- Email: [email protected]
Distortions in time perceptions during task switching
Computers in Human Behavior 80 (2018) 362e369
Contents lists available at ScienceDirect
Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh
Full length article
Distortions in time perceptions during task switching Shan Xu a, *, Prabu David b a b
School of Communication, The Ohio State University, Columbus, OH, USA College of Communication Arts and Sciences, The Michigan State University, East Lansing, MI, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Received in revised form 14 November 2017 Accepted 21 November 2017 Available online 22 November 2017
Perceived time passage and time duration were examined in a between-subjects design with four conditions: watching a sitcom, reading a journal article, occasional switching between sitcom and article, frequent switching between sitcom and article. Consistent with our prediction, time “flew by” in the high-entertainment condition that involved watching a sitcom, whereas time “dragged on” in the lowentertainment condition that involved reading a journal article. Switching between the two led to quicker passage of time than the low-entertainment condition, but not the high-entertainment condition. A different pattern was evident for duration estimation, with no difference between the low- and high-entertainment conditions, but a longer estimation of duration in the switching condition. Further, frequency of switching between the sitcom and article did not make a difference. These findings suggest that switching between tasks leads to overestimations of time spent on media. Published by Elsevier Ltd.
Keywords: Task switching Media multitasking Entertainment Time passage Time duration Media measure
1. Introduction Media multitasking has become a way of life for the younger generation as multitasking is facilitated by the omnipresent media on mobile devices that allow seamless integration of work, play, and social interaction (e.g., Rosen, Mark, & Cheever, 2013; Srivastava, 2013; David, Kim, Brickman, Ran, & Curtis, 2014; Carrier, Cheever, Rosen, Benitez, & Chang, 2009). A recent investigation in the U.S. revealed that half of the teens, when doing homework, “often” or “sometimes” watch TV (51%), use social media (50%), text (60%), and listen to music (76%) (Common Sense Census, 2015). Research has found that such multitasking decreases task performance (Ophir, Nass, & Wagner, 2009; Wang et al., 2012), and some studies have uncovered that multitasking also distorts users' self-perceived ability in that multitaskers tend to overestimate their ability to multitask even when their actual performance suffers from multitasking (Ran, Yamamoto, & Xu, 2016; Sanbonmatsu, Strayer, Medeiros-Ward, & Watson, 2013). Another intriguing finding is that media multitasking may also influence time perception. When participants were required to watch a commercial and simultaneously monitor another window on the same computer screen in which an “x” or “z” or a black circle
* Corresponding author. E-mail addresses: [email protected] (S. Xu), [email protected] (P. David). https://doi.org/10.1016/j.chb.2017.11.032 0747-5632/Published by Elsevier Ltd.
appeared, time appeared to be faster (Chinchanachokchai, Duff, & Sar, 2015). Chinchanachokchai and colleagues' article is probably the first published article that reveals the interesting relationship between media multitasking and perceived time passage in an experimental design, which sets a good foundation for future investigation into time perception and media use. However, as pointed out by Chinchanachokchai and colleagues, tasks manipulated in the study were low-level cognitive tasks (Chinchanachokchai et al., 2015, p. 189), whereas real-life multitasking is more complex and sophisticated. Therefore, it is necessary to look at the effects of more naturalistic everyday tasks on time perception. In addition to task structure, such as switching between tasks, time passage may also be influenced by task content, such as enjoyment of the task. Most of the studies on subjective passage of time have found task enjoyment to be a critical predictor. For example, if a task or an event is perceived as being enjoyable or entertaining, time passes quickly (Watt, 1991; Sucala, Scheckner, & David, 2010). Thus, the focus of this study is to investigate the influence of task content on time passage for high- and lowenjoyment tasks. Time perception involves not only time passage, but also another related yet distinct assessment known as time duration. While time passage is associated with the psychological perception of time passing quickly or slowly, time duration is the assessment of length of time interval. Time duration estimates are widely used in media
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
research and significant correlations between time spent on media and attitudinal or behavioral outcomes have been observed in various domains. For example, heavy consumption of television is correlated with a distorted view of reality conveyed in television dramas (Gerbner, Gross, Morgan, & Signorielli, 1994), and more time on mobile phones is correlated with disruption of everyday activities (David, et al., 2014). Despite the centrality of time as a key measure of media use, surprisingly little is known about the psychological process underlying estimation of time for media activities. The widely used measure of retrospective recall of time spent on media activities on a typical day or during a preceding interval is a daunting challenge. Consider an estimate of time spent on texting or email or social media within a 24-h period. With experiences dispersed throughout the day and interwoven into other activities, the temporal summation of those activities is likely to be contaminated by conjoint activities, typically known as media multitasking. With the pervasiveness of media multitasking or task-switching and the importance of time in media research, a better understanding of the psychological perception of time during media multitasking deserves attention. In this study, we examine two assessments of time: time passage and time duration. While the former may be tied to a time-based functional explanation of why people multitask or switch tasks, the latter may have important implications for the widely-used, self-reported measure of media use. The purpose of this study is threefold: first, to replicate a previous study on multitasking and time passage using a more naturalistic tasks; second, to examine how the entertainment value of tasks influences time passage and duration; third, to investigate how task switching impacts time passage and time duration estimates. 2. Time passage and time duration Researchers who examine the psychology of time make a distinction between prospective and retrospective estimations (Block, 1992; Zakay & Block, 2004). In prospective estimation, participants are cued in advance to attend to the duration of a stimulus and subsequently asked to estimate its duration. Participants offer multiple assessments for stimuli of different durations in a controlled laboratory setting, which are used to determine underlying psychometric functions. Such prospective estimations are explained using a hypothetical psychological clock that consists of a pacemaker that generates ticks, an accumulator that stores the clicks, and a switch that opens and closes the gate between the pacemaker and accumulator to mark the beginning and end of an assessment interval. Although concrete neural mechanisms that serve such functions have not been identified, prospective estimations of time appear to fit the predictions of this hypothetical model. Assessment of time in communication research, however, involves thinking back to prior experiences or estimating a typical experience, both involving retrospection. Under retrospective estimation, two distinct assessments of time have been examined e time duration and time passage. Duration is the subjective estimate of time interval in seconds, minutes or hours, and passage is the psychological perception of time along a bipolar continuum with flight (time flew by) and stagnation (time dragged on) as anchors. References to such phenomenological experiences of passage of time are found in tropes such as “a watched pot never boils” or “time flies when you are having fun.” Interestingly, time duration and passage have been found to be unrelated and explained through different mechanisms (Wearden, 2005). Consider this experiment in which one group of participants was assigned to watch a 9-min movie clip and the other group
363
assigned for the same duration to a waiting room condition without a specific task or activity. When asked to rate passage of time, time seemed to pass quickly in the video condition, whereas time seemed to pass slowly in the waiting room condition. After the assessment of time passage, participants in both conditions were assigned to a reading activity for 10 min, which served as a distractor. Then both groups were asked to think back to the first phase of the study and provide an estimate of time duration in minutes. Now estimates of time duration were found to be longer in the video condition than in the waiting-room condition. In other words, perceptions of time flying by did not result in estimates of shorter time durations. To account for these findings, researchers have advanced an attention explanation for time passage and a memory explanation for time duration (Wearden, 2005). The attention explanation is based on the reasoning that when sufficient attentional resources are available to think about time or to check the clock, time seems to crawl by slowly, as in the waiting room or “watched pot never boils” experience (Brown, 1997; Sucala et al., 2010). In contrast, when attentional resources are scarce, such as when watching a movie or exciting sporting event, time goes by quickly because of lack of attentional resources to allocate to the temporal dimension (Chinchanachokchai, et al., 2015; Wearden, 2005). The attentional explanation is not new and was noted by James (1890) who observed that time passes slowly when we are attentive to its passage. The attention model can also explain the “losing track of time” phenomenon that is central to flow experiences (Csikszentmihalyi, 1990). Flow is a subjective state that individuals report when they are completely absorbed in something to the point of forgetting time, fatigue, and everything else but the activity itself. The defining feature of flow is full attention involvement in momentto-moment activity (Csikszentmihalyi, 2014). Attention is fully invested in the task at hand, thus causing the “losing track of time” (Nakamura & Csikszentmihalyi, 2002). Further evidence for the attentional explanation is found in studies in which time appears to pass by quickly when engaged in a demanding activity that requires deeper processing than in a less demanding activity that can be accomplished with less cognitive effort (Sucala et al., 2010). In essence, the attentional model suggests that temporal and non-temporal information compete for limited attentional resources from a common pool. To the extent that a concurrent nontemporal task is demanding or vying for attention, less attentional resource is available for temporal processing, thus creating the perception of speedy passage of time (Sucala et al., 2010). Among other variables, the entertainment value of the activity is correlated with attention and has been highlighted as a predictor of time passage. When a task or an event is perceived as entertaining or enjoyable, time passes quickly (Watt, 1991; Sucala et al., 2010; Sackett, Meyvis, Nelson, Converse, & Sackett, 2010). Based on the foregoing review of passage of time, we predict that time will pass quickly when the participant is engaged in an entertaining activity, such as watching a sitcom, than when engaged in less entertaining activity, such as reading an abstract journal article. H1. Passage of time will be perceived to be quicker when experiencing a more entertaining media activity than a less entertaining activity. From the psychological perception of time passage, which is tied to the limited capacity of attention (Lang, 2000), we shift to retrospective estimation of time duration, also described as remembered time, which is related to memory (Block & Reed, 1978). For remembered time, cues stored in memory serve as temporal markers and retrospective estimates of time duration are based on a summation of these markers (Ornstein, 1969). Some of these markers, such as contextual changes in a stimulus, have rich
