DIVIDED ATTENTION IN MULTITASKING WITH

2 downloads 0 Views 4MB Size Report
... while others believe that they are actually the same action on different scales. ... match the cards, but are usually not given instructions on how to match them (Berg, .... Pew Research Center Survey (2012) found that television watchers are ..... weight and dimensions of the device were not a factor, which would lead the ...
 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES        

1  

Divided Attention in Multitasking with Mobile Devices   Spencer Castro, Travis Seymour   University of California, Santa Cruz     Author  Note   Spencer  Castro,  University  of  California,  Santa  Cruz   Dr.  Travis  Seymour,  University  of  California,  Santa  Cruz   Approved:  ______________________________  Date:_________   Second  Reader:  Dr.  Nicolas  Davidenko   Approved:  ______________________________  Date:_________   By  signing  above  I  hereby  confirm  that  the  following  paper  represents  Master’s  level   work.   Correspondence  concerning  this  article  should  be  addressed  to  Spencer  Castro  at   the  Department  of  Psychology,  University  of  California,  Santa  Cruz,  1156  High   Street,  Santa  Cruz,  CA,  95064   Contact:  [email protected]              

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES         Abstract  

2  

Screen  displays  today  seem  to  be  ubiquitous,  varied  in  appearance,  and  deployed  in   many  situations.  As  a  result  they  potentially  affect  people’s  abilities  to  attend  to  and   perform  everyday  tasks.  It  is  important  to  discuss  how  multiple  screen  displays  af-­‐ fect  attention  as  they  become  more  commonplace.  In  applied  domains,  screen  dis-­‐ plays  affect  human  performance  at  work,  the  ability  to  attend  to  multiple  stimuli  op-­‐ timally,  and  distractedness  while  operating  vehicles.  Two  ways  in  which  displays   vary  are  their  size  and  mobility.  Previous  research  has  shown  that  people  are  in-­‐ creasingly  using  multiple  displays,  including  mobile  devices,  simultaneously  and   that  this  split  in  attention  has  detrimental  effects  on  goal-­‐directed  behavior.  Other   research  suggests  that  a  larger  display  improves  performance  over  a  smaller  display   when  showing  textual  information.  More  studies  show  that  holding  a  display  will   alter  attention,  which  is  unique  to  mobile  devices.  In  order  to  address  these  ques-­‐ tions  participants  were  asked  to  perform  a  task  with  a  mobile  device  and  a  station-­‐ ary  device.  On  a  mobile  device  (foreground)  of  either  4.8  or  10.1  inches,  participants   maintained  a  jittering  ball  within  a  circle.  On  a  19in.  computer  display  participants   vocally  responded  to  a  choice  task  (background  task).  Data  analysis  suggests  that   those  using  a  smaller  foreground  device  achieved  higher  performance  on  the  back-­‐ ground  task  than  those  using  a  larger  device,  while  having  similar  performance  in   the  foreground  task.  Future  studies  are  needed  to  determine  the  mechanism  of  this   difference  in  performance.   Keywords:  display,  size,  attention,  hands,  multitasking

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES         Divided Attention in Multitasking with Mobile Devices    

3  

A recent study by CourseSmart and Wakefield Research questioned 500 college

students about their studying habits and found that 73% reported using some type of technology (Kessler, 2011). In addition, 38% reported not being able to go more than 10 minutes without switching attention to their laptop, smartphone, tablet or e-reader. Linda Stone, a former executive of Microsoft called our fixation with devices, ‘‘continuous partial attention’’(Rose, 2011). Data support the prevalence of this state of continuous partial attention. Neilson Media Research (2014) reported that 40% of mobile device users indicated that they used their device while watching television daily. Continuous partial attention may refer to one of two types of scenarios defined within the psychology literature: task switching and simultaneous multitasking. There are several theories that purport a distinction between switching between separate tasks and simultaneously performing two tasks, while others believe that they are actually the same action on different scales.   Task Switching   The  term  task  switching  was  first  used  by  Jersild,  (1927)  and  has  since  been   based  upon  the  findings  within  attention  literature  that  define  attention  as  a  lim-­‐ ited-­‐capacity  system  (see  Broadbent,  1958;  Triesman,  1964;  Kahneman,  1973).   Some  of  the  tasks  most  common  to  task  switching  are  variations  of  the  Stroop  task   (Stroop,  1935)  and  the  Wisconsin  Card  Sorting  Test  (WCST)  (Berg,  1948).  Examples   of  the  Stroop  task,  (1935)  would  be  quickly  naming  the  ink  color  of  letters  that  spell   out  a  different  color.  A  variant  of  the  Stroop  task  is  a  motor  response  with  cursor   location  to  the  color  of  a  word  on  the  screen,  or  to  the  color  that  the  word  named  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 4         (Durgin,  2000).  The  Wisconsin  Card  Sorting  Test  involves  a  set  of  cards  that  vary   along  the  shape,  color,  and  number  of  objects  depicted.  Participants  are  asked  to   match  the  cards,  but  are  usually  not  given  instructions  on  how  to  match  them  (Berg,   1948).  By  switching  between  variations  of  Stroop  or  WCST  tasks,  one  can  study  the   effects  of  task  switching  upon  performance  of  the  tasks.  The  multitasking  literature   as  a  whole  tends  to  report  decrements  in  performance  in  multitasking  and  task   switching  as  similar.  Allport,  Styles,  and  Hiesh  (1994)  specifically  showed  perfor-­‐ mance  decrements  in  switching  between  multiple  tasks.  They  found  a  decrement  in   performance  due  to  switching  tasks  in  a  study  using  the  Reverse  Stroop  (Durgin,   2000)  and  standard  Stroop  (Stroop,  1935)  tasks.  A  group  of  participants  either  per-­‐ formed  those  tasks  in  repetitive  or  alternating  task  block,  and  those  in  the  alternat-­‐ ing  condition  had  slower  reaction  times  after  switching  tasks.  Rubinstein,  Meyer  &   Evans  (2001)  also  have  shown  that  switching  tasks  comes  at  an  attentional  cost.   They  found  that  there  could  be  performance  decrements  up  to  1.4  seconds  when   alternating  visuo-­‐motor  tasks  were  compared  to  repeating  variations  of  the  Wiscon-­‐ sin  Card  Sorting  Test.   Simultaneous Multitasking    

