Notifications efficiency, impact and interference in
second-screen scenarios
ABSTRACT
The growing use of second-screen devices stresses the importance of finding a balance between engagement, distraction and disturbance of its users, while simultaneously watching television. In this framework, this paper reports on a study designed to analyze the efficiency, impact and interference of different type of notifications aiming to identify the best approach to be used when an alert is needed in second-screen scenarios.
A prototype able to deliver synchronized information related with TV content, with intervals of 10, 30 and 60 seconds, followed by individual or combined notifications (e.g. audio, visual and haptic - on the tablet and visual - on the TV) was developed. A laboratory adapted to replicate a living room was set-up and a test that included the task of watching 3 segments of a 20-minute film while using the prototype was carried with 30 participants, under a cognitive walk-through protocol.
Quantitative and qualitative results show that receiving notifications while watching TV is effective in alerting users about new information, despite its inherent cognitive disturbance for the TV viewing experience. It was also possible to highlight that in an HCI perspective, the most suitable strategy for integrating notifications (considering type and cadence) should be based on a combination of a visual notification (displayed only on the TV screen) along with a haptic notification (vibration on the tablet); spaced by an interval of at least 30 seconds.
Author Keywords
Notifications; television; second-screen applications; user experience
1. INTRODUCTION
The growing success of second-screen devices is changing the way users relate and interact with the television. Using second-screen devices while watching TV is an increasingly common activity: according to the March Nielsen report (2015), 62% of North Americans and 44% of European consumers used second screen devices while watching TV. As technological devices surrounding the television change, so do consumer’s habits (Abreu et al., 2013; Hess et al., 2012). Consumers are increasingly adopting a lean forward approach to the television experience, using connected devices as extensions of the programme they are watching (The Nielsen Company, 2014). Aware of these changes and trends, the iTV (interactive TV) industry began to focus its attention on the development of second-screen applications designed to provide additional information related with televised content (Tecnologia, 2014; Geerts et al., 2014). Nevertheless technological advancements, the user’s attention is still vulnerable to the excess of disturbances and
interruptions (Adamczyk and Bailey, 2004). The pervasiveness of broadband access and the proliferation of second-screen devices are raising the challenges in the area of divided attention (Iqbal and Horvitz, 2010), enhancing the importance of finding solutions able to balance the user’s attention between two or more sources (screens) of information (Chorianopoulos et al., 2010).
In this context, notifications – visual, audible or haptic signals designed to alert for the existence of new information – are often used as a mechanism to call the user’s attention to the second screen, and this paper reports on a study aiming to analyse the impact of its use in the TV ecosystem.
2. RELATED WORK
2.1. Multitasking, media and divided attention
When developing applications designed to promote the engagement between the consumer and the television content, designers must take into consideration the boundaries and limits of the consumer’s cognitive capabilities.
The pervasiveness of broadband access and the proliferation of companion devices are creating new challenges in the area of attention span (Iqbal et al., 2010). Defined as the process in which the assignment of cognitive resources is divided amongst multiple tasks (Srivastava, 2013), multitasking requires more mental and cognitive effort than the one required for completing a single task at a single time (Srivastava, 2013; David et al., 2013). As the individuals’ brain is vulnerable to the excess of disturbances and interruptions (Adamczyk and Bailey, 2004), media multitasking – the tendency of seeking out, combining and using multiple media simultaneously (Srivastava, 2010; Wang and Tchernev, 2012; Xu, 2008) – is still limited to their capacity for attending multiple tasks in a satisfying way (Srivastava, 2013).
In media multitasking environments, most activities involve the engagement with more than one information source at the same time, leading to the division of attention at the encoding phase (Srivastava, 2013). While combining tasks that share a common goal may contribute to achieve a balance between the task demands and the individual’s cognitive resources (David et al., 2013; Van Cauwenberge et al., 2014), viewers still need to split their cognitive resources (e.g. working memory, ability to process and understand content) between two different sources of information. Under certain circumstances, this situation may create a cognitive overload, leading to the declining of the individuals’ performance and loss of information (Van Cauwenberge et al., 2014).
2.2. Second-screen applications and attention management
With the growing adoption of second-screen devices as companion apps, the importance of finding solutions able to balance the user’s attention between two or more sources
(screens) of information (Chorianopoulos et al., 2010) becomes even more important.