364
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
semantic features that facilitate temporal summation (Block & Reed, 1978). Likewise, changes in environmental context, mood, and type of processing strategy also create memory traces or markers that facilitate retrospective recall (e.g., Block & Reed, 1978). Once encoded and stored in memory, these markers are available for subsequent retrieval and temporal summation (Tversky & Kahneman, 1973). If retrospective assessment of time duration is related to summation of memorable cues, a sitcom with concrete events, likeable characters, memorable jokes and punch lines has more built-in context changes and temporal markers compared to a academic article forced upon a student. Therefore, we predict that the estimated duration of the entertaining video activity will be longer than the duration for the less entertaining reading activity. H2. Estimates of time duration will be longer for a more entertaining media activity than for a less entertaining activity. Estimated time duration also can be influenced by segmentation of a stimulus. Segmentation markers can facilitate subsequent reconstruction of the interval duration and retrospective estimation of time duration is greater when an activity or stimulus is segmented (Poynter, 1989). The importance of segmentation as a contextual cue for time duration is relevant to this study because media multitasking involves activity-switching, which is the division of the media experience into narrower segments or activities. Next, we examine estimation of time passage and time duration in such media multitasking or activity-switching contexts. 3. Time passage and duration in media multitasking Multitasking is commonly defined as the performance of two or more independent activities at the same time (Jeong & Fishbein, 2007; Sanbonmatsu et al., 2013) in situations in which at least one activity or goal is accomplished through a media channel (e.g., television, internet, cellphone, etc.). Often, both task-switching and simultaneous media use are considered under the umbrella of media multitasking. In fact, these two types of multitasking behavior represent two ends of the multitasking continuum (Salvucci, Taatgen & Borst, 2009). At one end is simultaneous multitasking such as listening to music and doing homework at the same time; at the other end is task switchingdreading a book and checking text messages from time to time that involves relatively longer intervals between tasks, from minutes to hours (Xu & Wang, 2017). In real life, media multitaskers switch frequently from one activity to another, such as reading a printed article and watching television at the same time, both of which utilize visual recourses. Media multitasking has spurred a line of research focused on productivity, performance and persuasion as key outcomes (Jeong & Hwang, 2012, 2014) and another line of inquiry has focused on underlying motivations for media multitasking behaviors (e.g., Hwang, Kim, & Jeong, 2014; Wang & Tchernev, 2012; Yeykelis, Cummings, & Reeves, 2014). The latter group of studies report that multitasking or task-switching eases boredom, making routine and required activities more entertaining and enjoyable (Wang & Tchernev, 2012). Based on self-reports from student samples it is apparent that switching happens frequently when the student is engaged in a required activity, such as doing homework. Activities, such as watching television, exchanging text messages or checking on social media are pursued by students to make the primary activity of doing homework less laborious and more entertaining. In keeping with this line of reasoning, we predicted that switching between reading a academic article and watching a sitcom would be more entertaining than having to read a academic article without any breaks. Extending H1, we predict that passage of time will be quicker for the more entertaining activity involving
switching between video and reading than just the reading activity. H3. Passage of time will appear to be quicker when switching between less and more entertaining activities (article þ sitcom) than when engaged only in the less entertaining activity (reading article). In addition to making a less entertaining activity more entertaining, switching between activities also creates contextual changes that can serve as temporal markers for estimation of time duration. Switching involves changes in media environment (e.g., from computer to TV, or from computer to smart phone), processing strategies (watching passively vs. interactive media) and strategic allocation of perceptual and cognitive resources (e.g., reading involves decoding text and comprehending, playing a video game requires hand-eye coordination). Such changes are automatically coded as memory markers (Levin & Zakay, 1989). Moreover, switching between activities could induce different affect states, such as positive affect from comedy, arousal from music, or a sense of social belonging from social media, thus making these segments more memorable and available for recall during retrospective duration estimation (Block & Zakay, 1997). Hence the process of activity-switching, commonly referred to as media multitasking, could lead to estimates of longer time duration. H4. Estimation of time duration will be greater when switching between less and more entertaining activities (article þ sitcom) than when engaged in either those activities separately. Published findings suggest a significant effect of individual difference on media multitasking. Those who prefer doing one task at a time (monchrons) have been differentiated from (polychrons) (Zhang, Goonetilleke & Plocher, 2005) and a link between multitasking preference and performance has been reported (Poposki & Oswald, 2010). These scales, designed originally for work contexts, were adapted for media multitasking and introduced in a recent study (David, Kim, et al., 2014). If preference for media multitasking is correlated with performance, it could also likely influence individual perceptions of time passage and time duration. RQ1: What is the relationship between media multitasking preference and estimates of time passage and time duration in activity-switching contexts? 4. Method 4.1. Participants Participants were recruited from an introductory course in communication and received course credit for participation. A total of 108 students, between 18 to 34 years of age (M ¼ 20.4, SD ¼ 2.65) participated in the study and 69.2% were female. Three participants were removed from the analysis for not filling out the questionnaires. 4.2. Pre-experiment measures Before the experiment, all participants completed a survey that included an assessment of multitasking preference and media multitasking habits. Activity-Switching Preference. Activity-switching preference was measured using a scale with 16 items, such as, I am more efficient when I am multitasking, I try to multitask whenever possible, and I enjoy multitasking (David, Kim, et al., 2014). All items were rated on a 7-point scale (1 ¼ strongly disagree, 7 ¼ strongly agree). After reverse coding negatively worded items, the scale had good reliability and the 15 items were averaged to create a composite score
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
of activity-switching preference (M ¼ 3.58, SD ¼ 0.89, a ¼ 0.83). Activity-Switching Behaviors. Frequency of concurrent use of different types of media while studying was assessed using a sixitem activity-switching measure (David, Brickman, et al., 2014). All items were rated on a 7-point scale (1 ¼ never, 7 ¼ always). Each item followed the pattern, ‘When studying or doing home, I also _____.’ Switching activities used to fill in the blank included social media, texting, music, video, browsing, and gaming. Items were averaged to create a composite score for activity-switching habits (M ¼ 4.43, SD ¼ 1.08, a ¼ 0.75). 4.3. Post-experiment measures In the single-activity condition, assessments focused on the lowentertainment (reading article) or high-entertainment (watching sitcom) activity. In the activity-switching condition, separate assessments were obtained for the combined activity in addition to separate assessments for low- and high-entertainment activities. A 7-point scale (1 ¼ time dragged on, 7 ¼ time flew by) was used to assess time passage (Sucala et al., 2010) and duration was measured in minutes and seconds (Block & Zakay, 1997). To ensure that participants followed instructions and did not use time cues available in the environment, they were asked the following question: Did you use any electronic devices in the room (such as cellphone, clock, computer, watch, etc.) to help with time estimation? Enjoyment was evaluated using six items e enjoyable, engaging, boring, fun, happy, and entertaining, which were rated on a 7-point scale (1 ¼ not at all, 7 ¼ extremely) and with Cronbach's a ¼ 0.94. Four items were used to measure arousal e exciting, arousing, awake and stimulated, which were rated on a 7-point scale (1 ¼ not at all, 7 ¼ extremely), with Cronbach's a ¼ 0.80. Participants also provided age, gender, and year in college. 4.4. Design and stimuli A between-subjects design with 4 conditions was employed: low-entertainment only, high-entertainment only, low þ high entertainment with low switching, and low þ high entertainment with high switching. Participants were assigned randomly to one of four conditions. The high-entertainment stimulus was a 10-min storyline from an episode of the comedy series The Big Bang Theory. In the storyline, using operant conditioning in lab rats as the scientific principle, Sheldon tries to influence Penny's behavior by offering chocolate as a reward. The Big Bang Theory was chosen because of its high-entertainment value and popularity among students. The low-entertainment stimulus was a published academic article on international journalism. The length of the article was 1719 words and participants were given 10 min to read the article. In the low-switching condition, the video and reading stimuli were divided into two parts and task-switching was enforced by having participants alternate between the video and text (see Fig. 1). Participants switched between the video and text three times in the low-switching condition. In the high-switching condition, the video and text were divided into four chunks and participants switched seven times between video and reading. 4.5. Procedure Upon arrival at the lab, participants were assigned to one of four conditions. After informed consent, they were asked to set aside all distractions, including watches and mobile phones. No clocks or time cues were visible in the room and the time on the video play bar also was hidden.
