Usability and Acceptance Analysis of Wearable BCI Devices

Abstract

In the current scientific and technological scenario, wearable neuroimaging devices represent a revolution in neuroscience and wearable technology. These tools combine the features of neuroimaging technologies with the convenience of wearable devices, enabling real-time exploration of brain activity in real-world contexts. This convergence defines new perspectives in scientific research, medical diagnosis, and human performance analysis. Technologies such as EEG and fNIRS enable the non-invasive monitoring of brain activity without the need for heavy clinical equipment. Indeed, miniaturization, portability, wireless communication, and energy efficiency are key objectives in the design of advanced devices. In such a scenario, comfort is a key requirement to enable widespread use in different contexts, requiring the design of lightweight and minimally invasive wearable devices. The literature review examines the impact of wearable EEG and fNIRS devices on the user in real-life and laboratory environments in terms of usability and acceptability. The study presents evaluation and design factors—applied to laboratory testing—defined to improve the quality and perception of the user experience and to ensure the accuracy of cognitive load detection. These results will be useful in defining wearable devices, new applications, and future challenges for BCI.

1. Introduction

Neuroergonomics, an interdisciplinary field of study and research into the brain and human behavior, combines neuroscience with ergonomics (or human factors) to maximize the benefits of both [1]. Therefore, the goal is not only to investigate the anatomy and physiology of the brain but also to shed light on how behavior and brain function relate to one another in the workplace and in daily life [2].

Continuous innovation in hardware and software technologies such as sensors, displays, processors, data storage, and algorithms are essential elements in the paradigm shift of computing devices. Wireless connectivity has enabled the creation of a new generation of smart and connected objects with assistive capabilities, ranging from wearable computing devices to connected physical items in the environment like sensors and specialist displays [3,4]. Wearable devices are defined by Khakurel et al. [5] as smart electronic devices that come in different types and are placed on or close to the human body to record and analyze psychophysiological data, including blood pressure, heart rate, emotions, and movements, using applications that are installed on the devices or on external devices (such as smartphones that are connected to the cloud). The development of such devices has contributed significantly to the improvement of individual and societal quality of life; among the different types of devices, self-quantifying devices stand out as the most obvious example of technological progress [5]. The study of mental workload provides two primary functions in the field of human factors: (i) quantifying the interaction between operators and a set of task demands, technological systems, or operational protocols and (ii) estimating the possibility of performance compromise during operational scenarios, which may be necessary for safety.

To date, research on the measurement of mental workload has outgrown the development of theoretical models, and this can be analyzed and evaluated based on performance or through subjective self-assessment and combined with studies in psychophysiology or neurophysiology [6].

The development of advanced portable neuroimaging (or brain imaging) techniques that enable non-invasive examination of the brain at work [1] has led to a revolution in the field of understanding the neural mechanisms that are essential to human attention and performance over the past three decades [7]. Also, the market for neurofeedback systems is expected to expand at a Compound Annual Growth Rate (CAGR) of 7.72% from USD 1.40 billion in 2025 to USD 2.03 billion by 2030. The study of the neural basis of perceptual and cognitive processes, including vision, hearing, memory, decision-making, and planning, in relation to real-world technologies and environments is the focus of neuroergonomics [1]. The structure, function, or pharmacology of the nervous system can be directly or indirectly mapped [8] using brain imaging technologies.

Norman, in “The Design of Everyday Things”, emphasizes how all artificial things are designed, focusing on the interaction between technology and people to ensure that products meet human needs while being understandable and usable. Specifically, physiological sensing devices have been used to assess and control psychophysiological states [9,10,11], monitor personal health, evaluate stress [12], prevent brain injury [13], and control robots, robotic arms, and devices like wheelchairs by detecting eye movements and brain waves [14,15,16]. Additionally, they are employed in cognitive training, stress reduction, and emotion detection [7,17].

There are two types of neuroimaging techniques: those that detect neural activity directly and those that represent metabolic brain processes linked to neural activity [7]. The most popular in vivo neuroimaging technology is electroencephalography (EEG), which measures electrical activity in the brain from the scalp. Other options include electrocorticography (ECoG), functional magnetic resonance imaging (fMRI), and functional near-infrared spectroscopy (fNIRS) [18]. The sensitivity of each type of measurement with neuroimaging technologies may be variable depending on the level of workload experienced by the user [2] and the configuration of the wearable device used. Three criteria can be used to characterize the technology under consideration: (1) ease of use in terms of human factor/ergonomics; (2) temporal resolution in terms of determining the timing of neural processing; and (3) spatial resolution in terms of localizing neural activity throughout the brain [1].

