Method and control circuit for operating an operating interface of a device and correspondingly operable device, e.g. motor vehicle

Abstract

Method for operating an operating interface (16) of a device (10), wherein a user input (18) of a user (15) is recognized by a control circuit (19) on an input device (17) of the operating interface (16), and depending on the recognized user input (18), a system output (22) is output to the user (15) via an output device (21), wherein in an operating assistance procedure of the control circuit (19), at least one reaction signal (28) of a body of the user (15) generated by the user (15) after the system output (22) is detected, and the at least one reaction signal (28) is compared with a signal template (30) for user dissatisfaction (23), and upon detection of a match between one or more or each reaction signal (28) and the associated signal template (30), at least one adaptation measure (31) for changing the system output (22) and / or a future system output (22) is triggered,wherein the operating assistance procedure is only executed if, during and / or after the user input (18) and / or during the system output (22), a filter module of the control circuit (19) signals that the at least one reaction signal (28) is unaffected by at least one predetermined competing stimulus, wherein, on the basis of at least one observation sensor, the filter module detects as a competing stimulus an operation performed by the user (15) on at least one unit different from the operating interface (16).

Description

[0001] The invention relates to a method for operating a user interface of a device. Such a device can be, for example, a motor vehicle, in which case the user interface is a system for operating the motor vehicle, for example, the infotainment system (information entertainment system). The invention also includes a control circuit for operating the user interface of a device according to the method. Finally, the invention encompasses a device that can be configured, for example, as a motor vehicle in the manner described. The device can alternatively be configured, for example, as a mobile terminal, for example, a smartphone.

[0002] A user interface of a device can, for example, comprise a graphical output device (e.g., a pixel-based screen and / or a head-up display) and / or a voice-based output device (e.g., a voice dialogue system). To enable a user to trigger a function of the device, such as playing media content such as an MP3 file, an input device is also provided on such a user interface, which can, for example, provide a touchscreen and / or voice input.

[0003] Intuitive usability of such a user interface with a display / operating concept in a motor vehicle is a key prerequisite for safe and acceptable use by the user or driver. This affects not only motor vehicles, but also devices in general, including mobile devices such as smartphones. Devices with such a system (user interface) may encounter operating hurdles that prevent users from using the system intuitively (e.g., information cannot be found in the menu; the voice assistant executes a command incorrectly).

[0004] Different user characteristics (tech-savvy, tech-inexperienced, old, young, novice drivers, to name just a few) can also lead to these systems being operated with varying degrees of success by different users. This can lead to frustration, acceptance problems, or avoidance of future use if the system is not intuitive to use.

[0005] From a scientific publication by FREY, Jérémy [et al.]: Framework for electroencephalography-based evaluation of user experience. In: ACM: CHI Conference on Human Factors in Computing Systems - 7-12 May 2016 - San José, CA, USA, 2016, pp. 2283-2294. - ISBN 978-1-4503-3362-7. DOI: 10.1145 / 2858036.2858525. URL: https: / / dl.acm.org / doi / pdf / 10.1145 / 2858036.2858525 [accessed on 2020-08-14] It is known that an event-related potential (ERP) can be measured in a human electroencephalogram (EEG), which represents a person's cognitive response to whether or not the person expected a certain system output of a user interface.If a user interface responds with a system output that a user did not expect after performing an operating step, a corresponding reaction signal can be detected on the EEG, for example, which signals that the user perceives an interaction error when operating the user interface.

[0006] A scientific publication by TIWARI, Ravikumar K.; GIRIPUNJE, Shubhangi: Design approach for eeg-based human computer interaction driver monitoring system. In: International Journal of Latest Trends in Engineering and Technology, Vol. 3, 2014, No. 4, pp. 250-255. - ISSN 2319-3778 (P); 2278-621X (E). DOI: 10.21172. URL: https: / / www.ijltet.org / wp-content / uploads / 2014 / 04 / 42.pdf [accessed on 2020-09-17] describes how an EEG can be used to measure a driver's attention while performing their driving task (operating a motor vehicle). This can be used to prepare a driver assistance system for a possible intervention that might become necessary due to driver inattention.