Valuch et al. (2014) conducted a research focused on the effect of cinematic cuts on attention and studied attention management between two different screens. Authors state that including repeated visual elements between screens may ease the user experience and minimize the time and effort necessary for shifting attention between screens. Also in this field, Vatavu and Mancas (2014) studied the use of multi-screen TV layouts and its impact on visual attention and cognitive load, concluding that visual attention (measured in viewing time) is related not only to the size of the screen but also to the content being emitted.
In what comes to managing attention when using second-screen applications while watching TV, Basapur et al. (2012) developed an application that enabled the creation and consumption of content around and synchronised with TV shows, by the user’s social circle. The study included a four-week field trial, after which researchers concluded that participants both felt motivated to use the application and concerned regarding the created social dynamics. The trial revealed that the prototype provided a better connection with the TV show and enriched the participant’s social life, but it also contributed to increase distraction from the televised programme.
Holmes et al. (2012) studied how TV viewers’ manage visual attention between two television shows, while using a second-screen device. By analysing participant’s eye-movement patterns while interacting with synchronized second-screen applications, researchers found that considerable visual attention was gathered by the second screen, even in the absence of interactive or advertising content on the television screen. According to the study, the presence of the second-screen device significantly decreased the average gaze length on the TV. A similar study was conducted by Brown et al. (2014), who used two eye-trackers to monitor viewers shifting attention between a television and a tablet. Although the study results demonstrated that participants focused their eyes mainly on the TV screen, rather than on the tablet, it also revealed that attention was shifted to the tablet each time an update occurred, while audio cues (e.g. exclamations) seemed to bring back the viewer to the television screen.
Still in the field of attention management between multiple screens, Morales and Shekhawat (2013) conducted a study focused on the development and evaluation of a companion app able to provide information synchronised with the TV content; researchers found that, while using the app, consumers both felt better connected to and distracted from the TV show. Centieiro, Romão and Dias (2014) designed and evaluated an eyes-free interaction method for users to use a second-screen application during a real-time sport event without constantly shifting their attention from the TV screen to the mobile device. The study’s results indicate that second-screen interactions could increase user
engagement, especially if they integrate haptic and audio feedback in order to reduce the visual attention required for second-screen interaction. Finally, Geerts et al. (2014) developed a study based on viewers and producer perspectives on a companion application, studying how participants experienced the attention they had to pay to the second-screen app, and worked on finding a balance between engagement and distraction provided by the application. They found that, as participants applied a form of self-regulation in order not to get distracted, they managed to keep their main focus on the TV program. 2.3. Notification mechanisms and users’ perceptions When multitasking is seen as a process of switching rather than dividing attention, notification mechanisms emerge as an effective way of helping users to keep information awareness while removing some of the cognitive stress and effort (Technologia, 2014; Adamczyk and Bailey, 2004). Defined as “visual cues, auditory signals, or haptic alerts generated by an application or service that relays information to a user outside of the current focus of attention” (Iqbal and Horvitz, 2010), notifications are designed to alert the user about the existence of new information – e.g. a system or application updates, a new message, content updating.
According to Shirazi et al. (2014), notifications can be based: on the act of sharing an interaction that occurs on social media on gamification and presentation of extra content such as TV-Guides, editorially generated content; or based on expanded experiences such as music/artist identification, related news or related content. In the last category introduced by Shirazi et al. (2014) and named as expanded-experience apps, one can find applications designed to provide additional information (enhanced information) that is often synchronized with the content being displayed on the TV.
Notification systems are adopted as a way to alert the user to new events, a situation not always compatible with the users’ dedication to an on-going task. In fact, the user’s perception regarding notifications varies, with recent research pointing out that they are both valued by users and considered as a source of interruption of the on-going task (Adamczyk and Bailey, 2004; Iqbal and Horvitz, 2010; Narasimhan and Vasudevan, 2012). Adamczyk and Bailey (2004) measured the effects of interruption on task performance and its impacts on the user's emotional state and social attribution. Taking as starting point the idea that different interruption moments have different impacts on user emotional state, researchers introduce an attention manager system that attempts to identify opportune moments in a user’s task sequence for an interruption to occur. According to the researchers, such a system could significantly decrease the disruptive effects of interruption on users’ emotional state and social attribution. Iqbal and Horvitz (2010) studied how e-mail notifications influenced users’ quest for awareness and the impact of notifications on the overall task focus, concluding that users consider
notifications as mechanisms to provide passive awareness rather than a trigger to switch tasks. According to the researchers, users admit that notifications as disruptive but decide to keep them because of their perceived value in providing awareness.