365
After filling out the pre-questionnaire, participants in the textonly condition read the printed text for 10 min and those in video-only condition watched the video for 10 min. Participants in the activity-switching conditions were forced to alternate between text and video for 20 min. During the reading activity, the lights in the room came on, a large screen in the front of the room turned to black and the participants were prompted to read. Before switching to video, the lights in the room dimmed until it was too dark to read, the screen turned bright and the video resumed playing. Lights were dimmed in the room to block participants from reading the article when the video was playing. In the low-switching condition, participants watched the video on the big screen for 5 min and switched to reading for 5 min and this cycle was repeated twice. In the high-switching condition, participants engaged in short durations (1.5e3 min, see Fig. 1) of watching video followed by reading and were treated to four cycles of video followed by reading. At the end, participants in all conditions filled out the post-experiment questionnaire. 5. Results 5.1. Preliminary analysis One-way ANOVA was conducted with four conditions (low entertainment, high entertainment, low- and high-entertainment with low switching, and low- and high-entertainment with high switching) and with enjoyment, arousal, time passage and time duration as separate dependent variables. While the overall model was significant for each of the four dependent variables, in each case the difference between the low (3 switches) and highswitching (7 switches) conditions was not significant. To simplify the analysis, data from the low- and high-switching conditions were combined into an overall switching condition and subsequent ANOVAs were conducted with three conditions (low entertainment, high entertainment, switching between low and high entertainment). 5.2. Enjoyment and arousal One-way ANOVA with the three conditions explained above was examined with enjoyment as the dependent variable and planned contrasts were used to test for differences between conditions (see Table 1 for summary of means). The overall model was significant, F(2, 102) ¼ 54.73, p < 0.001, and not surprisingly, the highentertainment activity of watching a sitcom was rated more enjoyable (M ¼ 5.81, SD ¼ 1.01) than the low-entertainment activity of reading a academic article (M ¼ 2.70, SD ¼ 1.21). Enjoyment in the switching condition (M ¼ 4.51, SD ¼ 0.98) was less than the enjoyment in high-entertainment condition, but more than the enjoyment in the low-entertainment condition. Enjoyment from the three conditions were significantly different from one another. The results for arousal were similar to the results for enjoyment. The overall model was significant, F(2, 102) ¼ 21.20, p < 0.001, and differences between all three conditions were significant. Arousal in the high-entertainment condition (M ¼ 4.42, SD ¼ 1.19) was higher than in the low-entertainment condition (M ¼ 2.49, SD ¼ 1.22). Arousal in the switching condition (M ¼ 3.56, SD ¼ 0.91) was between arousal in the high- and low-entertainment conditions. 5.3. Time passage Next a one-way ANOVA was examined with time passage as the dependent variable. The model was significant, F(2, 104) ¼ 13.39, p < 0.001, par h2 ¼ 0.21. Scheffe's post hoc contrasts confirmed H1
366
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
Time No switching low-entertainment condition Watching Video 10 min 10 sec
Time No switching high-entertainment condition 5m
5 m 10 s
Time Low + high entertainment with low-switching condition
3m
1m 45s
2m 46s
2m 39s
Time Low + high entertainment with high-switching condition Fig. 1. Stimuli for four conditions.
Table 1 Summary means (SD) for key variables by experimental conditions.
Enjoyment Arousal Time Passage Duration Estimates
No Switching Low Entertainment n ¼ 25
No Switching High Entertainment n ¼ 26
Switching Between Low and High Entertainment n ¼ 54
2.70 (1.21) 2.49 (1.22) 3.35 (2.06) 0.83b (0.29)
5.81 (1.01) 4.42 (1.19) 5.37a (1.39) 1.07b (0.59)
4.50 (0.98) 3.56 (0.91) 5.01a (1.11) 1.45 (0.80)
Note. Entries that share subscripts are not significantly different at p 05. Duration Estimates refers to the ratio of the estimated duration to the actual duration.
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
and H3. As predicted in H1, passage of time was faster in the highentertainment, sitcom condition (M ¼ 5.37, SD ¼ 1.39) than in the low-entertainment, academic article condition (M ¼ 3.46, SD ¼ 2.01), was significant at p < 0.001. Also as predicted in H3, passage of time was faster in switching condition (M ¼ 5.02, SD ¼ 1.11) compared to the low-entertainment condition, at p < 0.001. But the difference between the high-entertainment (M ¼ 5.37, SD ¼ 1.39) and switching condition (M ¼ 5.02, SD ¼ 1.11) was not significant. 5.4. Time duration To test H2 and H4, estimated duration was transformed by dividing estimated duration by actual duration of each condition: 10 min for the less entertaining condition, 10 min 10 s for the high entertaining condition, and 20 min 10 s for the switching condition. This transformation was necessary to account for the fact that the actual duration was about 20 min in the activity-switching condition, but only about 10 min in the other two conditions (Block & Zakay, 1997). Then the data were analyzed using single-factor, between-subjects ANOVA with three levels (low-entertainment, high-entertainment, switching between low- and highentertainment). Main effect for condition was significant, F(2, 99) ¼ 8.03, p ¼ 0.001, par h2 ¼ 0.14, and Scheffe's post hoc contrasts confirmed that the difference between the low-entertainment condition (M ¼ 0.83, SD ¼ 0.29) and the activity-switching condition (M ¼ 1.45, SD ¼ 0.80) was significant at p ¼ 0.001, and the difference between the high-entertainment condition (M ¼ 1.07, SD ¼ 0.13) and the activity-switching condition (M ¼ 1.45, SD ¼ 0.80) was tended toward significance at p ¼ 0.06. These two contrasts lend support for H4. However, the results did not support H2 because the difference between the low-entertainment and highentertainment condition was not significant. 5.5. Individual differences Two individual differences, frequency of activity-switching during studying and preference for activity-switching, were examined through bivariate Pearson correlations with estimates of time passage and time duration (see Table 2). Both frequency (r ¼ 0.23, p < 0.05) and preference for task-switching (r ¼ 0.18, p ¼ 0.06) were negatively correlated with time duration, but not with time passage. The two dimensions of time, duration and passage, were not correlated. In light of the significant correlation, follow-up analysis was conducted to examine the role of individual differences in estimation of time duration. When both switching habit and switching preference were entered as covariates in the time duration analysis, a main effect was observed for the switching habit, but not switching preference. But the main effect for condition remained significant and the findings reported for time duration remained intact. Next, to examine the potential effect of gender and age, a 2
Table 2 Bivariate correlations between time variables and individual differences in ActivitySwitching.
1. 2. 3. 4.
Time Passage Duration Estimates Activity-Switching Habits Activity-Switching Preference
Note. yp < 0.10, *p < 0.05.
1
2
3
0.05 0.07 0.10
0.23* 0.18y
0.14
367
(male, female) x 3 (conditions: low-entertainment, high-entertainment, switching between low and high) ANOVA was examined with age as a covariate. This analysis was carried out separately for time passage and time duration. In both instances, gender and age were not significant. 6. Discussion Although time is an important variable in media and communication research, the psychological mechanism underlying the estimation of time spent on media activities has received surprisingly little attention from researchers. Building upon an existing study on media multitasking and time passage (Chinchanachokchai et al., 2015), our study examined two measures associated with the psychology of time, time passage and time duration, with more naturalistic task stimuli, reading an academic article and watching a sitcom, and further investigated how task switching influenced time perceptions. Passage of time is the psychological perception on a continuum that extends from time standing still to time flying by. We predicted that time would fly by when watching an entertaining situational comedy such as Big Bang Theory, and that time would be perceived to stagnate when asked to reading a less entertaining academic article. In line with our predictions, students assigned to the sitcom reported that time flew by, whereas students in the academic article condition reported that time dragged on. Although the finding that time passes quickly when engaged in a more entertaining activity than a less entertaining activity is predictable, the perceived passage of time when the two activities were combined offers a plausible, functional explanation for multitasking. Switching from the less entertaining activity of reading an academic article to the more entertaining activity of watching a sitcom significantly sped up passage of time when compared to uninterrupted involvement in the less entertaining reading activity. Previous studies, based on self-reports, have identified relational and emotional benefits (Hwang et al., 2014; Wang & Tchernev, 2012) and alleviation of challenge or boredom of the work or study task (Yeykelis et al., 2014) as a motivation for multitasking. In this study, we demonstrate this finding through an experiment that demonstrates that activity switching not only makes the less entertaining activity significantly more entertaining, but also speeds up passage of time to the level of the more entertainment activity. This finding suggests that sporadic relief from demanding or mundane activities through bursts of distractions may explain the appeal of media multitasking. And the potential dip in task performance from switching or multitasking may be offset by the persistence in the combined activity. In other words, the perception of the rapid passage of time in multitasking contexts may prolong goal pursuit, which may offset potential erosion in performance. When time passes more quickly, an individual may perceive the activity as interesting and persist longer toward success in the primary activity (Gable & Poole, 2012), such as in completing homework that one dislikes. In short, the findings associated with passage suggest that the benefit of multitasking is not limited to entertainment, but extends to the perception of flight of time that may translate into prolonged engagement in required activities, even if they are challenging or boring. While the findings associated with time passage offer a timebased explanation for multitasking, the findings associated with time duration have significant implications for measurement of media use. A media experience that involved switching between less and more entertaining activities led to an overestimation of time. Time duration was overestimated during activity switching in comparison to estimation of time duration for single activities.
