The Experts below are selected from a list of 174 Experts worldwide ranked by ideXlab platform
Raj M. Ratwani - One of the best experts on this subject based on the ideXlab platform.
-
Understanding Emergency Medicine Physicians MultiTasking Behaviors Around Interruptions.
Academic Emergency Medicine, 2018Co-Authors: Allan Fong, Raj M. RatwaniAbstract:BACKGROUND Interruptions can adversely impact human performance, particularly in fast-paced and high-risk environments such as the emergency department (ED). Understanding physician behaviors before, during, and after interruptions is important to the design and promotion of safe and effective workflow solutions. However, traditional human factors-based interruption models do not accurately reflect the complexities of real-world environments like the ED and may not capture multiple interruptions and multiTasking. METHODS We present a more comprehensive framework for understanding interruptions that is composed of three phases, each with multiple levels: interruption start transition, interruption engagement, and interruption end transition. This three-phase framework is not constrained to discrete Task transitions, providing a robust method to categorize multiTasking behaviors around interruptions. We apply this framework in categorizing 457 interruption episodes. RESULTS A total of 457 interruption episodes were captured during 36 hours of observation. The Interrupted Task was immediately suspended 348 (76.1%) times. Participants engaged in new self-initiated Tasks during the interrupting Task 164 (35.9%) times and did not directly resume the Interrupted Task in 284 (62.1%) interruption episodes. CONCLUSION Using this framework provides a more detailed description of physician behaviors in complex environments. Understanding the different types of interruption and resumption patterns, which may have a different impact on performance, can support the design of interruption mitigation strategies.
-
A spatial memory mechanism for guiding primary Task resumption
2008Co-Authors: Raj M. RatwaniAbstract:A SPATIAL MEMORY MECHANISM FOR GUIDING Task RESUMPTION Raj M. Ratwani, Ph.D. George Mason University, 2008 Dissertation Director: Dr. J. Gregory Trafton Theories accounting for the Task resumption process following an interruption have primarily been memory based accounts (Altmann & Trafton, 2002, 2007; Oulasvirta & Sarrlilouma, 2004). The purpose of this study was to examine the resumption process at the perceptual level to determine whether spatial memory processes are used to resume and to determine whether these processes can be directly integrated with an activationbased theoretical framework of goal memory (Altmann & Trafton, 2002). Based on previous literature two plausible hypotheses, a retrace hypothesis and a spatial memory hypothesis, were examined to account for the perceptual processes used to resume an Interrupted Task. Six eye movement studies, using two different Tasks that varied in Task structure, were conducted to distinguish between these two hypotheses. In Experiments 1 and 4, the pattern of eye movements upon resumption was examined to distinguish between the retrace and spatial memory hypotheses. In Experiments 2 and 5, an interrupting Task that required spatial working memory resources was shown to be more disruptive than a non-spatial interrupting Task. These results directly implicate spatial memory in the Task resumption process. In Experiments 3 and 6, interruption length was manipulated to determine whether spatial memory remains intact over longer interruption lengths. Together, the results of these experiments provide strong support for a spatial memory mechanism of Task resumption that can be directly integrated with the Altmann and Trafton (2002, 2007) memory for goals theory.
-
Goal and Spatial Memory Following Interruption
2007Co-Authors: Michel E. Brudzinski, Raj M. Ratwani, J. G. TraftonAbstract:Abstract : The process of resuming an Interrupted Task has been understood by Task level goals. Recent empirical evidence has implicated spatial memory as a component of the resumption process suggesting that spatial level representations are important as well. We collected eye track data in an interruptions paradigm to examine the perceptual processes involved in resumption. Four models were created to illustrate the importance of the role of spatial representations and further, to demonstrate how the Task level and spatial representations can be integrated.
