctuational motion that respond to differences between different cognitive states. One way of collecting reference data (as well as training data for machine learning algorithms, if used) is to implement parametric manipulation of the cognitive state, for example by giving different amounts of alcohol to the subjects or by inducing different levels of fatigue in the subjects. The machine learning model (or neural networks) may than then run on the data. Such a model may be specifically guided to explore the motion patterns during the highly fluctuational motion occurring within the primary movement, especially following changes to the response dynamics of the input user interface.

[0049] One example scenario in which such systems, methods, and computer program products may be used is for fleets managers (for any suitable type of vehicles,such as cars, trucks, scooters, heavy machinery) to verify, before permitting a user to drive a vehicle of the fleet, whether the intended driver is capable of driving. The verification in such a case should be quick in order to be commercially viable (and in some cases, to be safe). Nevertheless, since the verification process in such a scenario takes place prior to the beginning of the drive, it can include tasks which cannot be performed during driving (during which time the driver needs to concentrate on the driving itself). The systems, methods, and computer program products disclosed herein may utilize simple tasks (e.g., bringing a cursor to a target on a screen), which may be implemented more cheaply, require less computational power, and be more user friendly. Furthermore, the selected tasks do not need to involve specialty input user interface or user output interface, and optionally also not include dedicated sensors (e.g., using a preinstalled camera, or sensing capabilities of the touchscreen). It is nevertheless noted that more complicated tasks as well as tasks which include dedicated user interface may also be implemented if needed.

[0050] In some commercially viable scenarios, the task which is used to determine the cognitive state of the user should be very short (e.g., seconds and not minutes). In such cases, the implemented task should generate a very fluctuational motion (e.g., many submovements) in a relatively short time. Some of the tasks discussed below have this unique quality of enabling inducing of highly fluctuational motion (e.g., many submovements) in a very short time. Some of the ways a task can be used to create such highly fluctuational motion include either creating a sudden change in the target that the subject is required to reach, or using perturbation tasks, in which the target stays constant (location, shape, etc.), but the medium through which the movement occurs changes (esp. viscosity). The proposed systems, methods, and computer program products may use only perturbation tasks, or a combination of perturbation tasks and other tasks for the assessment of the cognitive state.

[0051] According to an aspect of the invention, there is disclosed a system for assessing a cognitive state of a user by monitoring execution of a task by the user, the system including: a. a hardware output user interface for providing to the user information indicative of progression of the user at performing the task;b. an input user interface controllable by the user for performing the task; the response dynamics of the input user interface to user instructions are changed by the system during the performance of the task, thereby resulting in highly fluctuational motion of at least one body part of the user; c. at least one processor; and d. at least one memory including computer program code, where the at least one memory and the computer program code are configured, with the at least one processor, to cause the system to at least: (i) receive over at least one hardware communication channel sensor based kinematic data, the kinematic data being indicative of kinematic parameters of the at least one body part at different times during the performance of the task; (ii) process the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and (iii) determine a cognitive state of the user based on the motion patterns.

[0052] According to a further aspect of the invention, the sensor may include at least one sensor selected from the group consisting of: a camera, an accelerometer, a RADAR, a touch screen, and a joystick.

[0053] According to a further aspect of the invention, the system may include a cognitive state-gated access authorization module which is operable to selectively grant users control of a requested system based on cognitive state assessment, and the cognitive state-gated access authorization module is configured to: (i) permit control by the user of the requested system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, (ii) repeat the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and (iii) prevent control by the user of the requested system in response to the determination of the second cognitive state.

[0054] According to a further aspect of the invention, the processor may be configured to determine the cognitive state of the user without accessing historical performance information of the user.

[0055] According to a further aspect of the invention, the input user interface may be further used to receive user instructions for operating of a machine to which access is selectively granted by the system based on the determined cognitive state.

[0056] According to a further aspect of the invention, the kinematic data may be indicative of motion patterns originating from a basal ganglia of the user which is stimulated by the performance of the task.

[0057] According to a further aspect of the invention, the processor may be configured to determine real-time modification parameters for modifying the response dynamics of the input user interface in response to the kinematic data.

[0058] According to a further aspect of the invention, the system may be a vehicle which includes: (a) an engine operable to provide power for propelling the vehicle, (b) a user controllable steering mechanism for controllably changing a propagation direction of the vehicle, and (c) input user interface for detecting user instructions for modifying a behavior of at least one module out of the engine and the steering mechanism; the processor in such case may be operable to selectively prevent controlling of performance of the at least one module by the input user interface based on the determined cognitive state.

[0059] According to a further aspect of the invention, the at least one memory and the computer program code may be configured, with the at least one processor, to cause the system to at least: process the kinematic data to identify submovements of the user during the performance of the task, and determine the cognitive state of the user based on analysis of the submovements.

[0060] According to an aspect of the invention, there is disclosed a method for assessing a cognitive state of a user, the method including: (i) receiving, over at least one hardware communication channel, sensor based kinematic data that is indicative of kinematic parameters of at least one body part of the user at different times during a performance by the user of a task that includes operating an input user interface whose response dynamics to user instructions change during the performance of the task, thereby resulting in highly fluctuational motion of the at least one body part; (ii) processing the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and (iii) determining a cognitive state of the user based on the motion patterns.

[0061] According to a further aspect of the invention, the determining may be followed by: (i) permitting controlling by the user of another system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, (ii) repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and (iii) preventing control by the user of the other system in response to the determination of the second cognitive state.

[0062] According to a further aspect of the invention, the determining of the cognitive state of the user may be based on decision model which is agnostic to historical performance information of the user.

[0063] According to a further aspect of the invention, the method may include selectively permitting the user to operate a vehicle via a plurality of input user interfaces which exclude any input user interface used for the performance of the task.

[0064] According to a further aspect of the invention, the method may include selectively permitting the user to operate a vehicle via a plurality of input user interfaces which includes the input user interface used for the performance of the task.

[0065] According to a further aspect of the invention, the kinematic data may be indicative of highly fluctuational motion originating from a basal ganglia of the user which is stimulated by the performance of the task.

[0066] According to a further aspect of the invention, the method may include modifying the response dynamics of the input user interface in response to the kinematic data.

[0067] According to a further aspect of the invention, the processing may include processing the kinematic data to identify submovements of the user during the performance of the task; the determining in such case may include determining a cognitive state of the user based on analysis of the submovements.

[0068] According to an aspect of the invention, there is disclosed a non-transitory computer-readable medium for assessing a cognitive state of a user, including instructions stored thereon, that when executed on a processor, perform the steps of:(i) receiving, over at least one hardware communication channel, sensor based kinematic data that is indicative of kinematic parameters of at least one body part of the user at different times during a performance by the user of a task that includes operating an input user interface whose response dynamics to user instructions change during the performance of the task, thereby resulting in highly fluctuational motion of the at least one body part; (ii) processing the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and (iii) determining a cognitive state of the user based on the motion patterns.

[0069] According to a further aspect of the invention, the instructions may further include instructions for execution after the determining of: (i) permitting controlling by the user of another system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, (ii) repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and (iii) preventing control by the user of the other system in response to the determination of the second cognitive state.

[0070] According to a further aspect of the invention, the instructions may include instructions for determining the cognitive state of the user may be based on decision model which is agnostic to historical performance information of the user.

[0071] According to a further aspect of the invention, the instructions may include instructions for selectively permitting the user to operate a vehicle via a plurality of input user interfaces which exclude any input user interface used for the performance of the task.

[0072] According to a further aspect of the invention, the instructions may include instructions for selectively permitting the user to operate a vehicle via a plurality of input user interfaces which includes the input user interface used for the performance of the task.

[0073] According to a further aspect of the invention, the kinematic data may be indicative of highly fluctuational motion originating from a basal ganglia of the user which is stimulated by the performance of the task.

[0074] According to a further aspect of the invention, the instructions may include instructions for modifying the response dynamics of the input user interface in response to the kinematic data.

[0075] According to a further aspect of the invention, the instructions may include instructions for processing the kinematic data to identify submovements of the user during the performance of the task; the determining in such case may include determining a cognitive state of the user based on analysis of the submovements.BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to understand the invention and to see how it may be carried out in practice, embodiments will now be described, by way of non-limiting examples only, with reference to the accompanying drawings, in which:

[0077] Figs. 1A and IB are a functional block diagram illustrating examples of a system for assessing a cognitive state of a user;

[0078] Fig. 2 illustrates an example of a system for assessing a cognitive state of a user;

[0079] Fig. 3 illustrates an example of a perturbation task and user interface for performing of the task;

[0080] Fig. 4 illustrates examples of change patterns of the response dynamics of an input user interface over time;

[0081] Fig. 5 is a flow chart illustrating a method for cognitive state assessment;

[0082] Fig. 6 illustrates a sub-method for identify submovements of one or more body parts of the user following a modification of the response dynamics;

[0083] Figs. 7A and 7B illustrate measurements of movement of a body part of the user in different cognitive states and subject to changes in response dynamics of the user interface;

[0084] Fig. 8 illustrates a method for assessing a cognitive state of a user;

[0085] Fig. 9 illustrates a method for selectively granting control of an asset based on assessment of a cognitive state of a user; and

[0086] Fig. 10 illustrates an examples of kinematic data and analysis of motion patterns in that kinematic data, in accordance with examples of the presently disclosed subject matter.

[0087] It will be appreciated that for simplicity and clarity of illustration and description, certain elements in the figures may not have been drawn to scale. This could include the exaggeration of certain element dimensions relative to others. Additionally, corresponding or analogous elements may be identified using repeated reference numerals in the figures.DETAILED DESCRIPTION OF EMBODIMENTS

[0088] In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. While specific details are provided to enable a thorough understanding of the invention, those skilled in the art will appreciate that the invention may be practiced without these details. Additionally, well-known methods, procedures, and components are not described in detail to avoid obscuring the invention. Finally, any reference to a method, system, or non-transitory computer readable medium should be interpreted as including related aspects of the invention.

[0089] The terms “computer”, “processor”, and “controller” should be expansively construed to cover any kind of electronic device with data processing capabilities, including, by way of non-limiting example, a personal computer, a server, a computing system, a communication device, a processor (e.g. digital signal processor, DSP), a microcontroller, a field programmable gate array (FPGA), cloud computing server, an application specific integrated circuit (ASIC), a smartphone, an electronic control unit (ECU) of a vehicle, an and so on. Unless stated otherwise, the terms “computer”, “processor”, and “controller” may also include a combination of several modules (e.g., several central processing units, CPUs), which operate together toward a goal. Unless specifically stated otherwise, as apparent from the following discussions, it is appreciated that throughout the specification discussions utilizing terms such as "processing", "calculating", “computing”, "determining", "generating", “setting”, “configuring”, “selecting”, “defining”, or the like, include actions and / or processes of a computer that manipulate and / or transform data into other data. That data isrepresented as physical quantities, e.g., such as electronic or electromagnetic quantities, and / or said data representing physical objects.

