Keywords

1 Introduction

Wearable technologies are versatile, accommodating requirements from multiple domains [4]. In healthcare and fitness, wearables have been largely explored, in particular to track users’ activities and monitor physiological signs in real time [13]. Despite the significant advances faced by wearable computing in the past decades, little is known about the potential of wrist-worn applications in educational settings [2], in special as assistive technologies to support students with intellectual and developmental disabilities.

To address this issue and leverage wearable technologies in special education, we employed a user-centered design approach and developed a cross-device application that integrates wrist-worn technologies with a mobile device to enhance the learning experiences of students with intellectual and developmental disabilities. We primarily elicited system requirements (functional and non-functional) and assessed the application needs in terms of technology. Then, we conducted interviews and five focus group sessions with students and staff members (assistants in the special education program of the university). Using Hierarchical Task Analysis (HTA) and Unified Modeling Language (UML), we formally specified the design decisions for the assistive system proposed. To create, develop and test the application, an Android phone (Nexus 5x) with a Sony Smartwatch SWR50 were employed. To gather early feedback about the design, a pilot application was implemented using a Pebble Time smartwatch and a set of graphic user interface sketches (paper-based and digital designs) were created, evaluated and discussed within the research team.

The system proposed, named Wearable Life (WL), provides a computational solution that integrates a mobile and a wearable platform to assist students with intellectual and developmental disabilities (IDD). WL promotes independent living by supporting students in their everyday activities in class and on campus. The educational activities supported involve interpersonal communication, mood assessment, collaborative work, event planning, location and reminders.

This paper describes the development process of the application and a feasibility assessment, including key findings of five focus group sessions and interviews, the list of requirements elicited (functional and nonfunctional aspects), the system specification (formal models, design decisions for hardware, software, user interfaces and interaction), as well as the challenges faced and lessons learned in the development process. We also derive and discuss design implications for future work on wrist-worn applications as assistive technologies for students with intellectual and developmental disabilities in special education settings.

2 Wearable Technologies

Wearable technologies are characterized by body-worn devices equipped with sensors and actuators that provide computational solutions to support users in their daily lives [12]. The versatility of wearable computers due to their various form factors, sensors and actuators, enables applications in diverse fields [6], spanning across healthcare, education and user interaction [20, 22, 25, 26]. Wearable technologies are able to continuously collect users’ data, process it, and offer prompt notifications, supporting users’ activities and providing services and information in (near) real time. The capability to inform and empower human users, augmenting and enhancing their senses, boost the potential of wearable solutions in numerous domains.

2.1 Wearables in Education

Wearables, thanks to their conventional look, small dimension and alternative placements for easy access, provide several opportunities for applications in teaching-learning environments [2]. Activities that can be potentially supported include student engagement, contextual learning, recording and sharing, evaluation and feedback, among others [4].

Existing applications of wearables in education include: the investigation of a wearable shirt to explore visualizations of body parts [17] and for teaching anatomy and physiology to children using e-textiles [16]. Head-mounted devices (such as Peer [19] and Google Glass [18]) have been studied to provide augmented reality experiences in classroom settings [1], for instance to teach physics to high school students [10].

Despite its large potential, the challenges related to the complexity of building wearable applications [6, 14], limit the existing work in wearable learning to an exploratory stage [2].

2.2 Wearables as Assistive Technologies

Research on wearable technologies as assistive technologies started to gain increasing attention more recently. Head-mounted displays, such as Google Glass, have been explored to address the needs of special populations, including: older adults [9], users who are color-blind [23], children with autism [24], persons with Aphasia [28], upper body motor impairments [11], and hearing loss [7].

Wrist-worn devices have been explored to detect activity patterns and repetitive movements of children with autism [8] and to support users with visual impairments to recognize faces [15], and to communicate using Braille [5]. To the best of our knowledge, the closest research on the domain of wearables and special education focused on the design of an independent behavior management application to help children manage problem behaviors with minimal supervision [27].