[0007] It is known from US 2010 / 0082516 A1 that a user can be enabled to signal their frustration with the operation verbally or by pressing a button, after which an operating concept can be switched within the user interface. Therefore, if a user notices that they are having difficulty operating such an interface, they must actively request a modification of the operating concept, for example, by activating a beginner mode.

[0008] US 2007 / 0 061 735 A1 discloses an adaptive user interface that detects a user's mood and adapts a user menu accordingly. User observation can be video-based, for which a filter component can also be used to filter the video signal.

[0009] US 10 170 111 B2 discloses using video surveillance to additionally check the user's mood during voice control to determine the system's response. Furthermore, the system monitors how distracting a current driving situation is for the user in order to delay the system's output in confusing driving situations.

[0010] US 2010 / 0 082 516 A1 discloses determining a user's frustration level while operating a user interface. This can then be used to predict in which operating situations the user will react with frustration in the future.

[0011] DE 10 2018 221 271 A1 discloses measuring an EEG in a motor vehicle driver to determine, for example, the driver's degree of fatigue. The EEG signal can be disrupted by, for example, muscle twitches or the interference from electrical amplifiers, which is why the EEG signal is corrected for such artifacts.

[0012] The invention is based on the object of making the operating interface of a device operable for a user with a low error rate.

[0013] This object is achieved by the subject matter of the independent patent claims. Advantageous embodiments of the invention are described by the dependent patent claims, the following description, and the figure.

[0014] The invention provides a method for operating a user interface of a device. A control circuit on an input device of the user interface detects a user input and, depending on the detected user input, a system output is output to the user via an output device. In other words, the user interface provides a coupling of input device (e.g., touchscreen) and output device (e.g.,Screen) is provided in which a user can specify their operating request or a selection of a function of the device to be activated on the input device as user input, and the operating interface then responds with the system output, which should preferably consist of the function desired by the user being activated or executed and the result of the function execution then being presented to the user as system output via the output device. For example, the user input can consist of the selection of a radio function or a media file, and the system output should then be the playback of a radio station or the media file.

[0015] However, this can lead to incorrect operation by the user or to incorrect recognition of the user input by the user interface. If the user selects the wrong submenu, for example, they cannot access the function that is suitable for fulfilling their operating request (e.g., the radio functionality or media playback). If the user executes the user input, for example, as an operating gesture on a touchscreen or as a voice input, the user input may be correct, but an incorrect recognition may occur on the user interface side, i.e., a different user input is recognized than the one actually performed by the user. Accordingly, this also results in incorrect system output. A "function" can also be understood here as switching between menus in a menu hierarchy, i.e.An incorrect system output can also mean that the user ends up in a submenu that does not contain the function he is looking for.

[0016] In order to technically further develop the user interface, the invention provides that during the user input and / or thereafter (during and / or after the system output) in an operating assistance procedure of the control circuit, at least one reaction signal from a user's body is detected and the at least one reaction signal is compared with a signal template for user dissatisfaction. Upon detection of a match between one or more or each reaction signal and the associated signal template, at least one adaptation measure for changing the system output and / or a future system output (if the same operating situation occurs again) is triggered. The operating assistance procedure can be designed as a program code or software module of the control circuit.They evaluate whether the user's response signal indicates that the system output does not correspond to the system output expected by the user in response to their input. The response signal can be a conscious response by the user or an unconscious response. It can include an externally visible action and / or a cognitive response and / or a vegetative response (e.g. pulse rate and / or skin conductance). The response signal can be a signal value or a time course of a signal recorded on the user's body. The respective signal template used to test the response signal can be determined on the basis of experiments with test subjects.In such test subjects, a respective reaction signal can be measured when the user interface responds as desired (system output matches user input or user expectations) and how they react when the users are dissatisfied because the system output does not match the user input or user expectations. The system output can be specifically controlled in experiments. Accordingly, exemplary signal curves arise that can be generalized or summarized into signal templates through parameterization and / or a pattern description (for example, using an HMM - Hidden Markov Model) and / or using an artificial neural network. This can then be used as a general representation for a reaction signal from a dissatisfied user.