[0090] It is appreciated that certain features of the presently disclosed subject matter, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the presently disclosed subject matter, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination. In embodiments of the presently disclosed subject matter one or more steps illustrated in the figures may be executed in a different order and / or one or more groups of steps may be executed simultaneously. The figures illustrate a general schematic of the system architecture in accordance with an embodiment of the presently disclosed subject matter. Each module in the figures can be made up of any combination of software, hardware and / or firmware that performs the functions as defined and explained herein. The modules in the figures may be centralized in one location or dispersed over more than one location.

[0091] Any reference in the specification to a method should be applied mutatis mutandis to a system capable of executing the method. Any reference in the specification to a method which can be executed by a computer should be applied mutatis mutandis to a non-transitory computer readable medium that stores instructions that once executed by a computer result in the execution of the method. All the details, variations, optional features, optional steps which are discussed with respect to a system are also applicable, mutatis mutandis, to such a corresponding method (and non-transitory computer readable medium, where applicable), and vice versa.

[0092] Fig. 1A is a functional block diagram illustrating an example of system 200, in accordance with the presently disclosed subject matter. System 200 is a system for assessing a cognitive state of a user by monitoring the execution of a task by the user. System 200 may use one or more different types of tasks in order to assess the cognitive state of the user. In at least one type of task which is used by system 200, the response dynamics of an input user interface which is used by the user for the performance of the task are modified (e.g., by system 200) during the performance of the task.

[0093] Within the context of the present disclosure, the term "cognitive state" should be expansively construed to include various neurological conditions which limits an ability of a user to perform an activity which the user is otherwise capable of performing (e.g., driving a car, operating a machine, babysitting a child, perceiving video context). Such cognitive states include, but are not limited to, tiredness, drowsiness, fatigue, distraction, confusion, stress, anxiety, intoxication, acute physical condition, exhaustion, attention withdrawal, inattention, neurodegenerative disorders, movement-related ailments, synaptic dysfunction conditions, basal ganglia-related disorders, motor system pathologies, and other factors that could impair the user's cognitive abilities. It should be noted that a system or a method according to the present invention may only detect some of these exemplary cognitive states, or even none (instead detecting other cognitive states). The cognitive states which are detectable by the present invention are characterized by distinctive corresponding brain states of the user, which cause the user to act in a detectably distinct fashion (e.g., having distinct motor abilities).

[0094] The term “task” in the context of the present disclosure pertains to an activity that the user attempts to perform which has an at least partly predictable goal and / or an at least partly predictable progression (e.g., path, pace). A system or a method according to the present teaching may utilized any suitable combination of the following types of tasks: a. Explicit task which is presented to the user explicitly (e.g., bring the cursor to a specific location on a monitor); b. Implicit task which is triggered implicitly (e.g., increasing a volume of the audio of an infotainment system such that the user is led to use the controls of the system to lover the volume); and c. A task which is initiated by the user, and whose beginning may be detected or predicted by the system (for example, when providing the user with a standard touch screen, it is expected that the user will attempt to zoom-out a display on the touchscreen using a pinching gesture; in another example, using a computer a user is expected to try to drag a file or an object across the screen using a mouse. Perturbation may optionally be inserted during such task, upon detecting that the user started such a task).

[0095] Herein below the definite phrase “the task” is referred to a task which is used to assess the cognitive state of a user and during which the response dynamics of an input user interface which is used by the user for the performance of the task are modified. A system may utilize one or more different types of tasks for assessing one or more cognitive states of the user using user performance in tasks in which the response dynamics of one or more input user interfaces which are used by the user for the performance of the respective task are changing.

[0096] System 200 includes one or more hardware output user interface 230 (also referred to as “output interface 230”, “interface 230”, “user interface 230”, “output UI 230”, and “feedback interface 230”) for providing to the user information indicative of progression of the user at performing the task (and possibly any other type of data, in addition). Output UI 230 referred to in this disclosure enables the user to interact with system 200, and it can be any suitable type of output interface that provides feedback to the user. This may include visual, auditory, tactile, or any other suitable form of feedback. Some examples of possible visual output interfaces include: computer monitor, touch screen, led light, led lights array, dial, and so on. Such a visual output interface could provide information by any suitable type of visibly discernable data such as text, image, video, light, color, position and orientation of components (e.g., dials, clock hands, needles), and so on. Some examples of possible auditory output interfaces include: speaker, buzzer, beeper, acoustic transducer (e.g., bell, a drum) which can be used to play sounds or voice prompts to guide the user. Some examples of possible tactile output interfaces include: vibration device, haptic feedback device, moving lever, moving apparatuses, changing texture feedback device, force feedback device, vibrotactile device, and so on. In the illustration, one visual output UI 230 is exemplified (represented by a screen), as well as one auditory output UI 230 (represented by headphones), and one tactile output UI 230 (embedded in the seat in the illustrated example). It is noted that the actual system 200 may include any combination of one or more output UIs 230 of any one or more types (e.g., of the one specified above). An example of a task which uses a monitor as its output UI 230 includes bringing a cursor to a target which is indicated on a screen, changing a position of a lever until two sound frequencies converge, and so on. Output UI 230 may optionally include any combination of software and / or firmware in addition to hardware. It is noted that a plurality of output UIs 230 may be used to present to theuser information required for the performance of the task (e.g., both audial and visual data).

[0097] System 200 includes input user interface 240 (also referred to as “input interface 240”, “interface 240”, “input UI 240”, and “user interface 240”) which is controllable by the user for performing the task. While not necessarily so, the same input UI 240 may be used for other uses, such as controlling a machine or another system. Input UI 240 may include any combination of one or more of hardware, software, and firmware. Input UI 240 referred to in this disclosure enables the user to interact with system 200, and it can be any suitable type of input interface that collects data from the user. This may include visual, auditory, tactile, electromagnetic, or any other suitable form of input data or instructions. Input UI 240 may be manipulated by touch (e.g., mouse, joystick) or remotely (e.g., camera, radar, microphone). Input UI 240 may include active instructions by the user (e.g., mouse movement, touchscreen, voice commands) or more passive forms of controlling the input UI, such as biofeedback. Some examples of possible input UIs 240 include: touch screen, mouse, joystick, camera (e.g., observing movement of the user or of one or more body parts thereof such as hand, head, eyes), radar (same), pressure sensors, accelerometers, EEG or Autonomous activity sensing system. Referring to the aforementioned tasks of bringing a cursor to a target and making sound frequencies converge, the control of the cursor or the modification of the sound frequencies (or of any other controllable aspect of the task) may be achieved by the user using any suitable type of input UI 240, such as a joystick, eye movements tracker, touch screen, breathing monitoring, and so on. Input UI 240 may optionally include any combination of software and / or firmware in addition to hardware. It is noted that a plurality of input UIs 240 may be provided to the user in order to control the performance of the task (e.g., both touch screen and speech recognition).

[0098] It is noted that the UI input interface 240 may be the same UI which is used for routine operation of an asset whose use is restricted by system 200 (e.g., vehicle, another machine, computerized system), but this is not necessarily so. For example, input UI 240 may be a steering wheel, pedals, or a touchscreen of an infotainment system of a car (or other vehicle) whose use is subjected to the determination of cognitive state by system 200. Referring to the methods discussed below (500, 600,900), it is noted that any of these methods may optionally include selectively permitting the user to operate a vehicle via one or multiple input user interfaces which includes the input user interface used for the performance of the task.

[0099] Alternatively, a different input UI 240 may be selected, which is not used for the control of the respective vehicle, machine, or system. Such an input UI 240 may nevertheless be incorporated into the respective vehicle, machine, or system (e.g., as a standalone UI), or be independent thereof (e.g., a smartphone of the user, a UI of a rental station of electric vehicles like bikes or scooters, also referred to as “docking station). Referring to the methods discussed below (500, 600, 900), it is noted that any of these methods may optionally include selectively permitting the user to operate a vehicle via a plurality of input user interfaces which exclude any input user interface used for the performance of the task.System 200 also includes one or more communication channels 212 for receiving sensor based kinematic data which is based on information collected by one or more sensors 210. The sensor based kinematic data may be received directly from the respective one or more sensors 210, but may also be preprocessed by another system (e.g., by a computer installed in the vehicle in which system 200 operates). The kinematic data received over the at least one communication channel 212 is indicative of kinematic parameters of a body portion of the user (also referred to in this disclosure as “at least one body part of the user”) at different times during the performance of the task (e.g., eyes, hands, thighs, lungs). Communication channels 212 may be wired or wireless, e.g., depending on the nature of the sensors 210 and their proximity to a processor 220 of system 200. Optionally, communication channels 212 may provide for continuous communication with the sensors 210 (and / or with the pre-processing intermediary system); alternatively (or additionally), ad-hoc communication may be used. Sensors 210 themselves may be part of system 200, or they may be external thereto, depending on the particular application of system 200. Regardless of the nature of sensors 210 or the communication channels 212 used to receive information from them, system 200 is designed to process and analyze the sensor based kinematic data, and to use that data to drive actionable insights and decisions. Optionally, communication channels 212 may be designed to enable the transfer of the sensor based kinematic data in real-time, ensuring that the information received is accurateand up-to-date. Optionally, system 200 may be designed to be flexible, allowing for the integration of additional sensors or modification of existing sensors as required. The communication channels can be customized to meet the needs of the application, and may be adapted to accommodate different types of sensors and data formats. The information received from the sensors can be analyzed and processed by the system in order to provide useful insights and inform decision-making. Overall, the system provides a powerful tool for monitoring and analyzing data from a range of different sources, allowing for more informed and efficient decision-making in a variety of applications. It is noted that any suitable type of sensor 210 may be implemented for detecting kinematic information pertaining to one or more body parts of the user. For example, system 200 may utilize (and optionally include) one or more sensors 210 selected from the group consisting of: a camera, an accelerometer, a RADAR, a touch screen, a joystick, gyroscope, magnetometer, a computer, or computer software responding to user action such as moving a mouse, etc.

[0100] Optionally, any two or three functionalities of sensor 210, input UI 240, and output UI 230 may be combined into a single device. For example, a touch screen implements functionalities of both input and output UI, and possibly of sensing movement of a body part of the user (e.g., finger). A joystick may combine the functionalities of input UI and sensor, and so on.