Concerning the design approaches for technological solutions in special education, according to Kientz et al. (2007), the following design consideration is key: one must fully understand the domain through user-centric approaches, seeking to facilitate the adoption of technology with solutions that are unobtrusive, easy to use and to adopt [8].

Wearable technologies have significantly evolved in the past decades, still few works focus on their potential in special education, in particular as assistive technologies to support students with intellectual and developmental disabilities (such as deficit of attention, Down syndrome and bipolar disorder) in their everyday activities.

3 Methods

To better understand how wrist-worn wearables can serve as assistive technologies in special education, we adopted a user-centered design approach, combining user studies (focus groups and interviews) with iterative prototyping sessions (user interface sketches). For the technical development of the application proposed, we integrated a wearable device (Sony Smartwatch) with a mobile platform (Nexus 5x). The wrist-worn device employed in the study was a Sony SWR50, a smart watch commercially available, and as mobile phone, we used a Nexus 5x smartphone with Android as operating system.

This study was conducted in a suburban Fairfax, Virginia four-year inclusive higher education (IHE) setting. The IHE has a postsecondary education (PSE) program for college students with intellectual and developmental disabilities (CSWSIDD). The PSE Program provides specialized instructions in skill development for academic learning, employment, and social integration with support from matriculating undergraduate and graduate students attending the same IHE. As part of their academic learning and integration in college academics, the CSWIDD at this IHE also participate in regular college courses (Exploration courses) alongside diverse peers without disabilities. The overarching issue related to this site is the limited resources available to provide individualized support services for CSWIDD in their Exploration classes. Although academic support service personnel (ASSP) attend classes with the CSWIDD, they provide the services quietly and are not able to effectively assist the CSWIDD with developing independence in navigating the college academics learning thereby limiting their benefit from the experience.

Seeking to elicit the system requirements and identify the application needs, we conducted firstly two semi-structured interviews with three staff members of the program, and secondly five focus group sessions with the university staff (support staff who assist the students), program administrators and students. Both functional and nonfunctional requirements were identified through qualitative analysis of the study contents and the technology needs were identified based on the analysis of the requirements.

Once the requirements were elicited, the research team conducted a needs’ assessment to select the devices and platforms that properly met the system requirements. The needs’ assessment was followed by a design and discussion session, to define the graphic user interfaces for the system. This session was followed by prototyping, implementation and tests of the application proposed. The study protocol received IRB-approval prior to data collection.

The study participants were recruited using purposeful sampling selection criteria [21]. Participants are of diverse race and ethnicity (African American, Caucasians, and Middle Eastern), gender (male and female), and age. All students are English-speaking young adults, who have IDD, and are over 18 years of age (M = 19.5, SD = 11.7). The students’ diagnosis, medical, and educational history were evaluated before admission to the Program and are determined as qualified participants for this study. All students have a modified high-school diploma and are non-matriculating students. They attend academic classes at the IHE from 9 a.m. to 3 p.m., Monday through Friday, with designated lunch period from 11:20 a.m. to 12 p.m. Within the 9 a.m. to 3 p.m. time frame, participants follow college-level (Exploration) courses in lieu of the special education classes. Each participant’s Exploration course is different and has different accommodations, such as preferred seating or extended time for coursework completion. The participants selected through the purposeful sampling procedure have experiences participating in multiple Exploration courses at this targeted IHE, have competent understanding and command of expressive language, are technology savvy, can willingly accept or decline participation in the study, and have some independence in academic learning. Their familiarity with the type of technology application being explored, ability to read and answer the interview questions, and knowledge of difficulties associated with participation in their Exploration courses makes them perfect fit for the study.

The focus group sessions were documented with audio and video after getting the informed consent of all participants, and the multimedia contents were transcribed verbatim for qualitative text analysis.

3.1 User-Centered Design

The research team, composed by two developers (computer scientists), one designer, and one HCI expert, met initially with faculty of the PSE Program who have extensive experience in special education for university students with learning, intellectual and developmental disabilities. The initial semi-structured interviews were conducted to assess the application needs and better understand the study context. A pilot demonstration of the application was implemented using a Pebble Time and a Nexus 5x phone to gather initial feedback.