Users tend to drop the task they are working on in order to check the received notification (Adamckyk and Bailey, 2004; Iqbal and Horvitz, 2010; Shirazi et al., 2014). Nevertheless, they value the awareness notifications provide and are comfortable to experience some disruption in order to maintain it (Iqbal and Horvitz, 2010). Amongst notifications benefits, there is the rapid availability of important information, the access to instantaneous communication and awareness of the availability of personal contacts (McCrickard et al, 2003).
Finally, the users’ perception regarding notifications can also depend on their current engagement and interest (Mashhadi et al., 2014). According to Shirazi et al. (2014), users value notifications from applications they use to communicate and interact with others, as well from calendar interfaces, but not from operating system or applications.
3. NOTIFICATION MECHANISMS IN SECOND-SCREEN SCENARIOS
One sensitive and crucial aspect when designing a second-screen companion application is to find a balance between the involvement with the TV experience and the distraction (Geerts et al., 2014) caused by inadequate notification systems.
When designing applications able to promote engagement between the consumer and the television content, developers must consider the boundaries and limits of the users’ cognitive capacities, as well as their reactions to disturbance and interruptions. Their impact in user’s attention and effort and the management of attention between multiple screens has been largely studied. However, few studies focus on the impact of notifications on second-screen applications, namely those designed to deliver synchronized content related with TV programmes. To study the impact, on users, of notification mechanisms during second-screen TV viewing experiences, 30 individual test sessions were conducted in laboratory settings aiming to identify: (1) the user’s perception on receiving notifications (of regular or specific content) while watching TV programs; and (2) the influence of notifications type (visual, audio or haptic - vibration) and cadence (i.e. frequency) in the user’s second-screen TV viewing experience.
Development of a prototype - In order to achieve the aforementioned research goals, the research team developed an agile prototype of a second-screen application that emulates the structure of the final application able to cope with the following scenario: When used while watching TV, the application running on the second-screen device
was able to identify content being emitted on the TV screen and automatically search for additional information; once the additional and related information was found, it was presented on the tablet screen (see fig. 1).
[FIGURE 1 HERE]
Content presented on the tablet was composed of an image and a short paragraph of text (hereafter referred to as marker). Whenever new content was identified and new information added to the tablet, the previously presented marker was miniaturized and saved in a timeline placed at the bottom of the tablet screen, creating a string of images/thumbnails - a timeline of the detected content (see fig. 2).
[FIGURE 2 HERE]
Each presented marker introduced aditional information related with the televised content: for instance, when a TV character was perfoming an action scene, the marker displayed on the second-screen device presented the user an image (identical or similar to the one of the TV scene) and a short text describing the identified place, actor, character, technology used, accordingly to a predefined set of categories.
The prototype included a notification system with visual, audible and haptic notifications. Whenever new TV content was detected and a new marker sent to the tablet, a notification was triggered. Visual notifications associated with the presentation of new markers could be set off on the Television, on the tablet or on both devices, and were presented through an horizontal rectangle with the message “Content with additional information” (fig. 3); audio and haptic notifications were triggered only in the tablet, being the used audio for these notifications the default acoustic signal of the Android 5.0.
[FIGURE 3 HERE]
Laboratory settings - All individual observation sessions were conducted in laboratory settings, with the laboratory adapted to replicate a living room with sofas, a table with snacks, and a television screen (40-inch, Full HD). The need for quietness and lack of disturbance was assured by placing folding screens between the simulated living room and the researchers’ observation point. The prototype was presented in a Nexus 9 (8.9-inch, 4:3 aspect ratio, 1536x2048p resolution) device, with the latest version of Android, 5.0 Lollipop.
The TV program plot has relative importance in the users’ viewing experience, with movies and dramas requiring a more dedicated attention than TV talent shows or news programs (Doughty et al., 2012; Geerts et al., 2008; Hawkins et al., 2005). For this study – focused on awareness, distraction and engagement provided by notifications – researchers looked for a scenario that
grabbed the dedicated attention and selected scenes from an action film (18 minutes total length, in English and subtitled in Portuguese - participants’ native language) to be watched by participants while using the second-screen application prototype.
Test design and specifications – At the beginning of each test session, the research team introduced the prototype and its main functionalities (notifications, timeline, notification system), after what participants were asked to answer a short questionnaire aiming to obtain information regarding general data (age, gender, academic degree, occupation) as well as TV viewing and second-screen usage habits. After that, participants were invited to watch the prepared short movie, while using the second-screen application prototype. They could keep the tablet on their hands, put it down, explore the new content, navigate between content and explore the timeline.
