Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 214
9 Strategic Task Management The previous chapter has described the importance of maintaining situ- ation awareness and effective decision making in both the pre- and post- transition period. These elements, however, represent only a part of the cognitive activity required by teams in periods of high workload. A critical activity that we address in this chapter is strategic task management. The operator, faced with a host of tasks competing for attention in periods of high workload, must decide which tasks to perform when, how frequently to switch from one task to another, and when it is appropriate to break off performance of an ongoing task to shift to one of higher priority. Further- more, to be effective, the shift from one activity to the next should be rapid. Some aspects of strategic task management are quite similar to those addressed in the previous chapter. For example, the decision made by a tank commander of whether to engage an enemy or retreat is clearly a choice between tasks. But issues of task scheduling and switching are not ones that are typically addressed in the study of decision making. This chapter addresses the issue of task switching and task management at two levels. In the first part, we consider basic laboratory research that has addressed the speed with which humans can initiate, change, and cease tasks. In the second part, we address research in more complex domains that has focused more on when or whether tasks of different priority are performed. COGNITIVE SWITCHING Task switching is a pervasive phenomenon in everyday life. Every day we engage in several tasks, switching from one to another with little thought 214
OCR for page 215
STRATEGIC TASK MANAGEMENT 215 and little apparent effort. We get up, bathe, dress, eat, and go to work. We come home, cook, eat, clean up, recreate, and go to bed. Each activity is focused on specific goals, often requiring that the same things be used in different ways (consider, for example, the different ways we use water in these activities). In each activity, our behavior could be modeled effec- tively as a special-purpose processor, dedicated to the one act and no other. At different times, we appear to be dressing machines, bathing machines, cooking machines, eating machines, and so on. However, we appear to be more than special-purpose machines, and the clues to our nature become clear at the transitions between activities: we appear to be collections of special-purpose machines, capable of changing from one to another on a moment's notice. The transitions are brief and hardly noticeable in every- day life, yet they are fundamentally important in understanding human be- havior. Strategy Switching One approach to the study of task switching has been through the study of strategy switching. A person's behavior in performing a task, whether simple or complex, can be described as a strategy. A strategy can be de- fined as an optional organization of cognitive processes that is designed to achieve some goal in a particular task environment (Logan, 1985a; Logan and Zbrodoff, 1982; Logan et al., 19839. The organization is optional be- cause people can configure their processes in several ways. Often, there are several ways to perform the same task; each one is a different strategy. The cognitive processes are organized because strategic behavior is coherent and planful. Strategic behavior is intentional organized around the achievement of goals. =~ r others will not. People choose the former. Some configurations of pro- cesses will achieve the coals better than others. Peoole trY to choose the , -, come conr~ura~ons of processes will achieve the seals anr1 better approaches. Having adopted a strategy, people will appear as special purpose processors dedicated to the task at hand. But this way of defining strategies highlights the general purpose nature of human activity and pro- vides a way to describe transitions from one strategy to another. The research of Logan, Zbrodoff, and their colleagues has revealed that people can be relatively rapid at adapting or changing the strategies with which they perform a task. For example, in one task, a variant of the Stroop task (Stroop, 1935), people can either attend to the position of a word (presented above or below a fixation point), or the word's semantic content ("above" or "below". Depending on the nature of the information pre- sented on a given trial, attending to the one attribute or the other (a differ- ence in attentional strategy) can speed or slow their performance. Logan and Zbrodoff (1982) found that subjects can adopt the required attentional
OCR for page 216