Five focus group sessions were conducted between June and December 2016. The three first sessions had 5 participants each, including 15 staff members (14 females, 1 male), assistants that support students in their daily activities. The two last focus group sessions were conducted with 9 students (3 females, 6 males), young adults with intellectual and developmental disabilities following a post-secondary education program in the university. All the sessions were documented in audio and video, and focused on eliciting requirements for the application, prioritizing those, anticipating potential concerns, and getting insights regarding technology acceptance. The sessions lasted around one hour each, were moderated by 2 researchers with Special Education and Human Computer Interaction background. The study protocol received IRB-approval and the multimedia contents (audio and video files) were transcribed verbatim for content analysis. Notes were taken during the focus groups and a summary of the key findings was generated immediately after the sessions were concluded. No incentives were provided for the study subjects to gather a more impartial perspective from potential users and as such prevent bias in the data collection.

The qualitative analysis of the contents aimed at defining the application scope, refining the list of requirements initially elicited, and listing the benefits and drawbacks of the application proposed, as described in the result section. Seeking to complete the requirements elicitation, the text contents were annotated, highlighting sentences according to their semantic association to requirement (functional and non-functional), relevance to the study, frequency, advantages and potential concerns.

3.2 Development

A set of graphic user interface sketches was initially generated, in a paper-based approach to gather initial feedback (pros and cons) about the application interfaces, icons, and layout. The development process followed an iterative incremental approach, in which high-priority requirements and functions were implemented and tested first. The platform chosen for implementing the app was Android Wear (version 1.5), and the Sony smart watch was connected with the Nexus 5x mobile phone using the Bluetooth protocol.

To start the prototype development, we focused on implementing the in-class subsystem requirements which include most of the interactions between the student and staff members. Initially, for pilot tests and demonstration, we used a Pebble Time smart watch along with a Nexus 5x Android phone. The key advantages of the Pebble Time watch included its affordable price (around US$70), long-lasting battery life (around one week) and waterproof feature. Due to the discontinuity of the production of the Pebble device announced in December 2016, and interruption in the warranty support, upgrades and maintenance, a new wearable platform had to be selected. To choose another smart watch to replace the Pebble watch, we carefully analyzed the application needs and list of requirements, considering the eventual costs (financial, development efforts, learning curve and training) and long-term benefits (including: platform stability, upgrades, maintenance, documentation and scalability). A needs assessment report was generated to document the analysis results.

In the focus group sessions conducted with the students and staff members, a demonstration application of the Pebble watch and Nexus phone was presented for feedback, showing a set of basic functions for the in-class activities, including: monitoring and moderating student attention levels and assisting student during question and answer sessions. This initial demo served as a pilot study, being relevant to communicate the design decisions for the application proposed (including graphic interfaces and interactive solutions) to the end users (students and staff), and as such get their feedback, advice and suggestions concerning the existing solution and additional requirements.

To select a novel wearable device, two main quality criteria were considered – the affordability of the device and the waterproof feature. Besides this, in the updated needs assessment, we also considered additional characteristics of the device, including: the battery life, embedded sensors, compatibility with mobile operating systems and flexibility for implementation according to key project requirements.

As the long-term goal of the project considers scaling the application for a larger number of students in the PSE program, we analyzed the financial costs of existing smartwatches. We immediately discarded watches that were too expensive (e.g. Apple Watch series 2 with the price of nearly 400 dollars). We also investigated other mainstream wearable solutions, selecting Android Wear to develop the wearable app. The Sony SmartWatch 3 was selected, given that its features meet the key requirements of the study. To specify, the price of Sony Smartwatch 3 (around 140 US dollars in February 2017) was reasonable among the other Android Wear watches analyzed. Besides this, the Sony Smartwatch 3 has battery life of two days, lasting longer than most Android Wear watches in general. The battery life was raised as a key concern during the focus group sessions, in which the staff members mentioned that a longer battery life could prevent students from forgetting to recharge their devices.