The  aforementioned  scenarios  (e.g.  using  a  cell  phone  and  watching  TV)  

could  actually  be  examples  of  Task  Switching  as  opposed  to  simultaneous  multitask-­‐ ing,  and  some  would  even  argue  that  simultaneous  multitasking  is  impossible  due  to   queuing  in  a  serial  representation  of  central  processing  (see  Pashler,  2000).  Howev-­‐ er,  others  argue  that  there  is  evidence  for  simultaneous  multitasking,  or  virtually   perfect  time-­‐sharing  (see  Schumacher,  Seymour,  Glass,  Fencsik,  Lauber,  Kieras  &  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 5         Meyer,  2001).    Both  of  these  explanations  support  similar  deficits  of  behavioral  per-­‐ formance  during  dual  tasking.  This phenomenon is known as dual task cost (DTC).   According to Kahneman (1973), human attention is inherently limited. Performing more than one task simultaneously often results in interference that degrades the performance of each. In one prevalent model of DTC, Pashler (1994) posits a bottleneck in processing information and subsequent action that inevitably leads to interference and performance decrements in goal-directed multiple-task behavior. However, attention is not consistently attenuated by two tasks; some tasks in combination show almost no performance decrement, while others create ample interference and lower performance (see Kahneman, 1973; Wickens, 1984). In contrast to Pashler’s (1994) proposal of a central cognitive decision-making resource that queues tasks, others argue that sufficient practice and motivation may reduce DTC in some cases (e.g., Meyer and Kieras, 1997a; 1997; 1999, Schumacher, Seymour, Glass, Fencsik, Lauber, Kieras, and Meyer, 2001). However, these papers focus upon very simple tasks to show that simultaneously performing two tasks is indeed possible, but it does not occur without motivation, training, or very simple cognitive tasks. When tasks are more complicated, there are multiple measures of bottlenecks, but the processing of these tasks can occur in parallel.  

At first these performance decrements may seem counterintuitive to the trends of

increasing monitors in the work place. Some businesses employ multiple monitors in order to display different sources of information, reasoning that having multiple monitors actually optimizes task completion and lessens cognitive load. According to Colvin, Tobler, & Anderson (2004), the main benefit of using multiple monitors is during taskswitching periods where a user must toggle a new window on a single screen. The re-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 6         searchers reported that across three types of editing tasks, when participants used multiple screens as opposed to a single screen they took less time, made fewer editing errors, and made more overall correct edits. They attributed the fact that people do not have to physically interact with the interface in order to visually assess information as a main factor in increased performance.    

There are other scenarios that may not see as much of a benefit as adding more

screens in the workplace. Many studies have focused upon the particular difficulty we have with simultaneously performing continuous visuo-motor activities. A salient example of this situation is while driving a motor vehicle. Research has shown that driving performance is significantly impaired by secondary tasks, with the most distracting common behavior being cell phone use (see Strayer & Johnston, 2001; Strayer, Watson, & Drews, 2011). Meanwhile car manufacturers are adding more and more complex “infotainment”  systems with touch-screen displays. For example, the Tesla Model S electric vehicle currently features a 17-inch touch-screen display along its center console. Considering that during any daylight hour 10% of drivers are using their cell phone, and car interiors are becoming increasingly complex, we may be creating increasingly unsafe cars that require more and more divisions of our attention (Glassbrenner, 2005; Cook and Jones, 2011).    

This problem becomes even more worrisome now that phones are capable of so

much more than just making calls. According to Cook and Jones (2011), 74.3% of young adults text and drive. Over half (51.8%) text weekly while some even engage in accessing the web (16.8%). Cook and Jones (2011) also found a positive correlation between young adults’  cell phone behaviors and crashes, as well as traffic citations. Handheld devices are

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 7         becoming ubiquitous due to their mobility and size, which inevitably has led to these situations with a mobile device and a secondary activity. People use their mobile phones to order something or look up a restaurant that they just saw on a television commercial. We text a friend updates on the game’s score or send pictures of news headlines from the T.V. (Neilson, 2014). In travel situations we look up destination addresses and then navigate within map applications. Why then, when people know that multitasking hurts performance, and can endanger their lives, do they continue to do it?   There are certain types of people who seem heavily prone to multitasking, and that these same people are the worst at simultaneously performing two tasks. Ophir, Nass, and Wagner, (2009) discovered that self-reported heavy media multi-taskers performed worse than light media multi-taskers in a laboratory test of multitasking. The researchers concluded that self-reported heavy media-multitaskers were worse at ignoring irrelevant information while attempting to detect changes in the orientation of a target bar amongst distractors.    

Even in high-stakes environments like driving, people seem unable to avoid mul-

titasking. However, there are situations like the workplace or at home, where dividing attention may be more acceptable. Again, there are more situations in which multitasking may occur due to technological availability. According to the Pew Research Center (2005), there was a 38% increase in the teens using the Internet from 2000 to 2005. In 2005, fourty-five percent of teens had a cell phone, and 33% had used it to send a text message. One in four cell phone-owning teens used their phone to connect to the Internet (Lenhart, Madden, & Hitlin 2005). Now, helped by the convenience and constant access provided by mobile devices, especially smartphones, 92% of teens report going online

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES       daily —  including 24% who say they go online “almost constantly,”  (Lenhart, 2015).

8  

There are now approximately 11 million teens who go online daily, as opposed to 7 million in 2000. This growth in mobile device use and Internet availability has increased the opportunity for the aforementioned multi-screen scenarios (Smith & Boyles, 2012). A Pew Research Center Survey (2012) found that television watchers are increasingly using secondary devices to react in real-time to things viewed during the course of television programs. For example, 38% of users said they used cellphones to keep occupied during commercial breaks. Similarly 28% posted a message to friends who were watching the same program. In the Ericsson Consumer Insight Summary Report (2013), 75% of people polled used their mobile device while watching TV. This type of situation has been increasing rapidly over the last 15 years and compels researchers to discern the implications of pervasive multitasking.    

