On the need for attention-aware systems: Measuring effects of interruption on task performance, error rate, and affective state

“Research has shown that interruptions lead to annoyance and anxiety (Bailey & Konstan, 2006), and interviewees who are annoyed may volunteer less information (Bull, 2010; Fisher, 2010).”

“With respect to interruptions, only few EEG studies have been conducted, but a large number of behavioral studies has also demonstrated the disruptive effects of interruptions on performance in a primary task (Gillie and Broadbent, 1989; Edwards and Gronlund, 1998; Altmann and Trafton, 2002; Speier et al., 2003; Baile…”

“Bailey and Konstan (2006) demonstrate that task interruptions and continuous task switching increase the stress and anxiety levels of individuals and lead to up to twice the number of errors committed across the tasks.”

“…the observations of position effects in the UNRAVEL paradigm and previous findings that interruptions are less disruptive if they occur after the completion of subtasks compared to the ones positioned within subtasks (Monk et al., 2002, 2004; Botvinick and Bylsma, 2005; Bailey and Konstan, 2006).”

“Enforced interruptions like these disrupt activity, incurring time costs and increasing the likelihood of errors [Bailey and Konstan 2006; Trafton et al. 2011].”

“If they choose to do the latter, then potential distractions force individuals to suspend their primary task, leading to a longer time to completion [16].”

“Previous work has shown that interrupting with notifications randomly can decrease performance[6, 13], yet it has remained unclear when to show the notifications to not affect performance negatively?”

“Studies have actually shown that interruptions at inopportune moments will not only slow down a developer’s work significantly, they will also lead to negative emotions and more errors in the code [88], [89].”

“…the right moments to interrupt a user has been an important problem in the attention-aware computing literature because interruptions at inopportune moments can adversely affect the cognitive state of the user [2], increase task completion times [30], and may lead to high task error rate [3].”

“While interruptions increase all types of errors [2], interruptions that occur just prior to the postcompletion step drastically increase the postcompletion error rate [9].”

“In the meanwhile, mistakes and the anxiety are increased twice with 31%–106% more annoyance [2], [3].”

“Research has shown that interruptions (such as TOR) are more cognitively demanding when interrupting an active sub-task, rather than when issued during breaks in between (Bailey & Konstan, 2006; Naujoks et al., 2017; Wintersberger et al., 2019).”

“To mitigate the disruption caused by the interruptions, deferring them to opportune moments, such as task breakpoints [34, 80] or rough boundaries during task execution [4] could largely mitigate the cost of interruption.”

“For instance, Bailey et al. [5] suggested that an attention-aware system that defers presenting peripheral information until coarse boundaries are reached during task execution could mitigate the negative impact of such information’s arrival.”

“For instance, it has been shown that during a human-computer interaction, the task boundaries represent suitable moments for interrupting the ongoing activity, enabling an easier resumption of the task at a later point ( 47 – 49 ).”

“They also cause affective discomfort, for example by increasing subjectively experienced annoyance and anxiety [4, 36] or by lowering motivational factors to continue the task [22].”

“Researchers in human-computer interaction have advocated for attention aware displays (Bailey & Konstan, 2006), systems which monitor, adapt themselves to, and manage the user ’ s attention.”

“For example, Bailey and Konstan found that opportune moments for information delivery occur naturally at the end of ongoing tasks in desktop work contexts [5].”

“This result is in line with previous work, which reported that interruptions affected task performance but the length of the interruption did not affect the encoded memory representation (27, 28).”

“Even though experiential effects of notifications such as annoyance and anxiety have been studied before (Bailey and Konstan, 2006; Iqbal and Bailey, 2008), the main body of earlier work focused on task performance rather than user experience.”

“A subject was typically performing a primary task before being interrupted by a secondary task, and the effects of the interruption on the primary task were analyzed (Iqbal and Horvitz 2007; Mark, Gudith, and Klocke 2008; Bailey and Konstan 2006).”

“It is proven, that notifications provoking task-switches work best at task boundaries or times of low mental workload [4], while random notifications lead to stress and cognitive overload (what is everything else then ideal in safety-critical driving situations).”