The waterproof feature of the Sony Smartwatch 3 is also convenient to the project needs, being emphasized as an important requirement by the staff in the focus group sessions. The GPS and Compass features of the Sony Smartwatch 3 were also pointed as relevant device features, being helpful to assist the students, especially freshman, in finding their location on campus (including: classrooms, library, dining rooms, sport center, etc.). Finally, for future releases of the application, the accelerometer and the gyroscope sensors combined with intelligent algorithms for pattern recognition could be explored to detect whether the student is sleepy in class by analyzing his/her movement and biosignals (as explored in [3]).

Concerning connectivity features, the smartwatch selected offers three options: Bluetooth, WiFi and NFC, which provides alternative connections to pair the watch with a phone or any other relevant device. On the aspect of the compatibility, the Sony Smartwatch 3 supports Android 4.3 and onwards as well as Android Wear 1.5. For display, the Sony Smartwatch 3 carries the \(1.8^{\prime \prime }\) transflective display with 320\(\,\times \,\)320 resolution.

Currently, to provide continuous assistance in class for students with intellectual and developmental disabilities, a staff member sits near the student. To reduce stigma and be as unobtrusive as possible, in the focus group the staff members mentioned that it would be preferable if they could actually sit in the back of the classroom. To consider this preference in the implementation of the application, we use the notification feature of the smartwatch. In this scenario, the staff sitting in the back of the class can easily use the application installed on the mobile phone to send notifications with text and vibration to the student’s smartwatch, remotely assisting them in class.

To develop the application, we used Android studio as a platform. For programming, we used the Android 7.0 Nougat along with Android Wear 1.5. On the phone side, the application provides the functions corresponding to the requirements, which represents the common routine of the staff while assisting the students. On the watch, the notifications are shown immediately to the student. The two devices used are paired via Bluetooth; with the maximum range reaching up to 100 m, the Bluetooth connection is sufficient for most regular classroom environments.

3.3 Design Decisions

The in-class functions implemented can be divided in two main categories according to their trigger events. The first ones are triggered by a clicking action of the staff (touch the phone screen to select a menu option) and the second ones are triggered by a logical condition, such as a given time or activity potentially detected (e.g. sleeping). To illustrate one example of application for the first category (action), if the staff observes that the student is talking too loud during a group discussion, the staff can select the “lower voice” button and immediately the phone will send a notification to the student watch with the following text message “Please lower your voice” or a corresponding graphic icon (e.g. silence sign). To call the student attention to check the message without being too obtrusive, the watch quickly vibrates and lightens itself up. For the second category (time-based events), in the case of the class break, the staff can set a countdown timer of 5 or 10 min, and when the time is over, the phone application automatically sends a notification to inform the student about the end of the break. Also, for the second category, to reduce the student anxiety due to a long class, a notification is sent to inform the student 10 or 15 min before the class is over. Once the time of the end of class is set, the reminder can be set to repeat according to the class schedule, using the alarm service of Android, and as such requiring less configuration efforts for the staff.

To define the notifications on the watch, most of them are set to be silent in order to avoid disturbing the instructor and other students in class. Noisy environments, such as a dance class scenario, could be handled exceptionally. Also, most notifications are just presenting the message sent by the staff without any specific requirements to get feedback from the student. In principle, the notification should be unobtrusive, avoiding too much explicit interaction of the student as this could distract the student in class. However, there are a few cases where the feedback from the student is needed. For instance, when the student receives the notification asking whether he/she feels sleepy, the student should confirm or deny, so that the staff knows the subjective feeling of the student. Thus, based on the staff’s observation and student’s feedback, the staff could make better decisions to help the student in class. This feedback feature of the notification is implemented by adding a specific action to the Android notification. Also, Android Wear provides features to separate the action of the watch notification from the action of the phone notification, so that we can develop different actions for the student and staff.

4 Results

The requirements elicited for the application focus on three phases of the student activities: before, during and after class. The 13 main requirements identified for the application proposed (functional requirements) are listed in Table 1.