[0101] As aforementioned, during the task which is used for assessing the cognitive state of the user, the response dynamics of at least one input user interface are changed during the performance of the task. The change may be initiated by system 200 (e.g., by a processor thereof), but may optionally coincide otherwise with the performance of the task (e.g., may be native to the user interface which may be dedicated to this assessment task). Changing in the response dynamic of a user interface means that the responsiveness of that user interface to user instructions is changed, e.g., to a different degree, rate, and so on. For example, the response dynamic of a joystick may change if the number of pixels in which a cursor is moved across a screen when the joystick tilts at a certain angle from its resting position changes (e.g., from N pixels / second to 2-N or 0.37 -N, when the joystick tilts N degrees). In another example, a pressuresensitive touchscreen may change a range of pressures in which it responds to movements of the user’s finger across the touchscreen, stopping responding topressures which in which it was previously responsive during the duration of the task. Especially, the response dynamics of the at least one user interface may result in highly fluctuational motion of the at least one body part. Especially, the response dynamics of the at least one user interface may be modified in a way which is known to result in highly fluctuational motion of the at least one body part. It is noted that the duration of the fluctuations may depend on various factors, e.g., such as (although not limited to) the characteristics of response dynamics of the at least one input user interface which are changed during the performance of the task, the characteristics of the task to be performed, etc. For example, the duration of the fluctuations may depend on a length of the movement which must be performed to complete the task, on a length of the distance that must be completed to reach the goal, on how many movements must be performed to complete the task, and / or how complex is the task that must be performed. In addition, the duration of the fluctuations may depend on the characteristics of the task itself. For example, if the task is dynamic in such a way that the target moves or changes while performing the task, the nature of the fluctuations in the movement towards the target will change according to changes in the location or size or identity of the target.

[0102] Reverting to processor 220, it is noted that processor 220 is coupled to at least one memory module 250 which includes computer program code. The at least one memory module 250 may also store any other data required for the operation of processor 220 and / or of any other component of system 200, such as measurement data, analysis data, processing results, decisions, and so on. The at least one memory module 250 may store the computer program code and any optional additional data in any suitable technology, such as (but not limited to): volatile memory (e.g., random access memory or RAM), non-volatile memory (e.g., read-only memory or ROM, flash memory, magnetic storage devices), optical storage devices (e.g., CD-ROM, DVD), and emerging memory technologies (e.g., ferroelectric RAM, phase-change memory). The at least one memory module 250 and the computer program code are configured, with the at least one processor 220, to cause system 200 to determine a cognitive state of the user based on motion patterns of a body potion of an examined user, as discussed below. Whenever processor 220 is discussed below as carrying any action or being operable to perform any action, the execution of such action may be based on a computer program code stored in the at least one memory module 250.

[0103] Processor 220 is operable to receive — over at least one hardware communication channel 212 — sensor based kinematic data, to process that kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion during the performance of the task and, and to determine a cognitive state of the user based on analysis of the aforementioned motion patterns. In addition to motion patterns which occur during the induced period of highly fluctuational motion (resulting from the changing of the response dynamics of the user interface during the task), processor 220 may additionally process and use information of motion patterns which occur following that highly fluctuational period. Likewise, processor 220 may also process and use information of motion patterns which occur before this highly fluctuational period. For example, in computing any of its aforementioned outputs, processor 220 may take into account a reaction time to the appearance of different stimuli and / or a reaction time to different changes in the task. For example, in computing any of its aforementioned outputs, processor 220 may use the characteristics of the movement in parts of the task where there are no changes in response dynamics (e.g., the movement speed or derivatives of the movement speed, or the movement angle, or the presence of sub-movements or the precision of the movement (for example if there is a requirement to move along a certain line)). Additionally or alternatively, in computing any of its aforementioned outputs, processor 220 may take into account movement characteristics of the body part used by the subject to perform the task can be measured (e.g., measured changes in the acceleration or angle of movement of the body part or of the device operated by the user). Additionally or alternatively, in computing any of its aforementioned outputs, processor 220 may take into account measurements of a stability of the grip on the device operated by the user (if it is a watch, then the stability of the hand on which the watch is placed).

[0104] Following the determining of the cognitive state, processor 220 may proceed to take any suitable action, e.g., depending on the type of system or its objective. For example, processor 220 may proceed with any one or more of the following actions: d. Triggering a message to the user, to another person or another system, indicative of the determined cognitive state (e.g., “intoxication detected”),on the meaning or the implications of the determined cognitive state (e.g., “please reduce your speed to under 70km / h”), and so on; e. Triggering an action which affects the user or an environment of the user (e.g., triggering emission of breathable materials for awakening the user, releasing a medication or electrical or magnetic pulse); f. Triggering an action which affects a machine operated by the user or operable by the user (e.g., slowing down of a car in which the user travels, limiting the options in which the user can control a computerized and / or mechanical system, up to preventing use of such a system by the user), and so on.

[0105] Referring to the aforementioned kinematic data, this is a data being indicative of kinematic parameters of at least one body part of the user at different times during the performance of the task (and possibly at other times as well). Kinematic parameters are parameters which are indicative of movement of the respective body part, and may include, for example, locations at different times, velocities (e.g., vector, or absolute value), accelerations (e.g., vector, absolute values), higher degree derivatives, as well as relative locations, relative velocities, relative accelerations, or higher derivatives of relative parameters which pertain to the relationship between different body parts. It is noted that optionally, the kinematic data may pertain to the same one or more body parts which are used for controlling input UI 240. However, processor 220 may optionally utilize for its decisions kinematic data which pertains to another body part. For example, the user may control a touch screen using a finger, and the kinematic data may pertain to pressure sensors located within a seat on which the user sits, under the thighs of the user. The kinematic data may be collected by one sensor or by a plurality of sensors of one or more types (e.g., two pressure sensors, or a camera and a touchscreen sensor).

[0106] It is noted that the processing may focus on identifying motion patterns of the user while executing the task which follow the changes in the response dynamics of the user interface (and are thus characterized, at least partly, in highly fluctuational motion), and may include differentiating between motion patterns which occur before such changes in the response dynamics to motion patterns which occur after such a change in the response dynamics of input UI 240 (all during the performance of thetask by the user). In an example, the processing may focus on identifying submovements of the user while executing the task which follow the changes in the response dynamics of the user interface, and may include differentiating between submovements which occur before such changes in the response dynamics to submovements which occur after such a change in the response dynamics of input UI 240 (all during the performance of the task by the user).

[0107] Further information pertaining to the use of information pertaining to motion patterns for determining cognitive states is provided below (e.g., as part of the discussion of the methods 500, 600, and 900 below). Such ways of utilizing motion patterns information for determining cognitive states as provided below may be incorporated to system 200, mutatis mutandis.

[0108] As discussed throughout this disclosure, the disclosed technologies for assessing a cognitive state of the user, and for controlling access to various assets (e.g., vehicles, machines, computerized systems) based on such determination of cognitive state may be implemented in many different fields. In such cases, system 200 may optionally be associated with, connected with, included in, including of, or having another relationship with the system, vehicle, or machine whose control depends on the determination of the cognitive state.

[0109] For example, system 200 may be a vehicle which includes: (a) an engine operable to provide power for propelling the vehicle, (b) a user controllable steering mechanism for controllably changing a propagation direction of the vehicle, and (c) input user interface for detecting user instructions for modifying a behavior of at least one module out of the engine and the steering mechanism. Such user interface may be manual (e.g., steering wheel, pedals), computerized (e.g., cruise control, navigation software, driving mode such as sport, comfort, or economic), or any other suitable user interface. Processor 220 in such a case may be operable to selectively prevent controlling of performance of the at least one respective module (engine or steering mechanism) by the input user interface, based on the determined cognitive state.

[0110] In another example, system 200 may be a safe which includes: (a) a compartment in which valuables or other items may be stored, (b) a door or other latch for granting access to the compartment, (c) a lock which selectively disable opening of the door, (d) user interface for unlocking the lock, such as keypad or fingerprintdetector. Processor 220 in such a case may be operable to selectively prevent unlocking (or, possibly, locking) of the safe by the user interface, based on the determined cognitive state. Similar principles of operation may be applied, for example, for unlocking a smartphone, a computer, or access to any other computerized system, mutatis mutandis.

[0111] It is noted that while processor 220 was discussed above as a part of a larger system 200 (which also include additional components such as a hardware output user interface and an input user interface), it is possible to plan, manufacture, sell, and promote processor 220 independently of such other components. For example, processor 220 may be offered as a standalone computer chip which may be later integrated into a system (e.g., a vehicle) which already include such additional component, or used with a set of user interface components which are selected by the client, based on the client’s needs. Furthermore, as discussed below in greater detail, it is possible to implement capabilities of processor 220 as a computer readable computer code, and install this code (e.g., from a tangible storage medium such as a memory stick or hard drive, or via an online connection) to such a “host system”. Such a computer executable code may include any functionality discussed above with respect to system 200 (and especially to processor 220), as well as any step or variations of the methods discussed below (500, 600, and 900).

[0112] Optionally, system 200 may include a cognitive state-gated access authorization module (e.g., as part of processor 220, or as an independent module) which is operable to selectively grant users control of a requested system based on cognitive state assessment. It is noted that the granting of the control of the requested system (e.g., vehicle, machine, computerized system) may be independent of the success level of the user at the specific task. That is, the user may fail the task and be granted access if the determined cognitive state is suitable for operating the requested system. On the other side, the user may complete the task successfully, but the analysis of the motion patterns of the at least one body part of the user during that successful completion of the task (especially but not necessarily exclusively during the period of highly fluctuational motion) may indicate that the user is in a cognitive state which is not suitable for the operation of the requested system, and access would be denied.

[0113] For example, the cognitive state-gated access authorization module may be configured to: (a) permit control by the user of the requested system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, (b) repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level, and (c) preventing control by the user of the requested system in response to the determination of the second cognitive state.

[0114] As discussed in greater detail, some or all of the processing by processor 220 (or other modules such as the aforementioned cognitive state-gated access authorization module) may be agnostic of user history and / or private data. For example, optionally processor 220 may be configured to determine the cognitive state of the user without accessing historical performance information of the user.

[0115] As noted in this disclosure, the input UI may be dedicated for system 200, but may also be shared with another system. Especially, the input UI 240 may optionally be used by the user to control a system to which access is selectively restricted by system 200. Optionally, input user interface 240 may be further used to receive user instructions for operating of a machine to which access is selectively granted by the system based on the determined cognitive state.