216 WORKLOAD TRANSITION strategy very rapidly (within a half second) if they are motivated to do so, and they will choose one strategy or the other on the basis of the probability that it will best serve their performance (Logan et al., 1984; Zbrodoff and Logan, 19869. Logan's research on strategy selection of attentional dimensions has parallels in Rabbitt's work (1989) in the selection of speed-accuracy tradeoff strategies. In tasks typifying those in which operators must make a series of rapid choice responses, Rabbitt notes the facility with which subjects seek the strategy that provides the optimal level of performance. For ex- ample, if accuracy is important, they will increase in speed until an error is made, and then immediately back off just enough to restore errorless perfor- mance on subsequent trials. Similarly, Wickens et al. (1985) have found that, within a dual task paradigm, subjects can shift resources from one task to another within a few seconds, in response to a sudden increase in re- source demand (difficulty) imposed by one of the tasks. The facility with which strategies can be chosen and modified is impor- tant and certainly testifies to the flexibility of human cognition. This is an important issue in workload transition periods, for it is evident that a change from pre- to post-transition conditions will normally require a host of dif- ferent strategies to be adopted, such as a shift to more rapid performance. It should be noted, however, that most studies of rapid strategy shifting have not been carried out in high-stress conditions, and so, to some extent, the generalizability of this finding of flexibility to the tank crew environ- ment must be treated cautiously. Also, shifting the strategy with which a given task is performed is not the same as shifting between two different tasks-an issue that we now address. 'Task Switching Research on the issue of task switching is made complex because of ambiguity in the issue of "What is a task?" A relatively unexplored re- search area concerns the hierarchical nature of complex tasks. High-level goals are often accomplished by the attainment of several subordinate low- level goals, and the high-level goals need to be kept in mind while dealing with the low-level goals. This is evident, for example, in navigation tasks in which higher-level goals (e.g., reach a given location) are accomplished by meeting lower-level goals (e.g., come to a given heading). The research most relevant to this issue is Vallacher and Wegner's (1987) action identification theory. It assumes that action is organized hierarchi- cally and concerns itself with discovering the level at which actions are construed and the circumstances under which people shift the level at which they describe their action. Vallacher and Wegner (1987) argue that the same act can be described at several hierarchical levels. For example, the same
OCR for page 217
STRATEGIC TASK MANAGEMENT 217 act may be described as pressing buttons, dialing a telephone, calling a friend, maintaining social contacts, or searching for meaning in life. Among other things, familiarity with the act encourages high-level descriptions. Vallacher and Wegner show that alcoholics describe imbibing as "getting smashed" whereas nonalcoholics will describe it as "having a drink." People drinking coffee from a 1-pound cup will describe their actions as "picking up a cup," "moving the cup to my lips" and so on, whereas people drinking coffee from normal cups describe their actions as "drinking coffee" or "get- ting a dose of caffeine." Vallacher and Wegner argue that levels of action description lower than the one chosen by the actor are performed automatically: the actor need only think of the goal at the level at which he or she describes the action. Lower-level activities will take care of themselves. They argue that people will maintain action descriptions at the highest possible level until some- thing goes wrong. In that case, they switch to a lower-level description to try to discover what went wrong. Thus, we think we are calling a friend until we discover that the number wasn't dialed correctly. Then we pay attention to button pressing, trying to discover whether we executed the sequence incorrectly or the telephone isn't working properly. Action identification theory is very new and has not been applied to many situations, but it seems very promising, especially in the context of workload transitions. At the very least, it provides a framework in which to understand the complex tasks encountered in workload transition situations. More likely, it will provide a rich source of hypotheses for future research (e.g., what happens to high-level goals when attention is called to low-level goals). The relatively scarce research that has been done on switching between tasks has considered the concept of task at a relatively low hierarchical level. It reveals that, like strategy switching, task switching can be carried our fairly rapidly. Jersild (1927) studied the time taken to alternate between similar (add- ing versus subtracting numbers) and dissimilar alternatives (addition or sub- traction versus naming opposites), and found that subjects could alternate between very dissimilar alternatives with virtually no cost (also see Biederman, 1973; Spector and B,iederman, 1976~. This suggests that strategy engage- ment time depends on the relation between the strategy to be engaged and the one that went before it. (This theme is discussed at greater length in the section below on disengaging strategies.) More recently, a series of studies on multichannel information processing have revealed that people are able to switch attention from one channel to another, on cue, within a few tenths of a second (Gopher, 1982; Sperling and Dosher, 1986~. Interestingly, Go- pher has found that individual differences in the speed of switching between two auditory channels provides a valid predictor of differences in perform- ing complex skills, such as those found in bus driving and aviation.