The previous studies establish that multitasking with mobile devices is a prevalent

activity, and work to address the nuances of multitasking scenarios. This research suggests that splitting one task visually seems to increase performance, and to switch between tasks or simultaneously multitask is detrimental to performance. In this study, we are interested in the scenario of simultaneous multitasking; what are we aware of while simultaneously performing two tasks? This study attempts to apply behavioral methods to determine how performance changes within a growing ‘in-the-wild’  situation; using mobile devices while attending to a second screen.   Factors Affecting the Use of Handheld Devices in Multitasking   Recent research has shown that the size of the screen we are searching affects visual attention and the performance of that search. Chen, Liao, and Yeh (2011) used Anne

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES       Treisman’s (1980) Visual Search Paradigm to distinguish two ends of the spectrum in

9  

allocating visual attention to search a screen; parallel and serial search. They found that larger screen size affected reaction time in conditions where the target symbol was almost indistinguishable from the distractors. They defined this condition as serial search. However, screen size had no effect on searches for clearly distinct objects, such as a “v”   among “o”s. They defined this condition as parallel search. According to this study, larger screens would take longer to search when trying to find something similar to everything else on the screen.    

Another factor that could potentially affect attention to multiple screens is their

location relative to each other. Colvin, Tobler, & Anderson (2004) had multiple displays set side by side at the same distance from the user. Other studies only use one screen with multiple stimuli (Allport, Styles, and Hiesh (1994) ; Chen, Liao, and Yeh, 2011). However, the applied literature in aviation (Wickens and Andre, 1990) and driver distraction (Strayer & Johnston, 2001) addresses a Near-Field (foreground) and Far-Field (background) presentation of relevant stimuli. According to the work of Wickens and Andre (1990) and Wickens, Martin-Emerson, and Larish (1993), as relative distances of relevant information decreases, dual task cost decreases, but does not disappear. However, in both the aviation and driving literature, the secondary “screen”  is really the Far-Field window that allows visual information to update where the vehicle needs to be maneuvered. In aviation, the use of a Head Up Display (HUD) allows for a more explicit second screen with co-located information, but it is still overlaid upon physically distant information. Instead, our questions centered around ubiquitous everyday interactions with mobile de-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 10         vices, which allow for novel orientations, but do not tend to involve the distances between relevant information involved in driving and flying.   Experiment 1: Effect of Size in a Non-Coplanar, Multi-Screen Task    

The goal of Experiment 1 was to recreate a common scenario involved with mo-

bile devices, such as attending to the mobile device while having a computer or television directly behind it. Participants performed a continuous tracking task on a mobile device while simultaneously vocally responding to a four-alternative choice task on a background screen. We varied the size of the mobile device. Based upon the previous literature on multitasking, we expected participants to perform more poorly on the background choice task during dual-task blocks compared to single-task blocks. We also predicted that participants would perform more poorly on the mobile device task during dual-task blocks compared to single-task blocks. In our study, the relevant information of the mobile device was held in a fixed location and the distance of information of the background task varied in height and lateral distance from the mobile device. Based upon previous work, (Wickens & Andre, 1990; Wickens, Martin-Emerson, & Larish, 1993) we predicted that reaction times to a change further from the mobile device would be slower. In addition, because Chen, Liao, and Yeh (2011) found no difference for the effect of screen size on reaction times in parallel search, but Bridgeman et al. (2003) and Bruijn, et al. (1992) found performance improvements with larger screens in tasks similar to the current experiment. Therefore, we predicted improved performance for the tablet over the phone.   One possible consequence of a study with different tasks is unequal allocation of attention to one of the tasks. If we think about the example of using a mobile device

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 11         while watching TV, then the relevance of replying to a friend might override watching a rerun of Seinfeld. However, if it’s the first time seeing a highly anticipated show or movie during a climactic point in the plot, then replying to the friend might have to wait. In our task, participants were not giving instruction as to a primary task and a secondary task and were encouraged through the instructions to attempt to do both tasks at the same time. There are also lower level considerations that influence attention across multiple screens. Mazza, Turatto, & Umilta, (2005) have shown that in images people are much quicker to notice small changes in the foreground and slower to notice large changes (such as the color) to the background. This research may point to the background task suffering the most in this foreground-background arrangement. Therefore, we predict that the effects of dual task cost will occur more in background tasks and the continuous foreground task will be prioritized.   Method   Participants    

46 undergraduates at the University of California, Santa Cruz (13 Males, 33 Fe-

males) enrolled in a psychology course, received course credit in exchange for their participation. Three participants were removed from the analysis due to technical malfunctions and resulting loss of data.   Materials For All Experiments   For the background task, a 19-inch Dell computer monitor (1280 x 1024 pixels) was positioned at approximately 36 inches from the participant. In the Foreground condition, participants used either a 5.4 x 2.8 inch (720 x 1280 pixels) Samsung Galaxy SIII smart phone or a 9.6 x 6.9 inch (1280 x 800 pixels) Samsung Galaxy Tab 3 tablet that

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 12         was held approximately 30 inches from the participant. The center point (target location) on both devices was held constant approximately 5 inches below the bottom-center of the Dell computer monitor. Vocal responses were recorded using a Psychology Software Tools Serial Response Box™ Voice Key with an Audio Technica® ATR20 microphone mounted on a 5 inch table stand. Background task stimuli were presented electronically using the E-Prime 2.0 software (Psychology Software Tools, Pittsburgh, PA), which also recorded reaction time and accuracy for each response. Foreground task stimuli were presented by a custom android application.   Procedure   Participants were randomly assigned to either use the phone or the tablet for the duration of the experiment. The experiment was conducted across 5 blocks: a single-task mobile device block (Foreground) for approximately 5 minutes, followed by a single-task choice task block (Background) for approximately 5 minutes. Afterwards, participants completed 3 dual task blocks, each lasting approximately 5 min (see Table 1). After all blocks were completed, participants completed a survey that included prior experience with multitasking, a self-efficacy measure, and technology use (see Appendix B).   Before starting the first single-task block, participants were handed the mobile device and asked to first pay close attention to instructions describing the mobile device task, and then to complete this task as accurately as possible. Next, participants received detailed instructions about the choice task (Background). The mobile device was taken from the participant during this single-task block. The participants completed the choice task in approximately 5 minutes. After the single task blocks participants were asked to complete three blocks in which they were to perform both tasks simultaneously to max-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 13         imize their performance in terms of accuracy, reaction time, and error score. A written version of the instructions are included in Appendix A.   A post-survey was administered to assess previous experience with technology, confidence (Appendix B). We used the General Self-Efficacy Questionnaire, adapted to compare confidence in abilities with confidence in multi-tasking (Chen, Gully, & Eden, 2001).   Tasks. In the Foreground task, participants controlled the movements of a blue ball by physically tilting the mobile device. Participants were asked to keep the ball within the center of a target circle. A weighted jitter, approximately the distance of the radius of the target circle, was applied to the ball. If the blue ball was allowed to slide outside of the circle, a beep would sound at 1-second intervals until the ball was back in the target circle. A numerical error score accumulated in the top left corner of the device. The Foreground task interface is shown in Figure 1.   In the Background task, participants were asked to vocally respond to the position of a change on a computer monitor. The computer would present one row at four possible heights to the participant. After a randomized interval between 3 and 10 seconds, a change would occur at one of four possible counterbalanced locations within that row. For the background there was a change in the direction of an arrow. One arrow out of four possible would switch direction from the previously uniform arrow set. This change was characterized as a discrete change.   On the majority of blocks, participants were asked to complete both tasks simultaneously. No ranking of importance was specified for the goals. Participants attempted