[0116] As discussed in greater detail with respect to the methods below, the modification of the response dynamics of the one or more input UI 240 during the performance of the task may be based on data collected during the task, such as the kinematic data, data collected by the input UI 240, or data related to the content of the task. For example, processor 220 may optionally be configured to determine real-time modification parameters for modifying the response dynamics of the input user interface in response to the kinematic data.

[0117] Fig. IB is a functional block diagram illustrating an example of system 200, in accordance with the presently disclosed subject matter. Optionally, processor 220 may include sensor interface module 221 for receiving the sensor based kinematic data over the at least one hardware communication channel 212 from the one or more sensors 210. Sensor interface module 221 may also be used for receiving andtransmitting additional information between the one or more sensors 210 and other parts of processor 220, such as instructions, functional data, and so on. Optionally, sensor interface module 221 (or another component of processor 220 or system 200) may control the operation of one or more of the at least one sensor 210, fully or partly.

[0118] Optionally, processor 220 may include memory interface module 222 for receiving and / or transmitting data from the at least one memory module 250. Such data may include, for example, detection data, kinematic data, processing data, temporary files and storage, computer code instructions, and any other types of data, whether digital or analogue. Optionally, memory interface module 222 (or another component of processor 220 or system 200) may control the operation of one or more of the at least one memory module 250, fully or partly.

[0119] Optionally, processor 220 may include user interface controller 223 which is operable to monitor and / or control operation of user output interface 230 and / or user input interface 240. Especially, user interface controller 223 may be operable to trigger changes in the response dynamics of input user interface 240, for inducing changes in the motion patterns of the user, e.g., for creating periods of highly fluctuational motion of the body portion.

[0120] Optionally, processor 220 may include bus 224 or any other suitable means of internal communication, for facilitating communication between the different modules of processor 220.

[0121] Optionally, processor 220 may include motion analysis module 225, operable to execute any one or more of the following: process sensor-based kinematic data, characterizing motion patterns of the at least one body part, identifying periods of highly fluctuational motion (or other periods defined by the rate of fluctuations or by any other suitable parameters), identifying submovement, identifying corrective submovements, analyzing submovements, analyzing corrective submovements, and so on.

[0122] Optionally, processor 220 may include cognitive state assessment module 226, operable to determine a cognitive state of the user based on the any one or more of the aforementioned outputs of motion analysis module 225.

[0123] Optionally, processor 220 may include cognitive state-gated access authorization module 227, is operable to selectively grant users control of a requested system (e.g., a vehicle, a computer, a printing press) based on cognitive state assessment. Cognitive state-gated access authorization module 227 in such case may optionally be configured to: (a) permit control by the user of the requested system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, (b) repeat the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and (c) prevent control by the user of the requested system in response to the determination of the second cognitive state.

[0124] Processor 220 may include any combination of one or more of the aforementioned optional components (221, 222, 223, 224, 225, 226, and 227). Additionally, additional modules may be implemented to implement any other functionality, such as any functionality discussed below with respect to processor 220, method 500, method 600, or method 900.

[0125] Fig. 2 illustrates system 200 and a user, in accordance with examples of the presently disclosed subject matter. In the illustrated example, the user is operating a computer, but it is noted that system 200 and the methods below may be implemented in any scenario in which the performance of a task by the user is subject to perturbations. Optionally (e.g., as exemplified in the diagram), sensors 210 may be connected to the processor via wired or wireless means, directly or indirectly (e.g., via a server, communication bus, cloud, intermediary system, etc.), in any suitable way. The at least one or more sensors may be located in any one or more locations such as (but not limited to): on the body or clothing of the user, otherwise carried, worn, or touched by the user, integrated into a system within or next to which the user is located, integrated into a system which is at least partly controlled by the user, and so on.

[0126] Fig. 3 illustrates a perturbation task and user interface for performing of the task by a user, in accordance with examples of the presently disclosed subject matter. The task requires the user to move the frog toward the fly on the computer monitor (denoted 230), by moving the joystick which serves as a user input interface 240.

[0127] Fig. 4 illustrates some of the ways in which the response dynamics of the input user interface 240 (the joystick in the illustrated examples) can change over time, in order to induce a period of highly fluctuational motion (e.g., including many corrective submovements) by the user who performs the task.

[0128] Optionally, the responsivity level of the input UI (e.g., the amount of force needed for achieving a certain velocity of the frog cursor, in the case of the joystick of the illustrated example) may change between different discrete levels, e.g., as exemplified in diagram 401. The triggering to switching between different levels of responsivity and the level of change may be determined, for example, based on a preexisting plan, in response to sensor 210 data, in response to input UI 230 data, in response to processing of processor 220 (e.g., based on any one or more of the previously mentioned types of data), and so on. Optionally, the responsivity level of the input UI may change continuously over a span of responsivity levels, e.g., as exemplified in diagram 402. The momentarily rate of change in the response dynamics of the input UI may be determined, for example, based on a preexisting plan, in response to sensor 210 data, in response to input UI 230 data, in response to processing of processor 220 (e.g., based on any one or more of the previously mentioned types of data), and so on. It is noted that a combination of discrete and gradual changes may be implemented at different times during the performing of the task. It is also noted that different aspects of the responsivity of the input UI (e.g., of the joystick) may change differently - e.g., in different times, to different degrees, and under different decisionmaking regimes. Diagram 403 illustrates an example of a second parameter of the responsivity of the same joystick (or another input UI) which may be modified independently from the first parameter, or in coordination with which. In the illustrated example, the resting position of the joystick (i.e., the angle of the joystick with respect to its base in which the cursor ceases to move) may also be changed with time. For simplicity of the diagram, this change is illustrated as a one-dimensional change over time (e.g., only angle), but a two-dimensional change (e.g., angle of diversion from the vertical Z axis, and direction in the horizontal XY plane) may also be implemented.

[0129] The number of changes in the response dynamics of the input UI interface during the duration of performance of the task by the user, their frequency, rate, degree, and so on, may all vary, depending for example on the type of task, on the cognitivestates which the system is designed to identify, on the type of input UI interface, optionally on user parameters, and so on. For example, in some tasks which were used by the inventor to identify different cognitive states such as intoxication and fatigue, the response dynamics of the input interface were changed at a rate of between 20 times per minute and 1 time per second. Other rates may also be implemented, such as more or less frequently than 20 times per minute, between 1 and 4 times a minute, and so on.

[0130] The rate of dynamic changes can be determined as a function of the passing time (say, several times a minute or a second, as discussed above), but it can also be determined based on other factors, in addition to — or instead of — temporal considerations. For example, the rate of dynamic changes can be determined as a function of the number of steps or number of initiated movements in a task (say, N>1 changes will appear in each step / movement or N>1 changes will appear in every three steps / movements). Optionally, the rate of dynamic changes can be predetermined and spread randomly along the assignment. Optionally, changes in the response dynamics of the input user interface may be induced in one or more different times with respect to the initiation of the movement (e.g., immediately following initiation, towards conclusion of the movement, in the middle of the movement). Optionally, changes in the response dynamics of the input user interface may be induced in one or more different times with respect to the initiation of the task and / or of the presenting of the task to the user (say, the target is flashing). Optionally, changes in the response dynamics of the input user interface may be induced based on the completion level of the task (e.g., the distance of the cursor from the location of the target, e.g., immediately before the target is reached).

[0131] Another way of inducing a period of highly fluctuational motion for the body portion (also referred to as “the at least one body part”) of the user includes having the user input interface 240 (possibly as part of a larger physical module) vibrate or otherwise mechanically move during the operation of the task, thereby resulting in similar or equivalent reactions in the brain of the user. Depending on the specific implementation, the triggering of the change in the mechanical motion of the user input interface 240 may be a change in the response dynamic of input user interface 240. In an example, of user input interface 240 is a touchscreen, a button, or a switch of asmartphone or another handheld electronic device operated by the user, using the vibration motor or haptic actuator of the device. Other types of changing mechanical movements may also be implemented, e.g., if system 200 is supported by another system or apparatus with controllable motion of any kind (e.g., vibration, reciprocating, linear, circular). The moving of user input interface 240 (e.g., the changing of the movement, of the acceleration), may be controlled and / or operated by system 200 (e.g., by processor 220, by user interface controller 223) or by any other capable system (which may trigger the change in viscosity or operate based on signals of system 200). Causing or changing the movement of the user input interface for inducing a period or highly fluctuational motion of the body portion and / or to induce change in the corrective submovement characteristics may also be implemented in methods 500, 600, and 900, mutatis mutandis.

[0132] Any suitable way of introducing perturbation to task may be implemented, including but not limited to any one or more of the following: g. Mechanical Perturbations: Mechanical perturbations are external disturbances applied to the body or an object being manipulated during a movement. These perturbations can be delivered through a variety of devices such as springs, motors, or air jets. One of the most common methods of mechanical perturbation is through the use of a robotic exoskeleton, which can apply precise and controlled perturbations to different joints of the body during a movement. h. Electrical Perturbations: Electrical perturbations involve delivering a brief electrical stimulation to a specific muscle or group of muscles to disrupt ongoing movement. This method can be used to produce perturbations at specific times during a movement, and the magnitude of the perturbation can be varied by adjusting the intensity and duration of the electrical stimulation. i. Visual Perturbations: Visual perturbations are a type of sensory perturbation that involves manipulating visual feedback during a movement. This can be achieved through the use of virtual reality or computer graphics that alter the visual feedback of a movement in realtime. For example, researchers can use a visual display to manipulate thesize or shape of a target, which can lead to changes in movement trajectory or direction. j. Auditory Perturbations: Auditory perturbations are a type of sensory perturbation that involve manipulating auditory feedback during a movement. This can be achieved by altering the timing or loudness of a sound cue that is associated with a specific movement. For example, a sound cue can be used to indicate when a movement should be initiated, or a sudden loud noise can be used to disrupt ongoing movement. k. Gravitational Perturbations: Gravitational perturbations involve manipulating the direction or magnitude of gravitational forces during a movement. This can be achieved through the use of tilting platforms or harnesses that can be used to alter the orientation of the body during a movement. For example, researchers can use a tilting platform to induce a sudden change in body orientation, which can lead to changes in movement trajectory or balance.

[0133] It is noted that while past studies examined association between corrective submovements and cognitive changes, the novel systems and methods disclosed in the present disclosure demonstrate an effective way to use dedicated perturbation in order to create the characteristics and motion patterns in the highly fluctuational motion that are sensitive to and indicative of different cognitive states. The studies we conducted show that the use of perturbations to produce highly fluctuational motion produces unique characteristics that are sensitive to cognitive changes.