[0017] If a response signal matches a signal template for user dissatisfaction, it is detected or recognized that the system output does not correspond to the user request or intention as expressed by the user input, and / or the user expectation. Accordingly, the control circuit triggers the adaptation measure to change the system output. Therefore, if user dissatisfaction is detected based on a signal curve or a signal value of at least one response signal, it can be assumed that the user is frustrated and / or that the user interface is reacting incorrectly, and the control circuit triggers the adaptation measure accordingly.

[0018] The invention provides the advantage that the user interface reacts independently to a preferably unconscious and / or cognitive reaction signal of the user's body in such a way that the last system output is revised or changed and / or a behavior of the user interface is changed in the future because it is recognized that the user was dissatisfied with the system output.

[0019] According to the invention, said operating assistance procedure is only executed if, during and / or after the user input and / or during the system output, a filter module of the control circuit signals that the at least one reaction signal is unaffected by at least one predetermined competing stimulus. In a further development, said operating assistance procedure is only executed if, during and / or after the user input and / or during the system output, a filter module of the control circuit signals that the at least one reaction signal is free of at least one predetermined artifact. In other words, a check is carried out to determine whether the at least one reaction signal could also signal user dissatisfaction for another reason, i.e., not due to the system output.If the user only operates the user interface casually and then immediately after the user input no longer concentrates on the user interface and / or performs an activity that could also influence the reaction signal, for example a cough or a sneeze, this can also result in a match between the respective reaction signal and the respective signal template for user dissatisfaction, even though this match occurs independently of the system output of the output interface. In order to avoid misinterpretation of at least one reaction signal, it is detected whether an artifact could be present in the reaction signal and / or whether the user is confronted with at least one competing stimulus in addition to the system output. An artifact is a signal component of the respective reaction signal that is independent of the system output.A competing stimulus is a user-perceivable event that is independent of the system output. How to detect a corresponding artifact and / or competing stimulus depends on the specific artifact and / or stimulus, so those skilled in the art can conduct appropriate tests to exclude the influence of corresponding artifacts and / or stimuli and provide appropriate criteria in the control circuitry for detecting an artifact and / or stimulus.

[0020] The invention also includes embodiments which provide additional advantages.

[0021] There are particularly preferred embodiments for detecting an artifact and / or stimulus. If the user performs a movement, this typically also influences a sensor signal recorded on the user's body, in that the response signal may contain an artifact, i.e., a signal component that is not part of the response to the system output itself and yet still fits the signal template.

[0022] In one embodiment, the filter module, based on at least one observation sensor, detects as an artifact a movement on a body part of the user that has a jerk value greater than a predetermined threshold and / or detects as a competing stimulus an operation performed by the user on at least one unit other than the operating interface. The jerk is the time derivative of the acceleration. Detecting a movement with a jerk value greater than a predetermined threshold has the advantage of detecting a sudden change in the user's movement, which is highly likely to generate an artifact in at least one response signal. If it is detected that the user is operating a unit other than the operating interface, it can be assumed that a competing or different system output on this unit may also dissatisfy the user.

[0023] One embodiment provides that the device is designed as a motor vehicle and an operation of a respective unit for longitudinal and / or lateral guidance of the motor vehicle and / or a driving task to be carried out due to a deactivated state of a driver assistance device of the motor vehicle is detected as a competing stimulus. In the event that the device is a motor vehicle, a competing stimulus can arise from the driving task itself, i.e. from driving the motor vehicle by steering and / or braking and / or accelerating, because the user, for example, pays attention to and / or reacts to road traffic. This can be recognized by corresponding sensory detection of operating actions on the steering wheel and / or brake pedal and / or accelerator pedal.If an autonomous driving function (autopilot) is provided in the vehicle but is deactivated, this can also be seen as an indication that the user is confronted with a competing stimulus. In general (not only for an autopilot), the deactivated state of a driver assistance system (e.g., parking assistance and / or lane keeping assistance) can be interpreted as an indication that a competing stimulus may be present.