Test sessions included three different tests.
The first test aimed to identify what type of notification was more efficient in balancing awareness, distraction and engagement. On the first test, participants were invited to watch a segment of the film – 10 minutes length – during which visual, audio and haptic notifications were triggered on the television and/or on the tablet. Notifications were triggered independently (e.g. only on the TV, only on the tablet, only visual, only haptic, only audio) or simultaneously combined (e.g. visual notifications on the TV along with audio notification on the tablet or audio and haptic notification on the tablet). Notifications were randomly ordered and, as they were triggered when new markers were added, had no regular time intervals. Participants were instructed to, whenever they perceived a notification, press a button integrated in the prototype interface and therefore acknowledging its reception. As the application recorded the timestamps of each notification and corresponding reaction (process invisible to the participant), it was possible to collect each participant’s response time. At the end of the test, they were asked to answer a small questionnaire with a 5-point Likert scale (1=totally disagree, 5=totally agree), aiming to collect information regarding the experience they just had. Participants were asked: if receiving notifications while watching the film was annoying; if it disturbed the TV experience; if it led him/her to deviate his/her attention to the tablet; if it alerted to the existence of new information; and if it had improved the experience of watching TV. Test 2 aimed to analyse if the users’ perceptions regarding notifications changed when notifications were triggered only for specific/personalised content. Before starting to watch the following 3 minutes of the film, participants were invited to select two of three content categories for which they wanted to be alerted for, e.g. actors, places, or technology. During this test, markers would still be added to the timeline but notifications would be triggered only for
content associated with the selected categories. As for notification, during this test they were simultaneously visual, audio and haptic, and triggered only on the tablet (cf. figure 2). Participants were invited to watch the movie while the application was running, not having to press any button or to perform any other action. At the end of the test, participants were asked to answer the same questionnaire they answered after test 1.
In order to study the interference of notifications cadence (i.e. frequency) in the user’s TV experience, the third test consisted in the visualization of the remaining 5 minutes of the film while receiving notifications in three different time intervals: every 1-minute, every 10-seconds and every 30-seconds. Similarly to test 2, participants were invited to watch the movie while the application was running, not having to press any button or to perform any other action. At the end of the test session participants were asked to answer another small questionnaire, regarding the experience they just had. Participants were asked if receiving notifications within each one of the time intervals (10, 30 and 60 seconds) allowed them to: read the information presented on the tablet; to manage attention between the movie and the information presented on the tablet; if notifications were annoying; if it made him/her lose the engagement with the movie; if it made it difficult to follow the content on the TV. To collect participants’ opinion, the same 5-point Likert scale was adopted (1=totally disagree, 5=totally agree).
All three tests were conducted under a cognitive walk-through methodology. Nevertheless, at the end of the test session participants were invited to give an overall opinion about the experience they just had (i.e. receiving notifications while watching TV) through a final questionnaire, which included questions about each type of notifications: if the notifications were successful in alerting about new content; if it interfered with the TV experience; if it was annoying. The same 5-point Likert-scale was adopted.
4. RESULTS
Aiming to study the effects of notifications in second-screen scenarios, individual observation sessions were conducted. 30 individuals (18 male, 12 female) participated in this study, 15 aged between 18 and 30 and 15 from 31 to 40 years old. In what concerns occupations, 10 were students, 4 were researchers, 10 lecturers, and 6 programmers and ICT technicians.
4.1. TV and second-screen habits
Regarding TV and second-screen habits, 80% of the participants are used to watch TV alone, 77% with family and 40% with friends. Only 7% of the participants mentioned to watch TV with strangers.
As for second-screen devices, 63% of the participants use it while watching TV to check on social networks, 47% to check their e-mail and 37% to perform random web searches. 30% of the participants use their second-screen
devices to look for information related with the TV program they are watching, and only 10% use it for engaging with apps related with/developed for the TV show. 33% of the participants claimed not to use second-screen devices while watching TV.
When watching TV, participants aged between 18 and 30 years old use their second-screen devices mostly to check on social networks (23% do it while watching TV alone, 13% while watching with family and 7% while watching with friends).