Concerning non-functional requirements for the application, the three major features identified in the user studies were: (a) Ease of use: the solutions for user interfaces, interaction and navigation must be simple, light weighted, minimalistic, meaningful, and intuitive; (b) Sturdy: the device should resist shocks and humidity, as it is expected to be worn continuously in a daily basis; (c) Responsive: to provide meaningful feedback for users, in a timely and quickly fashion (in real time or near real time).

Table 1. List of key system requirements
Full size table

4.1 Requirement Elicitation

In the requirements elicitation, three major stages in the special education setting were identified regarding the potential for an assistive system to support. As Table 1 shows, the assistive system can support users before, during and after class. In the requirements elicitation, we also identified two main system perspectives – the assistant and the student view. The former, implemented as a mobile application, corresponds to the assistant view, so that he/she can mediate the communication with the student and his/her behavior. The later, the student view, corresponds to the wrist-worn application. The mobile app, developed for Android using a Nexus 5x device, combined with the wrist-worn application is in close contact with the student, providing prompt notifications and reminders whenever necessary.

The watch application was initially developed using a Pebble Time (for demonstration purposes in the pilot study) then an updated version was implemented and tested using a Sony Watch. The devices are connected via Bluetooth with a 100-meter range of reach, which is suitable for most classroom environments.

From the materials collected in the focus group sessions with the staff, four reference documents were employed to derive system requirements: (i) the checklist for courses and (ii) the checklist for skills are applied before class, (iii) the zones of regulation for mood assessment are applied during class, and (iv) the frame protocols are used after class. These documents are described as follows.

  • Course Checklist: refers to the mediation of the students behavior right before class starts, verifying whether the student arrived on time, brought the necessary materials, switched off the phone (silence mode), sit down, and is ready to listen actively and take notes.

  • Skills Checklist: refers to the activities that the student should master right before a new semester begins, including Blackboard and email access, basic computer usage, and Internet access.

  • Zones of Regulation: refers to the ability of the students to self-regulate and control their emotions. It includes four actions to take, namely resting, proceeding, slowing down (calming down) and stopping (in case of inappropriate behavior or extreme emotional state).

  • Frame: refers to the study guide prepared by the staff to help students to learn and/or memorize class contents. It generally includes four main ideas with four sub items (detailed information describing a main idea).

Once the requirements were elicited a set of formal models were generated to specify the system requirements, and also a set of user interface prototypes were built to discuss the system implementation, design decisions and assess those according to their usefulness. A combination of graphic and haptic modality was used to provide feedback for users and different layouts and graphics for the contents were assessed (e.g. grid, tab, buttons-menu and icons).

4.2 Technical Implementation

Models. To formally specify the tasks involved in the assistive system, we created three diagrams for hierarchical task analysis (HTA), focusing on activities that take place before, during and after class. Also, to capture the dynamic behavior of the users and design the app, we used UML diagrams (use case and sequence diagrams). The Use Case Diagrams were created to specify the system requirements considering user behaviors of three specific actors – students, staff and instructor.

Fig. 1.
figure 1

Use case diagram of in-class assistive system

Full size image

Figure 1 illustrates the use case diagram for the in-class features. It provides an overview of 12 functions related to in class scenarios, including the main events and their relationship with the system. The behavior corresponding to “Maintaining student’s attention and alertness” is triggered by a staff input event. This behavior includes “Send notification”, an event which sends the notification to the watch of student, corresponding for instance to the routine when the staff observes the student noting he/she is absent-minded in class in the traditional special education scenario. Also, for the “Coordinate student’s presentation” behavior, the staff helps the student to present a seminar in class, reminding the student for instance to advance the slides, as well as informing the student of the remaining time to the end of his/her presentation. This behavior includes the “Send notification” and “Present countdown” events. The “Present countdown” also sends notifications to the student temporally, including thus a “Send notification” event too. As we mentioned in Sect. 3.2, once the ending time of a class has been set, the behavior “Remind of the end of class” will be repeated weekly using the “Set alarm” event which sends the notification to the student 10 or 15 min before the end of the class in case they get impatient, anxious or bored.