368
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
When both the less and more entertaining conditions are combined, we predicted that the cognitive segmentation markers between switches would serve as memory cues that facilitate temporal summation, thus resulting in the estimates of longer duration in the switching condition compared to the single-task condition. This hypothesis was confirmed, which exposes the potential danger of relying on retrospective self-reports of time spent on multitasking media activities. Contrary to our prediction, duration estimation for the more interesting task was not greater than the duration estimation for the less entertaining task. The prediction, drawn from an earlier study, was based on the reasoning that more interesting content would leave a memory trace with stronger cues than less interesting content, which will improve recall, in turn increasing estimates of time duration (Wearden, 2005). We did not find support for this hypothesis. Results of this study suggest a potential further explication and examination on the memory-based theoretical model involving summation of temporal markers, i.e., the greater number of contextual cues offered by the experience, longer the retrospective duration is estimated (Block & Reed, 1978). However, contextual cues may have different dimensions and each dimension may influence duration estimation differently, for example, segmentation markers may have a stronger effect on duration estimation than the semantic cues. As demonstrated in this study, on one hand, switching between the high entertainment and low entertainment tasks offers segmentation markers. On the other hand, the high entertainment task, that is watching a sitcom, also offers memorable attributes in the storyline as semantic markers, such as the concrete events, likeable characters, memorable jokes and punch lines in a sitcom. Results from this study suggested that segmentation markers served as more memorable contextual cues than the semantic markers, as participants significantly overestimated the experience duration in the switching condition, but not in the high entertainment condition. Future study can further explore different dimensions of contextual cues and how they may impact duration estimation differently. A limitation of this study is the artificiality of the media experience in which participants were forced to switch between activities in the laboratory setting, which is unlike everyday media experiences that are more fluid. Further, participants in the study were college students, limiting the generalizability of results to the general population. Despite these limitations, the overestimation of time duration in activity-switching contexts deserves attention from media researchers and advertisers. When students report more than 40 h of activities in a 24-h day (David, et al., 2014), it is important to understand the underlying psychology of time estimation. The absence of a correlation between time passage and time duration should be noted because it supports findings in the literature that these assessments are tied to different psychological processes, with time passage tied to attention and time duration tied to memory. Also, the negative correlation between individual differences and time duration offers avenues for future research. Both task-switching preference and task-switching habits were negatively correlated with time duration, which can be interpreted as attenuation in overestimation among heavy multitaskers. With experience, it is likely that multitaskers may develop heuristics or cognitive strategies to correct for the additional segmentation markers created during activity-switching. It is also possible that as fluency of switching increases, segmentation markers may become less salient and less likely to be coded in memory. These competing explanations can be addressed in future research. In summary, the findings from this study offer new insight into the psychology of time perception that is critical to a better understanding of time
estimates of media multitasking. The attention and memory models associated with passage and duration respectively offer a platform for future extensions both on the underlying mechanisms of time perception during media experiences and on measurement of media use. References Block, R. A. (1992). Prospective and retrospective duration judgment: The role of information processing and memory. In Time, action and cognition (pp. 141e152). Springer Netherlands. https://doi.org/10.1007/978-94-017-3536-0. Block, R. A., & Reed, M. A. (1978). Remembered duration: Evidence for a contextualchange hypothesis. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 656. https://doi.org/10.1037//0278-7393.4.6.656. Block, R. A., & Zakay, D. (1997). Prospective and retrospective duration judgments: A meta-analytic review. Psychonomic Bulletin & Review, 4(2), 184e197. https:// doi.org/10.3758/BF03209393. Brown, S. W. (1997). Attentional resources in timing: Interference effects in concurrent temporal and nontemporal working memory tasks. Perception and Psychophysics, 59, 1118e1140. https://doi.org/10.3758/BF03205526. Carrier, L. M., Cheever, N. A., Rosen, L. D., Benitez, S., & Chang, J. (2009). Multitasking across generations: Multitasking choices and difficulty ratings in three generations of Americans. Computers in Human Behavior, 25(2), 483e489. https:// doi.org/10.1016/j.chb.2008.10.01. Chinchanachokchai, S., Duff, B. R. L., & Sar, S. (2015). The effect of multitasking on time perception, enjoyment, and ad evaluation. Computers in Human Behavior, 45, 185e191. https://doi.org/10.1016/j.chb.2014.11.087. Common Sense Census. (2015). The common sense Census: Media use by tweens and teens. Retrieved from: www.commonsensemedia.org/census. Csikszentmihalyi, M. (1990). Flow the psychology of optimal experience. New York: Harper-Perennial. Csikszentmihalyi, M. (2014). Flow and the foundations of positive psychology. Dordrecht: Springer. David, P., Brickman, J. S., Silva, D. E., Xu, S., & Ran, W. (2014). Beyond time and frequency: Attending to attention in media multitasking. International Communication Association 2014 Annual Convention in Seattle. David, P., Kim, J.-H., Brickman, J. S., Ran, W., & Curtis, C. M. (2014). Mobile phone distraction while studying. New Media & Society. https://doi.org/10.1177/ 1461444814531692. Gable, P. A., & Poole, B. D. (2012). Time flies when you're having approachmotivated fun effects of motivational intensity on time perception. Psychological Science. https://doi.org/10.1177/0956797611435817. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with television: The cultivation perspective. In J. Bryant, & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 17e41). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Hwang, Y., Kim, H., & Jeong, S.-H. (2014). Why do media users multitask? Motives for general, medium-specific, and content-specific types of multitasking. Computers in Human Behavior, 36, 542e548. https://doi.org/10.1016/ j.chb.2014.04.040. James, W. (1890). The principles of psychology (Vol. 2). MacMillan. Jeong, S. H., & Fishbein, M. (2007). Predictors of multitasking with media: Media factors and audience factors. Media Psychology, 10(3), 364e384. https://doi.org/ 10.1080/15213260701532948. Jeong, S.-H., & Hwang, Y. (2012). Does multitasking increase or decrease persuasion? Effects of multitasking on comprehension and counterarguing. Journal of Communication, 62(4), 571e587. https://doi.org/10.1111/j.14602466.2012.01659.x. Jeong, S.-H., & Hwang, Y. (2014). Multitasking and persuasion: The role of structural interference. Media Psychology, 1e24. https://doi.org/10.1080/ 15213269.2014.933114. Lang, A. (2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46e70. https://doi.org/10.1093/joc/50.1.46. Levin, H., & Zakay, D. (1989). Judging the duration of time intervals: A process of remembering segments of experience. Time and human cognition: A life-span perspective. Elsevier. Nakamura, J., & Csikszentmihalyi, M. (2002). The concept of flow. Handbook of Positive Psychology, 89e105. Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. Proceedings of the National Academy of Sciences, 106(37), 15583e15587. https:// doi.org/10.1073/pnas.0903620106. Ornstein, R. E. (1969). On the experience of time. Harmondsworth, UK: Penguin. Poposki, E. M., & Oswald, F. L. (2010). The multitasking preference inventory: Toward an improved measure of individual differences in polychronicity. Human Performance, 23(3), 247e264. https://doi.org/10.1080/08959285.2010.487843. Poynter, D. (1989). Judging the duration of time intervals: A process of remembering segments of experience. Advances in Psychology, 59, 305e331. https:// doi.org/10.1016/S0166-4115(08)61045-6. Ran, W., Yamamoto, M., & Xu, S. (2016). Media multitasking during political news consumption: A relationship with factual and subjective political knowledge. Computers in Human Behavior, 56, 352e359. https://doi.org/10.1016/ j.chb.2015.12.015.
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369 Rosen, L. D., Mark Carrier, L., & Cheever, N. A. (2013). Facebook and texting made me do it: Media-induced task-switching while studying. Computers in Human Behavior, 29(3), 948e958. https://doi.org/10.1016/j.chb.2012.12.001. Sackett, A. M., Meyvis, T., Nelson, L. D., Converse, B. A., & Sackett, A. L. (2010). You’re having fun when time flies the hedonic consequences of subjective time progression. Psychological Science, 21(1), 111e117. https://doi.org/10.1177/ 0956797609354832. Salvucci, D. D., Taatgen, N. A., & Borst, J. P. (2009). Toward a unified theory of the multitasking continuum: From concurrent performance to task switching, interruption, and resumption. In Proceedings of the SIGCHI conference on human factors in computing systems (pp. 1819e1828). New York: ACM. Sanbonmatsu, D. M., Strayer, D. L., Medeiros-Ward, N., & Watson, J. M. (2013). Who multi-tasks and why? Multi-tasking ability, perceived multi-tasking ability, impulsivity, and sensation seeking. PloS One, 8(1), e54402. https://doi.org/ 10.1371/journal.pone.0054402. Srivastava, J. (2013). Media multitasking performance: Role of message relevance and formatting cues in online environments. Computers in Human Behavior, 29(3), 888e895. https://doi.org/10.1016/j.chb.2012.12.023. Sucala, M., Scheckner, B., & David, D. (2010). Psychological time: Interval length judgments and subjective passage of time judgments. Current Psychology Letters, 26(2), 2011. Retrieved from: https://cpl.revues.org/4998. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207e232. Watt, J. D. (1991). Effect of boredom proneness on time perception. Psychological
369
Reports, 69(1), 323e327. https://doi.org/10.2466/pr0.1991.69.1.323. 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), 493e513. https://doi.org/10.1111/j.14602466.2012.01641.x. Wang, Z., David, P., Srivastava, J., Powers, S., Brady, C., D’Angelo, J., & Moreland, J. (2012). Behavioral performance and visual attention in communication multitasking: A comparison between instant messaging and online voice chat. Computers in Human Behavior, 28(3), 968e975. https://doi.org/10.1016/ j.chb.2011.12.018. Wearden, J. (2005). Origins and development of internal clock theories of psychological time. Psychology Francaise, 50, 7e25. Xu, S., & Wang, Z. (2017). Does multitasking really exist? In P. Vorderer (Ed.), Permanently online - permanently connected (pp. 72e82). London: Routledge. Yeykelis, L., Cummings, J. J., & Reeves, B. (2014). Multitasking on a single device: Arousal and the frequency, anticipation, and prediction of switching between media content on a computer. Journal of Communication, 64(1), 167e192. https://doi.org/10.1111/jcom.12070. Zakay, D., & Block, R. A. (2004). Prospective and retrospective duration judgments: An executive-control perspective. Acta Neurobiologiae Experimentalis, 64(3), 319e328. https://doi.org/10.3758/BF03209393. Zhang, Y., Goonetilleke, R. S., Plocher, T., et al. (2005). Time-related behavior in multitasking situations. International Journal of Human-computer Studies, 62, 425e455. https://doi.org/10.1016/j.ijhcs.2005.01.002.