OCR for page 218
218 WORKLOAD TRANSITION A substantial research program on task switching has been undertaken by Logan and his colleagues, who have examined, independently, the two components of task switching: stopping the first activity and changing to the second activity. Action can be stopped relatively easily. Highly skilled actions, such as speaking and typewriting, can be stopped very quickly, within a syllable or two or keystroke or two of the stop signal. The same results can be found with experimenter-imposed stop signals and with naturally occurring stop signals such as errors in speaking or typing (for a review, see Logan and Cowan, 19841. Inhibition is difficult only in pathological cases. Schachar and Logan (1990) found that hyperactive children were deficient in their ability to inhibit action, compared with other disturbed children and to normal control children. But even their inhibitory ability can be improved by administration of stimulant medication (methylphenidate) which improves behavioral symptoms of hyperactivity (Tannock et al., 1989~. Action appears to be controllable up to the point of execution. There is no apparent "point of no return" after which action becomes ballistic. This was suggested by behavioral experiments (Logan, 1981; Osman et al., 1990) and confirmed with psychophysiological measures (de Jong et al., 1990~. Nevertheless, stopping is strategic. People stop action more quickly when they expect to stop (i.e., when stop signals occur on a greater proportion of trials; Logan, 1981; Logan and Burkell, 1986~. It takes time to stop action, but not much. Actions that are simple and complex, skilled and unskilled, can be stopped in 200-400 milliseconds (for a review, see Logan and Cowan, 1984~. The time to stop action increases somewhat as the difficulty of performing that action increases (Logan et al., 1984, 19871. A variation of the stop-signal paradigm, relevant to the discussion of activity switching, is the change-task paradigm, in which the signal to stop one activity serves as the stimulus to start a different, changed activity. The change-task signal is a type of procedural bridge between the stop task and traditional dual tasks, in which subjects respond overtly to two signals. It provides a simple analog of workload transition situations, in which current (low-workload) activities have to stop suddenly to make way for new, more important (high-workload) activities. Stopping performance in the change-task paradigm is similar to stop- ping performance in the stop task. The time to stop action may be a little longer in the change task, but the pattern of results is essentially the same. The most interesting results in the change task involve the overt response to the stop (change) signal analyzed contingent on whether or not the subject successfully inhibited the first activity. If subjects fail to inhibit their re- sponses to the first task and therefore respond overtly to the first task and the stop signal, then overt responses to the change signal are delayed as are
OCR for page 219
STRATEGIC TASK MANAGEMENT 219 dual-task reaction times. However, if subjects successfully inhibit their responses to the first task, there is no competition (Logan, 1985b; Logan and Burkell, 1986; also see Logan, 1982~. This suggests that successful inhibition "clears the system," removing residual effects of the inhibited response so that the new response can begin without interference. The lack of interfering effects from inhibited actions has important implications for workload transitions: it suggests that there should be little carryover across the transition, unless, of course, the operator must pick up and complete the stopped task at a later time a common occurrence in the real world but rare in laboratory tasks. Provided that people inhibit thei pretransition actions, there should be little interference with performance of post-transition tasks. However, the change task has not been explored very broadly, so it is not clear how far the lack of refractory effects will general- ize. More research is needed. In particular, it will be important to vary the demands of the post-transition task, as post-transition tasks have been rela- tively simple in previous research (but see Logan, 19833. In conclusion, the data suggest that actions can be stopped very easily and rapidly. The difficulty of a stopping action depends somewhat on the complexity of the action being stopped and on the selectivity of the stop- ping response. Studies of the change task suggest that successful inhibition has few carryover effects for subsequent actions, which is good news for workload transition situations. Implications for Workload Transition While the research on task engagement, change, and task stopping has been carried out in fairly basic laboratory paradigms, it is possible to specu- late to some extent on generalizations that might be made to more complex high-workload environments. Engaging Tasks Engaging tasks take time and resources. Task engagement can be rela- tively leisurely in low-workload situations, but when workload suddenly increases, time pressure increases substantially and there may not be enough time for engagement. Research suggests that the time for strategy engage- ment can be reduced in several ways. First, the engagement process can be automatized (see also Chapter 3), the more practice, the faster the engage- ment process. Again, automaticity is specific, so it will be important to train engagement of each strategy or task separately. In principle, engage- ment is different from choice, and availability of alternatives does not imply the ability to engage them quickly. Trainees should practice rapid engage ment.