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 14         to keep the blue ball within the circle while responding as quickly and accurately as possible to the position of the background stimulus change.   Measures. For the mobile device task, we measured the accumulated error score across each block. Each participant received an error score for single task blocks and dual task blocks. For the background task, we measured reaction time (RT) and accuracy for each question. Each participant received an average reaction time for single task blocks and dual task blocks. Accuracy in the background task was not significantly different across groups or substantially different from 100% for any participant. Therefore, it was excluded as a measure. Finally, we subtracted performance in the single task conditions from performance in dual task conditions to assess Dual Task Cost (DTC). Statistical analyses were performed using an independent samples t-test to assess differences between phone and tablet conditions.   Results   Single Task Trials   An independent-samples t-test was conducted to compare mobile device error, and background reaction time in phone and tablet conditions. In single task trials mobile device error score and background choice task reaction time were calculated and compared, but no statistical differences were observed between phone and tablet conditions, t(42) = 1.06, p = .28; t(42) = 1.39, p = .17 (see Table 3).   Dual Task Trials   Mobile device error per block and mean reaction times per block were measured for dual task trials (see Table 3). An independent-samples t-test was conducted to compare mobile device error and background reaction time in phone and tablet conditions.

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 15         There was a not a significant difference in the mobile device error scores for phone (M = 39.72, SD = 35.66) and tablet (M = 50.37, SD = 32.7) conditions; t(42) = 1.04, p = .3. However, there was a significant difference in the background choice task reaction times (milliseconds) for phone (M = 896.18, SD = 177.54) and tablet (M = 1106.85, SD = 264.84) conditions; t(42) = 3.15, p = .003; 95% CI[75.21, 346.14].   Dual Task Cost   Dual Task Cost was calculated by subtracting the average single task trial scores from the average dual task trial scores for each participant. There was a not a significant difference in the mobile device error scores for phone (M = -.23, SD = 25.06) and tablet (M = 1.83, SD = 32.92) conditions; t(42) = .18, p = .8. However, there was a significant difference in the background choice task reaction times (milliseconds) for phone (M = 111.09, SD = 266.54) and tablet (M = 283, SD = 310.09) conditions; t(42) = 2.00, p = .026; 95% CI[27.31, 345.71] (see Table 3).   Discussion    

In Experiment 1 we conducted analyses for assessing factors of performance in a

background/foreground context with a mobile device and a free-standing screen. From the previous literature we have established that there is a DTC associated with multitasking and task switching (Pashler, 1994; Rubinstein, Meyer, & Evans, 2001), and have built on that research to show how design factors manipulate this cost. We have also shown that multiple displays are being used simultaneously and in a novel configuration with increasing frequency (Ericsson Consumer Lab, 2013) and that this paradigm should be studied further (see also Colvin, Tobler, & Anderson, 2004; Cook and Jones, 2011; Strayer, Watson, & Drews, 2011). Finally, we have identified a few factors that potential-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 16         ly influence performance in this new context, including screen size and hands in the visual field (see Bruijn et al., 1992; Reed et al., 2006). From this foundation we conducted a study to assess how size of the device in the visual field influences performance across both devices, biases visual attention toward one device over the other in this foreground/background configuration, and how these factors influence each other. Although we predicted that a larger screen size would increase performance of the tasks overall, we found that performance for the phone condition did not show as much of a detriment in the background choice task. We found that the slower reaction times in the tablet condition apparent in the background task speak to some performance detriment inherent in the comparison of the two devices. This could be an interesting finding regarding screens in that previous research often involved reading and finding information or errors in walls of text on coplanar screens. This finding suggests that the effect of screen size on performance may be tied to perceiving, processing, and responding to information in the foreground task differently, leading to an observable decrement in performance on the background task.   However, in this analysis we identified a few possible improved measurements. One result that seemed apparent is that there was a practice effect for the Dual Task Cost measures. Having only one single task practice block for the DTC measure meant that participants performed more poorly at the beginning of the study, and therefore showed negative dual task cost in some conditions. This confound was corrected by having sufficient practice for the tasks and interleaving the blocks in Experiment 2.   Experiment 2: Influence of Non-Attentional Size and Dimensions of Mobile Device  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 17           In the proposed study we have created a design that allows for practice to be accounted for in the data by counterbalancing the order of single and dual task blocks, thereby giving a more accurate assessment of the measures of dual task cost than Experiment 1 (see Table 2). In addition, more trials per participant and more total participants than Experiment 1 allowed for more reliable comparisons between phone and tablet conditions.    