As for participants aged between 31 and 40 years old, they use their second-screen devices while watching TV mostly for checking their e-mail and social networks (37% while with family, 27% while alone and 13% while with friends). 4.2. Notifications in second-screen scenarios
In order to identify what type of notification was more efficient in balancing the user’s awareness, distraction and engagement, a set of data was collected from the participants, namely their response time to notifications (automatically collected by the application system); their responses to the questionnaires; and their opinion, expressed at the end of the test sessions.
[FIGURE 4 HERE – SPANNING 2 COLUMNS] Efficiency of Notification Systems - Figure 4 shows the participants’ response time to notifications (collected during test 1), with the Y-axis indicating the total number of participants (N=30) and the X-axis indicating the notifications number and type (24 notifications, of different types and combinations). In order to analyse the data collected by the timestamps saved by the prototype, the response time values were grouped in: 3 seconds or less; between 4 and 5 seconds; 6 or more seconds, with timestamps being rounded to the next integer value. In the figure, mean time between notifications is equally distributed, i.e., it does not reflect the notifications’ cadence.
Data shows that visual notifications on the TV (either alone or combined with other notifications) had a faster response time. Of the 12 notifications with “TV” (i.e. visual on the TV) stimuli, 9 (corresponding to 75%) were acknowledged by half of the participants in 3 seconds or less. As for audio and haptic notifications, when triggered alone (n14 and n17 for audio, and n6 and n18 for haptic), were acknowledged within 3 seconds by 12 and 13 participants, and by 14 participants, respectively. When combined with TV notifications, audio and haptic notifications were acknowledged within 3 seconds by more than half of the participants.
Notifications 9, 10, 11 and 12 had the higher response (i.e. acknowledged within 3 seconds by the higher number of participants) rate. All of them were a combination of TV notifications along with audio notifications (n9, n10 and n
11), haptic notifications (n10 and n11) and visual notifications on the tablet (n12). When triggered alone, visual notifications on the tablet (n1 and n23) had low response rates (more than 50% of the participants acknowledge these notifications six or more seconds after they occurred).
The impact of notifications - Data collected through questionnaires revealed that receiving notifications while watching TV was annoying only for 33% of the participants, but disturbing for the TV viewing experience for 73%. 23% of the participants answered that receiving notifications helped to improve the TV viewing experience (see figure 5, where the X-axis indicates the total number of participants). All participants agreed that receiving notifications alerted for new information on the tablet, and 93% answered that notifications made them turn attention from the TV to the tablet.
[FIGURE 5 HERE]
Asked to share their opinion regarding the experience they just had, 3 participants mentioned that receiving notifications (namely the audio ones) made them loose the engagement with the film, especially during the action scenes. Visual notifications displayed on the TV were considered as the less intrusive. When displayed on the tablet, visual notifications were often unnoticed.
Receiving notifications only for selected categories of content was equally annoying but somehow less disturbing to the TV viewing experience (in this scenario, only 56% of the participants considered that notifications were disturbing). In this scenario, receiving notifications improved the TV viewing experience for 40% of the participants, an increase of 17% when related with receiving notifications for all content added to the tablet. As for the interference – if any – of the cadence of notifications (i.e. frequency) in the participant’s TV viewing experience, data collected after test 3 revealed that it was easier for participants to read the information presented in the tablet when notifications were triggered with a 30-seconds or 1-minute interval (see figure 6, where the X-axis indicates the total number of participants).
[FIGURE 6 HERE]
Managing attention between the film and the information presented on the tablet was also easier when notifications were triggered every 30-seconds or 1-minute. Notifications became more tiring when triggered every 10 seconds, but easier to deal when set in larger time intervals. As it is possible to see in figure 6, only 6 of the 30 participants (corresponding to 20%) considered notifications tiring when prompted every 30-seconds, and only 3
(corresponding to 10%) felt the same when receiving notifications every 1-minute. As for the engagement with the televised content, when prompted in short time intervals notifications disturbed the TV viewing experience (70% of the participants totally agreed or agreed with the sentence), situation that is less recurrent when notifications are triggered with a 30-seconds or 1-minute interval. 63% of the participants totally agreed or agreed that it was difficult to follow content on the TV when notifications were triggered every 30 seconds, and more than 50% considered that it was easier when they were prompted in larger time intervals.
Balancing awareness, engagement and distraction - One sensitive and crucial aspect when designing a second screen companion application is to achieve a balance between engagement (with the TV program) and distraction, namely the one caused by inappropriate notifications systems. At the end of the test sessions, participants were asked to answer a questionnaire focused on the notifications type, i.e. asking participants about their opinion regarding the four different types of notifications (visual on the TV, visual on the tablet, audio and haptic). According to the collected data, visual notifications displayed on the TV, audio and haptic notifications were efficient in alerting participants for the existence of new content on the tablet (see figure 7). As for visual notifications on the tablet, only 53% of the participants considered it as being useful for alerting about the existence of new content.