OCR for page 220
220 WORKLOAD TRANSITION Second, the engagement process can be sped up by reducing the degree of choice; the fewer the alternatives, the faster the engagement. The tank commander can reduce the degree of choice for subordinates (i.e., by order- ing specific alternatives). That would make engagement easier for the sub- ordinates but it may substantially increase the workload of the commander. A clear chain of command can reduce the degree of choice for commanders and subordinates alike; neither will have to ponder suggestions from unau- thorized sources (e.g., "Is that a command or a suggestion?") or decide between conflicting directives (e.g., "Should I obey the sergeant or the lieutenant?". Third, the engagement process can be sped up by making the conditions for engagement highly discriminable. Fortunately or unfortunately, there may not be much to do here. On one hand, high-workload situations are compelling perceptually (to say the least) and therefore are easily discriminable from low-workload situations. On the other hand, many of the high-workload situations we are considering are either adversarial (onset of combat) or noncooperative (fires or medical emergencies) and therefore cannot be engi- neered to make conditions more discriminable. One has to take what one gets. However, discrimination can be improved by practice; the performer can be engineered even if the situation cannot. Disengaging Tasks Disengaging tasks should not be much of a problem in workload transi- tions. Tasks can be disengaged very quickly whether the task is simple or complex, whether the performer is novice or skilled. Stopping time de- pends on the discriminability of the stop signal, but discriminability should be very clear in workload transition situations (e.g., it should be easy to tell when a battle begins). Disengagement may be difficult when some activi- ties must be stopped while others continue (i.e., when stopping is selective), but even then, stopping times are typically less than one second. Studies of the change task suggest that it should be relatively easy to stop one task and begin another, though existing research has not investigated changes from low- to high-workload tasks. In general, more research would be desirable, particularly in workload transition situations, but the data so far are encour- ag~ng. The high stress in workload transition situations may make disengage- ment difficult. Following the Easterbrook (1959) hypothesis, the high arousal associated with high stress may narrow the range of cue utilization and render the performer insensitive to stop signals. Indeed, case studies of real-world accidents and incidents reveal many examples of such undesir- able cognitive tunneling (e.g., Three Mile Island Rubinstein and Mason, 1979~.
OCR for page 221
STRATEGIC TASK MANAGEMENT 221 TASK PRIORITY MANAGEMENT The research of Logan and others on task switching described above has laid the foundation for an understanding of the time course and relative ease of activity switching, but it has intentionally focused on research para- digms in which the destination activity of the switch was well defined and the overall time frame of the switching process was short, within a second or so. We now discuss data in which both the identity and the time of the particular tasks to be performed are less constrained, and the time frame of performance is on the order of minutes or longer. Our interest here is in the nature of human operator task management strategies: the appropriateness of human behavior in selecting what task to do when (Adams et al., 1991~. Turning to the operational community, one can find numerous examples of aircraft accidents (in National Transporta- tion Safety Board (NTSB) reports) and incidents (in NASA Aviation Safety Reporting System (ASRS) data) that have resulted from failures of effective task management. Here, for example, one might consider a pilot so preoc- cupied with geographic orientation that the key task of flight control is neglected and the aircraft is stalled. In this case the optimal task manage- ment strategy dictated by the adage "aviate-navigate-communicate" is vio- lated. The account of the Eastern Airlines L1011 crash into the Everglades in 1972, when the flight crew, preoccupied with a landing gear problem, failed to monitor their altitude, is another example (Wiener, 1977~. Indeed, significant altitude deviations resulting from task neglect reflect a major concern in the aviation industry because of their growing frequency of oc- currence (Granda et al., 1991~. Two general features would seem to be heavily responsible for the success or failure of individuals to appropriately manage tasks in situations such as those described above. First, it is apparent that good situation awareness is a key component. Situation awareness describes an under- standing not only of geographic orientation, as described in Chapter 7, but also of the state of systems under the operator's control (a breakdown here was partially responsible for the Three Mile Island incident discussed in Chapter 8) and the current responsibilities assumed by equipment automa- tion and by other members of the team. Good situation awareness provides a context or background within which different tasks can be more appropri- ately selected or dropped. The concept of situation awareness remains somewhat ill-defined, but recent theoretical treatments by Sarter and Woods (1991) and by Adams et al. (1991) have made major advancements in link- ing the concept both to existing empirical data and to psychological con- cepts from cognitive psychology. A second feature critical to effective task management is the concept of task priority, and the implicit or explicit assumption that the optimal task