It is also possible that the various aspects of holding a larger compared to a small-

er device could have non-attentional effects. In order to account for possible nonattentional effects a third device was created with the screen size of the phone and the weight and dimensions of the tablet. We referred to this device as the Phony Tablet (PT). If the confound of weight and dimension are eliminated and the effects are attentional, previous literature on attention may be updated to account for this new context. In accordance with the previous literature on screen size and attention (see Bruijn, Mul, and Oostendorp, 1992; Bridgeman, Lennon, and Jackenthal, 2003), we predicted that the weight and dimensions of the device were not a factor, which would lead the phony tablet to perform similarly to the phone and differently from the tablet. Also, in past research with multiple mounted displays, researchers have not accounted for the possible effect of physically holding a device on a person’s attention. However, there is a plethora of research demonstrating that hands within our visual field affect performance in reaction time tasks (see Reed, Grubb, & Steele, 2006; Dufour & Touzalin, 2008; Tseng, Bridgeman, and Juan, 2012; Qian, Al-Aidroos, West, Abrams, & Pratt, 2012). Hand distance from target stimuli varies by the nature of differing sizes between devices in the current study. Therefore, we must attempt to account for its effect upon performance in our non-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 18         coplanar multi-screen scenario. The visibility of the hands within the current study is manipulated with the addition of the third condition that holds the hands the same distance from target information as the tablet, but mirrors the screen size of the phone. Based upon the aforementioned hand-related literature, we would predict that hands closer to target information in the visual field will facilitate attention and lead to better performance than the far-hands condition.   Method   Participants    

123 undergraduates at the University of California, Santa Cruz (36 Males, 87 Fe-

males) enrolled in a psychology course, received course credit in exchange for their participation. The experimental design consisted of a comparison of means in phone, tablet, and phony tablet conditions. Three participants were removed from the analysis due to technical malfunctions and resulting loss of data.   Materials   The technology in Experiment 1 was used for the current study (see Experiment 1). However, additional information was taken into account for the weight and dimensions of the technology. For example, the 5.4 x 2.8 inch (720 x 1280 pixels) Samsung Galaxy SIII smart phone weighs approximately 4.7 ounces while the 9.6 x 6.9 inch (1280 x 800 pixels) Samsung Galaxy Tab 3 tablet weighs approximately 18 ounces. A 9.6 x 6.9 inch press board with a 5.4 x 2.8 inch cutout was built to accommodate the smart phone and added together weigh a total of 18 ounces.   Procedure  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 19         The study consisted of three between-subjects conditions: Phone, Tablet, and Phony Tablet (PT). The experiment was conducted across 8 blocks: a single-task mobile device block (Foreground) for approximately 10 minutes, followed by a single-task choice task block (Background) for approximately 10 minutes. Afterwards, participants completed 2 dual task blocks, each lasting approximately 10 min (see Table 2). These blocks were then repeated, with a single task mobile, background task choice task, and two dual task blocks. After all blocks were completed, participants completed the same survey as Experiment 1 (see Appendix B).   Before starting the first single-task block, participants were handed the mobile device and asked to first pay close attention to instructions describing the mobile device task, and then to complete this task as accurately as possible. Next, participants received detailed instructions about the choice task (Background). The mobile device was taken from the participant during this single-task block. The participants used the choice task for approximately 10 minutes. After the single task blocks participants were asked to complete two blocks in which they were to perform both tasks simultaneously to maximize their performance in terms of accuracy, reaction time, and error score. A written version of the instructions provided to participants are included in Appendix A.   The same post-survey from Experiment 1 was administered to assess selfefficacy, specifically with regard to multitasking, familiarity with technology, and video game use (Chen, Gully, & Eden, 2001).   Tasks. The tasks in Experiment 1 were used for the current study (see Experiment 1).  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 20         Measures. For the Background task, we measured reaction time (RT) and accuracy for each question. For the mobile device task, we measured the accumulated error score across each block.   Results   Single Task Trials    

A one-way analysis of variance was conducted in order to compare mobile device

error and background reaction time in phone, tablet, and phony tablet conditions. In single task trials mobile device error score and background choice task reaction time were calculated, but no statistical differences were observed between phone and tablet conditions, t(120) = 1.07, p = .29; t(120) = 1.38, p = .19 (see Figure 2).   Dual Task Trials    

Mobile device error per block and mean reaction times per block were measured

for dual task trials. A one-way between subjects ANOVA was conducted to compare the effect of device type on mobile error and background reaction time in the phone, tablet, and PT conditions. No main effect of device was found for mobile device error score F(2, 2

120) = 0.64, p = .53, ηp = .01. However, there was a significant effect of reaction time (milliseconds) on device type at the p < .05 level for the three conditions, F(2, 120) = 2

6.38, p < .001, ηp = .10 (see Figure 3). Post Hoc pairwise comparisons using the Tukey Honest Significant Differences test indicated that the mean score for the phone condition (M = 970.30, SD = 217.74) was significantly different from the tablet condition (M = 1102.87, SD = 265.28), but not the PT condition (M = 929.27, SD = 200.82). However, PT was significantly different from the tablet condition. Taken together, these results

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 21         suggest that the device with the weight and dimensions of the tablet, but the screen size of the phone (PT) performed similarly to the phone and differently than the tablet.   Dual  Task  Cost   Dual Task Cost was calculated by subtracting the average single task trial scores from the average dual task trial scores for each participant. A one-way between subjects ANOVA was conducted to compare the effect of device type on mobile error and background reaction time in the phone, tablet, and PT conditions (see Figure 4). There was not a significant difference in the mobile device error scores for phone, tablet, or PT, 2

conditions, F(2, 120) = 1.17, p = .31, ηp = .02. There was a significant effect of device type on background reaction time (milliseconds) at the p < .05 level for the three condi2

tions, F(2, 120) = 5.71, p < .01, ηp = .09. Post Hoc pairwise comparisons using the Tukey Honest Significant Differences test indicated that the mean reaction time (milliseconds) for the phone condition (M = 124.49, SD = 212.03) was significantly different from the tablet condition (M = 271.43, SD = 266.55), but not the PT condition (M = 125.51, SD = 192.18). The PT condition was also significantly different from the tablet condition, suggesting that the phone and PT conditions reacted more quickly to background stimuli than the tablet condition. These results suggest that the tablet condition had a higher DTC than the phone or PT conditions.   Discussion   Experiment 2 showed that when accounting for practice, DTC occurs as expected for a multitasking experiment, replicating previous DTC studies (e.g. Pashler, 1994). This study also shows that as people attend to a foreground screen of various sizes, how they attend to that foreground screen changes. With regard to the size and location of the ef-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 22         fect, it appears that it occurs for information beyond the foreground screen. It also seems that larger screens, although possibly priming more attentional focus to that location, do not necessarily enhance performance for that task, and cause detriments to attention allocated to background tasks.    