[FIGURE 7 HERE]
Audio notifications are more disturbing to the TV viewing experience, as sustained by 80% of the participants, while visual notifications on the TV were considered as the least intrusive. These notifications were also considered as the less annoying, followed by visual notifications on the tablet. Audio notifications were considered as annoying by 70% of the participants.
5. DISCUSSION
The study described in this paper aimed to analyse the user’s perceptions about receiving notifications (from broad or pre-selected categories of content) from a second-screen application while watching TV; to identify what type of notification (visual, audio or haptic) is more efficient in balancing awareness, distraction and engagement; and to analyse the interference, if any, of notifications cadence (i.e. frequency) in the user’s TV viewing experience. Collected data revealed that receiving notifications while watching TV is not necessarily annoying but it is disturbing for the TV viewing experience: “notifications are less disturbing when triggered in the movie quiet moments. When in action scenes, they were somehow disturbing” (participant #1); “notifications interrupted the film” (participant #5); “with notifications, I ended up by losing the engagement with the film” (participant #12).
Notifications are effective in alerting users about new information, as pointed out by all participants. When associated only with content categories previously selected by the user, notifications are less disturbing than when prompted for all new content, but still annoying. As mentioned by participant #6, “if I received notifications only for specific categories, maybe I would be more confortable. But they would still be annoying. Yes.”. Regarding the notification type, the data analysis, presented in the previous section, makes it possible to sustain that the visual notifications on the TV are the most efficient on alerting the user about the existence of new information, especially when combined with audio and haptic stimuli. Although audio and haptic notifications are equally effective, data collected through the questionnaires revealed that audio notifications are more annoying and disruptive to the TV viewing experience. By contrast, haptic notifications are less annoying and interfere less with the TV experience. It is therefore possible to claim that the best way to achieve a balance between engagement, distraction and awareness when using a second-screen application while watching TV, would be to create a notification system based on a combination of visual notifications (displayed on the TV) and haptic notifications (prompted on the tablet).
Finally, and regarding the interference of notifications cadence in the user’s TV viewing experience, data analysis shows that when notifications are triggered with short time intervals it is difficult for the user to manage attention between the TV and the tablet. It is also possible to sustain that longer periods between stimuli – 30-seconds and 1-minute intervals – increase the users’ ability to read content on the tablet and to manage attention between the two devices, making it easier to follow content on the TV. 6. CONCLUSIONS
The growing adoption of second screen devices as companion apps stresses the need to find a balance between engagement, distraction and disturbance in second-screen applications. Notifications – visual, audible or haptic – are often used to call the user’s attention to the second screen. But, they can be also a source of stress, distraction and annoyance. When developing applications designed to promote engagement between the consumer and the television content, one must have into consideration the boundaries and limits of the users’ cognitive capacities, as well as their reactions to disturbance and interruptions. The study described in this paper aimed to analyse the users’ perceptions regarding notifications in a second-screen scenario. According to some authors (Adamczyk and Bailey, 2004; Iqbal and Horvitz, 2010; Narasimhan and Vasudevan, 2012), user’s perception regarding notifications diverges: while alerting for the existence of new content or a new event, and therefore easing the shift of attention from one source of information to another, they are considered by users as sources of interruption of the on-going task, disturbing and even a source of stress.
During test sessions conducted in laboratory settings, 30 participants interacted with the prototype. The analysis of data collected during the tests and discussed in the previous section indicates that, overall, notifications are somehow annoying and disturbing to the TV viewing experience. Nevertheless, they are useful in alerting the user for the existence of additional information on the second-screen device. When displayed on the TV, visual notifications are not only easily perceived by users but also considered as non-intrusive and, when combined with haptic stimuli, they are efficient in alerting the user for new content without disturbing the TV experience. Audio notifications, although effective in calling for the users’ attention, were considered as the most annoying and disturbing ones.
During the test sessions, participants also interacted with the prototyped application receiving notifications in difference cadences. When trigged in short time intervals (every ten seconds), notifications became tiring and made it difficult for participants to read the information displayed on the tablet. According to collected data, notifications emitted with a 30-seconds or 1-minute interval were the most appreciated by participants, as it allowed them to manage attention between screens, keeping the engagement with the film playing on TV.