“The goal was to mimic a real-world situation that would cause the walker to temporarily suspend attention to the primary task (walking and memorizing the path) [4].”

“Several researchers have investigated the effect of interruption caused by notification under various situations [3, 7, 34].”

“…findings that responding to an ESM prompt can influence individuals’ affective and cognitive states (e.g., increase in stress 43 , annoyance 44 , and anxiety 45 ), we clarified the extent to which answering a given ESM prompt caused participants’ emotions to be positive or negative on a…”

“] and users’ affective state [2, 7].”

“Incidental interruptions usually hinder one’s progress, leading to unpleasant feelings (Bailey and Konstan, 2006; Speier et al., 2003; Carver, 2006).”

“Past studies have revealed several negative influences of in-terruptive notifications, such as on productivity [3, 4, 6, 20, 32, 39], emotional and social attribution [3], and psycho-physiological states [39].”

“They can furthermore cause discomfort by increasing subjectively experienced annoyance and anxiety [29] and increase the total workload [30] (especially for complex tasks [31]).”

“…and Rigotti 2013; Einstein et al. 2003), lengthening task times (Bailey and Konstan 2006; Eyrolle and Cellier 2000), increasing errors (Bailey and Konstan 2006; Oulasvirta and Saariluoma (Oulasvirta and Saariluoma, 2004) and decreasing performance satisfaction in general (Baethge and Rigotti…”

“Bailey and his colleagues (Bailey et al. 2006; Bailey and Konstan 2006; Iqbal et al. 2005) represent tasks as two-level hierarchies composed of coarse events which are further split into fine events, and demonstrate that interruptions are less disruptive when presented at coarse breakpoints,…”

“While the focus of this paper is squarely on the visual method used to grab users’ attention, this work complements prior research studying why interruptions happen [8, 9]; the effects of interruptions on performance, resumption speed, and long-term stress [3, 7, 23]; and when to schedule interruptions [1, 11, 20].”

“Typical notification systems that deliver notification immediately make various negative affect to user’s work productivity and even their emotion according to previous studies[1, 2, 5, 14, 8, 4].”

“For example, Bailey and Konstan (2006) found task execution time increased by 3% to 27% due to interruptions.”

“The corpus of research on multitasking with digital media has found deleterious relationships with successful task completion (Bailey and Konstan 2006; Wijekumar et al. 2006; Wilson 2005).”

“Researchers have documented the harmful effects of interruptions and information overload on productivity, memory, and emotions (Bailey and Konstan 2006).”

“The underlying idea is, that users should be interrupted only in “opportune moments” [48], for example at times of low mental workload, or at the presence of task boundaries [3], to reduce the negative effects of so-called “interruption” and “resumption lags”, which describe the time and effort needed to rehearse and…”

“These opportunities can increase stress (Bailey & Konstan, 2006), reduce productivity (Mark, Gudith, & Klocke, 2008) and alter perceptions of personal freedom (Leshed & Sengers, 2011).”

“This additional effort is likely to be the cause of the prolonged processing time, leading to more errors [7,23,25].”

“…Broadbent, 1989; Kapista and Blinnikova, 2003; Gluck et al., 2007; Basoglu et al., 2009; Mirhoseini et al., 2020), interruptions can also have different negative psychological consequences (Adamczyk and Bailey, 2004; Bailey and Konstan, 2006; Dodhia and Dismukes, 2009; Feldman and Greenway, 2021).”

“They can result in slower task completion time, increased error rates and additional job stress (e.g., Bailey and Konstan, 2006; Cellier and Eyrolle, 1992; Czerwinski et al., 2000; Van Bergen, 1968).”

“Overly-distracting notifications have been shown to induce higher levels of stress [27], mistakes [5], and productivity loss [14].”

“This affords flexibility to model tasks with high interest (Horrey and Wickens 2006; Iani and Wickens 2007), tasks with substructures (Bailey and Konstan 2006; Monk et al. 2004), as well as complex interruption and resumption costs (Trafton et al.”