-
Now, where was I? Examining the Perceptual Processes while Resuming an Interrupted Task
2006Co-Authors: Raj M. Ratwani, J. G. TraftonAbstract:Now, where was I? Examining the Perceptual Processes while Resuming an Interrupted Task Raj M. Ratwani (rratwani@gmu.edu) George Mason University, Department of Psychology Fairfax, VA J. Gregory Trafton (trafton@itd.nrl.navy.mil) Naval Research Laboratory Washington, D.C applied to interruptions (e.g., Long Term Working Memory, (Ericsson & Kintsch, 1995; Oulasvirta & Sarrlilouma, 2004)), however, these theories do not make clear predictions about the specific processes used to resume an Interrupted Task. Altmann and Trafton (2002) have put forward an activation based memory model specific to the resumption of an Interrupted Task. This theory, called Memory for Goals, suggests that an interrupting secondary Task results in a suspension of the current subgoal of the primary Task. The resumption lag is a consequence of the time it takes to retrieve the suspended subgoal after completing the interrupting Task. The most active goal is the goal that will be retrieved and the goal that will be selected to drive behavior. The theory suggests there are two determinants of goal activation, and consequently, two determinants of what goal will be retrieved when resuming the primary Task. First, the activation of a goal is based on its history; for example how frequently the goal has been retrieved and how recently the goal was retrieved will impact goal activation. In addition, the activation is based on context and the environment. The context provides priming of the suspended goal resulting in a boost in its activation. Several things should be noted about this description. First, the theory does not make any specific predictions about the perceptual processes used or needed to resume the primary Task. Different environmental cues have been shown to facilitate resumption lag (Trafton, Altmann, & Brock, 2005), but the interaction between perceptual processes and environmental cues is currently unspecified. Second, the memory for goals theory (and others) make the assumption that resuming a Task is, in large part, a matter of determining what had been done previously. For many Tasks (e.g., computer based interactions), however, determining where in the Task or interface pre-interruption work had occurred is just as important as determining what had been done previously. The purpose of this paper is to examine the specific perceptual processes involved in resuming the primary Task during the resumption lag. In order to examine the perceptual processes during the resumption lag a spreadsheet Task was selected that had a flat goal structure. This flat goal structure allowed us to focus on where the resumption point should be rather than on what had been Abstract Several empirical papers have demonstrated that interruptions are disruptive and that after being Interrupted it takes some time to resume the primary Task. This study examined the cognitive processes, specifically at the perceptual level, that were used to resume a Task after being Interrupted. Eye movement data showed that participants were able to use spatial memory to return to the general area where they were Interrupted. This spatial heuristic was used for interruptions that occurred both early and late in the primary Task, however, participants were more imprecise when returning to the Task after a late interruption. Introduction There is a great deal of evidence suggesting that in most instances interruptions are disruptive. Several empirical studies have demonstrated how detrimental an interruption can be to primary Task performance (Altmann & Trafton, 2004; Monk, Boehm-Davis, & Trafton, 2004; Trafton, Altmann, Brock, & Mintz, 2003). Being Interrupted can result in more errors on the primary Task, a longer time to complete the primary Task and greater feelings of stress and anxiety when performing the Task (Adamcyzk & Bailey, 2004; Speier, Vessey, & Valacich, 2003). One dependent measure that has been used to examine how disruptive an interruption can be is the resumption lag (Altmann & Trafton, 2004; Trafton et al., 2003). The resumption lag has been operationally defined as the time interval between the completion of the secondary (interrupting) Task and the first action back on the primary Task. The resumption lag is essentially the time it takes to resume the primary Task after completing the interrupting Task. For example, while working on a paper (the primary Task) a student may stop by (interrupting Task) to talk about research ideas. Once the student leaves, the time it takes to focus one’s thoughts back on the paper and actually resume writing the paper is the resumption lag. How does one go about resuming the primary Task after being Interrupted? Several studies have illustrated a significantly longer resumption lag after being Interrupted as compared to a control condition (Altmann & Trafton, 2004; Monk et al., 2004; Trafton et al., 2003). However, most of the research on interruptions has dealt with reducing or changing the resumption lag, not on the processes used to resume the primary Task. Several general memory theories have been
Frédéric Vivien - One of the best experts on this subject based on the ideXlab platform.
-
Scheduling independent stochastic Tasks on heterogeneous cloud platforms
2019Co-Authors: Yiqin Gao, Louis-claude Canon, Yves Robert, Frédéric VivienAbstract:This work introduces scheduling strategies to maximize the expected number of independent Tasks that can be executed on a cloud platform within a given budget and under a deadline constraint. The cloud platform is composed of several types of virtual machines (VMs), where each type has a unit execution cost that depends upon its characteristics. The amount of budget spent during the execution of a Task on a given VM is the product of its execution length by the unit execution cost of that VM. The execution lengths of Tasks follow a variety of standard probability distributions (exponential, uniform, halfnormal, etc.), which is known beforehand and whose mean and standard deviation both depend upon the VM type. Finally, there is a global available budget and a deadline constraint, and the goal is to successfully execute as many Tasks as possible before the deadline is reached or the budget is exhausted (whichever comes first). On each VM, the scheduler can decide at any instant to interrupt the execution of a (long) running Task and to launch a new one, but the budget already spent for the Interrupted Task is lost. The main questions are which VMs to enroll, and whether and when to interrupt Tasks that have been executing for some time. We assess the complexity of the problem by showing its NPcompleteness and providing a 2-approximation for the asymptotic case where budget and deadline both tend to infinity. Then we introduce several heuristics and compare their performance by running an extensive set of simulations.