OCR for page 222
222 WORKLOAD TRANSITION manager will possess a mental priority scale that can provide the basis for appropriately shedding tasks when workload becomes excessive and resum- ing (or assuming) tasks when workload is relieved. In spite of the impor- tance of this concept, there appears to be few data to indicate the effective- ness of subjective priority in driving task management in operational environments. What evidence is available comes from a set of partially, but not entirely, relevant research domains. First, we may refer to the National Transportation Safety Board and the Aircraft Safety Reporting System reports on maladaptive task management strategies (Williams et al., 1992~. These reports are important because they illustrate the existence of a problem that has major implications for air safety (e.g., the Everglades crash referred to above). Yet, like all accident and incident reports, these accounts provide a tenuous base from which to draw firm conclusions regarding the precise causes of the failure to priori- tize tasks. Was it a lack of training? A high level of stress? Poor cockpit design? Poor procedures? Accidents and incidents are usually ambiguous as to their cause. A second relevant domain is the laboratory research that has been car- ried out regarding the manipulation of task priorities in dual task perfor- mance, following the pioneering work of Gopher and his colleagues (Go- pher and Navon, 1980; North and Gopher, 1976; Wickens and Gopher, 1977; see also Goettl, 1991; Sperling and Dosher, 1986; Tsang and Wickens, 1988; Vidulich, 1988~. This work generally reveals that, under controlled labora- tory circumstances, subjects can fairly accurately modulate the performance level of the tasks in proportion to the implicit or explicit priority given to the task. Furthermore, there appears to be recent evidence that the skill of priority-based performance modulation learned in the laboratory is one that transfers to the operational environment of the aircraft cockpit (Gopher, 1991~. Yet even in this domain there are hints of occasional failures. For example, Wickens and Liu (1988) have pointed to a phenomenon of "pre- emption" in which a lower-priority discrete auditory task may interrupt or preempt the performance of a higher-priority visual one, a preemption that will not take place if the discrete task is visual. Cumming and Croft (1973) note that people may sometimes fail to recognize or anticipate a change in task demands. Clearly, altitude busts in aviation represent failures to appro- priately allocate resources to this monitoring task; in the hospital trauma center, one can imagine circumstances in which monitoring of key aspects of a patient's status may be neglected because of high attention demands associated with other aspects of the trauma. A third domain of data derives from studies that have been carried out from an engineering perspective, based heavily on the formal treatment provided by queuing theory (see Kleinmann, 1991; Moray, 1986; Moray et al., 1991 for good reviews). While much of this work describes the sam
OCR for page 223
STRATEGIC TASK MANAGEMENT 223 pling of visual channels (i.e., optimality in a perceptual sense), studies by Tulga and Sheridan (1980), Pattipati et al. (1983), and Moray et al. (1991) considered optimality of task selection using a paradigm in which tasks were represented as boxes moving across a computer screen timeline. Tulga and Sheridan's conclusions regarding the shortcomings in human anticipa- tion when tasks become too numerous are important, as is Moray et al.'s (1991) recent work examining performance when optimal schedules become too complex. However, it is not immediately clear how generalizable data are from paradigms in which tasks are visibly viewable and symbolically represented as boxes in a displayed queue, to those circumstances in which tasks may have various forms of visible or nonvisible evidence for their existence. That is, a waiting task may be an action to be performed that is not triggered by a perceptual event, thus introducing prospective memory (Meacham and Leiman, 1972) as an added (confounding) factor. Finally, there are a few studies that have examined task management strategies in complex environments with real tasks (Adams et al., 1991~. Puffer (1989) studied how students managed the completion of assigned tasks having varying attributes (boredom, difficulty) over the course of a semester. She found that earlier completion of tasks resulted in superior performance, and that more difficult tasks were completed later. However, the time scale of one course semester is far longer than the scale of minutes or a few hours involved in most operational tank engagements. Chou and Funk (1990) and