Contents lists available at ScienceDirect
Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh
Full length article
Distortions in time perceptions during task switching Shan Xu a, *, Prabu David b a b
School of Communication, The Ohio State University, Columbus, OH, USA College of Communication Arts and Sciences, The Michigan State University, East Lansing, MI, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Received in revised form 14 November 2017 Accepted 21 November 2017 Available online 22 November 2017
Perceived time passage and time duration were examined in a between-subjects design with four conditions: watching a sitcom, reading a journal article, occasional switching between sitcom and article, frequent switching between sitcom and article. Consistent with our prediction, time “flew by” in the high-entertainment condition that involved watching a sitcom, whereas time “dragged on” in the lowentertainment condition that involved reading a journal article. Switching between the two led to quicker passage of time than the low-entertainment condition, but not the high-entertainment condition. A different pattern was evident for duration estimation, with no difference between the low- and high-entertainment conditions, but a longer estimation of duration in the switching condition. Further, frequency of switching between the sitcom and article did not make a difference. These findings suggest that switching between tasks leads to overestimations of time spent on media. Published by Elsevier Ltd.
Keywords: Task switching Media multitasking Entertainment Time passage Time duration Media measure
1. Introduction Media multitasking has become a way of life for the younger generation as multitasking is facilitated by the omnipresent media on mobile devices that allow seamless integration of work, play, and social interaction (e.g., Rosen, Mark, & Cheever, 2013; Srivastava, 2013; David, Kim, Brickman, Ran, & Curtis, 2014; Carrier, Cheever, Rosen, Benitez, & Chang, 2009). A recent investigation in the U.S. revealed that half of the teens, when doing homework, “often” or “sometimes” watch TV (51%), use social media (50%), text (60%), and listen to music (76%) (Common Sense Census, 2015). Research has found that such multitasking decreases task performance (Ophir, Nass, & Wagner, 2009; Wang et al., 2012), and some studies have uncovered that multitasking also distorts users' self-perceived ability in that multitaskers tend to overestimate their ability to multitask even when their actual performance suffers from multitasking (Ran, Yamamoto, & Xu, 2016; Sanbonmatsu, Strayer, Medeiros-Ward, & Watson, 2013). Another intriguing finding is that media multitasking may also influence time perception. When participants were required to watch a commercial and simultaneously monitor another window on the same computer screen in which an “x” or “z” or a black circle
* Corresponding author. E-mail addresses: [email protected] (S. Xu), [email protected] (P. David). https://doi.org/10.1016/j.chb.2017.11.032 0747-5632/Published by Elsevier Ltd.
appeared, time appeared to be faster (Chinchanachokchai, Duff, & Sar, 2015). Chinchanachokchai and colleagues' article is probably the first published article that reveals the interesting relationship between media multitasking and perceived time passage in an experimental design, which sets a good foundation for future investigation into time perception and media use. However, as pointed out by Chinchanachokchai and colleagues, tasks manipulated in the study were low-level cognitive tasks (Chinchanachokchai et al., 2015, p. 189), whereas real-life multitasking is more complex and sophisticated. Therefore, it is necessary to look at the effects of more naturalistic everyday tasks on time perception. In addition to task structure, such as switching between tasks, time passage may also be influenced by task content, such as enjoyment of the task. Most of the studies on subjective passage of time have found task enjoyment to be a critical predictor. For example, if a task or an event is perceived as being enjoyable or entertaining, time passes quickly (Watt, 1991; Sucala, Scheckner, & David, 2010). Thus, the focus of this study is to investigate the influence of task content on time passage for high- and lowenjoyment tasks. Time perception involves not only time passage, but also another related yet distinct assessment known as time duration. While time passage is associated with the psychological perception of time passing quickly or slowly, time duration is the assessment of length of time interval. Time duration estimates are widely used in media
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
research and significant correlations between time spent on media and attitudinal or behavioral outcomes have been observed in various domains. For example, heavy consumption of television is correlated with a distorted view of reality conveyed in television dramas (Gerbner, Gross, Morgan, & Signorielli, 1994), and more time on mobile phones is correlated with disruption of everyday activities (David, et al., 2014). Despite the centrality of time as a key measure of media use, surprisingly little is known about the psychological process underlying estimation of time for media activities. The widely used measure of retrospective recall of time spent on media activities on a typical day or during a preceding interval is a daunting challenge. Consider an estimate of time spent on texting or email or social media within a 24-h period. With experiences dispersed throughout the day and interwoven into other activities, the temporal summation of those activities is likely to be contaminated by conjoint activities, typically known as media multitasking. With the pervasiveness of media multitasking or task-switching and the importance of time in media research, a better understanding of the psychological perception of time during media multitasking deserves attention. In this study, we examine two assessments of time: time passage and time duration. While the former may be tied to a time-based functional explanation of why people multitask or switch tasks, the latter may have important implications for the widely-used, self-reported measure of media use. The purpose of this study is threefold: first, to replicate a previous study on multitasking and time passage using a more naturalistic tasks; second, to examine how the entertainment value of tasks influences time passage and duration; third, to investigate how task switching impacts time passage and time duration estimates. 2. Time passage and time duration Researchers who examine the psychology of time make a distinction between prospective and retrospective estimations (Block, 1992; Zakay & Block, 2004). In prospective estimation, participants are cued in advance to attend to the duration of a stimulus and subsequently asked to estimate its duration. Participants offer multiple assessments for stimuli of different durations in a controlled laboratory setting, which are used to determine underlying psychometric functions. Such prospective estimations are explained using a hypothetical psychological clock that consists of a pacemaker that generates ticks, an accumulator that stores the clicks, and a switch that opens and closes the gate between the pacemaker and accumulator to mark the beginning and end of an assessment interval. Although concrete neural mechanisms that serve such functions have not been identified, prospective estimations of time appear to fit the predictions of this hypothetical model. Assessment of time in communication research, however, involves thinking back to prior experiences or estimating a typical experience, both involving retrospection. Under retrospective estimation, two distinct assessments of time have been examined e time duration and time passage. Duration is the subjective estimate of time interval in seconds, minutes or hours, and passage is the psychological perception of time along a bipolar continuum with flight (time flew by) and stagnation (time dragged on) as anchors. References to such phenomenological experiences of passage of time are found in tropes such as “a watched pot never boils” or “time flies when you are having fun.” Interestingly, time duration and passage have been found to be unrelated and explained through different mechanisms (Wearden, 2005). Consider this experiment in which one group of participants was assigned to watch a 9-min movie clip and the other group
363
assigned for the same duration to a waiting room condition without a specific task or activity. When asked to rate passage of time, time seemed to pass quickly in the video condition, whereas time seemed to pass slowly in the waiting room condition. After the assessment of time passage, participants in both conditions were assigned to a reading activity for 10 min, which served as a distractor. Then both groups were asked to think back to the first phase of the study and provide an estimate of time duration in minutes. Now estimates of time duration were found to be longer in the video condition than in the waiting-room condition. In other words, perceptions of time flying by did not result in estimates of shorter time durations. To account for these findings, researchers have advanced an attention explanation for time passage and a memory explanation for time duration (Wearden, 2005). The attention explanation is based on the reasoning that when sufficient attentional resources are available to think about time or to check the clock, time seems to crawl by slowly, as in the waiting room or “watched pot never boils” experience (Brown, 1997; Sucala et al., 2010). In contrast, when attentional resources are scarce, such as when watching a movie or exciting sporting event, time goes by quickly because of lack of attentional resources to allocate to the temporal dimension (Chinchanachokchai, et al., 2015; Wearden, 2005). The attentional explanation is not new and was noted by James (1890) who observed that time passes slowly when we are attentive to its passage. The attention model can also explain the “losing track of time” phenomenon that is central to flow experiences (Csikszentmihalyi, 1990). Flow is a subjective state that individuals report when they are completely absorbed in something to the point of forgetting time, fatigue, and everything else but the activity itself. The defining feature of flow is full attention involvement in momentto-moment activity (Csikszentmihalyi, 2014). Attention is fully invested in the task at hand, thus causing the “losing track of time” (Nakamura & Csikszentmihalyi, 2002). Further evidence for the attentional explanation is found in studies in which time appears to pass by quickly when engaged in a demanding activity that requires deeper processing than in a less demanding activity that can be accomplished with less cognitive effort (Sucala et al., 2010). In essence, the attentional model suggests that temporal and non-temporal information compete for limited attentional resources from a common pool. To the extent that a concurrent nontemporal task is demanding or vying for attention, less attentional resource is available for temporal processing, thus creating the perception of speedy passage of time (Sucala et al., 2010). Among other variables, the entertainment value of the activity is correlated with attention and has been highlighted as a predictor of time passage. When a task or an event is perceived as entertaining or enjoyable, time passes quickly (Watt, 1991; Sucala et al., 2010; Sackett, Meyvis, Nelson, Converse, & Sackett, 2010). Based on the foregoing review of passage of time, we predict that time will pass quickly when the participant is engaged in an entertaining activity, such as watching a sitcom, than when engaged in less entertaining activity, such as reading an abstract journal article. H1. Passage of time will be perceived to be quicker when experiencing a more entertaining media activity than a less entertaining activity. From the psychological perception of time passage, which is tied to the limited capacity of attention (Lang, 2000), we shift to retrospective estimation of time duration, also described as remembered time, which is related to memory (Block & Reed, 1978). For remembered time, cues stored in memory serve as temporal markers and retrospective estimates of time duration are based on a summation of these markers (Ornstein, 1969). Some of these markers, such as contextual changes in a stimulus, have rich