-
Scheduling stochastic Tasks on heterogeneous cloud platforms under budget and deadline constraints
2019Co-Authors: Yiqin Gao, Louis-claude Canon, Frédéric Vivien, Yves RobertAbstract:This work introduces scheduling strategies to maximize the expected number of independent Tasks that can be executed on a cloud platform within a given budget and under a deadline constraint. The cloud platform is composed of several types of virtual machines (VMs), where each type has a unit execution cost that depends upon its characteristics. The amount of budget spent during the execution of a Task on a given VM is the product of its execution length by the unit execution cost of that VM. The execution length of Tasks follow an exponential, uniform or lognormal probability distribution whose mean and standard deviation both depend upon the VM type. Finally, there is a global available budget and a deadline constraint, and the goal is to successfully execute as many Tasks as possible before the deadline is reached or the budget is exhausted (whichever comes first). On each VM, the scheduler can decide at any instant to interrupt the execution of a (long) running Task and to launch a new one, but the budget already spent for the Interrupted Task is lost. The main questions are which VMs to enroll, and whether and when to interrupt Tasks that have been executing for some time. We assess the complexity of the problem by showing its NP-completeness and providing a 2-approximation for the asymptotic case where budget and deadline both tends to infinity. We introduce several heuristics and compare their performance by running an extensive set of simulations.
-
CLUSTER - Scheduling independent stochastic Tasks on heterogeneous cloud platforms
2019 IEEE International Conference on Cluster Computing (CLUSTER), 2019Co-Authors: Yiqin Gao, Louis-claude Canon, Yves Robert, Frédéric VivienAbstract:This work introduces scheduling strategies to maximize the expected number of independent Tasks that can be executed on a cloud platform within a given budget and under a deadline constraint. The cloud platform is composed of several types of virtual machines (VMs), where each type has a unit execution cost that depends upon its characteristics. The amount of budget spent during the execution of a Task on a given VM is the product of its execution length by the unit execution cost of that VM. The execution lengths of Tasks follow a variety of standard probability distributions (exponential, uniform, half-normal, etc.), which is known beforehand and whose mean and standard deviation both depend upon the VM type. Finally, there is a global available budget and a deadline constraint, and the goal is to successfully execute as many Tasks as possible before the deadline is reached or the budget is exhausted (whichever comes first). On each VM, the scheduler can decide at any instant to interrupt the execution of a (long) running Task and to launch a new one, but the budget already spent for the Interrupted Task is lost. The main questions are which VMs to enroll, and whether and when to interrupt Tasks that have been executing for some time. We assess the complexity of the problem by showing its NP-completeness and providing a 2-approximation for the asymptotic case where budget and deadline both tend to infinity. Then we introduce several heuristics and compare their performance by running an extensive set of simulations.
Srikanth V. Krishnamurthy - One of the best experts on this subject based on the ideXlab platform.
-
CoNEXT - Computing while charging: building a distributed computing infrastructure using smartphones
Proceedings of the 8th international conference on Emerging networking experiments and technologies - CoNEXT '12, 2012Co-Authors: Mustafa Y. Arslan, Indrajeet Singh, Harsha V. Madhyastha, Krishnakumar Sundaresan, Shailendra Singh, Srikanth V. KrishnamurthyAbstract:Every night, a large number of idle smartphones are plugged into a power source for recharging the battery. Given the increasing computing capabilities of smartphones, these idle phones constitute a sizeable computing infrastructure. Therefore, for an enterprise which supplies its employees with smartphones, we argue that a computing infrastructure that leverages idle smartphones being charged overnight is an energy-efficient and cost-effective alternative to running Tasks on traditional server infrastructure. While parallel execution and scheduling models exist for servers (e.g., MapReduce), smartphones present a unique set of technical challenges due to the heterogeneity in CPU clock speed, variability in network bandwidth, and lower availability compared to servers. In this paper, we address many of these challenges to develop CWC---a distributed computing infrastructure using smartphones. Specifically, our contributions are: (i) we profile the charging behaviors of real phone owners to show the viability of our approach, (ii) we enable programmers to execute parallelizable Tasks on smartphones with little effort, (iii) we develop a simple Task migration model to resume Interrupted Task executions, and (iv) we implement and evaluate a prototype of CWC (with 18 Android smartphones) that employs an underlying novel scheduling algorithm to minimize the makespan of a set of Tasks. Our extensive evaluations demonstrate that the performance of our approach makes our vision viable. Further, we explicitly evaluate the performance of CWC's scheduling component to demonstrate its efficacy compared to other possible approaches.