Also in Experiment 2, it is clear that the Phony Tablet (PT) acts more like a phone

to participants than it does a tablet. There are a few theories that address this outcome. First, it could have to do with the type of attention being deployed. From the three main types of visual attention listed by Carrasco (2011), our task requires overt and/or covert spatial attention, and could be deployed as feature-based attention (FBA) or object-based attention. We have no measures, such as eye tracking, to discern between overt and covert attention, as we were interested in change in performance, and not individual differences in strategy. However, we can still reach some conclusions about the nature of attentional resources being allocated to each of the two tasks (foreground and background). It first requires that we view the brain as a limited capacity system and attention as a selective process from a finite set of options. If there is a fixed amount of overall energy available for the brain, and this metabolic energy is thought of as constant, then the firing rates of neurons responding to visual information quickly saturate this capacity (Attwell & Laughlin, 2001). This limited resource view is fairly well established by the previous research (see Broadbent, 1958; Treisman, 1960). Secondly, simultaneous stimuli within receptive fields interact in a mutually competitive way, also known as lateral inhibition. There is neural research to show that responses to simultaneous stimuli are a weighted average of an individual response within a receptive field (see Reynolds, Chelazzi, & Desimone, 1999). The Eriksen Flanker Task shows us that there is a minimum size our

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 23         attentional field might be (Eriksen & Eriksen, 1974), and Posner, (1980) has shown how we might voluntarily adjust the attended region independently from our focal point. However, Eriksen (1974) and subsequent work (see Faber, Maurits, & Lorist, 2012) based upon the Eriksen Flanker task show how mental fatigue lowers our ability to suppress irrelevant stimuli. The findings most relevant to our current manipulation are that attention distributed over a larger region of the visual field has a lower spatial resolution and less processing efficiency for a given subregion (Castiello & Umilta, 1990, 1992; Eriksen, 1990 in Carrasco, 2011). For a Post Hoc hypothesis, it would make sense now that the larger screen, which would require more attentional resources and activate a larger area of the visual cortex, would leave fewer resources under a cognitive workload for subsequent change detection in the background choice task. So in our original hypothesis we predicted that a larger screen would lead to a more diffuse attentional field, but we did not allow for the trade-off between size, spatial resolution, and loss of processing efficiency. This led us to wrongly hypothesize that the allocation of attention to the background task would be greater overall than if it was completely outside of the attentional field with the smaller screen. It seems that the smaller screen allowed for the capacity of greater change detection in the periphery. For neural evidence of this phenomenon, Muller, Bartlet, Donner, Villringer, and Brandt (2003) showed that as attention is distributed more widely, the amount of activated retinotopic visual cortex increases, but the relative level of any subregion of activity decreases when compared to a less distributed attentional field.   It is possible that there are other aspects of this task that modulate attention to the devices differently. In terms of the hands’ effect on visual attention, there are recent stud-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 24         ies showing that hands can facilitate attentional focus near the palms. Supporting data shows that the effect disappears when monitor adjacent hands are arranged with outwardfacing palms (Tseng, Bridgeman, & Juan, 2012). Also the grasp position similarly modulates this effect. In Experiment 2 we can ask if this is meaningfully different in two screen displays. Thomas, (2012), shows that having the hands in a pincer grasp position with the thumb and forefinger resting on the display will nullify the effect of responding faster to targets nearest to the hand. Participants responded to stimuli on a device screen while holding their hands in adjacent positions with different hand shapes. This suggests that how you hold the mobile device could modulate your tendency to pay attention to it. These results also demonstrate that changes in visual processing near the hands rely on the hands’ posture. Although hands positioned to afford power grasps facilitate a faster reaction time, a pincer grasp posture that affords more precise action does not. In addition, having an expensive and complex piece of technology and paying attention to it may have to do with not dropping it, the device being close to you, or it getting your attention through sound and haptic feedback. The current study allows us to suggest that hand distance from relevant stimuli may not have as noticeable of an effect on performance as screen size with our task.   Limitations and Future Directions    

There are many questions still to be asked of this novel ‘in-the-wild’ scenario. For

instance, stimuli cannot be too easy or too difficult in order to maintain a desired multitasking environment. Tasks that people perform on their mobile devices now vary widely, from small updates or alarms to in-depth conversations via text, email, or voice command. Also, the weight and dimensions of the devices still have some relevance with re-

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 25         gard to our hand comparisons. In future iterations of this study, we may want to focus upon the difference between two continuous tasks running simultaneously instead of the current continuous Foreground and discrete Background. This proposed change should lessen the strategic variability across the two tasks and make it easier to discern differences across conditions.   In future research within the context of Foreground mobile devices and Background monitors we would like to look at individual differences in strategy as well. Effects may be obscured by a large difference in performance strategies of participants. They may attempt to switch tasks (Allport, Styles, & Hsieh, 1994; Rubinstein, Meyer, & Evans, 2001), attend to both tasks simultaneously (Schumacher et al., 2001) or ignore one task completely due to engagement differences in the two tasks. Further analyses could compare participants who performed well in one task against participants who performed well in the other, and then overall performance across these two groups. We could correlate this type of analysis in future studies with survey questions about participants’ strategy. Research also suggests that there may be an age preference for mobile device screen size. There was an interesting preference for types of mobile devices between age groups. Nearly half of people 18-24 years of age used their phone while watching television daily, while people 55-64 years of age were most likely to use tablets during daily television viewings (Neilson Media Research, 2014). However, our sample only contained the former group, and age did not correlate with performance on phone or tablet. This could potentially lead to a familiarity bias for the phone against the tablet, but post-survey results indicate no correlation in device familiarity with performance.   Conclusions  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 26         In these studies we’ve seen evidence that suggests that when holding a mobile device, larger screens may decrease our abilities to attend to information beyond the device. In this experiment, it seems as if UCSC undergraduate students can attend to background information on a computer with very minimal performance loss to foreground or background tasks if the foreground screen is small. However, once the device is changed to the larger tablet, dual task cost and reaction times for the background task both increase. This finding could be attributed to familiarity with the phone over the tablet in young people (see Neilson, 2014), but evidence suggests that the more visual area taken up activates more cortical neurons, which would leave fewer resources for processing other information. It is possible that eliminating this configuration of multiple displays from one’s routine at work, in the car, or at home could become a proven method for individuals to bolster their efficiency and performance at certain tasks. However, due to the previously shown effects of training and DTC (see Schumacher et al. 2001), we may only need time to adjust to this new configuration of displays developing all around us. Even now participants have shown that the device they are more likely to be familiar with does not always show a reliable dual task cost for simultaneously attending to background information. Further study is required to answer these questions.