364
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
semantic features that facilitate temporal summation (Block & Reed, 1978). Likewise, changes in environmental context, mood, and type of processing strategy also create memory traces or markers that facilitate retrospective recall (e.g., Block & Reed, 1978). Once encoded and stored in memory, these markers are available for subsequent retrieval and temporal summation (Tversky & Kahneman, 1973). If retrospective assessment of time duration is related to summation of memorable cues, a sitcom with concrete events, likeable characters, memorable jokes and punch lines has more built-in context changes and temporal markers compared to a academic article forced upon a student. Therefore, we predict that the estimated duration of the entertaining video activity will be longer than the duration for the less entertaining reading activity. H2. Estimates of time duration will be longer for a more entertaining media activity than for a less entertaining activity. Estimated time duration also can be influenced by segmentation of a stimulus. Segmentation markers can facilitate subsequent reconstruction of the interval duration and retrospective estimation of time duration is greater when an activity or stimulus is segmented (Poynter, 1989). The importance of segmentation as a contextual cue for time duration is relevant to this study because media multitasking involves activity-switching, which is the division of the media experience into narrower segments or activities. Next, we examine estimation of time passage and time duration in such media multitasking or activity-switching contexts. 3. Time passage and duration in media multitasking Multitasking is commonly defined as the performance of two or more independent activities at the same time (Jeong & Fishbein, 2007; Sanbonmatsu et al., 2013) in situations in which at least one activity or goal is accomplished through a media channel (e.g., television, internet, cellphone, etc.). Often, both task-switching and simultaneous media use are considered under the umbrella of media multitasking. In fact, these two types of multitasking behavior represent two ends of the multitasking continuum (Salvucci, Taatgen & Borst, 2009). At one end is simultaneous multitasking such as listening to music and doing homework at the same time; at the other end is task switchingdreading a book and checking text messages from time to time that involves relatively longer intervals between tasks, from minutes to hours (Xu & Wang, 2017). In real life, media multitaskers switch frequently from one activity to another, such as reading a printed article and watching television at the same time, both of which utilize visual recourses. Media multitasking has spurred a line of research focused on productivity, performance and persuasion as key outcomes (Jeong & Hwang, 2012, 2014) and another line of inquiry has focused on underlying motivations for media multitasking behaviors (e.g., Hwang, Kim, & Jeong, 2014; Wang & Tchernev, 2012; Yeykelis, Cummings, & Reeves, 2014). The latter group of studies report that multitasking or task-switching eases boredom, making routine and required activities more entertaining and enjoyable (Wang & Tchernev, 2012). Based on self-reports from student samples it is apparent that switching happens frequently when the student is engaged in a required activity, such as doing homework. Activities, such as watching television, exchanging text messages or checking on social media are pursued by students to make the primary activity of doing homework less laborious and more entertaining. In keeping with this line of reasoning, we predicted that switching between reading a academic article and watching a sitcom would be more entertaining than having to read a academic article without any breaks. Extending H1, we predict that passage of time will be quicker for the more entertaining activity involving
switching between video and reading than just the reading activity. H3. Passage of time will appear to be quicker when switching between less and more entertaining activities (article þ sitcom) than when engaged only in the less entertaining activity (reading article). In addition to making a less entertaining activity more entertaining, switching between activities also creates contextual changes that can serve as temporal markers for estimation of time duration. Switching involves changes in media environment (e.g., from computer to TV, or from computer to smart phone), processing strategies (watching passively vs. interactive media) and strategic allocation of perceptual and cognitive resources (e.g., reading involves decoding text and comprehending, playing a video game requires hand-eye coordination). Such changes are automatically coded as memory markers (Levin & Zakay, 1989). Moreover, switching between activities could induce different affect states, such as positive affect from comedy, arousal from music, or a sense of social belonging from social media, thus making these segments more memorable and available for recall during retrospective duration estimation (Block & Zakay, 1997). Hence the process of activity-switching, commonly referred to as media multitasking, could lead to estimates of longer time duration. H4. Estimation of time duration will be greater when switching between less and more entertaining activities (article þ sitcom) than when engaged in either those activities separately. Published findings suggest a significant effect of individual difference on media multitasking. Those who prefer doing one task at a time (monchrons) have been differentiated from (polychrons) (Zhang, Goonetilleke & Plocher, 2005) and a link between multitasking preference and performance has been reported (Poposki & Oswald, 2010). These scales, designed originally for work contexts, were adapted for media multitasking and introduced in a recent study (David, Kim, et al., 2014). If preference for media multitasking is correlated with performance, it could also likely influence individual perceptions of time passage and time duration. RQ1: What is the relationship between media multitasking preference and estimates of time passage and time duration in activity-switching contexts? 4. Method 4.1. Participants Participants were recruited from an introductory course in communication and received course credit for participation. A total of 108 students, between 18 to 34 years of age (M ¼ 20.4, SD ¼ 2.65) participated in the study and 69.2% were female. Three participants were removed from the analysis for not filling out the questionnaires. 4.2. Pre-experiment measures Before the experiment, all participants completed a survey that included an assessment of multitasking preference and media multitasking habits. Activity-Switching Preference. Activity-switching preference was measured using a scale with 16 items, such as, I am more efficient when I am multitasking, I try to multitask whenever possible, and I enjoy multitasking (David, Kim, et al., 2014). All items were rated on a 7-point scale (1 ¼ strongly disagree, 7 ¼ strongly agree). After reverse coding negatively worded items, the scale had good reliability and the 15 items were averaged to create a composite score
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
of activity-switching preference (M ¼ 3.58, SD ¼ 0.89, a ¼ 0.83). Activity-Switching Behaviors. Frequency of concurrent use of different types of media while studying was assessed using a sixitem activity-switching measure (David, Brickman, et al., 2014). All items were rated on a 7-point scale (1 ¼ never, 7 ¼ always). Each item followed the pattern, ‘When studying or doing home, I also _____.’ Switching activities used to fill in the blank included social media, texting, music, video, browsing, and gaming. Items were averaged to create a composite score for activity-switching habits (M ¼ 4.43, SD ¼ 1.08, a ¼ 0.75). 4.3. Post-experiment measures In the single-activity condition, assessments focused on the lowentertainment (reading article) or high-entertainment (watching sitcom) activity. In the activity-switching condition, separate assessments were obtained for the combined activity in addition to separate assessments for low- and high-entertainment activities. A 7-point scale (1 ¼ time dragged on, 7 ¼ time flew by) was used to assess time passage (Sucala et al., 2010) and duration was measured in minutes and seconds (Block & Zakay, 1997). To ensure that participants followed instructions and did not use time cues available in the environment, they were asked the following question: Did you use any electronic devices in the room (such as cellphone, clock, computer, watch, etc.) to help with time estimation? Enjoyment was evaluated using six items e enjoyable, engaging, boring, fun, happy, and entertaining, which were rated on a 7-point scale (1 ¼ not at all, 7 ¼ extremely) and with Cronbach's a ¼ 0.94. Four items were used to measure arousal e exciting, arousing, awake and stimulated, which were rated on a 7-point scale (1 ¼ not at all, 7 ¼ extremely), with Cronbach's a ¼ 0.80. Participants also provided age, gender, and year in college. 4.4. Design and stimuli A between-subjects design with 4 conditions was employed: low-entertainment only, high-entertainment only, low þ high entertainment with low switching, and low þ high entertainment with high switching. Participants were assigned randomly to one of four conditions. The high-entertainment stimulus was a 10-min storyline from an episode of the comedy series The Big Bang Theory. In the storyline, using operant conditioning in lab rats as the scientific principle, Sheldon tries to influence Penny's behavior by offering chocolate as a reward. The Big Bang Theory was chosen because of its high-entertainment value and popularity among students. The low-entertainment stimulus was a published academic article on international journalism. The length of the article was 1719 words and participants were given 10 min to read the article. In the low-switching condition, the video and reading stimuli were divided into two parts and task-switching was enforced by having participants alternate between the video and text (see Fig. 1). Participants switched between the video and text three times in the low-switching condition. In the high-switching condition, the video and text were divided into four chunks and participants switched seven times between video and reading. 4.5. Procedure Upon arrival at the lab, participants were assigned to one of four conditions. After informed consent, they were asked to set aside all distractions, including watches and mobile phones. No clocks or time cues were visible in the room and the time on the video play bar also was hidden.