[0024] One embodiment provides that the control circuit receives a respective reaction signal from a sensor device different from the input device, an EEG signal from an electroencephalograph and / or an emotion signal from a speech analysis device and / or from a camera-based facial expression evaluation, and / or a corrective user input from the input device, which causes a revision of the system output. An EEG signal from an electroencephalograph has proven to be a particularly random signal source for generating a meaningful reaction signal. This allows a user's disappointment and / or annoyance and / or dissatisfaction in response to a system output to be detected based on a change in the temporal progression of the signal components of the EEG signal.Using a speech analysis device as a signal source, a change in tone of voice (speech melody and / or speech speed) and / or word choice can be mapped or interpreted in such a way that a change in the user's mood can be inferred. Corresponding speech analysis devices are available in the state of the art. A facial expression analysis of a camera image using a corresponding software module can, for example, detect whether the user is furrowing their eyebrows and / or frowning and / or rolling their eyes and / or opening their mouth, which can also be signaled as disappointment or user dissatisfaction. A corresponding signal source with facial expression analysis can, for example, be based on a camera and a downstream image processing device provided by software that can segment the user's face and analyze facial expressions.Another signal source can be the input device itself, which can detect at least one additional user input after the system output has been output. If the additional user input causes the system output to be revised or reversed—for example, if the user returns to the parent menu after calling up a submenu—this can also be interpreted as a reaction signal indicating user dissatisfaction.

[0025] One embodiment provides that a reaction signal is received from different signal sources, and the at least one adaptation measure is only triggered if more than one or each of the reaction signals matches the respective signal template. In other words, multiple signal sources are used to perform a plausibility check. If, in particular, an EEG signal is used as a reaction signal, at least one of the reaction signals from the emotion signal of the speech analysis device and / or the facial expression evaluation and / or the corrective user input can also be evaluated. The adaptation measure is only triggered if more than one reaction signal, in particular the combination of the EEG signal and at least one further reaction signal, indicates user dissatisfaction (matching the respective signal template).The advantage resulting therefrom is that the error rate in detecting user dissatisfaction is reduced.

[0026] One embodiment provides that the system output includes a speech statement of a speech assistant and includes at least one adjustment measure, that the speech statement is revised and / or an apology is issued, and / or wherein the system output includes a warning and includes at least one adjustment measure, that the warning is suppressed in the future. The adjustment measures described herein have the advantage that they contribute to reducing user dissatisfaction. If a user is annoyed by a speech statement or a warning or considers it unnecessary, this is prevented with these embodiments.

[0027] One embodiment provides that the operating interface provides an operating path with one or more operating steps for triggering a predetermined function, and the at least one adaptation measure comprises changing the operating path with respect to the included operating steps (with associated system output) and / or at least one provided operating instruction. If the multiple operating steps consist, for example, in the user having to navigate through a menu hierarchy or a menu structure using consecutive operating steps, the change in the operating path can consist of moving a menu entry for a function that the user intended to activate through user input up in the menu hierarchy, so that the operating path is shortened for future operation.If, when calling up or operating the user interface based on at least one reaction signal, it turns out that the user is dissatisfied when entering the user input and / or when receiving the system output, for example because another submenu repeatedly appears instead of the desired function, this user dissatisfaction can be avoided in the future by adapting the operating path.

[0028] The change in the operating path can be continued or adapted iteratively, for example with each user input, and in this case, for example, the minimization of user dissatisfaction (as determined by the at least one reaction signal) can be sought using a minimization function. To this end, a change can be made and, if user dissatisfaction increases again, this change can be reversed and another change in the operating path can be made. In the manner described, a menu hierarchy can be changed and / or a word selection of a voice output can be changed, for example with regard to vocabulary, and / or a number of operating instructions and / or a rate of operating instructions can be changed.