Despite the technological development and the evolution of (second screen) devices, the human mind is still vulnerable to excessive interruptions and disturbance. In the second-screen applications scenario, namely those designed for specific TV programs, notifications are often used to call the users’ attention to the second-screen device. Attending to the method and procedures adopted during the study, it is believable that its results may bring positive insights to the second-screen applications’ design field.
ACKNOWLEDGMENTS
This paper is a result of the “XXXXX” project, funded by XXXXX (grant agreement no. XXXXX).
The authors would like to acknowledge XXXXXXXX. REFERENCES
Abreu, J., Almeida, P., Teles, B., Reis, M. (2013). Viewer Behaviors and Practices in the (New) Television
Environment. In Proceedings of the 11th european conference on Interactive TV and video - EuroITV '13, (5-12). June 24–26, Como, Italy.
Adamczyk, P., & Bailey, B. (2004). If Not Now, When?: The Effects of Interruption at Different Moments Within Task Execution. In Proceedings of the SIGCHI conference on Human factors in computing systems, 6 (271–278). Bailey, B.P., Konstan, J.A. and Carlis, J.V. (2000). Measuring the effects of interruptions on task performance in the user interface. 2000 IEEE International Conference on Systems, Man, and Cybernetics, 2. pp 757-762
Basapur, S., Mandalia, S., Chaysinh, S., Lee, Y.,
Venkitaraman, N. and Crysta Metcalf (2012). FANFEEDS: evaluation of socially generated information feed on second screen as a TV show companion. In Proceedings of the 10th European conference on Interactive tv and video (EuroiTV '12). ACM, New York, NY, USA, 87-96.
http://doi.acm.org/10.1145/2325616.2325636
Brown, A., Evans, M., Jay, C., Glancy, M., Jones, R. and Harper, S. (2014). HCI over multiple screens. In CHI '14 Extended Abstracts on Human Factors in Computing Systems (CHI EA '14). ACM, New York, NY, USA, 665-674. DOI=10.1145/2559206.2578869
Chorianopoulos, K., Fernández, F., Salcines, E. and Lozano, C. (2010). Delegating the visual interface between a tablet and a TV. In Proceedings of the International Conference on Advanced Visual Interfaces (AVI '10), Giuseppe Santucci (Ed.). ACM, New York, NY, USA, 418-418. DOI=10.1145/1842993.1843096
http://doi.acm.org/10.1145/1842993.1843096
David, P., Xu, L., Srivastava, J., Kim, J. (2013). Media multitasking between two conversational tasks. Computers in Human Behavior, Volume 29 (4) 1657-1663.
http://dx.doi.org/10.1016/j.chb.2013.01.052.
Doughty, M., Rowland, D., and Lawson, S. (2012). Who is on your sofa?: TV audience communities and second screening social networks. In Proceedings of the 10th European conference on Interactive TV and video (EuroiTV '12). ACM, New York, NY, USA, 79-86.
Geerts, D., Leenheer, D., Heijstraten, S. & Negenman, J. (2014). In Front of and Behind the Second Screen: Viewer and Producer Perspectives on a Companion App. In Proceedings of the TVX’14 Conference, (95102). June 25 -27, 2014, Newcastle Upon Tyne, United Kingdom.
Geerts, D., Cesar, P., and Bulterman, D. (2008). The implications of program genres for the design of social television systems. In Proceedings of the 1st international conference on Designing interactive user experiences for TV and video (UXTV '08). ACM, New York, NY, USA, 71-80.
Hawkins, R.P., Pingree, S., Hitchon, J., Radler, B., Gorham, B., Kahlor, L., Gilligan, E., Serlin, R.C., Schmidt, T., Kannaovakun, P., and Kolbeins, H. (2005). What produces television attention and attention style?: Genre, situation, person, and media perceptions as predictors. Human Communication Research. 31, 162-187
Holmes, M.E., Josephson, S. and Carney, R.E. (2012). Visual attention to television programs with a second-screen application. In Proceedings of the Symposium on Eye Tracking Research and Applications (ETRA '12), Stephen N. Spencer (Ed.). ACM, New York, NY, USA, 397-400. http://doi.acm.org/10.1145/2168556.216864 Iqbal, S. and Bailey, P. (2010). Oasis: A framework for linking notification delivery to the perceptual structure of goal-directed tasks. ACM Trans. Computer-Human
Interaction 17 (4), Article 15, 28 pages. DOI=10.1145/1879831.1879833
Iqbal, S. and Horvitz, E. (2010). Notifications and awareness: a field study of alert usage and preferences. In Proceedings of the 2010 ACM conference on Computer supported cooperative work (CSCW '10). ACM, New York, NY, USA, 27-30. DOI=10.1145/1718918.1718926
Hess, J. Ley, B., Ogonowski, C., Wan, L. and Wulf, V. (2012). Understanding and supporting cross-platform usage in the living room, Entertainment Computing, 3 (2), 37-47,
http://dx.doi.org/10.1016/j.entcom.2012.04.001.
