• Embed Doc
  • Copy Link
  • Readcast
  • Collections
  • CommentGo Back
 
Abstract
Knowledge workers increasingly use multiple devicessuch as desktop computers, laptops, cell phones, andPDAs for personal information management (PIM)tasks. The use of several of these devices togethercreates higher task difficulty for users than when usedindividually (as reported in a recent survey weconducted). Prompted by this, we are conducting anexperiment to study mental workload in multi-devicescenarios. While mental workload has been shown todecrease at sub-task boundaries, it has not beenstudied if this still holds for sub-tasks performed ondifferent devices. We hypothesize that the level of support provided by the system for task migrationaffects mental workload. Mental workloadmeasurements can enable designers to isolate criticalsub-tasks and redesign or optimize the user experienceselectively. In addition, we believe that mentalworkload shows promise as a cross-tool, cross-taskmethod of evaluating PIM tools, services and strategies,thus fulfilling a need expressed by several researchersin the area of personal information management. Inthis paper, we describe our ongoing experiment of measuring mental workload (via physiological as well assubjective measures) and its implications for users,designers and researchers in PIM.
Keywords
Personal Information Management, Mental Workload,Multiple Devices
ACM Classification Keywords
H.5.2 Information Interfaces and Presentation: Userinterfaces – Evaluation/ methodology
Introduction & Motivation
As we amass vast quantities of personal information,managing it has become an increasingly complexendeavor. The emergence of multiple informationdevices and services such as desktops, laptops, cellphones, PDAs and cloud computing adds a level of complexity beyond simply the use of a single computer.In traditional single terminal computer systems, themajority of a user’s attentional and cognitive resourcesare focused on the terminal while performing a specifictask. However, in an environment where multipledevices require intermittent attention and presentuseful information at unexpected times, the user issubjected to different mental workload.
Mental Workload in Multi-DevicePersonal Information Management
Manas Tungare
Dept. of Computer Science, Virginia Tech.Blacksburg, VA, USA.manas@tungare.name
Manuel A. Pérez-Quiñones
Dept. of Computer Science, Virginia Tech.Blacksburg, VA, USA.perez@cs.vt.edu
Copyright is held by the author/owner(s).
CHI 2009
, April 4 – 9, 2009, Boston, MA, USAACM 978-1-60558-246-7/09/04.
 
In an earlier study we conducted [15], usersconsistently reported difficulties in performinginformation tasks with multiple devices, especially whentransitioning between/among devices. From theresponses we received, we observed (from a contentanalysis of free-form responses) that users’ adoption of various technological alternatives is guided by an innatesense of certain specific factors. We noted that severalof these factors constitute mental workload, e.g.frustration level, temporal demand, and mental effort.In systems where users lacked the freedom of choice,they turned to solving problems by adoptingworkarounds motivated by one or more of thesefactors.It has been shown that an operator’s task performanceis inversely correlated with high levels of mentalworkload [12]. Thus, we set out to explore if mentalworkload estimates could be used to compare taskdifficulty in PIM tasks. Prior work in mental workloadmeasurement has established that physiologicalmeasures such as changes in pupillary diameter(known as Task-Evoked Pupillary Response [3]) can beused to estimate mental workload. Such continuousmeasures of mental workload can help locate sub-tasksof high task difficulty. Iqbal et al. [8] demonstrated thatwithin a single task, mental workload decreases at sub-task boundaries. A fundamental goal of our research isto examine if their finding still applies when the lattersub-task is performed on a different device than theformer. Our contrary hypothesis is that mental workloadrises just before the moment of transition, and returnsto its normal level a short duration after the transitionis complete.Systems differ in the level of support they provide forpausing a task on one device, and resuming it onanother [13]. A related goal of our research is toexamine if the increase in mental workload at the pointof transition is correlated with the level of systemsupport available for the sub-task of transitioning. I.e.,if the system incorporates full support for taskmigration, we hypothesize that mental workload will beless than in case of another system where such supportis lacking.In addition, there has been no standard way tocompare the effectiveness of tools, services, andtechniques developed independently at differentresearch labs. Kelly [9] notes the methodologicaldifficulties in studying PIM because of its highlypersonal nature, leading to challenges in developing aset of reference tasks or cross-tool cross-task metrics.In several other task domains, workload assessmentssuch as NASA TLX [6] have been administered insteadof direct measurement of task performance metrics forseveral reasons: chief among them is that subjectiveworkload assessments require less effort andinstrumentation of the task, and are easier toadminister. If mental workload in PIM tasks can beshown to be inversely correlated with task performance(as has already been shown in several other domains[12, 2, 5]), such a measure can be used to comparethe effectiveness of these tools across varying tasks.Thus, a tertiary goal of our research is to examinewhether mental workload estimates captured using theNASA TLX scale can serve as a predictor of taskperformance for personal information managementtasks.
Related Prior Work
Mental workload is an important, practically relevant,and measurable entity [6]. The NASA Task Load Index(NASA TLX) [6] is a multi-dimensional subjective
 
workload assessment technique that has been appliedin studies of airline cockpits [2], navigation [14], and inthe medical field [5]. It combines information aboutspecific sources of workload weighted by theirrelevance, thus reducing the influence of those areexperimentally irrelevant, and emphasizing thecontributions of others that are experimentallyrelevant. This reduces between-subject variability forthe measure as compared to other subjective scales.Physiological measures such as changes in pupillarydiameter (known as Task-Evoked Pupillary Response)have been shown to be responsive to changes inmental workload [3] and used as a physiologicalmeasure of mental workload in several studies [7, 1].Within a single task, mental workload decreases at sub-task boundaries [8]. Such continuous measures of mental workload can help locate sub-tasks of high taskdifficulty.As the problem of information overload has worsenedover the years, human attentional resources havestayed constant [11]. The issue of informationfragmentation across multiple devices (the condition of having a user’s data in different formats, distributedacross multiple locations, manipulated by differentapplications, and residing in a generally disconnectedmanner [4]) threatens the effectiveness of users aswell as of our tools and systems.An understanding of mental workload in PIM tasks isnot only expected to lead to a better understanding of why a particular tool causes high frustration or mentaldemand in users, but also can be used to isolate criticalsub-tasks and for comparing different tools against oneanother.
Results from Preliminary Studies
Experimental tasks for the current study were chosenfrom among the most common representative tasksidentified in an exploratory survey study [15] andanother ethnographic investigation [16] (reportedelsewhere).File management across multiple machines stood out asthe most reported problematic task. 12 out of 79survey users said that they encountered difficultieswhile syncing data between multiple machines, 11reported unexpected deletion of their data whilecopying across machines, and 6 reported havingtrouble with managing conflicting versions of files thatwere copied manually. Based on these findings, our firstexperimental task involves managing files across adesktop and a laptop, with and without support forautomatic synchronization.From the ethnographic investigation of calendar use[16], we found that paper calendars were actively usedby a majority of interviewees despite the widespreadprevalence of electronic calendars (corroborating thefindings reported in previous studies). 35% of participants reported printing their electronic calendarfor offline use. Based on this, our second experimentaltask is calendar management, and involves managingschedules using an online calendar and papercalendars.From the survey, we also found that several devices areoften used in groups, e.g. laptops and cell phones(reported by 52 participants), and integrated multi-function portable devices such as Palm Treos,Blackberries and Apple iPhones have begun to replacesingle-function devices for communication (e.g. email
of 00

Leave a Comment

You must be to leave a comment.
Submit
Characters: ...
You must be to leave a comment.
Submit
Characters: ...