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES      

27  

References   Allport, D. A., Styles, E. A., & Hsieh, S. (1994). Shifting intentional set: Exploring the dynamic control of tasks. Attention and performance 15: Conscious and nonconscious information processing. Attention and performance series, (421-452). Cambridge, MA, US: The MIT Press, xix, 945 pp. Berg, E. A. (1948). A simple objective technique for measuring flexibility in thinking. Journal of General Psychology, 39, 15-22. Braune, R., & Wickens, C. D. (1986). Time sharing revisited: Test of a componential model for the assessment of individual differences. Ergonomics, 29, 1399– 1414. Brujin, D. D., Mul, S. D., & Oostendorp, H. V. (1992). The influence of screen size and text layout on the study of text. Behaviour & Information Technology, 11(2), 71-78. Carrasco, M. (2011). Visual attention: The past 25 years. Vision research, 51(13), 14841525. Castiello, U., & Umiltà, C. (1990). Size of the attentional focus and efficiency of processing. Acta psychologica, 73(3), 195-209. Chen, I. P., Liao, C. N., & Yeh, S. H. (2011). Effect of display size on visual attention. Perceptual and motor skills, 112(3), 959-974   Colvin, J. Tobler, N. & Anderson, J.. (2004). Productivity and Multi-Screen Computer Displays. The Rocky Mountain Communication Review 2(1) 31-53.   Cook, J. L., & Jones, R. M. (2011). Texting and accessing the web while driving: Traffic citations and crashes among young adult drivers. Traffic injury prevention, 12(6), 545-549.   Dufour, A., & Touzalin, P. (2008). Improved visual sensitivity in the perihand space. Experimental Brain Research, 190(1), 91-98. Ericsson  ConsumerLab  (2013).  TV  and  media-­‐identifying  the  needs  of  tomorrow’s   video  consumers.  An  Ericsson  Consumer  Insight  Summary  Report.  Retrieved April 17, 2015, from http://www.ericsson.com/res/docs/2013/consumerlab/tv-and-mediaconsumerlab2013.pdf   Eriksen, B.A., & Eriksen, C.W. (1974). Effects of noise letters upon the identification of a target in a nonsearch task. Perception & Psychophysics, 16, 143–149.   Faber L.G., Maurits N.M., Lorist M.M. (2012). Mental fatigue affects visual selective attention. PLoS ONE 7(10): e48073. doi:10.1371/journal.pone.0048073   Glassbrenner, D. (2005). Traffic safety facts research note: Driver cell phone use in 2005 - Overall results. (DOT HS 809 967). Washington, DC: National Center for Statistics and Analysis, National Highway Traffic Safety Administration. Jersild,  A.T.  (1927).  Mental  set  and  shift.  Archives  of  Psychology  (Whole  No.  89,  pp.  5– 82).   Kaiser Family Foundation. (2010). Generation M2: Media in the Lives of 8 to 18 year olds. Program for the study of Media and Health.  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 28         Kessler, S. (2011). 38% of college students can’t go 10 minutes without tech [STATS].     Mashable Tech. . Lenhart, A., Madden, M., & Hitlin, P. (2005). Teens and technology: Youth are leading the transition to a fully wired and mobile nation. Retrieved May 7, 2015, from www.perwinternet.org Lenhart, A. (2015). Teens, social media & technology overview 2015. Retrieved May 7, 2015, from www.perwinternet.org Link, M. (2010). Out of Home Television and Other Video Viewing Behaviors of U.S. Adults. Results from the Council for Research Excellence,Video Consumer Mapping Study.   McCrae, R. & John, O. (1992). An Introduction to the five-factor model and its applications. Journal of Personality and Social Psychology, 83, 1456-1468. Muller, N., Bartelt, O., Donner, T., Villringer, A., & Brandt, S. (2003, May). A physiological correlate of the "Zoom Lens" of visual attention. The Journal of Neuroscience, 23(9), 3561-5. National Highway Traffic Safety Administration (2014). 2012 'Overview' Traffic Safety Fact Sheet. (DOT HS 812 016). Washington, DC: National Center for Statistics and Analysis. Ophir, E., Nass, C., & Wagner, A. D. (2009). Cognitive control in media multitaskers. Proceedings of the National Academy of Sciences, 106(37), 1558315587.   Pashler, H. (1994). Dual-task Interference in simple tasks: data and theory. Psychology     Bulletin,116:220–244.     Pashler, H. (2000). in Attention and Performance XVIII: Control of Mental Processes, eds Monsell S, Driver J (MIT, Cambridge, MA), pp 277–309.   Qian, C., Al-Aidroos, N., West, G., Abrams, R. A., & Pratt, J. (2012). The visual P2 is attenuated for attended objects near the hands. Cognitive neuroscience, 3(2), 98-104. Reynolds,  J.  H.,  Chelazzi,  L.,  &  Desimone,  R.  (1999).  Competitive  mechanisms  sub-­‐ serve  attention  in  macaque  areas  V2  and  V4.  The  Journal  of  Neurosci-­‐ ence,  19(5),  1736-­‐1753.   Rose, E. (2011). Continuous partial attention teaching and learning in the age of   interruption, Antistasis, 17-19.   Rubinstein, J., Meyer. D. & Evans, J. (2001). Executive Control of Cognitive Processes in Task Switching. Journal of Experimental Psychology: Human Perception and Performance, 27(4), 763-797.   Schumacher, E. H., Seymour, T. L., Glass, J. M., Fencsik, D. E., Lauber, E. J., Kieras, D.   E., & Meyer, D. E. (2001). Virtually perfect time sharing in dual-task per     formance: Uncorking the central cognitive bottleneck. Psychological Sci     ence, 12, 101–108. Smith,  A.,  &  Boyles,  J.  L.  (2012).  The  rise  of  the  “connected  viewer”.  Pew  Internet  &   American  Life  Project.   Strayer, D.L., & Johnston, W.A. (2001). Driven to distraction: Dual-task studies of simulated driving and conversing on a cellular phone. Psychological     Science, 12, 462–466). doi: 10.1111/1467-9280.00386  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 29         Strayer, D. L., Watson, J. M., & Drews, F. A. (2011). 2 Cognitive Distraction While Multitasking in the Automobile. Psychology of Learning and MotivationAdvances in Research and Theory, 54, 29.   Thomas, L.E. (2013) Grasp posture modulates attentional prioritization of space near the hands. Front. Psychol. 4(312).   Treisman, A. M., Gelade, G. (1980). A feature-integration theory of attention. Cognitive     psychology 12, 97–136. doi:10.1016/0010-0285(80)90005-5   Tseng, P., Bridgeman, B. & Juan. (2012). Take the matter into your own hands: A brief review of the effect of nearby-hands on visual processing. Vision Research, 72, 74-77.      