Mashhadi, A., Mathur, A. and Kawsar, F. (2014). The myth of subtle notifications. In Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct Publication (UbiComp '14 Adjunct). ACM, New York, NY, USA, 111-114.
DOI=10.1145/2638728.2638759
McCrickard, D.S., Catrambone, R., Chewar, C.M. and Stasko, J.T. (2003). Establishing tradeoffs that leverage attention for utility: empirically evaluating information display in notification systems. International Journal of Human Studies, 58 (5), 547-582. DOI=10.1016/S1071-5819(03)00022-3
Morales, G., Shekhawat, A. (2013). The Future of Second Screen Experience. In workshop on Exploring and Enhancing the User Experience for Television, CHI 2013. Narasimhan, N. and Vasudevan, N. (2012). Descrambling the social TV echo chamber. In Proceedings of the 1st ACM workshop on Mobile systems for computational social science (MCSS '12). ACM, NY, USA, 33-38. NPD Group. (2013). Digital Video Outlook Second Screens Report. Retrieved December 28, 2014, from
https://www.npd.com/wps/portal/npd/us/news/press- releases/divided-viewing-second-screens-vying-for-tv-viewers-attention/
Shirazi, A., Henze, H., Dingler, T., Pielot, M., Weber, D. and Schmidt, A. (2014). Large-scale assessment of mobile notifications. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '14). ACM, New York, NY, USA, 3055-3064.
DOI=10.1145/2556288.2557189 9
Srivastava, S. (2013). Media multitasking performance: Role of message relevance and formatting cues in online
environmentsç Computers in Human Behavior, Volume 29 (3), 888-895. http://dx.doi.org/10.1016/j.chb.2012.12.023. Srivastava, J. (2010). Media multitasking and role of visual hierarchy and formatting cues in processing of web content. Dissertation. Retrieved December, 13, 2014, from
https://etd.ohiolink.edu/!etd.send_file?
accession=osu1281718384&disposition=inline
The Nielsen Company. (2015). Screen wars: the battle for eye space in a TV-everywhere world,Shifts in viewing: The Cross-Platform ReportMarchQ2 20154. Retrieved July 13, 2015, from http://www.nielsen.com/us/en/insights/reports/2015/screen- wars-the-battle-for-eye-space-in-a-tv-everywhere-world.htmlhttp://www.nielsen.com/content/dam/corporate/u s/en/reports-downloads/2014%20Reports/q2-2014-cross-platform-report-shifts-in-viewing.pdf
Valuch, C., Ansorge, U., Buchinger, S., Patrone, A.R. and Scherzer, O. (2014). The effect of cinematic cuts on human attention. In Proceedings of the 2014 ACM international conference on Interactive experiences for TV and online video (TVX '14). ACM, New York, NY, USA, 119-122. DOI=10.1145/2602299.2602307
Van Cauwenberge, A., Schaap, G., Roy, R. (2014). “TV no longer commands our full attention”: Effects of second-screen viewing and task relevance on cognitive load and learning from news. Computers in Human Behavior, Volume 38, 100-109.
Vatavu, R. and Mancas, M. (2014). Visual attention measures for multi-screen TV. In Proceedings of the 2014 ACM international conference on Interactive experiences for TV and online video (TVX '14). ACM, New York, NY, USA, 111-118. DOI=10.1145/2602299.2602305
Wang, Z., & Tchernev, J. M. (2012). The ‘‘myth’’ of media multitasking: Reciprocal dynamics of media multitasking, personal needs, and gratifications. Journal of
Communication, 62(3), 493–513.
Xu, L. (2008). Impact of simultaneous collaborative multitasking on communication performance and experience. Dissertation. Retrieved November 23, from
https://etd.ohiolink.edu/rws_etd/document/get/osu12185930 84/inline