In addition to being widely employed in psychological, neurological, and cognitive research, as well as in the development of brain–computer interfaces, functional neuroimaging is used to identify metabolic disorders and injuries. Recent years have seen notable advancements in the field of research known as neuro-engineering, or brain–computer interface (BCI). Numerous studies have focused on the possibility of using EEG for applications in everyday contexts [7,18] and BCI for controlling exoskeletons, robotic arms, cursors, and computer spelling [15,19]. Several prototypes of EEG-based BCI have already been validated in real-world contexts; however, there are still problems that limit the development and use of such technologies, such as their ease of use and user acceptability.

Although brain–computer interface (BCI) technology has advanced significantly, there is still a lot to learn about how end users see and use these devices in practical settings. To maximize user experience, enhance device accessibility, and guarantee the smooth integration of BCI technology into several contexts, these gaps must be filled. This study proposes to investigate, through the performance and usability analysis of case studies, wearable devices that integrate neuroimaging technologies currently on the market, evaluating their strengths and possible transversality in various intended uses.

2. Usability and Acceptability of Brain–Computer Interface Devices

Usability, interoperability, accessibility, portability, and a lack of standards (devices, performance, clinical, and end-user metrics) are some of the primary issues that have been found in the deployment of these devices to end users [19]. For example, current commercial EEG headsets and bands are either cost-prohibitive if they return good data, or they are affordable but lack interoperability—the seamless communication and collaboration between various systems or devices—or do not return enough data to be used in different fields, such as BCI or studies of cognitive overload—the status when the brain is overloaded with information, efficiency, and decision-making are reduced—and mental fatigue [15,20]. EEG tends to have low spatial resolution and a high frequency of disruptions, including motion artifacts and eye movements, despite its superior temporal resolution [21]. Early in the development phase, the decision to adopt such on-chip real-time signal approaches should be made because many of the widely used signal de-noising techniques are not appropriate for real-time or mobile applications [22]. Furthermore, the user experience is greatly impacted by the device’s design, usability, and dependability. Adapting to various head shapes and sizes, hair types, user preferences [19], and the movements the user is required to do [6] is a problem when developing an EEG headset.

Although a one-size-fits-all configuration is preferable, the adaptability of the system must be considered early in the design process and tested thoroughly.

It is well known that high contact pressure can cause pain or discomfort; in fact, according to Smulders et al. [23], examining pressure sensitivity—which is defined as the pressure discomfort threshold of various areas—may result in new specifications and/or design guidelines. Therefore, the contact pressure characteristics of a product on the skin that may affect the user’s comfort experience must be taken into consideration as one of the relevant variables in the design process of gadgets that are worn, touched, and with which users have physical interaction [23].

The results of a 3D Pressure Discomfort Threshold (PDT) mapping study of the head, face, and neck conducted on 28 users are presented by Smulders et al. [23]. In contrast to other research, this work presented a PDT map that had a high density of landmarks and a novel technique that merged 3D landmark positions with pressure pain threshold data. According to the findings, the areas with the lowest pressure tolerance are the nose, front of the neck, mouth, jaw, cheeks, and cheekbones; the areas with the medium pressure tolerance are the forehead, temples, and back of the neck; and the areas with the highest-pressure tolerance are the scalp and rear of the head. However, these regions represent the main areas of interest for BCI devices, and for this reason, these must involve in-depth studies to ensure comfort and usability. Norman [24] outlines a three-level model for proper product design. This model includes the visceral level, which is the most fundamental and immediate level and deals with our initial responses to the product’s visual or sensory features (such as its quality and appearance). The reflexive level, which encompasses conscious cognition, comes after the behavioral level, which is related to the product’s usability features. In a broader sense, the reflexive level describes how well the product aligns with the user’s image and addresses social expectations in addition to mental and emotional factors [24].