[0029] One embodiment provides that the input device of the user interface comprises at least one of the following input components: a multi-finger touch gesture recognition system, a speech recognizer, a gesture recognizer (static and / or dynamic gestures executed freely in space), and an operating menu with a multi-level menu hierarchy. The automated triggering of adaptation measures has proven particularly effective for the aforementioned input components of a user interface.

[0030] The invention also provides a control circuit for a user interface of a device, wherein the control circuit is configured to carry out an embodiment of the method according to the invention. The control circuit can comprise a data processing device or a processor device configured to carry out an embodiment of the method according to the invention. For this purpose, the processor device can comprise at least one microprocessor and / or at least one microcontroller and / or at least one FPGA (Field Programmable Gate Array) and / or at least one DSP (Digital Signal Processor). Furthermore, the processor device can comprise program code configured to carry out the embodiment of the method according to the invention when executed by the processor device. The program code can be stored in a data memory of the processor device.The control circuit can be coupled to the input interface and the output interface of the operating interface, for example via cables and / or conductor tracks of a circuit board and / or a data bus (for example MOST - Media Oriented Systems Transport, CAN - Controller Area Network).

[0031] The invention also provides a device with a user interface comprising an input device for user input and an output device for outputting a system output, and an embodiment of the control circuit according to the invention. Such a device can be designed to be portable, for example, as a smartphone or vehicle.

[0032] One embodiment of the device provides that the device is configured as a motor vehicle, a mobile device, or a desktop computer. The motor vehicle according to the invention is preferably configured as a motor vehicle, in particular as a passenger car or truck, or as a passenger bus or motorcycle. As a mobile device, the device can be configured, for example, as a smartphone, tablet PC, or smartwatch. As a desktop computer, the device can be configured as a so-called PC (personal computer). The user interface can be implemented, for example, by application software.

[0033] The invention also includes implementations which each have a combination of the features of several of the described embodiments, unless these embodiments are described as mutually exclusive.

[0034] Exemplary embodiments of the invention are described below. The single figure shows: Fig. a schematic representation of an embodiment of the device according to the invention.

[0035] The exemplary embodiments explained below are preferred embodiments of the invention. In the exemplary embodiments, the described components of the embodiments each represent individual features of the invention that can be considered independently of one another, each of which also develops the invention independently of one another. Therefore, the disclosure is intended to encompass combinations of the features of the embodiments other than those shown. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

[0036] In the figure, the same reference symbols denote elements with the same function.

[0037] The figure shows a device 10, which can be a motor vehicle, a mobile device, or a desktop computer. The device 10 can provide a range of functions 11 with one or more device functions 12, 13. The functions are symbolically designated F1, F2, and possible additional functions are symbolized by ellipsis 14.

[0038] To trigger or select the device functions 12, 13, a user 15 can operate an operating interface 16 of the device 10, for example, by activating a user input 18 by voice or with one hand on an input device 17 of the operating interface 16. The operating interface 16 can, for example, represent an operating menu, and the user input 18 can navigate through a menu hierarchy of the operating menu to select a menu entry that should trigger the activation or selection of one of the device functions 12, 13.

[0039] After the user input 18, one of the device functions 12, 13 can be selected by a control circuit 19 of the operating interface 16 based on an actually recognized user input 20 (i.e. the result of the measurement or recognition procedure of the user input 18) and a corresponding result for the user input 18 can be presented on an output device 21. It should be noted that the user input 18 is the true user input that the user 15 wanted or intended to activate, whereas the recognized user input 20 can differ from this, for example in the case of an incorrect recognition. Another source of error can be that the user 15 makes a user input 18 that is also correctly recognized as recognized user input 20, but the user 15 has made an operating error.

[0040] Based on the system output 22 presented by the user interface 16 on the output device 21, user 15 can determine whether or not the operation was successful. If unsuccessful, the user will subconsciously experience user dissatisfaction 23, which is represented here by a "thumbs down."