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES       Tables and Figures  

30  

Table 1. Experiment 1 Procedure Design  

    Table 2. Experiment 2 Procedure Design

      Table  3.  Experiment  1  Results     Error (score)  

 

Mean  

SD  

Single Phone  

39  

37  

Single Tablet  

52  

32  

Dual Phone  

39  

35  

Dual Tablet  

50  

32  

DTC Phone  

-0.2  

25  

DTC Tablet  

-1.8  

32  

 

Reaction Time (ms)   p   .28  

.30  

.80  

Mean  

SD  

785  

285  

823  

192  

896  

177  

1106  

264  

111  

266  

283  

310  

p   .17  

.003**  

.02*  

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES      

31  

 

  Figure  1.  Mobile  device  foreground  and  computer  display  background  task  example.  

Figure  2.  Experiment  2  single  task  mobile  device  error  score  and  background  reac-­‐ tion  time.      

 

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES        

32  

  Figure  3.  Experiment  2  dual  task  mobile  device  error  score  and  background  reaction   time.          

Figure  4.  Experiment 2 dual task cost mobile device error score and background reaction time.            

 

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES           Appendix A   1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

17.

33  

Welcome to the Mobile Multitasking Game! Thank you for participating in this study. Please read the instruction screen carefully, and then tell the researcher you are ready to move forward. In this task you will be asked to use a mobile device and make decisions about information on a computer screen simultaneously. The mobile device must be held in both hands with arms resting upon the counter. On the mobile device there is a game which consists of keeping a ball in the center of a circle. If the ball is inside of the circle the mobile device will not beep. If the ball is outside of the circle the mobile device will beep until the ball is back in the circle. You can see the number of times the ball has left the circle in the top left corner of the screen. Let’s practice the mobile task now. The experimenter will hand you the mobile device and position your arms on the table. Remember that in order to keep the ball in the center of the circle, you must keep the device at the correct angle. When you are ready, press START on the mobile device Now let’s look at the background task. You are going to be making decisions about arrows. In this task you will be asked to respond to the location of the changed arrow. First, you will see a fixation cross. Then you will see a row of 4 arrows appear. At some point one of the arrows will change direction. When you notice the change, speak loudly and clearly into the microphone. Speak the number that corresponds with the position of the change. Let’s practice the computer task now. The experimenter will show you where to position your hands and feet relative to the table. Remember that in order to record a correct response, you must say, loudly and clearly, the number that corresponds with the location of the change. When you are ready, let the experimenter know you wish to begin.

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES      

34  

Appendix B

1502 Survey in this survey you will be asked questions about the experiment you just completed. It is very important that you answer honestly so take your time with each question. After you have completed the questions, let the experimenter know that you are ready to continue. 1. Session Number 2. Participant Age 3. Which gender do you identify with? Activities This section of the survey asks about your experience with various activities. Please answer the questions to the best of your ability. Never to Every Day 6-point Likert 4. How often do you play first person shooters? 5. How often do you play Massive Multiplayer Online Games? 6. How often do you go without using any type of screen display? 7. How often do you use a mobile touchscreen device specifically for games? 8. How often do you use a touchscreen tablet? 9. How often do you use a touchscreen phone? 10. How often do you use your computer specifically for games? 11. How often do you use your mobile phone or tablet for something other than calls? (e.g. surfing the Web, checking email, etc.) Abilities The next set of questions ask about your belief in your ability to complete tasks and accomplish goals. Rate the degree to which you agree or disagree with the following statements: Strongly Disagree to Strongly Agree 6-point Likert 12. I will be able to achieve most of the goals that I have set for myself. 13. When facing difficult tasks, I am certain that I will accomplish them. 14. In general, I think that I can obtain outcomes that are important to me 15. I believe I can succeed at most any endeavor to which I set my mind. 16. I will be able to successfully overcome many challenges. 17. I am not confident that I can perform effectively on many different tasks. 18. Compared to other people, I can do most tasks very well. 19. When things are tough I struggle. 20. I can perform most activities while doing other things at the same time. 21. When facing multiple difficult tasks, I am confident I can do them both. 22. In general, I think that I can balance multiple life goals well. 23. I don’t believe I can succeed at most endeavors where I have to keep track of multiple things. 24. I can successfully overcome many challenges.

 DIVIDED ATTENTION IN MULTITASKING WITH MOBILE DEVICES 35         25. I am confident that I can perform effectively on many different tasks. 26. Compared to other people, I can’t hold as many ideas in my head at once. 27. Even when things are complicated, I can do more than one at the same time. Single Task Questions In this Section, you will be asked questions about performing each of the tasks individually Not at all difficult to Very difficult 6-point Likert 28. How hard did you find the mobile task by itself? 29. How well do you feel you performed on the mobile task by itself? 30. How hard did you find the computer task by itself? 31. How well do you feel you performed on the computer task by itself? 32. Compared to others how well do you think you did on the mobile task by itself? 33. Compared to others how well do you feel you did on the computer task by itself? Multitasking 34. How how easy was it to do two tasks at the same time? 35. Compared to others, how good do you consider yourself at multitasking? 36. While balancing the ball on the mobile device, how hard did you find detecting a change in the direction of an arrow You have completed the questionnaire and are free to go. Thank you very much for your participation.