These elements have an impact on how consumers evaluate a device’s acceptance. A number of studies have looked at models of device acceptance that are based on the Expectation Confirmation Theory (ECT), the Unified Theory of Acceptance and Use of Technology (UTAUT), and the Technology Acceptance Model (TAM). In order to establish recommendations for examining the motivational elements influencing users’ intention to use smart wearable devices, Park’s model [25] synthesizes ECT and TAM.

This model (Figure 1) analyses the direct and indirect effects of device use by assessing factors related to (a) service and system quality, (b) perceived ease of use, (c) perceived usefulness, (d) perceived pleasure, (e) cost, and (f) satisfaction with use. This study confirms the effectiveness and validity of ECM and TAM in understanding user behavior in relation to smart wearable devices.

3. Materials and Methods

The review process used to choose the papers for this study is explained in this section. The selection criteria for eliminating papers that are not pertinent to this inquiry are explained after the review methodology. An analysis of the findings considering the chosen papers is finally provided.

Publications from well-known scientific databases were selected, such as IEEE Xplorer (https://ieeexplore.ieee.org/ (accessed on 31 January 2025)), Springer Link (https://link.springer.com (accessed on 31 January 2025)), Elsevier (https://beta.elsevier.com (accessed on 31 January 2025)), PsycINFO (https://www.apa.org/ (accessed on 31 January 2025)), Scopus (https://www.scopus.com/ (accessed on 31 January 2025)).

The keywords used to conduct the search were: “Human–machine” OR “Human–robot” AND “Interaction” OR “Interface” AND “Occupational” OR “Work*” AND “Safety” AND “Cognitive” OR “Mental Stress” OR “Fatigue” AND “Neuroergonomics” AND “Brain–Computer Interface” AND “Smart” OR “Intelligent” AND “Wearable Device” OR “PPE” OR “Healthcare Device” AND “Usability” OR “Comfort Evaluation”; studies published between 2009 and 2024 were considered. The citations (title and abstract) found in every source were independently reviewed by two reviewers. The final collection of eligible studies was then determined by reviewing full-text papers. By talking with the other authors, disagreements were settled. To identify the variables to be extracted, a data extraction form was created. The devices (year of manufacturing, technology, number of electrodes, electrode type, weight, connection method, and battery life) and publications (authors, year of publishing) were identified by these elements. A systematic review in accordance with PRISMA-ScR criteria is conducted [26]. The following figure provides a qualitative description of the included studies. Tables and diagrams are used to summarize the main results. A subdivision of the case studies identified (Figure 2) was carried out according to the formal characteristics of the devices, and in particular, two configurations were identified to encompass the different types of devices: cap device and headband/crown device.

4. Wearable Neuroimaging Devices Usability Factor Identification

In the last decade, there has been growing interest in wearable devices integrated with consumer-level EEG and fNIRS technologies: small, lightweight, battery-powered devices with relatively affordable costs [44]. To date, such neuroimaging technologies are the least invasive and have good temporal and spatial resolution [30]. The main applications of these devices include neuromarketing, brain–computer interfaces (BCIs), and neurofeedback for focus-related activities such as meditation or cognitive load [44]. Wu et al. [9] stated in their study that the accuracy of emotion recognition with neuroimaging techniques, for example, was about 85%, which was higher than that achieved with other assessment methods (about 77.57%).

Also, the advancement of dry and hybrid electrode technology—which either eliminates the need for gels or utilizes other conductive solutions—has facilitated the diffusion of such devices in the industrial and research consumer market as well [42]. Wearable neuroimaging devices come in different formal configurations, such as headbands and caps, capable of adapting to different uses. The cost-effectiveness, the quality of the data produced, and the portability of these devices have created new opportunities [44] for neuroimaging technologies, becoming accessible to non-expert users. New EEG and fNIRS device configurations integrate the electrodes in a single adjustable helmet or headset that is easy to place on the head, in contrast to previous models in which it was necessary to place the electrodes separately on the scalp [35], and differ in the number of channels and electrode placement [51], however, resulting in a non-modifiable predisposition and the possibility of interpreting different mental states with different accuracy as the subject changes. Therefore, among the main criticisms of mobile and dry systems are the placement of electrodes limited to the forehead, or generally on hairless areas, and the use of non-standard, fixed, and unchangeable electrode configurations [42].