Fig. 2.
figure 2

Use case diagram of off-class assistive system for staff and students

Full size image

Figure 2 shows the use case diagram created to define the key events and features of the off-class application. It includes off-class system requirements, covering events that take place either before or after a class. Before class and after class requirements are integrated in the same system mainly because these features have overlapping functions (e.g. reminder and learning vocabulary through flash cards). Figure 2 shows time-based events related to reminders, such as “Remind of Class information” which includes the location of the classroom and required materials to bring to the class, and “Remind of silence phone mode” presented right before a class starts. Compared to the in-class system, we note less off-class interaction between actors. The main role of the Staff and/or Instructor consists in uploading the content of vocabulary with flash cards, listing assignments and so on. The role of student involves checking the list of necessary materials and be reminded at a certain time through a watch and/or a phone notification.

Fig. 3.
figure 3

Communication between the Phone and Watch in Class via Notifications. (a) Shows the simple notification presenting the message and alert sent from the staff. And (b) shows the notification which also gets the feedback from the student besides sending the simple notification

Full size image

Architecture. To illustrate the interaction between the staff and the student of the in-class system, Fig. 3 shows the communication between the phone of the staff and the watch of student in class. The staff and the student interact with each other through notifications sent from the phone to the watch via Bluetooth connection, which implemented by using notification API provided by Android and Android Wear.

As illustrated in the Fig. 3, two kinds of notifications are used for communication. One of them (Fig. 3a) is the notification which just presents to the student a message and an alert, and the other one (Fig. 3b) is the notification to which the student should reply. As mentioned in Sect. 3.2, to keep the student focused in the class rather than excessively interacting with the watch, most of the notifications sent by the staff to the student in class belong to the simple notification kind, except for those notifications with the goal of confirming the mood and emotional state of the student (such as: bored, anxious and sleepy), which belong to the second kind of notification (with a reply option). To specify, the notification with a reply option is sent when the staff observes the that student has a tendency to sleep and then the notification will be sent to verify whether the observation is valid. Hence, it will guide the staff to offer personalized interventions to the student (e.g. go for a walk, drink water, etc.). For the Android smartwatch, by sliding the notification horizontally (from right to left), the action button will appear on the entire screen of the watch enabling the student to send a reply message.

User Interfaces. Following an iterative and incremental life cycle, we designed and built a set of prototypes seeking to get the feedback and evaluation from the staff and the student since early development stages, and as such, being able to refine the design and adjust the resulting prototypes according to the feedback received. As previously mentioned, we developed the first prototype for the in-class scenario using a Pebble Time watch and a Nexus 5x phone with Android, to gather early feedback. Because Pebble services were discontinued, we continued the project development using Android and Android Wear platform, but developing the application for a novel device—the Sony Smartwatch 3 (still paired with a Nexus 5x Android phone).

Figure 4 shows a set of prototype interfaces, including two screen shots of the phone (top) and six pictures of the notifications received on the Sony Smartwatch (bottom and right column). The first column of the Fig. 4(a) illustrates a scenario of a group conversation in class, in which the staff thought the student engaged very well, so he/she clicked on the “Well Done” button (menu item) to encourage the student, as shown on the phone (screen shot in the upper part of the first column). Meanwhile, a notification with the applause and encouragement message is displayed on the student’s smartwatch (bottom part of the first column).

Furthermore, with respect to the middle part of the Fig. 4(b), it illustrates the option of “reminding the student of the end of the class break” by using a countdown. The corresponding phone screen shots are shown on the upper part of the Fig. 4(b). Once the class break begins, the staff can input the number of minutes (for example 10 min) depending on how long the break was given by the instructor and a notification is then sent automatically to the student when the countdown is completed. A stop and notify button is provided in case the break ends earlier than expected. The staff can also use the notify button if they notice that the student has not come back to the classroom after the first call. The bottom part of the Fig. 4(b) shows the notification that the smartwatch receives at the end of the class break.

Fig. 4.
figure 4

Eight screenshots illustrating the user interfaces of the mobile and the wearable application.