[0041] The device 10 can be provided to detect this user dissatisfaction 23 during and / or after the output of the system output 22. For this purpose, at least one signal source 24 can be provided. One signal source 24 can be an electroencephalograph 25 (EEG). Another signal source 24 can be, for example, a camera 26 with a downstream image analysis device and / or a speech recognition device. A respective reception area 27 of the respective signal source can be directed toward the body of the user 15.

[0042] The respective signal source 24 can generate a respective user reaction signal 28 to the system output 22 and / or the effort required to enter the user input 18. The control circuit 19 can use signal templates 30 to check the respective reaction signal 28 to determine whether it meets a criterion for user dissatisfaction 23. Corresponding signal templates 30 can be generated or determined with the help of test subjects, and a respective signal source 24 can be provided in an adapted manner.

[0043] If user dissatisfaction 23 is detected by means of the at least one reaction signal 28, at least one adaptation measure 31 can be triggered in the user interface 16 by the control circuit 19. One of those already described can be provided as a possible adaptation measure 31.

[0044] Overall, the result is that in the device 10, the user interface 16 adapts over time to the needs of the user 15 by triggering the at least one adaptation measure 31 upon at least one user input 18, so that the probability of user dissatisfaction 23 when operating the device 10 via the user interface 16 is reduced.

[0045] According to this idea, EEG detection is used for error / frustration recognition within a system that learns for specific interactions in the motor vehicle using additional sensors.

[0046] EEG can be used for error and / or frustration detection. The use of other sensory inputs for frustration detection, such as speech recognition or recognition of user actions (scrolling, multiple selections, etc.), allows for plausibility checks or even independent error and / or frustration detection without EEG.

[0047] This provides a way to record and process EEG-based signals in motor vehicles. The electroencephalogram enables the recording of event-related potentials (EKPs), which are wave signals associated with an observable event. The advantage of this for use in motor vehicles is the ability to adapt the forms of interaction (e.g., speech, prompts, menu navigation). The use of other sensory signals for frustration detection, such as speech recognition and the recognition of user actions (scrolling, multiple selection), enables the plausibility of the recognition results from the EEG. The control circuit preferably uses learning over time to adapt the user interface.

[0048] This allows a system to be made adaptable, especially an electroencephalogram (EGG)-based system, by detecting when a user, e.g., a driver, is dissatisfied with the interaction. The system can be adapted accordingly, e.g., by changing its menu structure, offering help, and recognizing operating rules.

[0049] The preferred use of EEG EKPs is for adapting the interaction / operating interface to a display / operating concept in the motor vehicle. EKPs are potential shifts that can be measured in the EEG before, during, or after a sensory, motor, or psychological event. In the event of errors, i.e., a mismatch between an expected, correct, and an actual reaction (error), certain patterns in the EEG become visible via EKPs. This makes it possible to determine at which point in an interaction with a system (e.g., vehicle menu, assistants) an error is perceived by the driver or user. Therefore, those EKPs that indicate an unexpected stimulus, a surprise, or a detected error should be recorded. This should then lead to intelligent, automatic adaptation of content in the motor vehicle. For example, following a perceived error: - Warning from the motor vehicle which will be suppressed in the future, - Statement of the voice assistant these are revised (correct / apologize), - Operating action / menu structure will be supported in the future (e.g. display of help buttons).

[0050] In particular, it is not intended that the entire interaction with the vehicle / menu be controlled via this method, but rather that EKPs are used specifically as input for detecting the driver's error perception and adapting the systems accordingly. The following EKPs are particularly suitable for this purpose: Error-Related Negativity, Mismatch Negativity, N200, and Error Positivity.

[0051] It is conceivable that the interpreted EKP information is enriched with further input parameters such as analysis of voice / utterances; emotion recognition through driver observation and the so-called EKP signal P300.

[0052] The advantages include, in particular, automated adaptation of the user interface to create an intuitive system in the vehicle, the avoidance of frustration and operating errors, increased passenger comfort, the creation of confidence in the vehicle, and individual adaptation to the user.