[0134] Fig. 5 is a flow chart illustrating method 500 for cognitive state assessment, in accordance with examples of the presently disclosed subject matter. Referring to the examples set forth with respect to the previous drawings, method 500 may optionally be executed by system 200. Any variation, option, embodiment, or implementation discussed with respect to system 200 may be applied to method 500 (e.g., as a method step), mutatis mutandis. Method 500 can be executed using any suitable hardware, software, or combination thereof, as may be desired by the user. This flexibility allows the method to be adapted to a wide range of applications and user preferences, without sacrificing its core functionality. Furthermore, the method can be easily integrated intoexisting systems and workflows, making it a highly versatile and user-friendly solution.

[0135] Method 500 starts with step 510 of identifying a triggering event, in which assessment of the cognitive state of the user is triggered. Such a triggering event may be triggered in many different ways, such as any of the following examples: l. By the user (explicitly or implicitly, e.g., by attempting to control a machine such as a car, computer, etc.). m. By the system executing method 500. n. By a request from another system (e.g., whose control by the user is conditioned by the cognitive state of the user). o. Based on a schedule (e.g., a predetermined or an adaptive schedule). p. In response to kinematic data of the user, or to other data indicative a state of the user.

[0136] Following the triggering, method 500 continues with step 520 of presenting to the user a task (e.g., a “viscosity task”) and a UI for performing the task. As aforementioned, the task may be a stand-alone task or a part of a broader interaction between the user and the system, but it must include a stage in which the response dynamics of the input UI used by the user are changed (in step 530). In many implementations, the user is not given a head warning before the modification of the response dynamics, even though such a heads-up notice is optional. The response dynamics may be changed more than once during the task, in which case user data pertaining to either one or more of these response dynamics changes is being used. It is noted that either abrupt and / or gradual changes in the response dynamics may be implemented. It is also possible to implement method 500 with the changes in the response dynamic being executed by another system, and not controlled by the system which executes method 500. In such case, the system which executes method 500 may learn of such modifications to the response dynamic in any suitable way, such as directly from the modifying system, by analysis of any other data (e.g., pertaining to the kinematics of the user, other user parameters, pertaining to the respective input UI, etc.).

[0137] It is noted that a task during which the response dynamics of the input user interface is modified may also be referred to as “perturbation task”, “viscosity task”, “changing viscosity task”, or similar terms. Within the context of the present disclosure, terms such as "viscosity change" or “modifying viscosity” refer to any situation in which an unexpected gap is intentionally created between the user’s expectations regarding how an object in the task should respond to user instructions provided via the respective input UI and the actual response of the object. For example, terms such as "viscosity change" or “modifying viscosity” may refer to any situation in which an unexpected gap is intentionally created between the subject's expectations regarding how an object should be moved in a medium (e.g., cursor across a computer screen) and the actual movement of the object in the medium.

[0138] In step 540 of method 500, kinematic data of the user is obtained. Different aspects of kinematic data and its collection by one or more sensors are discussed above. It is noted that while stage 540 is illustrated as following stage 530, the monitoring of user’s kinematics may optionally start before the modification of the UI response dynamics. Use of kinematic data pertaining to the time before the modification of the UI response dynamics, if collected, in the processing of stage 550 is optional. While processing of such data may be useful, the invention may also be implemented using only kinematic data which follows the modification of the UI response dynamics.

[0139] As mentioned above in greater detail, asking the user to perform a task, and modifying the response dynamics of the one or more input UI used by the user to perform the task, may induce activity in the Basal Ganglia of the user. Referring to method 500 as well as to system 200 and to methods 600 and 900, it is noted that optionally, the kinematic data may be indicative of submovements or other motion patterns — especially highly fluctuational motion patterns — originating from a basal ganglia of the user, which is stimulated by the performance of the task.

[0140] Method 500 continues with step 550 in which the sensor-based kinematic data is processed to characterize motion patterns of the at least one body part during the highly fluctuational motion (especially — but necessarily only — following the changing of the response dynamics), for example, step 550 may include identifying and / or characterizing motion patterns of the at least one body part during the highly fluctuational motion (e.g., identifying submovements of one or more body parts of theuser) following the modification of the response dynamics. The identification of such highly fluctuational motion patterns by processing of the kinematic data may be achieved in many different ways, and the following discussion does not intend to exhaust all the different options.

[0141] As discussed above, method 500 may optionally include identifying submovements of one or more of the body parts of the user following the modification of the response dynamics based on processing of the kinematic data. Fig. 6 illustrates a sub-method 550’ for identify submovements of one or more body parts of the user following the modification of the response dynamics based on processing of the kinematic data, in accordance with examples of the presently disclosed subject matter. Sub-method 550’ may be used for implementing step 550, but other implementations may also be used.

[0142] Step 551 of method 550 includes obtaining values of at least one kinematic parameter of the user for a continuous duration during the performing of the task. Such one or more kinematic parameters may be indicative, for example, of the position of the body part (e.g., with respect to the rest of the body, to input UI 240, to sensor 210, to the vehicle or machine for which access is requested, and so on), of the orientation of the body part, of the velocity of the body part (scalar or vector), and so on. It is noted that the kinematic parameter may be related to the kinematics of the body part directly (e.g., a camera or a pressure sensor) or indirectly (e.g., internal joystick sensors which measure directly the angle of the joystick handle and which are indicative of the position of the hand of the user indirectly, or even cursor position on the screen, which is another indirect indication of the location of the hand). Continuing the example of Fig. 3 in which the user operates a cursor on a computer screen with a joystick, the obtained parameter may simply be cursor position data, joystick output data, or joystick internal sensors data (indicative of joystick orientation). Step 551 may be implemented for the entire duration of the task, to a shorter duration (e.g., every millisecond), for a predefined duration following each modification of the response dynamics of the input UI (e.g., for 500ms following such modification), or for any other suitable duration.

[0143] Optional step 552 includes processing the at least one kinematic parameter to generate values of a derived parameter for the continuous duration. For example, step552 may include generating a trace of velocity values over time by calculating the first derivative (differentiating) of the position data obtained in step 551. Of course, in some cases such processing is not required (e.g., if the sensor 210 is operable to provide the velocity parameter directly).

[0144] Step 553 includes identifying a beginning (onset) of a principal movement based on the kinematic data and / or derived parameter. The identifying of step 553 may be implemented using any suitable type of processing, such as meeting a predefined criterion. For example, principal movement onset may be identified as the moment when a parameter (e.g., linear velocity, angular velocity) surpasses a predefined threshold (e.g., a pixels-per-second rate, or 5% of the peak angular velocity) and stays above for a predefined duration (e.g., 100 milliseconds).

[0145] Step 554 includes identifying an end (offset) of the principal movement based on the kinematic data and / or derived parameter. The identifying of step 554 may be implemented using any suitable type of processing, such as meeting a predefined criterion. For example, principal movement offset may be identified as the moment following the onset in which a parameter (e.g., linear velocity, angular velocity) does not vary by 5° for 0.4 seconds. The timing of the primary movement is then defined to be between the onset time and the offset time.

[0146] We use the term “principal movement” to refer to the entire motion from point A to point B (e.g., when a person moves his hand to grab a mug from a table, the entire movement between the starting position of the hand until it reaches the mug and stops is the principal movement). This so-called principal movement can be separated into two types of movements: (a) the primary movement which is the main motion that initiates and drives the body part towards its target position (location and orientation). This movement may include in many cases utilization of larger muscle (or muscle groups) to accomplish the intended task; and (b) corrective submovements, which are smaller, more refined adjustments made during the primary movement (or even shortly after the primary movement, for example in the case of an overshoot). These adjustments are controlled by the brain's continuous feedback system, which detects inaccuracies in the motion and corrects them to ensure that the intended motion is carried out (e.g., the hand successfully reaches the mug). Corrective submovements may include, for example, changes in speed, direction, or grip strength, and are crucialfor fine-tuning and optimizing the overall movement. While not necessarily so, the corrective submovements may include activating synergist muscles to assist the agonist muscle in performing the desired movement. Such synergist muscles help to support and fine-tune the motion, as well as prevent undesired or excessive movement in other directions.

[0147] Step 555 includes defining submovements within the principal movement duration. Step 555 may include, for example, dividing each principal movement into a primary movement and any submovements. By way of example, such division may be achieved by analyzing the third derivative of position (commonly referred to as “jerk”) and locating all zero crossings between onset and offset of the principal movement. The number of zero crossings is halved to find the total number of movements (velocity peaks). Subtracting the primary movement from this count gives the number of submovements. The overlap of submovements with the primary movement may be estimated based on the method by Tunik et al. (2009). Other parameters which may also be detected in step 555 include, for example, peak velocity and peak acceleration of the primary movement, and the time until the first submovement. These parameters may be determined from the velocity and acceleration trace. Principal movement amplitude may be defined as the angular displacement between the starting angle and the maximum displacement in the trial.

[0148] It is noted that while sub-method 550’ was discussed as being implemented for identifying submovements of one or more body parts of the user following the modification of the response dynamics based on processing of the kinematic data, the same sub-method 550’ may be implemented — mutatis mutandis — for identifying motion patterns of other patterns. Fig. 7A illustrates measurements of movement of a body part of the user (in this example, a hand) when the user is in low fatigue level and in high fatigue level, in accordance with examples of the presently disclosed subject matter. The amplitudes of movements are normalized. Diagram 701 shows movement of the body part when the user is in low fatigue level and performing a low viscosity task. Diagram 703 shows the changed motion patterns of that low fatigue user when they are suddenly subject to high viscosity conditions for the same type of task. As can be seen, in duration 708 the motion of the body part following the change to high viscosity in the low fatigue user is characterized in highly fluctuational motioninduced by the change in viscosity, e.g., in comparison to the motion behavior of the corresponding diagram 701. Diagram 702 show movement of the body part when the user is in high fatigue level and performing a low viscosity task. Diagram 704 shows the changed motion patterns of that low fatigue user when they are suddenly subject to high viscosity conditions for the same type of task. As can be seen, in duration 709 the motion of the body part following the change to high viscosity in the high fatigue user is characterized in highly fluctuational motion induced by the change in viscosity, e.g., in comparison to the motion behavior of the corresponding diagram 701. As can be seen, the motion patterns during durations 708 and 709 in diagrams 702 and 704 (respectively) are different. Method 500 may be implemented to analyze such different motion patterns (by way of example) for distinguishing between high fatigue users and low fatigue users. By way of comparison, the motion patterns of diagrams 701 and 703 in the corresponding durations may be insufficient for such determination of cognitive state.