-
Computing while charging: Building a Distributed Computing Infrastructure Using Smartphones
8th international conference on Emerging networking experiments and technologies, 2012Co-Authors: Mustafa Y. Arslan, Indrajeet Singh, Harsha V. Madhyastha, Krishnakumar Sundaresan, Shailendra Singh, Srikanth V. KrishnamurthyAbstract:Every night, a large number of idle smartphones are plugged into a power source for recharging the battery. Given the increasing computing capabilities of smartphones, these idle phones consti- tute a sizeable computing infrastructure. Therefore, for an enter- prise which supplies its employees with smartphones, we argue that a computing infrastructure that leverages idle smartphones be- ing charged overnight is an energy-efficient and cost-effective alter- native to running Tasks on traditional server infrastructure. W`hile parallel execution and scheduling models exist for servers (e.g., MapReduce), smartphones present a unique set of technical chal- lenges due to the heterogeneity in CPU clock speed, variability in network bandwidth, and lower availability compared to servers. In this paper, we address many of these challenges to develop CWC—a distributed computing infrastructure using smartphones. Specifically, our contributions are: (i) we profile the charging be- haviors of real phone owners to show the viability of our approach, (ii)we enable programmers to execute parallelizable Tasks on smart- phones with little effort, (iii) we develop a simple Task migration model to resume Interrupted Task executions, and (iv) we imple- ment and evaluate a prototype of CWC (with 18 Android smart- phones) that employs an underlying novel scheduling algorithm to minimize the makespan of a set of Tasks. Our extensive eval- uations demonstrate that the performance of our approach makes our vision viable. Further, we explicitly evaluate the performance of CWC’s scheduling component to demonstrate its efficacy com- pared to other possible approaches.
Sophie Leroy - One of the best experts on this subject based on the ideXlab platform.
-
Tasks Interrupted: How Anticipating Time Pressure on Resumption of an Interrupted Task Causes Attention Residue and Low Performance on Interrupting Tasks and How a “Ready-to-Resume” Plan Mitigates the Effects
Organization Science, 2018Co-Authors: Sophie Leroy, Theresa M. GlombAbstract:This paper explores the attention regulation challenges brought by interruptions. In contrast to much of the research on interruptions that looks at the effects on the Interrupted Task, this paper examines the difficulty of focusing attention and performing well on interrupting Tasks. Integrating research on attention residue, time pressure, and implementation intention, we predict that when people anticipate resuming their Interrupted work under time pressure, they will find it difficult to switch their attention to the interrupting Task, leading to attention residue and low performance. A ready-to-resume intervention, in which one briefly reflects on and plans one’s return to the Interrupted Task, mitigates this effect such that attention residue is reduced and performance on the interrupting Task does not suffer. Data collected across four studies support these hypotheses. The e-companion is available at https://doi.org/10.1287/orsc.2017.1184.
-
The effect of regulatory focus on attention residue and performance during interruptions
Organizational Behavior and Human Decision Processes, 2016Co-Authors: Sophie Leroy, Aaron M. SchmidtAbstract:Abstract This paper explores how regulatory focus affects transitions between Tasks following interruptions. Consistent with the research on attention residue (Leroy, 2009), we argue that in order to be cognitively available and perform well on an interrupting Task, people must cognitively disengage from the Task that is Interrupted—that is they must fully switch their attention to the interrupting demand. Integrating the research on regulatory focus (Higgins, 1997) and attention residue (Leroy, 2009), we predict that both the framing of the initial/Interrupted Task and the framing of the interrupting Task interact to affect how well people switch their attention to and perform on an interrupting Task. This investigation allows the identification of when attention residue is most likely to occur, hindering performance on the interrupting Task and how attention residue can be prevented or mitigated. Data across three studies support our predictions.
Vivien Frédéric - One of the best experts on this subject based on the ideXlab platform.
-
Scheduling independent stochastic Tasks on heterogeneous cloud platforms
IEEE, 2019Co-Authors: Gao Yiqin, Canon Louis-claude, Robert Yves, Vivien FrédéricAbstract:International audienceThis work introduces scheduling strategies to maximize the expected number of independent Tasks that can be executed on a cloud platform within a given budget and under a deadline constraint. The cloud platform is composed of several types of virtual machines (VMs), where each type has a unitexecution cost that depends upon its characteristics. The amount of budget spent during the execution of a Task on a given VM is the product of its execution length by the unit execution cost of that VM. The execution lengths of Tasks follow a variety of standard probability distributions (exponential, uniform, halfnormal, etc.), which is known beforehand and whose mean and standard deviation both depend upon the VM type. Finally, there is a global available budget and a deadline constraint, and the goal is to successfully execute as many Tasks as possible before the deadline is reached or the budget is exhausted (whichever comes first). On each VM, the scheduler can decide at any instant to interrupt the execution of a (long) running Task and to launch a new one, but the budget already spent for the Interrupted Task is lost. The main questions are which VMs to enroll, and whether and when to interrupt Tasks that have been executing for some time. We assess the complexity of the problem by showing its NPcompleteness and providing a 2-approximation for the asymptotic case where budget and deadline both tend to infinity. Then we introduce several heuristics and compare their performance by running an extensive set of simulations