365
After filling out the pre-questionnaire, participants in the textonly condition read the printed text for 10 min and those in video-only condition watched the video for 10 min. Participants in the activity-switching conditions were forced to alternate between text and video for 20 min. During the reading activity, the lights in the room came on, a large screen in the front of the room turned to black and the participants were prompted to read. Before switching to video, the lights in the room dimmed until it was too dark to read, the screen turned bright and the video resumed playing. Lights were dimmed in the room to block participants from reading the article when the video was playing. In the low-switching condition, participants watched the video on the big screen for 5 min and switched to reading for 5 min and this cycle was repeated twice. In the high-switching condition, participants engaged in short durations (1.5e3 min, see Fig. 1) of watching video followed by reading and were treated to four cycles of video followed by reading. At the end, participants in all conditions filled out the post-experiment questionnaire. 5. Results 5.1. Preliminary analysis One-way ANOVA was conducted with four conditions (low entertainment, high entertainment, low- and high-entertainment with low switching, and low- and high-entertainment with high switching) and with enjoyment, arousal, time passage and time duration as separate dependent variables. While the overall model was significant for each of the four dependent variables, in each case the difference between the low (3 switches) and highswitching (7 switches) conditions was not significant. To simplify the analysis, data from the low- and high-switching conditions were combined into an overall switching condition and subsequent ANOVAs were conducted with three conditions (low entertainment, high entertainment, switching between low and high entertainment). 5.2. Enjoyment and arousal One-way ANOVA with the three conditions explained above was examined with enjoyment as the dependent variable and planned contrasts were used to test for differences between conditions (see Table 1 for summary of means). The overall model was significant, F(2, 102) ¼ 54.73, p < 0.001, and not surprisingly, the highentertainment activity of watching a sitcom was rated more enjoyable (M ¼ 5.81, SD ¼ 1.01) than the low-entertainment activity of reading a academic article (M ¼ 2.70, SD ¼ 1.21). Enjoyment in the switching condition (M ¼ 4.51, SD ¼ 0.98) was less than the enjoyment in high-entertainment condition, but more than the enjoyment in the low-entertainment condition. Enjoyment from the three conditions were significantly different from one another. The results for arousal were similar to the results for enjoyment. The overall model was significant, F(2, 102) ¼ 21.20, p < 0.001, and differences between all three conditions were significant. Arousal in the high-entertainment condition (M ¼ 4.42, SD ¼ 1.19) was higher than in the low-entertainment condition (M ¼ 2.49, SD ¼ 1.22). Arousal in the switching condition (M ¼ 3.56, SD ¼ 0.91) was between arousal in the high- and low-entertainment conditions. 5.3. Time passage Next a one-way ANOVA was examined with time passage as the dependent variable. The model was significant, F(2, 104) ¼ 13.39, p < 0.001, par h2 ¼ 0.21. Scheffe's post hoc contrasts confirmed H1
366
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
Time No switching low-entertainment condition Watching Video 10 min 10 sec
Time No switching high-entertainment condition 5m
5 m 10 s
Time Low + high entertainment with low-switching condition
3m
1m 45s
2m 46s
2m 39s
Time Low + high entertainment with high-switching condition Fig. 1. Stimuli for four conditions.
Table 1 Summary means (SD) for key variables by experimental conditions.
Enjoyment Arousal Time Passage Duration Estimates
No Switching Low Entertainment n ¼ 25
No Switching High Entertainment n ¼ 26
Switching Between Low and High Entertainment n ¼ 54
2.70 (1.21) 2.49 (1.22) 3.35 (2.06) 0.83b (0.29)
5.81 (1.01) 4.42 (1.19) 5.37a (1.39) 1.07b (0.59)
4.50 (0.98) 3.56 (0.91) 5.01a (1.11) 1.45 (0.80)
Note. Entries that share subscripts are not significantly different at p 05. Duration Estimates refers to the ratio of the estimated duration to the actual duration.
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
and H3. As predicted in H1, passage of time was faster in the highentertainment, sitcom condition (M ¼ 5.37, SD ¼ 1.39) than in the low-entertainment, academic article condition (M ¼ 3.46, SD ¼ 2.01), was significant at p < 0.001. Also as predicted in H3, passage of time was faster in switching condition (M ¼ 5.02, SD ¼ 1.11) compared to the low-entertainment condition, at p < 0.001. But the difference between the high-entertainment (M ¼ 5.37, SD ¼ 1.39) and switching condition (M ¼ 5.02, SD ¼ 1.11) was not significant. 5.4. Time duration To test H2 and H4, estimated duration was transformed by dividing estimated duration by actual duration of each condition: 10 min for the less entertaining condition, 10 min 10 s for the high entertaining condition, and 20 min 10 s for the switching condition. This transformation was necessary to account for the fact that the actual duration was about 20 min in the activity-switching condition, but only about 10 min in the other two conditions (Block & Zakay, 1997). Then the data were analyzed using single-factor, between-subjects ANOVA with three levels (low-entertainment, high-entertainment, switching between low- and highentertainment). Main effect for condition was significant, F(2, 99) ¼ 8.03, p ¼ 0.001, par h2 ¼ 0.14, and Scheffe's post hoc contrasts confirmed that the difference between the low-entertainment condition (M ¼ 0.83, SD ¼ 0.29) and the activity-switching condition (M ¼ 1.45, SD ¼ 0.80) was significant at p ¼ 0.001, and the difference between the high-entertainment condition (M ¼ 1.07, SD ¼ 0.13) and the activity-switching condition (M ¼ 1.45, SD ¼ 0.80) was tended toward significance at p ¼ 0.06. These two contrasts lend support for H4. However, the results did not support H2 because the difference between the low-entertainment and highentertainment condition was not significant. 5.5. Individual differences Two individual differences, frequency of activity-switching during studying and preference for activity-switching, were examined through bivariate Pearson correlations with estimates of time passage and time duration (see Table 2). Both frequency (r ¼ 0.23, p < 0.05) and preference for task-switching (r ¼ 0.18, p ¼ 0.06) were negatively correlated with time duration, but not with time passage. The two dimensions of time, duration and passage, were not correlated. In light of the significant correlation, follow-up analysis was conducted to examine the role of individual differences in estimation of time duration. When both switching habit and switching preference were entered as covariates in the time duration analysis, a main effect was observed for the switching habit, but not switching preference. But the main effect for condition remained significant and the findings reported for time duration remained intact. Next, to examine the potential effect of gender and age, a 2
Table 2 Bivariate correlations between time variables and individual differences in ActivitySwitching.
1. 2. 3. 4.
Time Passage Duration Estimates Activity-Switching Habits Activity-Switching Preference
Note. yp < 0.10, *p < 0.05.
1
2
3
0.05 0.07 0.10
0.23* 0.18y
0.14
367
(male, female) x 3 (conditions: low-entertainment, high-entertainment, switching between low and high) ANOVA was examined with age as a covariate. This analysis was carried out separately for time passage and time duration. In both instances, gender and age were not significant. 6. Discussion Although time is an important variable in media and communication research, the psychological mechanism underlying the estimation of time spent on media activities has received surprisingly little attention from researchers. Building upon an existing study on media multitasking and time passage (Chinchanachokchai et al., 2015), our study examined two measures associated with the psychology of time, time passage and time duration, with more naturalistic task stimuli, reading an academic article and watching a sitcom, and further investigated how task switching influenced time perceptions. Passage of time is the psychological perception on a continuum that extends from time standing still to time flying by. We predicted that time would fly by when watching an entertaining situational comedy such as Big Bang Theory, and that time would be perceived to stagnate when asked to reading a less entertaining academic article. In line with our predictions, students assigned to the sitcom reported that time flew by, whereas students in the academic article condition reported that time dragged on. Although the finding that time passes quickly when engaged in a more entertaining activity than a less entertaining activity is predictable, the perceived passage of time when the two activities were combined offers a plausible, functional explanation for multitasking. Switching from the less entertaining activity of reading an academic article to the more entertaining activity of watching a sitcom significantly sped up passage of time when compared to uninterrupted involvement in the less entertaining reading activity. Previous studies, based on self-reports, have identified relational and emotional benefits (Hwang et al., 2014; Wang & Tchernev, 2012) and alleviation of challenge or boredom of the work or study task (Yeykelis et al., 2014) as a motivation for multitasking. In this study, we demonstrate this finding through an experiment that demonstrates that activity switching not only makes the less entertaining activity significantly more entertaining, but also speeds up passage of time to the level of the more entertainment activity. This finding suggests that sporadic relief from demanding or mundane activities through bursts of distractions may explain the appeal of media multitasking. And the potential dip in task performance from switching or multitasking may be offset by the persistence in the combined activity. In other words, the perception of the rapid passage of time in multitasking contexts may prolong goal pursuit, which may offset potential erosion in performance. When time passes more quickly, an individual may perceive the activity as interesting and persist longer toward success in the primary activity (Gable & Poole, 2012), such as in completing homework that one dislikes. In short, the findings associated with passage suggest that the benefit of multitasking is not limited to entertainment, but extends to the perception of flight of time that may translate into prolonged engagement in required activities, even if they are challenging or boring. While the findings associated with time passage offer a timebased explanation for multitasking, the findings associated with time duration have significant implications for measurement of media use. A media experience that involved switching between less and more entertaining activities led to an overestimation of time. Time duration was overestimated during activity switching in comparison to estimation of time duration for single activities.