“As for the latter, there is a large body of literature on interruptions and their detrimental effects on productivity (e.g., Bailey & Konstan, 2006; Benbunan-Fich & Truman, 2009; Czerwinski, Cutrell, & Horvitz, 2000, pp. 71 e 76; Gonzalez & Mark, 2004; McFarlane & Latorella, 2002; Monk, 2004; Monk,…”

“However, most studies on interruption have focused on the effects of interruption on performance using experimental research design (e.g., Bailey & Konstan, 2006; Calderwood, Green, Joy-Gaba, & Moloney, 2016; Lin, Mills, & Ifenthaler, 2016；Speier, Vessey, & Valacich, 2003), which confines our…”

“Furthermore, the sensitivity of NIRS data affords the opportunity to determine the optimal time to provide notifications or interruptions along the course of a primary task [28], which has otherwise been shown to increase error, time to completion, annoyance, and anxiety [29].”

“In experimental studies, participants subjected to interruptions while performing tasks experienced increased stress, annoyance and frustration [10, 12].”

“It is known that task interruptions are less detrimental when they happen at task boundaries (i.e., small breaks between subtasks) or when humans are relaxed [6, 37], and the application of AUI concepts have shown to be beneficial in various settings from office work [6, 38, 39] to driving [84, 85].”

“Researchers suggest that interruptions on a current task can result in negative emotions such as annoyance and anxiety (Bailey & Konstan, 2006).”

“…2016), disrupted information processing (Carr, 2010; Levine et al., 2007; Poldrack & Foerde, 2008), longer completion times for cognitive tasks (Bailey & Konstan, 2005; Flanigan & Babchuk, 2015; Fox et al., 2009), and decreased learning outcomes, such as lower test scores and grades (e.g.,…”

“Based on research done by Bailey and Konstan (2006), we consider the best opportunities for raising attention occur at the boundaries between consecutive tasks, such as editing a piece of text, moving an object, giving a comment, and others.”

“…a majority of prior interruption research has focused on gaining an understanding of how interruptions affect task performance (e.g. Adamczyk and Bailey 2004; Bailey and Konstan 2006; Zijlstra et al. 1999) and how to manage interruptions efficiently (e.g. Grandhi and Jones 2015; McFarlane 2002).”

“Bailey and Konstan [40] report that interruptions have a disruptive impact on completion time and error rate.”

“The well-studied negative consequences of interruptions on performance in cognitive tasks (Bailey & Konstan, 2006; Baron, 1986; Eyrolle & Cellier, 2000; Hodgetts & Jones, 2006) are in contrast to an as yet insufficient understanding of the cognitive processes required to switch between the…”

“Bailey and his colleagues [5, 6] represent tasks as two level hierarchies composed of coarse events further split into fine events and demonstrate that interruptions are less disruptive when presented at coarse breakpoints, corresponding to the completion of coarse events.”

“Task interruptions increase feelings of annoyance and anxiety (Bailey & Konstan, 2006; Bailey, Konstan, & Carlis, 2001), and they increase task-completion time (Bailey & Konstan, 2006; Cauchard, Cane, & Weger, 2012; Hodgetts, Vachon, & Tremblay, 2014).”

“Although such studies demonstrated a link between eye movements and errors, they focused on user-generated errors, which are likely due to inattention, cognitive load, and/or interruptions [3,11,12,30,40].”

“…indicates that task interruptions are disruptive (Trafton & Monk, 2007) and are generally considered to exert negative effects on performance and increase the completion time and error rate of primary tasks (Bailey & Konstan, 2006), which results in a higher cognitive workload (Lin et al., 2013).”

“Users also feel more stressed and cannot focus well if they are frequently interrupted by secondary tasks [4,24].”

“Maintaining the attention on task related AOIs is interpreted as the concentration on a specific task [14], or on session related tasks in general.”

“Studies such as [3], [17]reveal that interruptions at wrong times might increase the users stress, reduce a workers performance, and increase chances of error.”