[0149] Fig. 7B illustrates measurements of movement of a body part of the user (in the illustrated example, a hand) when the user is in low level of intoxication and in high level of intoxication, in accordance with examples of the presently disclosed subject matter. The amplitudes of movements are normalized. Diagram 711 show movement of the body part when the user is in low level of intoxication and performing a low viscosity task. Diagram 713 shows the changed motion patterns of that low- intoxication user when they are suddenly subject to high viscosity conditions for the same type of task. As can be seen, in duration 718 the motion of the body part following the change to high viscosity in the low intoxication user is characterized in highly fluctuational motion induced by the change in viscosity, e.g., in comparison to the motion behavior of the corresponding diagram 711. Diagram 712 shows movement of the body part when the user is in high level of intoxication and performing a low viscosity task. Diagram 714 shows the changed motion patterns of that low intoxication user when they are suddenly subject to high viscosity conditions for the same type of task. As can be seen, in duration 719 the motion of the body part following the change to high viscosity in the high intoxication user is characterized in highly fluctuational motion induced by the change in viscosity, e.g., in comparison to the motion behavior of the corresponding diagram 711. As can be seen, the motion patterns during durations 718 and 719 in diagrams 712 and 714 (respectively) aredifferent. Method 500 may be implemented to analyze such different motion patterns (by way of example) for distinguishing between users subject to different levels of intoxication. By way of comparison, the motion patterns of diagrams 711 and 713 in the corresponding durations may be insufficient for such determination of cognitive state.

[0150] Reverting to step 550, it is noted that the characterization of motion patterns in step 550 may be implemented in many different ways. The characterization of the motion patterns (e.g., during the period of highly fluctuational motion) may pertain, for examples, to uniform subdivisions of the period of highly fluctuational movement, to individual submovements, to groups of submovements, to data-derived sub-periods (e.g., between consecutive peaks, between consecutive crossing of zero, between consecutive crossing of zero of the nthderivative), and so on. The characterization of motion pattern may be done in any suitable way and using any suitable parameters, such as intensity, time, steepness, amplitude, rate of change, frequency spectrum, autocorrelation, cross-correlation, phase, damping factor, periodicity, power spectral density, signal-to-noise ratio, peak-to-peak distance, waveform envelope, harmonic distortion, crest factor, modulation index, fractal dimension, entropy, wavelet coefficients, inter-arrival times, zero-crossing rate, statistical moments (mean, variance, skewness, kurtosis), bispectrum, and empirical mode decomposition.

[0151] Method 500 may include, for example, dividing the period of highly fluctuational motion into many portions or “bits”, and within these segments analyze the motion or its derivatives and / or other characterizing parameters.

[0152] It is noted that in addition to motion patterns which occur during the induced period of highly fluctuational motion (resulting from the changing of the response dynamics of the user interface during the task), method 500 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patterns which occur following this highly fluctuational period. Likewise, method 500 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patterns which occur before this highly fluctuational period.

[0153] Once the motion patterns have been characterized in step 550, method 500 continues with step 560 of processing the characterized motion patterns (e.g., based on motion pattern reference data) to identify presence of motion patterns indicative ofrecognized cognitive states. For example, step 550 may include processing kinematic patterns of the identified submovements based on submovements reference data to determine kinematic patterns indicative of recognized cognitive states.

[0154] It is noted that kinematic patterns of the user change as a result of changes applied to the UI response dynamics, and that this change depends on the cognitive state of the user at the time. The manipulation of the response dynamics of the user interface to user instructions has a great effect on the error control system of the brain of the user, and as a result this manipulation has a great effect on various characteristics of both the primary movement and highly fluctuational motion patterns (whether corrective submovements or other). For example, the very fact that the user expects the appearance of the manipulation (in cases in which the user knows about the modification of the response dynamic in advance) means that the user is required to control the primary movement more carefully before and during its execution. This control can be characterized by a more careful movement, especially around the points in the route or at the time when the manipulation is expected. The characteristics of the strict control depend on the type of change in the response dynamics and its timing in relation to the beginning of the movement, or in relation to the appearance of the change in the target stimulus. These characteristics can be expressed, for example, in the reaction time, in the rate of change of speed, in the course of the movement, in the rate of change in the course of the movement, in the length of the movement, in the characteristics of the strength of the start of the movement, and in the characteristics of the strength of the stop of the movement.

[0155] The calculations made on these characteristics can pertain to any one or more of the following: q. Averages of characteristics over the course of the task (for example, the average response time); r. The change in a specific characteristic over the course of the task (variability of the response time); s. The ratio between the averages of different characteristics over the course of the task (for example, the ratio between the average reaction time for the main movement and the average reaction time for the response to aspecific motion pattern, characteristic movement, submovement, or to a number of those); t. The ratio between the changes in the averages of specific characteristics throughout the task (for example the ratio between the variability in the reaction time of the main movement to the variability in the reaction time of a specific part of the highly fluctuational movement (e.g., specific submovement or number of submovements, times between consecutive peaks, and so on).

[0156] Beyond the effect of the expectation of the user of manipulation on brain error control, from the moment the manipulation appears, the brain error control system is required to intervene in the movement in order to correct the error created as a result of the manipulation. At this stage, motion patterns appear whose purpose is to correct the error or errors. Here too, the characteristics of these motion patterns depend on various factors, including: the type of manipulation, the timing in relation to the beginning of the movement or in relation to the appearance of the target stimulus. The characteristics can be expressed, for example, in the reaction time, the rate of change of speed, the trajectory of the movement, the rate of change of the trajectory of the movement, the length of the movement and the characteristics of the strength of the beginning of the movement and the strength of the stopping of the movement. The calculations made on these characteristics can pertain to any one or more of the following: u. Averages of characteristics over the course of the task (for example, the average response time); v. The change in a specific characteristic over the course of the task (variability of the response time); w. The ratio between the averages of different characteristics over the course of the task (for example, the ratio between the average reaction time for the main movement and the average reaction time for the response to a motion pattern, characteristic movement, submovement, or to a number of those);x. The ratio between the changes in the averages of specific characteristics throughout the task (for example the ratio between the variability in the reaction time of the main movement to the variability in the reaction time of a motion pattern, characteristic movement, submovement, or to a number of those).

[0157] The strengthening of brain error control activity following the expectation of manipulation or following the appearance of actual manipulation makes the brain's error control system more vulnerable when the person's mental or cognitive state goes awry due to various factors such as drinking alcohol, fatigue, and drug use. Thus, the changes in the primary movement and / or the motion patterns in the highly fluctuational period which follows the changing of the UI response dynamics become more noticeable and easier to identify and measure.

[0158] The identification of the cognitive state in stages 560 may be used in different ways, some of which were discussed above. One optional way of utilizing the identification of the cognitive state of the user is exemplified by the optional steps 570, 580, and 590. In decision box 570 method 500 continuous with determining whether the identified cognitive state limits the performance capabilities of the user (e.g., with respect to a specific activity which the user may commence or otherwise engage in), such as whether the user is drunk, tired, intoxicated, etc.

[0159] If the cognitive state identified in step 560 is determined to limit the performance capabilities of the user, different remedial actions may be executed or triggered (e.g., for execution by another system) in step 580, such as those discussed above. On the other hand, if the cognitive state identified in step 560 is determined to limit the performance capabilities of the user (at least non to a threshold-crossing degree), method 500 may continue with step 590 of facilitating user control of a system whose control is conditioned on the cognitive state of the user. Other actions may also be taken (e.g., informing the user or another party about the findings of method 500).

[0160] It is noted that some or all of the cognitive states identifiable in method 500 (or the other methods and systems disclosed in this document) may be ones which do not limit the capabilities of the user, but rather indicate a state of improved neurological capabilities, such as (but not limited to): flow state (i.e., “being in the zone”), meditative state, runner’s high, eustress, hypnagogia, post-meditation clarity,psychedelic experiences, nootropic effects, alertness. The ability to identify when a user is in such cognitive states may be useful, for example, for selecting to which person to assign a very sensitive task, or at what times should such a task be presented to a specific user.

[0161] Referring to method 500 as a while, it is noted that when there is an interest in analyzing highly fluctuational motion induced by change in response dynamics of a UI during the performance of a task in order to identify a cognitive state, different methods of machine learning may be used in synergistical fashion, in order to identify unique features in the motion patterns that correspond to fluctuations or changes in the cognitive state. For example, this may be facilitated by a parametric manipulation of the cognitive state of various people, for example by giving different amounts of alcohol to the subjects or creating different levels of fatigue in the subjects. A machine learning model (or neural networks) may then be run on the data. The machine learning model should be guided to explore the motion patterns occurring within the primary movement.

[0162] Fig. 8 illustrates method 600 for assessing a cognitive state of a user, in accordance with examples of the presently disclosed subject matter. Referring to the examples set forth with respect to the previous drawings, method 600 may optionally be executed by system 200 as a whole or by its processor 220. Method 600 may be implemented by any suitable combination of software, hardware, and firmware. Optionally, method 600 may be a computer-implemented method for assessing a cognitive state of a user, comprising executing on a processor the steps of method 600. It is noted that steps and variations discussed with respect to method 500 may also be implemented as a part of method 600 (where applicable), mutatis mutandis, and vice versa.

[0163] Step 610 of method 600 includes receiving, over at least one hardware communication channel, sensor based kinematic data that is indicative of kinematic parameters of at least one body part of the user at different times during a performance by the user of a task that includes operating an input user interface whose response dynamics to user instructions change during the performance of the task. Referring to the examples set forth with respect to the previous drawings, the communication channel may be communication channel 212, and the one or more sensors may besensors 210. The sensors may belong to the same system which executes method 600, but this is not necessarily so. As discussed above, the kinematic data may pertain to the same body part used by the user for controlling the input UI, but this is not necessarily so.

[0164] Method 600 may include the optional step of modifying the response dynamics of the input user interface (denoted 605). There are many optional ways for implementing modification of the UI response dynamics, such as (but not limited to) any one or more of the following: (a) based on predetermined plan which precedes the execution of the specific task by the specific user; (b) based on the way in which the user performs the task (e.g., how fast does the user moves a cursor); (c) based on the way the user interacts with the input UI (e.g., how hard does the user push the UI, the consistency of the force applied by the user); (d) based on the response of the user to previous modifications of the response dynamics (either of the same input UI or of another UI); (e) based on real time analysis of the kinematic data collected by one or more sensors.

[0165] The modification of the response dynamics of the input user interface, whether done as part of method 600 or by a process executed by another entity results in highly fluctuational motion of the at least one body part.

