Complexity of interruptions: Evidence supporting a non‐interruption‐based theory

Abstract

Interruptions are known to be pervasive and harmful for concentration. It is also known that many factors can influence their effect, such as the complexity of the interrupting task. Less, however, is known about the processes underlying the resumption of the primary task, especially if the interrupting task is complex. Many hypotheses have been proposed but, as yet, it has not been possible to distinguish between their various qualities. In the current study, we carried out two experiments in which we manipulated the complexity of the interrupting task (either simple or complex) and the time between the end of the interrupting task and the resumption of the primary task (i.e., a break). Results showed that complex interrupting tasks led to longer resumption lags (RL) when there was no break, as has already been shown many times. However, while RL after simple interrupting tasks did not change as the break before resumption lengthened up to 1,500 ms, those after complex interrupting tasks decreased. Moreover, after a break of 3,000 ms, RL were longer overall. These results can only support the hypothesis based on task-switching theories, but not hypotheses derived from interruption-based models. Our findings therefore provide empirical evidence in support of explanations of complexity effects and the implications for research and application are discussed.

Interruption in visual search: a systematic review

“…The impact of interruptions on search resumption was influenced by the nature of the interrupting event. Longer ( Labonté and Vachon, 2021 ) and more difficult ( Brazzolotto and Michael, 2020 ) interruptions significantly delayed the resumption time compared to shorter or simpler ones. Similarly, interruptions involving emotionally charged stimuli extended resumption time more than neutral interruptions ( Brazzolotto and Michael, 2021 ).…”

“…Most studies carried out in natural environments (6/10) allowed for search resumption, however this was always dependent on the completion of a task during the interruption ( Williams and Drew, 2017 ; Drew et al, 2018 ; Wynn et al, 2018 ; Brazzolotto and Michael, 2020 , 2021 ; Radović et al, 2022 ). In natural scenarios where the interruption ended the search, participants were required to complete a task in two studies ( Rice and Trafimow, 2012 ; Nachtnebel et al, 2023 ) while in one study, they were not required to perform any task post-interruption ( Rieger et al, 2021 ).…”

“…When we assessed whether the interruption event was considered a task or not, we observed that 14 studies did not involve a task as the interruption ( Olds et al, 2000a , 2000b , 2000c , 2001 ; Olds and Punambolam, 2002 ; Lleras et al, 2005 , 2007 ; Van Zoest et al, 2007 ; Jungé et al, 2009 ; Lleras and Enns, 2009 ; Thomas and Lleras, 2009 ; Godwin et al, 2013 ; Mereu et al, 2014 ; Rieger et al, 2021 ); and all but one of these studies ( Rieger et al, 2021 ) were conducted in artificial environments. In contrast, in the 13 studies where the interruption was a task, five were conducted in artificial environments ( Beck et al, 2006 ; Ratwani and Trafton, 2008 ; Höfler et al, 2011 ; Alonso et al, 2021 ; Labonté and Vachon, 2021 ) and eight in natural ones ( Rice and Trafimow, 2012 ; Williams and Drew, 2017 ; Drew et al, 2018 ; Wynn et al, 2018 ; Brazzolotto and Michael, 2020 , 2021 ; Radović et al, 2022 ; Nachtnebel et al, 2023 ). Within these 13 studies, two involved interrupting event tasks that were directly related to the interrupted search, requiring participants to use information obtained during the incomplete search to complete their tasks ( Beck et al, 2006 ; Rice and Trafimow, 2012 ) whereas the tasks in the remaining nine studies were search-unrelated.…”

“…Tasks unique to natural environments included number discrimination ( Radović et al, 2022 ), a prompt to respond to the interrupted search ( Rice and Trafimow, 2012 ), a search for a letter ( Wynn et al, 2018 ), the timeout and accompanying end of the search ( Rieger et al, 2021 ), a hybrid task requiring either an over the phone diagnosis or the completion of a demographic survey ( Drew et al, 2018 ) or an event that was either the momentary disappearance of the search display or passive viewing of a natural scene ( Shen and Jiang, 2006 ). Finally, two interrupting events where common to both environments: arithmetic tasks ( Williams and Drew, 2017 ; Brazzolotto and Michael, 2021 ; Labonté and Vachon, 2021 ; Nachtnebel et al, 2023 ) and a memory tasks ( Beck et al, 2006 ; Brazzolotto and Michael, 2020 ).…”

“…Frontiers in Psychology 13 frontiersin.org 2021; Nachtnebel et al, 2023) and a memory tasks (Beck et al, 2006;Brazzolotto and Michael, 2020).…”