The topic of usability applied to research in the field of BCI and cognitive load has been addressed in several studies, especially in order to specify design requirements so as to ensure technology transfer and improve acceptability [33]. Most of the research work on BCI proposed in the literature is limited to laboratory use or, in general, conducted in controlled environments [45]. The use of traditional EEG and fNIRS systems connected by cables to an amplifier or a computer may not be optimal for contexts other than the laboratory. This is exacerbated when such systems are integrated with portable devices, such as in the case of NirSport, which is configured as a device backpack that impedes and restricts the user’s movements [42], overloading physical exertion. Thus, portability and adaptability to real-life scenarios constitute challenges in the field of BCI and cognitive studies.

Another limitation of those systems concerns applications in contexts where dedicated personnel cannot be relied upon to configure the measurement equipment [30,35,42]. In order to increase the signal-to-noise ratio, research is removing the use of wet EEG electrodes that need a conductive gel filling and instead focusing on comfort, usability, and wearability aspects [33,35,42] in order to guarantee the BCI system’s usability even in settings where noise and disturbance levels are uncontrollable [33,45]. Therefore, there is a need to find mobile and lightweight solutions that can change the sensing matrix and are able to meet the users’ needs [42].

In this paper, several studies on devices integrating both EEG and fNIRS technologies are compared, comparing their usability and user acceptability aspects. Furthermore, two devices, specifically Unicorn Hybrid Black (UHB) and Diadem, are investigated during laboratory assembly activities with a collaborative robot. A self-assessment questionnaire of system usability and acceptability is administered to the sample of users for each device used.

These studies show that caps with gel-based electrodes induced general discomfort, although they remain the most popular option in terms of adapting to changes in shape and head size. In addition to signal quality issues, there are difficulties with the fit of the cap in terms of the adaptability of the devices to different head sizes, comfort, and preference of the device by subjects [30,35].

Therefore, for the evaluation of an EEG/fNIRS device from the perspective of system and user usability and acceptability, the two device configurations, i.e., caps and crowns, were compared. The following aspects were identified for the evaluation:

(1)

Adaptability and fit: the caps generally have a mechanism to adapt to the shape and size of the head, unlike crowns, which, being configured with rigid or semi-rigid structures, do not always manage to adapt to different head conformations;

(2)

Comfort: the caps, which are usually made of elastic and lightweight fabric, have a greater ability to adapt to the user, improving perceived comfort. Conversely, the rigid or semi-rigid configuration of the crowns may cause discomfort due to the lack of adaptability to the head, resulting in a feeling of instability and a more uneven weight distribution;

(3)

Wearability (ability of the device to be worn): crowns are usually quicker to put on than caps, which may require assistance, especially when supplemented with wet gel electrodes. In addition, crowns allow for one-handed use, which is essential for users with disabilities or operators in work contexts;

(4)

Formal acceptability: the configuration of crown devices is defined to have a small footprint and a desirable appearance in terms of aesthetics. In contrast, caps are often configured with a wiring system that is less practical and formally cumbersome;

(5)

Electrode setting: caps have guiding elements applied to the fabric, which allows correct positioning of the electrodes, according to the international standard 10–20. In contrast, crowns may cause electrode displacements in different sensing sessions and with respect to different configurations of the user’s head;

(6)

Data quality: caps, unlike crowns, can help maintain the correct positioning of the electrodes on the scalp, decreasing noise and disturbance due to small vibrations and/or movements. This ensures a better resolution of the data for their synthesis and interpretation;

(7)

Interoperability: the devices, regardless of their cap or crown configuration, can be integrated with other technologies. To ensure this, it is essential that the device can communicate with non-device software and allow direct access to data;

(8)

Ease of cleaning and disinfection: caps are configured to be disinfected by surface cleaning, whereas crowns require immersion in a disinfection solution for several minutes following removal of the electrodes. However, this also differs depending on the type of electrodes the device is equipped with (dry, semi-dry, and salt-based).