Full size image

The third column of the Fig. 4(c) shows one example of the notifications with replies for behavior mediation. The first user interface (top) asks the students whether they feel tired or not, and the following screens show three reply options by sliding the interface from right to left (yes, no, maybe).

Specifically, these notifications on the watch exemplify a mood mediation scenario and can be sent by the staff for instance when they notice that the student is feeling tired. To confirm his/her observation, the staff can ask the student the corresponding question using a watch notification. Figure 4(c) asks whether the student feels tired, and the following three notifications provides the student with three default answers (“YES”, “MAYBE” and “NO”). The bottom of each interface shows a list of dots, indicating that after the question is shown on the initial screen of the watch, the user can slide it right to see the reply choices on the following interfaces. At last, the answer of the student will show on the staff’s phone in the form of Android “Toast” (a short notification message that is placed on top of the interface).

As Fig. 4 illustrates, we employ different background images according to the notification types. On the unlocked watch screen, the “Well Done” message is placed over a thumbs-up background, while the “end of the class break” one uses the background of a classroom, and the “behavior mediation for feeling tired” one uses the background of a tired character in class. To gather further insights on the background images, we are running iterative design sessions along with a survey, seeking to identify the specific preferences of the students and staff. With sketching drafts designed, we will continue to evaluate and create an icon set and a library of images exclusive for the Wearable Life application.

5 Discussion

This paper describes the development of an assistant application to help students with intellectual and developmental disabilities aimed at enhancing their learning experiences, promoting independence and reducing stigmas. The first version of the application demonstrates the technical feasibility of mediating the student-staff communication through a wrist-worn application. To implement the application requirements for the in-class scenario, two key challenges emerged. Firstly, the development of a dual-control watch app (which is more complex than just sending a notification to the watch from phone) because the watch and the phone normally belong to a single owner. In the use case scenarios for the study, we often have two users in control—the student and the staff member. Secondly, the notifications for in-class scenarios face a major trade-off: how to notify the student without disturbing the class dynamics, i.e. with minimal impact to the student attention in class, and no (or low) interference to the colleagues nearby.

Further assessments are needed to address these issues, as well as to identify additional benefits of facilitating the communication between staff and student with wearable computing solutions. We hypothesize that the wearable app has potential to be less obtrusive than the current approach in which the assistant calls the student attention and talks to him/her in class. The wrist-worn wearable, by being in close contact to the student, can provide him/her a quick and unobtrusive notification without distracting and disturbing other students located near the user. Besides conducting further evaluations to quantify and assess the impact of the solution concerning students’ attention and interruptions, in the future, we also plan to observe, identify and model the specific events that trigger the interaction between staff and student, so that we can fully automate the mediation process based on information sensed from the users’ contexts, such as: audio from the professor (through speech recognition combined with natural language processing solutions), time of the lecture towards the end or a break, or inappropriate behaviors of the student.

Although the initial results show that both students and staff are enthusiastic about using novel technologies in classroom settings and willing to accept the solution proposed, further efforts are needed to properly assess how the application performs when deployed in a larger scale.

6 Conclusion

Smart watch applications hold a promising potential as assistive technologies to support students with intellectual and developmental disabilities in educational settings, contributing to the students’ autonomy and reducing potential stigmas associated to having a personal assistant to continuously follow, closely monitor and mediate the student activities. This paper presents the design and development of Wearable Life app, demonstrating the feasibility of employing a cross-platform solution that integrates a wrist-worn wearable (Sony smart watch) and a smart phone (Nexus 5x) to assist students with special needs in their learning activities. The application focuses primarily on mediating students’ communication and moderating their behavior. Initial results indicate that students are enthusiastic to adopt novel technologies and demonstrate high acceptance levels to use emerging technologies as assistive solutions for education. Results also show that the student assistants find the application to be helpful, useful and relevant to support their work and daily activities. As future work, we will continue to design and evaluate alternative user interfaces, besides also conducting further assessments of the technologies to better explore how they can successfully contribute to assist students in a more automated and a larger scale approach.