368
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369
When both the less and more entertaining conditions are combined, we predicted that the cognitive segmentation markers between switches would serve as memory cues that facilitate temporal summation, thus resulting in the estimates of longer duration in the switching condition compared to the single-task condition. This hypothesis was confirmed, which exposes the potential danger of relying on retrospective self-reports of time spent on multitasking media activities. Contrary to our prediction, duration estimation for the more interesting task was not greater than the duration estimation for the less entertaining task. The prediction, drawn from an earlier study, was based on the reasoning that more interesting content would leave a memory trace with stronger cues than less interesting content, which will improve recall, in turn increasing estimates of time duration (Wearden, 2005). We did not find support for this hypothesis. Results of this study suggest a potential further explication and examination on the memory-based theoretical model involving summation of temporal markers, i.e., the greater number of contextual cues offered by the experience, longer the retrospective duration is estimated (Block & Reed, 1978). However, contextual cues may have different dimensions and each dimension may influence duration estimation differently, for example, segmentation markers may have a stronger effect on duration estimation than the semantic cues. As demonstrated in this study, on one hand, switching between the high entertainment and low entertainment tasks offers segmentation markers. On the other hand, the high entertainment task, that is watching a sitcom, also offers memorable attributes in the storyline as semantic markers, such as the concrete events, likeable characters, memorable jokes and punch lines in a sitcom. Results from this study suggested that segmentation markers served as more memorable contextual cues than the semantic markers, as participants significantly overestimated the experience duration in the switching condition, but not in the high entertainment condition. Future study can further explore different dimensions of contextual cues and how they may impact duration estimation differently. A limitation of this study is the artificiality of the media experience in which participants were forced to switch between activities in the laboratory setting, which is unlike everyday media experiences that are more fluid. Further, participants in the study were college students, limiting the generalizability of results to the general population. Despite these limitations, the overestimation of time duration in activity-switching contexts deserves attention from media researchers and advertisers. When students report more than 40 h of activities in a 24-h day (David, et al., 2014), it is important to understand the underlying psychology of time estimation. The absence of a correlation between time passage and time duration should be noted because it supports findings in the literature that these assessments are tied to different psychological processes, with time passage tied to attention and time duration tied to memory. Also, the negative correlation between individual differences and time duration offers avenues for future research. Both task-switching preference and task-switching habits were negatively correlated with time duration, which can be interpreted as attenuation in overestimation among heavy multitaskers. With experience, it is likely that multitaskers may develop heuristics or cognitive strategies to correct for the additional segmentation markers created during activity-switching. It is also possible that as fluency of switching increases, segmentation markers may become less salient and less likely to be coded in memory. These competing explanations can be addressed in future research. In summary, the findings from this study offer new insight into the psychology of time perception that is critical to a better understanding of time
estimates of media multitasking. The attention and memory models associated with passage and duration respectively offer a platform for future extensions both on the underlying mechanisms of time perception during media experiences and on measurement of media use. References Block, R. A. (1992). Prospective and retrospective duration judgment: The role of information processing and memory. In Time, action and cognition (pp. 141e152). Springer Netherlands. https://doi.org/10.1007/978-94-017-3536-0. Block, R. A., & Reed, M. A. (1978). Remembered duration: Evidence for a contextualchange hypothesis. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 656. https://doi.org/10.1037//0278-7393.4.6.656. Block, R. A., & Zakay, D. (1997). Prospective and retrospective duration judgments: A meta-analytic review. Psychonomic Bulletin & Review, 4(2), 184e197. https:// doi.org/10.3758/BF03209393. Brown, S. W. (1997). Attentional resources in timing: Interference effects in concurrent temporal and nontemporal working memory tasks. Perception and Psychophysics, 59, 1118e1140. https://doi.org/10.3758/BF03205526. Carrier, L. M., Cheever, N. A., Rosen, L. D., Benitez, S., & Chang, J. (2009). Multitasking across generations: Multitasking choices and difficulty ratings in three generations of Americans. Computers in Human Behavior, 25(2), 483e489. https:// doi.org/10.1016/j.chb.2008.10.01. Chinchanachokchai, S., Duff, B. R. L., & Sar, S. (2015). The effect of multitasking on time perception, enjoyment, and ad evaluation. Computers in Human Behavior, 45, 185e191. https://doi.org/10.1016/j.chb.2014.11.087. Common Sense Census. (2015). The common sense Census: Media use by tweens and teens. Retrieved from: www.commonsensemedia.org/census. Csikszentmihalyi, M. (1990). Flow the psychology of optimal experience. New York: Harper-Perennial. Csikszentmihalyi, M. (2014). Flow and the foundations of positive psychology. Dordrecht: Springer. David, P., Brickman, J. S., Silva, D. E., Xu, S., & Ran, W. (2014). Beyond time and frequency: Attending to attention in media multitasking. International Communication Association 2014 Annual Convention in Seattle. David, P., Kim, J.-H., Brickman, J. S., Ran, W., & Curtis, C. M. (2014). Mobile phone distraction while studying. New Media & Society. https://doi.org/10.1177/ 1461444814531692. Gable, P. A., & Poole, B. D. (2012). Time flies when you're having approachmotivated fun effects of motivational intensity on time perception. Psychological Science. https://doi.org/10.1177/0956797611435817. Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1994). Growing up with television: The cultivation perspective. In J. Bryant, & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 17e41). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Hwang, Y., Kim, H., & Jeong, S.-H. (2014). Why do media users multitask? Motives for general, medium-specific, and content-specific types of multitasking. Computers in Human Behavior, 36, 542e548. https://doi.org/10.1016/ j.chb.2014.04.040. James, W. (1890). The principles of psychology (Vol. 2). MacMillan. Jeong, S. H., & Fishbein, M. (2007). Predictors of multitasking with media: Media factors and audience factors. Media Psychology, 10(3), 364e384. https://doi.org/ 10.1080/15213260701532948. Jeong, S.-H., & Hwang, Y. (2012). Does multitasking increase or decrease persuasion? Effects of multitasking on comprehension and counterarguing. Journal of Communication, 62(4), 571e587. https://doi.org/10.1111/j.14602466.2012.01659.x. Jeong, S.-H., & Hwang, Y. (2014). Multitasking and persuasion: The role of structural interference. Media Psychology, 1e24. https://doi.org/10.1080/ 15213269.2014.933114. Lang, A. (2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46e70. https://doi.org/10.1093/joc/50.1.46. Levin, H., & Zakay, D. (1989). Judging the duration of time intervals: A process of remembering segments of experience. Time and human cognition: A life-span perspective. Elsevier. Nakamura, J., & Csikszentmihalyi, M. (2002). The concept of flow. Handbook of Positive Psychology, 89e105. Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. Proceedings of the National Academy of Sciences, 106(37), 15583e15587. https:// doi.org/10.1073/pnas.0903620106. Ornstein, R. E. (1969). On the experience of time. Harmondsworth, UK: Penguin. Poposki, E. M., & Oswald, F. L. (2010). The multitasking preference inventory: Toward an improved measure of individual differences in polychronicity. Human Performance, 23(3), 247e264. https://doi.org/10.1080/08959285.2010.487843. Poynter, D. (1989). Judging the duration of time intervals: A process of remembering segments of experience. Advances in Psychology, 59, 305e331. https:// doi.org/10.1016/S0166-4115(08)61045-6. Ran, W., Yamamoto, M., & Xu, S. (2016). Media multitasking during political news consumption: A relationship with factual and subjective political knowledge. Computers in Human Behavior, 56, 352e359. https://doi.org/10.1016/ j.chb.2015.12.015.
S. Xu, P. David / Computers in Human Behavior 80 (2018) 362e369 Rosen, L. D., Mark Carrier, L., & Cheever, N. A. (2013). Facebook and texting made me do it: Media-induced task-switching while studying. Computers in Human Behavior, 29(3), 948e958. https://doi.org/10.1016/j.chb.2012.12.001. Sackett, A. M., Meyvis, T., Nelson, L. D., Converse, B. A., & Sackett, A. L. (2010). You’re having fun when time flies the hedonic consequences of subjective time progression. Psychological Science, 21(1), 111e117. https://doi.org/10.1177/ 0956797609354832. Salvucci, D. D., Taatgen, N. A., & Borst, J. P. (2009). Toward a unified theory of the multitasking continuum: From concurrent performance to task switching, interruption, and resumption. In Proceedings of the SIGCHI conference on human factors in computing systems (pp. 1819e1828). New York: ACM. Sanbonmatsu, D. M., Strayer, D. L., Medeiros-Ward, N., & Watson, J. M. (2013). Who multi-tasks and why? Multi-tasking ability, perceived multi-tasking ability, impulsivity, and sensation seeking. PloS One, 8(1), e54402. https://doi.org/ 10.1371/journal.pone.0054402. Srivastava, J. (2013). Media multitasking performance: Role of message relevance and formatting cues in online environments. Computers in Human Behavior, 29(3), 888e895. https://doi.org/10.1016/j.chb.2012.12.023. Sucala, M., Scheckner, B., & David, D. (2010). Psychological time: Interval length judgments and subjective passage of time judgments. Current Psychology Letters, 26(2), 2011. Retrieved from: https://cpl.revues.org/4998. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207e232. Watt, J. D. (1991). Effect of boredom proneness on time perception. Psychological
369
Reports, 69(1), 323e327. https://doi.org/10.2466/pr0.1991.69.1.323. 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), 493e513. https://doi.org/10.1111/j.14602466.2012.01641.x. Wang, Z., David, P., Srivastava, J., Powers, S., Brady, C., D’Angelo, J., & Moreland, J. (2012). Behavioral performance and visual attention in communication multitasking: A comparison between instant messaging and online voice chat. Computers in Human Behavior, 28(3), 968e975. https://doi.org/10.1016/ j.chb.2011.12.018. Wearden, J. (2005). Origins and development of internal clock theories of psychological time. Psychology Francaise, 50, 7e25. Xu, S., & Wang, Z. (2017). Does multitasking really exist? In P. Vorderer (Ed.), Permanently online - permanently connected (pp. 72e82). London: Routledge. Yeykelis, L., Cummings, J. J., & Reeves, B. (2014). Multitasking on a single device: Arousal and the frequency, anticipation, and prediction of switching between media content on a computer. Journal of Communication, 64(1), 167e192. https://doi.org/10.1111/jcom.12070. Zakay, D., & Block, R. A. (2004). Prospective and retrospective duration judgments: An executive-control perspective. Acta Neurobiologiae Experimentalis, 64(3), 319e328. https://doi.org/10.3758/BF03209393. Zhang, Y., Goonetilleke, R. S., Plocher, T., et al. (2005). Time-related behavior in multitasking situations. International Journal of Human-computer Studies, 62, 425e455. https://doi.org/10.1016/j.ijhcs.2005.01.002.