5. Results

5.1. State of the Art

The development of advanced neurofeedback technologies will impact multiple fields of application, from improving neurological and laboratory studies to optimizing the design of real-world environments (e.g., industrial environments and human–machine interaction) [30]. Since brain data are sensitive and could be misused if not handled properly, privacy and security issues arise. Furthermore, technical hurdles such as ensuring accuracy, avoiding signal interference, and improving ease of use often limit smooth integration into applications, especially in the case of specific uses, such as industrial applications, which require adequate and timely training of operators in the use of such devices. In the literature, the EPOCx (Emotiv Inc., San Francisco, California, U.S.A.) device appears to be the most widely used in experiments. Depending on the circumstances, the 14 electrodes that make up the EPOC main unit can adopt anywhere from 10 to 20 predefined settings. Because the electrodes are positioned on semi-rigid plastic arms that are attached to a frame that encircles the head and incorporates a battery/transmitter unit, this system has a limited capacity to adjust and apply to varying head sizes. Overall, the equipment does not provide much room for small or large head sizes; however, the flexible arms holding the electrodes can be adjusted to small distances. In fact, in that case, the electrodes would not contact the scalp or would exert too much pressure. In fact, in this case, the electrodes would not contact the scalp or would exert too much pressure. This causes, in some cases, instability and uneven weight distribution, altering the signal recording and especially preventing the user from moving comfortably during the session. Following a long period of use, the constant pressure could cause discomfort to the subject [30], altering the result of the detection test. In contrast, the B-Alert x10 (Advanced Brain Monitoring, Faraday Ave, Carlsbad, California, U.S.A.) device, despite having the predetermined configuration for electrode positioning, being composed of an overall light plastic structure, ensures less discomfort to pressure.

Current research includes the study of anthropometric data as a reference for the design of products in which fit, comfort, functionality, and safety must be optimized. To date, there are mainly two different approaches to the management of anthropometric dimensions and variables: one approach consists of offering headphones in different sizes to fit a wide range of users; another approach consists of offering one size with adjustments, but not always adaptable to all users.

From the trial questionnaires, it emerged that the EPOCx device scored worst in terms of comfort, even though it was the one preferred for aesthetics and ease of use [35]. From the point of view of signal quality, EPOCx adopts passive gold-plated electrodes that touch the scalp through felt pads soaked in saline solution, which, however, causes signal obstruction. In addition, coupled with the critical issues concerning the adaptability of the system to the different head conformations of users, it is unable to guarantee continuity of data transmission during use [28]. As Raduntz & Meffert [35] state, this shows that developers have placed more emphasis on the design of EEG devices than on user-centered design. This can lead to device configurations that are not always user-friendly or comfortable. In these cases, the visual appearance and behavior of the device can influence a person’s well-being.

Dry or, if necessary, gel-based electrodes, such as those adopted by the Unicorn Hybrid Black (UHB) (g.tec medical engineering GmbH, Schiedlberg, Austria) device, are designed to work efficiently despite direct contact with hair, without requiring any special pressure. In fact, the direct contact between the sensor and the scalp is guaranteed by the presence of conductive rubber electrodes that do not require any other substances to read the signal. This facilitates the application and disinfection of the system and keeps the user’s head clean.

In contrast to the NirSport (NIRx Medical Technologies, Berlin, Germany) device, which connects the cap to an external amplification system via a cable, the UHB device’s built-in Bluetooth technology enables flexible electrode placement rather than a set configuration (like the Emotiv EPOCx, Emotiv Isight, Mindwave, Muse, Mindo, and Crown systems). Additionally, it enables direct access to streaming data through the Matlab and Python application programming interfaces, extending beyond coupling with the manufacturer’s software. Another advantage of the system is its non-prohibitive price for application in field, clinical, or educational environments.

In its basic configuration, the UHB uses the active hybrid electrodes of the g.SAHARA system—configurable with 8, up to 64 electrodes mounted on a conventional EEG headset—which allows data to be collected using electrodes connected to the amplification system with a magnet located in the central part of the neck. This makes the headset lighter, allowing a stable grip and freedom of movement on the part of the user. In addition, the system features adhesive ECG electrodes that are applied on the mastoids, which are useful for better characterizing the EEG signal, and an accelerometer and gyroscope system. A similar example is the Mindwave device, which has only one EEG electrode on the forehead, thus limiting its applicability. While UHB addresses significant drawbacks of existing systems, the restricted number of accessible channels may not be optimal for some applications [42].

Devices using conventional EEG headgear, such as the SAHARA, were substantially less favored than the others, according to overall design ratings [35]. For the most part, the Unicorn Hybrid Black devices are successful in being adaptable and delivering high-quality data outside of conventional laboratory settings [42].