[0053] For the technical implementation, reliability can be increased through the aforementioned plausibility checks. Detection occurs in real time, as the EKPs occur approximately 50-500 milliseconds after the presentation of the error stimulus. Sensors that can be attached regardless of the driver's or user's head shape can be placed, for example, in a cap and / or headphones and / or glasses.

[0054] Overall, the examples show how event-related potentials can be used to adapt display content to user-perceived operating errors and system errors.

Claims

[1] Method for operating an operating interface (16) of a device (10), wherein a user input (18) of a user (15) is detected by a control circuit (19) on an input device (17) of the operating interface (16) and, depending on the detected user input (18), a system output (22) is output to the user (15) via an output device (21),wherein, in an operating assistance procedure of the control circuit (19), at least one reaction signal (28) of a body of the user (15) generated by the user (15) after the system output (22) is detected, and the at least one reaction signal (28) is compared with a signal template (30) for user dissatisfaction (23), and upon detection of a match between one or more or each reaction signal (28) and the associated signal template (30), at least one adaptation measure (31) for changing the system output (22) and / or a future system output (22) is triggered, wherein the operating assistance procedure is only executed if, during and / or after the user input (18) and / or during the system output (22), a filter module of the control circuit (19) signals that the at least one reaction signal (28) is unaffected by at least one predetermined competing stimulus,wherein the filter module detects, on the basis of at least one observation sensor as a competing stimulus, an operation performed by the user (15) on at least one unit different from the operating interface (16). [2] Method according to claim 1, wherein the operating assistance procedure is only carried out if during and / or after the user input (18) and / or during the system output (22) the filter module of the control circuit (19) signals that the at least one reaction signal (28) is free of at least one predetermined artifact. [3] The method of claim 2, wherein the filter module detects, on the basis of at least one observation sensor, as an artifact on a body part of the user (15), a movement having a jerk value greater than a predetermined threshold. [4] Method according to one of the preceding claims, wherein the device (10) is designed as a motor vehicle and an operation on a respective unit for longitudinal guidance and / or transverse guidance of the motor vehicle and / or a driving task to be carried out due to a deactivated state of a driver assistance device of the motor vehicle is detected as a competing stimulus. [5] Method according to one of the preceding claims, wherein a respective reaction signal (28), an EEG signal from an electroencephalograph (25) and / or an emotion signal from a speech analysis device and / or from a camera-based facial expression evaluation and / or a corrective user input (18) which causes a revision of the system output (22), is received from the input device (17) by the control circuit (19) from a sensor device different from the input device (17). [6] Method according to claim 5, wherein a respective reaction signal (28) is received from different predetermined signal sources (24) and the at least one adaptation measure (31) is only triggered if more than one or each of the received reaction signals (28) matches the respective signal template (30). [7] Method according to one of the preceding claims, wherein the system output (22) comprises a voice statement from a voice assistant and the at least one adaptation measure (31) comprises that the voice statement is revised and / or an apology is issued, and / or wherein the system output (22) comprises a warning and the at least one adaptation measure (31) comprises that the warning is suppressed in the future. [8] Method according to one of the preceding claims, wherein the operating interface (16) provides an operating path with one or more operating steps for triggering a predetermined function and the at least one adaptation measure (31) comprises that the operating path is changed with respect to the included operating steps and / or at least one provided operating instruction. [9] Method according to one of the preceding claims, wherein the input device (17) of the operating interface (16) comprises at least one of the following input components: a multi-finger touch gesture recognition, a speech recognizer, a gesture recognizer, an operating menu with a multi-level menu hierarchy. [10] Control circuit (19) for an operating interface (16) of a device (10), wherein the control circuit (19) is configured to carry out a method according to one of the preceding claims. [11] Device (10) with an operating interface (16) having an input device (17) for a user input (18) and an output device (21) for outputting a system output (22), and a control circuit (19) according to claim 10. [12] Device (10) according to claim 11, wherein the device (10) is designed as a motor vehicle or as a mobile terminal or as a desktop computer.

Images