[0166] Step 630 of method 600 includes processing the kinematic data to characterize motion patterns of the at least one body part of the user during the highly fluctuational motion (during the performance of the task), including at least motion patterns during the period of highly fluctuational motion induced and directly following the modification of the response dynamics of the user interface (also referred to as “viscosity change”). Step 630 may include identifying specific motion pattern (e.g., based on prior training data) which occur when the user is required to operate the input user interface (e.g., the task may include instructing the user to follow a moving target) and when the response dynamics of the input user interface changes, optionally frequently (e.g., the amount of force that the user is required to exert changes frequently). For example, step 630 may include identifying corrective submovements, which are the submovements that occur when the user is required to operate the input user interface (e.g., the task may be to follow a moving target) and the response dynamics of the input user interface changes, optionally frequently (e.g., the amountof force that the user is required to exert changes frequently). Corrective submovements may include, for example, subtle changes in body part angles, angular velocities, and angular accelerations that are not visible to the eye. The processing of stage 630 may be focused on identifying submovements (e.g., corrective submovements) following changes of the UI response dynamics (for one or more such modifications), but may also include identifying submovements (e.g., corrective submovements) prior to such modification of UI response dynamics. Such data may be used, by way of nonlimiting example, for comparison or calibration of the postchange submovements data.

[0167] Optionally, the identifying of cognitive- state-indicative motion patterns in step 630 may include a preliminary substep of processing the kinematic data (directly, or a preprocessed version thereof) for identifying principal movements (e.g., spanning seconds) and the beginning and ending timing of such principal movement. This substep may then be followed by identifying submovements or other identifiable motion patterns within the timing of such larger movements. Some ways of achieving these were discussed with respect to sub-method 550’, which may be incorporated, fully or partly, as part of step 630.

[0168] It is noted that the processing of step 630 may be preceded by one or more steps of preprocessing the kinematic data before it is used for the following stages of the method (collectively denoted 620). For example, the obtained kinematic data (or derived parameters derived therefrom, as discussed with respect to step 552 above) may be preprocessed to remove noise and outliers. If data from multiple sensors is obtained (or if a single sensor provides multidimensional data), step 620 may also include synchronizing data from different sensors or reducing the number of dimensions in data provided by a single sensor. The preprocessing step 620 may also include feature extraction, which involves selecting relevant features from the raw data that are used for the analysis. The features may include, but are not limited to, body angles, angular velocities, and angular accelerations.

[0169] It is noted that in addition to motion patterns which occur during the induced period of highly fluctuational motion (resulting from the changing of the response dynamics of the user interface during the task), method 600 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patternswhich occur following this highly fluctuational period. Likewise, method 600 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patterns which occur before this highly fluctuational period.

[0170] Following the analysis (e.g., characterization) of motion patterns in step 630 (e.g., corrective submovements), method 600 continuous with step 640 of determining a cognitive state of the user based on the aforementioned motion patterns (e.g., based on analysis of these motion patterns, whether corrective submovements or other patterns as discussed above).

[0171] Following step 640, method 600 may optionally continue with: (a) the equivalents of steps 570, 580, and 590, mutatis mutandis, (b) the equivalents of steps 970, 980, and 990, mutatis mutandis, or (c) a combination thereof. Other ways of following up may also be implemented, in method 600 as well as in methods 500 and 900. For example, the determining of the cognitive state may be followed by generating notification data indicative of the determined cognitive state. This may be useful, for example, when the user (or another entity such as a supervisor, health care provider, family member, etc.) may simply be interested in knowing what the cognitive state of the user is. For example, some people may want to use an app on their smartphone which implements method 500, 600, or 900 (fully or partly) before entering into a negotiation, argument, decision-making process, or any other situation in which they want to be in their best.

[0172] Fig. 9 illustrates method 900 for selectively granting control of a machine (or another type of user controllable system) based on assessment of a cognitive state of a user, in accordance with examples of the presently disclosed subject matter. Referring to the examples set forth with respect to the previous drawings, method 900 may optionally be executed by system 200 as a whole or by its processor 220. Method 900 may be implemented by any suitable combination of software, hardware, and firmware. Optionally, method 900 may be a computer-implemented method for selectively granting control of a machine (or another type of user controllable system), comprising executing on a processor the steps of method 900. It is noted that steps and variations discussed with respect to methods 500 and 600 may also be implemented as a part of method 900 (where applicable), mutatis mutandis, and vice versa.

[0173] Step 910 of method 900 includes identifying a situation in which the user intends to drive a vehicle (e.g., car, truck, ship, scooter) car or operate another machine such as another vehicle, a crane, a press, a bandsaw, and so on. It is noted that method 900 may also be applied to controlling access of users to any other type of user controllable systems, such as computerized system, voting systems, and so on. Furthermore, method 900 (as well as methods 500 and 600 above) may also be adapted to use for controlling access to a location, facility etc., when used by a service provider (e.g., a banker, a notary etc.), mutatis mutandis.

[0174] Step 920 of system 900 includes presenting to the user a task and a UI for performing the task. In but a few examples, such a task may be presented using a UI of the car or other system / machine, or another UI (e.g., smartphone of the user). While the user performs the task, step 930 includes modifying the response dynamics of the UI, once or more, instantly or gradually.

[0175] Step 940 includes obtaining kinematic data of the user, collected by one or more sensors (e.g., positioned within the car / on the machine / at the smartphone). This kinematic data is processed in step 950 for identifying motion patterns of one or more body parts of the user following (e.g., resulting from) the modification of the response dynamics. Step 950 may optionally include specifically identifying corrective submovements. Kinematic patterns (or other parameters) of the identified motion patterns are processed in step 960 based on motion-patterns reference data to determine kinematic patterns indicative of recognized cognitive states. As in the rest of the systems and methods disclosed in this disclosure, the level of detailing of identified cognitive states may be selected to suit the requirements of the respective system method (e.g., 200, 500, 600, 900). For example, for deciding whether to permit the user to operate a car it may not matter if the user is hallucinating or intoxicated (e.g., because control of the car should not be granted in any of these cases), while in other scenarios (e.g., autodialing to a selected contact person based on the determined cognitive state) such a distinction may be valuable and required. It is noted that in addition to motion patterns which occur during the induced period of highly fluctuational motion (resulting from the changing of the response dynamics of the user interface during the task), method 900 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patterns which occur following thishighly fluctuational period. Likewise, method 900 may also include sensing, obtaining, analyzing, processing, and / or utilizing information of motion patterns which occur before this highly fluctuational period.

[0176] Once one or more cognitive states are identified, method 900 continues with determining (in step 970) whether the identified cognitive state limits user’s performance (e.g., to an unpermitted level). If the answer is “YES”, method 900 continues with step 980 of preventing usage of the requested system / machine, or triggering execution of another remedial action (e.g., contact a preselected contact person). In case the answer is “NO”, method 900 may continue to step 990 of permitting control of the car or another system / machine (possibly subjective to other requirements or qualifications which may also be tested for).

[0177] Referring to system 200 and to methods 500, 600, and 900, it is noted that optionally, the decision of how to proceed after the determination of the cognitive state of the user may be independent from how well the user succeeded in the task (if at all). That is, the user may have succeeded in the task despite the changes in the response dynamics of the UI (e.g., brought a cursor to a marked location on the screen within a predetermined time threshold) and nevertheless control of the car, machine, or other system would be denied due to kinematic patterns following the change of response dynamics (also referred to as viscosity change). On the other side, the user may fail the designated task and nevertheless be granted control of the car. In one example, the determination of the cognitive state of the user may conclude during the performance of the task based on collection of sufficient kinematic data, in which case there may be no further need to wait for the completion of the task. In another example, the failing of the task is not indicative of the capability of the user to control the requested asset (e.g., vehicle, machine, system), and the kinematic data is indicative of a cognitive state which does not in itself hinder use of the requested asset.

[0178] Referring to methods 500, 600, and 900, optionally the determining of the cognitive state may be followed by: (a) permitting controlling by the user of another system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level (e.g., failing or low performance), (b) repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematicdata collected during performance of a second task by the user at a second success level that is higher than the first success level (e.g., success or high performance); and preventing control by the user of the other system in response to the determination of the second cognitive state (even though the success level in this case was lower).

[0179] Referring to system 200 and to methods 500, 600, and 900, it is noted that optionally, the determining of the cognitive state of the user may be based on decision model which is agnostic to historical performance information of the user. That means in such cases there is no need to save sensitive personal data of the user, and that the respective system or method could also be used for anonymous users, unregistered users, casual users, guest users, drop-in users, one-time users, etc. While not necessarily so, other steps of the respective methods (as well as processing capabilities of system 200) may also be agnostic to historical performance information of the user, or to other data about the user. For example, the processing of the kinematic data to identify, characterized, analyze, compare, or otherwise process motion patterns of one or more body parts of the user following the modification of the response dynamics may also be agnostic to historical performance information of the user, or to other data about the user. In addition or instead, the processing of the kinematic patterns of the identified motion patterns to determine motion patterns indicative of recognized cognitive states may also be agnostic to historical performance information of the user, or to other data about the user. It is noted that user-agnostic operation is just one alternative, and that in other implementations different levels of user information may be used for different aspects of the processing. Such user information may range from very general (e.g., age and gender) to the very specific (e.g., prior recordings of kinematic data of the same user). In some implementations, tailoring the disclosed technology to specific users may be especially beneficial (for example, when specific individuals are tasked with making very significant decisions).

[0180] Fig. 10 illustrates examples of kinematic data and analysis of motion patterns in that kinematic data, in accordance with examples of the presently disclosed subject matter. Such analysis may include, for example, various segmentation techniques, such as (but not limited to) techniques which are based on the differentiation of the sub-movements. In Fig. 10, the measured / sensed motion pattern MP(t) is shown represented by a curve Cl corresponding to direction changes of a continuous handmotion. Curve C2 represents motion acceleration changes. While the processing of the motion pattern by system 200 and / or in any one of methods 500, 600, and 900 does not necessarily involves identifying, distinguishing, or processing submovements in the motion patterns of the respective body portion, the following discussion explains the specific motion of the hand of the user in such terms, for clarity. Segment A of the curve Cl presents initiation of goal-directed movement; segment B presents undershoot corrective sub-movement immediately after pick acceleration; segment C corresponds to overshoot corrective sub-movement immediately after pick acceleration; segment D presents undershoot corrective sub-movement immediately before movement termination; segment E presents overshoot corrective submovement immediately before movement termination; and segment F presents termination of the goal-directed movement. Segments A, B, C, D, E, and F are just one of the possible many ways to analyze the kinematics of the hand during the motion.