The discomfort caused by pressure on specific scalp points, the overall weight or constriction of the unit, and discomfort during application and cleaning are the three main categories into which the system comfort issues reported by the participants in the reported studies can be divided [30]. On the other hand, necessities from a design and data collection point of view are the lack of interoperability, and thus the ability to dialogue with other systems and software not belonging to the parent company; signal quality due to numerous artifacts and a sometimes insufficient number of electrodes for a given application. Furthermore, factors and aspects related to emotional design are often overlooked: a discrete headphone design could have more potential to address different individual preferences, increasing visual pleasantness so as to entice the user to use the product [35]. For example, the visual appearance of the EPOCx device positively influenced users’ comfort ratings. The advancement of sensor miniaturization techniques and the use of lightweight materials has led to significant improvements over the years in the design and perceptual aspect of wearable devices, especially in the field of neuroimaging devices, enabling them to be used even by non-specialized users.

As shown in Figure 3, improvements in the design of the crowns focused on lightness, to the detriment of signal quality and the number of data recorded, focusing on aspects related to portability and formal acceptability. The Mindwave device (NeuroSky, San Jose, California, U.S.A.) year of production 2010, is, in fact, lightweight, allows dialogue with other technologies and software, but returns values recorded by a single sensor placed on the forehead. This implies its use for personal purposes only and not in industrial and/or medical and research contexts.

Regarding conventional neuroimaging devices, the changes were substantial. Comparing NirSport (Nirx) and UHB, there is a clear improvement in perceived comfort levels. UHB, with its eight dry, conductive rubber electrodes connected to a small, lightweight, portable amplifier hooked directly onto the elasticized cloth cap, does not exert great pressure on the user’s head, makes it easier to wear, and allows the electrodes to be positioned correctly. As a result, the device solves a lot of the problems discussed in this study and provides a reliable way to obtain high-quality EEG readings. Its portability also presents a number of potentials for future research into its uses outside of lab settings. However, the problem remains that, being a dome device, it needs to be assembled and prepared before it can be worn. This implies a time commitment for its assembly (which may require the help of another person) and its disinfection and cleaning. On the other hand, devices such as crowns do not allow for a proper fit, a critical factor that affects the performance, usability, and comfort of the system since most devices of this type on the market do not fit as well as traditional shells.

In order to accommodate the wide range of differences in the biometrics of the human body, cotton caps continue to be the most common solution. However, their formal acceptability remains inadequate. Quick, easy, and intuitive fitting that the user can accomplish themselves, as well as one-handed interaction with the device, continue to be crucial requirements in mobile device design.

5.2. Questionnaire Results

In order to evaluate two different EEG devices in terms of system, usability, and user acceptability, the (i) UHB headset device and the (ii) Diadem crown device, used to monitor the cognitive state of users during assembly activities with collaborative robots, were compared by means of point-in-time laboratory tests. Specifically, seven subjects (five males and two females) were evaluated, and while this experimentation aimed to deepen understanding of neural responses to specific stimuli during the execution of assembly tasks with a collaborative robot arm, it also provided a clear performance analysis of the systems. The experimental tasks were repeated with both devices on different days, and users were subsequently asked to evaluate the two devices based on the parameters identified by the state of the art. As Diadem has a rigid structure and, therefore, a fixed sensor arrangement, the electrode arrangement for UHB was designed to mirror as closely as possible the same configuration as Diadem so that the same pressure points were stimulated (see Figure 4). The sensors for Diadem are 14, Fp1, Fp2, F3, F4, F7, F8, P3, P4, P7, P8, O1, and O2, and the feedback sensor is a clip sensor that is required to be placed on the left lobe. For UHB, the sensors are 8, Fp1, Fp2, F3, F4, P3, P4, O1, and O2, and the feedback sensors are two ECG adhesive clips placed on the mastoid bone on either side of the neck.

The dry electrodes’ shape must (a) permit passage through the hair layer, (b) make dependable contact with the scalp, (c) adjust to the curve of each individual’s skull, and (d) distribute pressure uniformly while avoiding excessive local pressure points [52]. The UHB device, although elastic, has hybrid sensors—both dry and saline—with conductive rubber large enough to touch the scalp and easily pass through the hair. The Diadem device, on the other hand, has short and slightly pointed dry sensors, so a lot of pressure had to be applied to touch the scalp, which caused a sense of discomfort. The so-called “flower electrode” on the UHB has a larger contact area with the scalp and is more adaptable to the shape of each head because of its size, shape, angle, and pin arrangement. When the electrode is pressed against a surface, its design expands the contact area between the electrode and the scalp and restores mechanical stability [52]. The contact area is increased when the electrode is placed on the head because the electrode pins bend in contact with the scalp after passing through the hair. Expanding the contact area enhances comfort in addition to the electrical connection. Questionnaire results show that the elastic headset device is preferred, albeit slightly, in terms of comfort, whereas in terms of formal acceptability, all users showed a strong preference for the Diadem device (see Figure 5). In particular, there was a clear difference between male and female subjects: male users preferred the Diadem device in terms of comfort, reporting that it was less painful, and female users reported that the UHB device adapted better to the shape of the head, allowing contact between the sensors and the scalp through the hair without too much pressure (despite the larger size of the electrode), improving overall comfort. These considerations are also supported by the less deep and less visible skin marks of the flower electrodes in most subjects.

These considerations are in line with the study by Smulders et al. [23], whose results indicate that women are less sensitive than men (higher PDT) in terms of absolute PDT values, especially on the scalp, and who hypothesizes that this is due to the hair, which can dampen and diffuse the applied pressure. As an example, the results suggest that hair density could decrease vibration sensitivity in less sensitive parts of the head and marginally block vibration signals from reaching the scalp [53].

From a data quality perspective, the processing of the raw data using Python software makes it possible to extrapolate color maps showing the presence of artifacts. Figure 6 shows the strong presence of noise in the left hemisphere, probably due to the movement of the head towards the collaborative robot. UHB device and the flower electrodes produced fewer artifacts than the Diadem (Figure 6). Despite the ability to remove artifacts using Independent Component Analysis (ICA), the presence of strong noise can contaminate or distort the recorded signal. In general, such noise can come from a variety of sources, such as electrical interference, patient movement, or environmental noise. As a result, the EEG data may appear distorted or unclear, making it difficult or impossible to correctly analyze brain activity. In extreme cases, it may be necessary to repeat the recording to obtain reliable data.

6. Conclusions

Thus, hooded EEG devices offer several advantages over rigid EEG devices, making them the preferred choice in many applications, although the formal design of rigid devices is generally more popular. The main difference is in comfort and ease of use. EEG headsets are generally lighter and more flexible, allowing them to better conform to the shape of the user’s head and reduce discomfort during prolonged use. This is particularly important in clinical or research settings where participants may need to wear the device for prolonged periods of time. In addition, the flexibility of the textile headset allows for better contact of the electrodes with the scalp, ensuring more stable and accurate signal detection even with slight head movements. Crown devices are also easy to put on and take off, reducing the time needed to prepare participants and increasing the efficiency of recording sessions. The combination of comfort, ease of use, and signal quality makes textile headsets particularly suitable for long-term studies, applications in dynamic environments, and situations where minimizing user discomfort is essential for reliable and accurate results. These characteristics have a significant impact on the mechanical controls, the HW and SW system, and the overall form factor. This review opens up new opportunities for future research into the configuration of hybrid systems that combine the advantages of hooded and rigid layouts. Hybrid solutions may offer the best possible compromise between the signal quality required for accurate neuroimaging, user comfort, and adaptability to different head sizes and shapes. Also, improving the design methods of the neuroimaging headset is still an important issue, especially in medical and professional settings, where adoption is influenced by different standardizations. Their practical use could be greatly enhanced by addressing the issues of long-term wearability, durability, and ease of maintenance while maintaining adaptability in a variety of use cases.

Therefore, the design process of the device intended for direct consumer use should consider and optimize general usability factors, including overall weight and adaptability, clarity and operational accuracy, interoperability with other technology and software, user comfort, and aesthetics. In addition, it is necessary to ensure good fixation of the scalp and skin electrodes to reduce contact impedance, thereby improving signal-to-noise ratio and providing valid and reliable data. To date, therefore, no device has been found on the market that meets the usability requirements identified for the design of neuroimaging systems and that can be used in different areas, from industrial to personal.