[0181] The characteristics and prevalence of sub-movements can result from different task constraints. As previously mentioned, in the context of the current application, the relevant motion patterns may be interpreted as corrective adjustments resulting from changes of the response dynamic of the input user interface during the performance of the task, possibly resulting from the aforementioned mechanisms in the brain of the user.

[0182] However, there are other motion pattern interpretations and even other submovements interpretations. For example, sub-movements may be interpreted as a property of movement control. More specifically, sub-movements may be movement primitives used as building blocks of normal movements, thus having no direct relation to accuracy requirements. Also, many sub-movements may represent irregular velocity fluctuations, emerging due to noise in the kinematic output (i.e., muscle elasticity, coactivation, non- smooth activation of motor units, and noise in the neural circuitry involved in movement control).

[0183] Thus, it might be difficult to distinguish corrective and non-corrective submovements based on kinematic analyses. Corrective and non-corrective submovements have similar kinematic characters, reflected by velocity profile modulations, usually measured by zero crossings of the first three or four displacement derivatives. The distinction may be based on fitting movement trajectory with seriesof bell-shaped functions of scaled duration and amplitude. However, submovements extracted according to these methodologies can be either corrective or non-corrective. According to some other approaches, if the sub-movement brings the trajectory closer to the target it is a corrective one. However, noisy target-aimed motions may have the same or similar characteristics as a series of corrective submovements. According to yet further approaches, motion termination may cause submovements because it requires dissipation of movement mechanical energy and stabilization of the arm at the target. In discrete movements, motion termination results in complete halt of both velocity and acceleration. However, in continuous movements that reverse without residing on target, only the velocity, and not acceleration, is abolished at the target. The stabilization of the limb at the target in discrete motions may cause submovements, absent in continuous movements. Accordingly, submovements revealed with the lower derivatives (gross submovements) are often caused by motion termination in discrete motions but not in continuous motions. Conversely, submovements revealed with higher derivatives of motion (fine submovements) might be more related with corrective maneuvers associated with higher accuracy demands and occur in both discrete and continuous motions. However, during cyclical movements, incidence of fine submovements depends on cyclic frequency (frequency of periodic movement) and not on accuracy demands. Hence, slow movements may be prone to irregularities observed as fine submovements, and since highly accurate motions are also slower, these movements are characterized by non-corrective fine submovements.

[0184] Alternatively or additionally, the motion command data being measured / sensed may include a force (or any derivative thereof such as pressure) applied by the body part (e.g., on a certain device) when performing the action. In general, the force applied when erroneous action is performed is different from the force applied when a non-erroneous action is performed. The applied force can be measured on the individual's body part which is applying the force or on the device on which the force is being applied. Similar to the acceleration and the derivative of acceleration / deceleration in relation to time, the rate of change in the applied force can be calculated and used as an indication of an erroneous action.

[0185] The motion command data being measured / sensed may also include time to lift parameter being measured as a time interval before the body part is lifted from a certain device on which the action is applied, or a time interval before the pressure applied on the device is alleviated, or a time interval before an electric circle closed by the action, opens again. For example, a time to lift period can be measured from the moment of initial contact of a body part with the device until the body part is lifted from the device. In general lifting time shortens when the action is a result of an erroneous command as compared to an action which results from a non-erroneous command.

[0186] The sampling frequency used in measurements of error detection-related kinematics and other information related to the measurement of motion command data is preferably above 50 Hz and, preferably, above 100 Hz.

[0187] Referring to system 200 as well as to methods 500, 600, and 900, it is noted that the present invention may be implemented using various approaches, including but not limited to: machine learning techniques, heuristic algorithms, expert systems, statistical methods, fuzzy logic systems, or other suitable computational strategies.

[0188] Referring to the option of utilizing machine learning in system 200 and / or for the execution of methods 500, 600, and 900, it is noted that machine learning algorithms may be used for analyzing the kinematic data and / or the preprocessed data and to infer from it a person's cognitive state. The algorithm may include supervised or unsupervised learning methods, such as decision trees, random forests, neural networks, or support vector machines. Such a machine learning algorithm may be trained on a dataset of labeled data that includes information about the person's cognitive state and their corresponding body movements during the perturbation task.

[0189] Referring to any one of methods 500, 600, and 900, it is noted that the respective method may be executed by executing by one or more processors the steps of the respective method which are stored as a computer readable code on a non- transitory computer-readable medium.

[0190] According to an aspect of the invention, there is disclosed a non-transitory computer-readable medium for assessing a cognitive state of a user, comprising instructions stored thereon, that when executed on a processor, perform the steps of method 500 (in any one or more of its variations discussed above). According to anaspect of the invention, there is disclosed a non-transitory computer-readable medium for assessing a cognitive state of a user, comprising instructions stored thereon, that when executed on a processor, perform the steps of method 600 (in any one or more of its variations discussed above). According to an aspect of the invention, there is disclosed a non-transitory computer-readable medium for selectively granting control of a machine (or another type of user controllable system) based on assessment of a cognitive state of a user, comprising instructions stored thereon, that when executed on a processor, perform the steps of method 900 (in any one or more of its variations discussed above). According to an aspect of the invention, disclosed is a program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method for assessing a cognitive state of a user comprising the steps of any one or more variations of methods 500, 600, or 900.

[0191] While the embodiments described above are provided as examples, it should be understood that various modifications and substitutions may be made without departing from the scope of the invention as defined in the appended claims.

Claims

CLAIMSWhat is claimed is:

1. A system for assessing a cognitive state of a user by monitoring execution of a task by the user, the system comprising: a hardware output user interface for providing to the user information indicative of progression of the user at performing the task; an input user interface controllable by the user for performing the task; wherein response dynamics of the input user interface to user instructions are changed by the system during the performance of the task, thereby resulting in highly fluctuational motion of at least one body part of the user; at least one processor; and at least one memory including computer program code, where the at least one memory and the computer program code are configured, with the at least one processor, to cause the system to at least: receive over at least one hardware communication channel sensor based kinematic data, the kinematic data being indicative of kinematic parameters of the at least one body part at different times during the performance of the task; process the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and determine a cognitive state of the user based on the motion patterns.

2. The system according to claim 1, wherein the sensor comprises at least one sensor selected from the group consisting of: a camera, an accelerometer, a RADAR, a touch screen, and a joystick.

3. The system according to any one of claims 1 and 2, comprising a cognitive stategated access authorization module which is operable to selectively grant users control of a requested system based on cognitive state assessment, wherein the cognitive state-gated access authorization module is configured to:permit control by the user of the requested system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level; repeat the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and prevent control by the user of the requested system in response to the determination of the second cognitive state.

4. The system according to any one of claims 1—3, wherein the processor is configured to determine the cognitive state of the user without accessing historical performance information of the user.

5. The system according to any one of claims 1-4, wherein the input user interface is further used to receive user instructions for operating of a machine to which access is selectively granted by the system based on the determined cognitive state.

6. The system according to any one of claims 1-5, wherein the kinematic data is indicative of motion patterns originating from a basal ganglia of the user which is stimulated by the performance of the task.

7. The system according to any one of claims 1-6, wherein the processor is configured to determine real-time modification parameters for modifying the response dynamics of the input user interface in response to the kinematic data.

8. The system according to any one of claims 1-7, wherein the system is a vehicle which comprises: (a) an engine operable to provide power for propelling the vehicle, (b) a user controllable steering mechanism for controllably changing a propagation direction of the vehicle, and (c) input user interface for detecting user instructions for modifying a behavior of at least one module out of the engine and the steering mechanism, wherein the processor is operable to selectively prevent controlling of performance of the at least one module by the input user interface based on the determined cognitive state.

9. The system according to any one of claims 1-8, where the at least one memory and the computer program code are configured, with the at least one processor, to cause the system to at least: process the kinematic data to identify submovements of the user during the performance of the task, and determine the cognitive state of the user based on analysis of the submovements.

10. A method for assessing a cognitive state of a user, the method comprising: receiving, over at least one hardware communication channel, sensor based kinematic data that is indicative of kinematic parameters of at least one body part of the user at different times during a performance by the user of a task that includes operating an input user interface whose response dynamics to user instructions change during the performance of the task, thereby resulting in highly fluctuational motion of the at least one body part; processing the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and determining a cognitive state of the user based on the motion patterns.

11. The method according to claim 10, wherein the determining is followed by: permitting controlling by the user of another system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and preventing control by the user of the other system in response to the determination of the second cognitive state.

12. The method according to any one of claims 10 and 11, wherein the determining of the cognitive state of the user is based on decision model which is agnostic to historical performance information of the user.

13. The method according to any one of claims 11-12, comprising selectively permitting the user to operate a vehicle via a plurality of input user interfaces which exclude any input user interface used for the performance of the task.

14. The method according to any one of claims 11-13, comprising selectively permitting the user to operate a vehicle via a plurality of input user interfaces which includes the input user interface used for the performance of the task.

15. The method according to any one of claims 11-14, wherein the kinematic data is indicative of highly fluctuational motion originating from a basal ganglia of the user which is stimulated by the performance of the task.

16. The method according to any one of claims 11-15, comprising modifying the response dynamics of the input user interface in response to the kinematic data.

17. The method according to any one of claims 11-16, wherein the processing comprises processing the kinematic data to identify submovements of the user during the performance of the task; wherein the determining comprises determining a cognitive state of the user based on analysis of the submovements.

18. A non-transitory computer-readable medium for assessing a cognitive state of a user, comprising instructions stored thereon, that when executed on a processor, perform the steps of: receiving, over at least one hardware communication channel, sensor based kinematic data that is indicative of kinematic parameters of at least one body part of the user at different times during a performance by the user of a task that includes operating an input user interface whose response dynamics to user instructions change during the performance of the task, thereby resulting in highly fluctuational motion of the at least one body part; processing the sensor-based kinematic data to characterize motion patterns of the at least one body part during the highly fluctuational motion; and determining a cognitive state of the user based on the motion patterns.

19. The non-transitory computer-readable medium according to claim 18, wherein the determining is followed by:permitting controlling by the user of another system in response to determination of the cognitive state which is based on kinematic data collected during performance of the task by the user at a first success level, repeating the steps of receiving, processing, and determining to determine a second cognitive state of the user based on other kinematic data collected during performance of a second task by the user at a second success level that is higher than the first success level; and preventing control by the user of the other system in response to the determination of the second cognitive state.

20. The non-transitory computer-readable medium according to any one of claims 18 and 19, comprising modifying the response dynamics of the input user interface in response to the kinematic data.

Images