Ruffle-Smith (1979) have examined task scheduling and shedding in high-workload aviation simulation environments, but their re- search did not extract general principles regarding the form taken by this scheduling. Two investigations by Wickens and his colleagues examined the degree to which aviators optimally employed task priority as a means of scheduling tasks. That is, were higher-priority tasks consistently performed over those of lower priority when a conflict of demands was present? In both studies, prioritization was examined in simulations in which sudden increases in workload were imposed by unexpected time pressure. To investigate these issues, Segal and Wickens (1990) required licensed pilots to fly a series of simulated low-level flight segments on a computer- based helicopter simulator. Subjects needed to perform tasks of three well- identified levels of priority: the primary task of flying an accurate altitude and navigating accurately, a secondary task of solving spatial location prob- lems, and a tertiary low-priority task involving discrete keyboard entry to be performed when time allowed. Collectively, the results of the experiment indicated that pilots per- formed reasonably optimally in their task management strategies. The pri- mary tasks of both vertical and lateral flight path control were protected from the degrading effects of higher-workload demands. The secondary
OCR for page 224
224 WORKLOAD TRANSITION side task was degraded in its accuracy (but not postponed) during higher- workload periods, while the tertiary housekeeping task was simply not per- formed (or performed less often) when workload was high. Still, two note- worthy departures from optimal performance were identified. First, when subjects were given the opportunity to reschedule performance of lower- priority tasks to avoid high-workload periods, they did not avail themselves of the opportunity to do so. Instead, they allowed performance on those tasks to suffer during high-workload periods. Secondly, there was some tendency to procrastinate in the performance of the discrete secondary and tertiary tasks; pilots delayed the performance of these tasks rather than performing them early in each flight. One noteworthy aspect of the experiment was observed in the benefits of preview. Those pilots who were given reliable information regarding the existence of upcoming workload transitions (increases or decreases) per- formed better on many aspects of the task ensemble than did pilots without such preview, a finding with relevance to the importance of reliable forecast ntormatlon. The value of previewed intelligence information in strategic task man- agement should not, however, always be accepted uncritically, particularly if the preview is unreliable. Some studies have shown that preview in task management and scheduling can be detrimental if it is not entirely reliable; or if it is presented in such a way that the visual attention required to integrate and interpret the preview information competes with processing information regarding present system status (Sanderson, 1990; Wickens et al., 1991~. Raby et al. (1989) also performed an aviation-related study of strategic task management in which pilots flew an instrument flight simulation through airport landings at low, medium, and high levels of workload. As in the Segal and Wickens study, continuous flight performance measures were recorded, as well as discrete tasks of various levels of priority, as assessed by an independent group of observers. As in the helicopter simulation study, here too pilots adhered reasonably well to the imposed priority scheme. However, their adherence was imperfect, as they did allow flight perfor- mance to suffer as workload increased but did not allow their performance on high-priority discrete tasks to deteriorate. In fact, the amount of time devoted to the latter tasks actually increased as time pressure (and workload) increased. Within the set of discrete tasks, pilots showed reasonable optimality, devoting progressively less time to lower-priority tasks at higher levels of workload. Also, in agreement with the results of the Segal and Wickens (1990) study, it was observed that subjects failed to reschedule those tasks of lower priority. They simply abandoned their performance or did them less accurately. Although the levels of stress in these two simulations imposed by time
OCR for page 225
STRATEGIC TASK MANAGEMENT 225 pressure and high workload were far less than that imposed in battlefield conditions, the simulations do suggest that appropriate task priorities can be maintained under sudden increases in workload. But they suggest a failure to appropriately reschedule when lower-priority tasks are performed around intervals of lower workload. Such a finding has implications both for the possible utility of task management training and the implementation of computer models of human task scheduling (Shanker, 1991), arguing that these sched- uling routines need not be too complex. Equally important are the implica- tions of the preview benefits observed by Segal and Wickens and reinforc- ing the points made in Chapter 3. The preview can be valuable, if it is reliable and well formatted. Yet it remains evident that full confidence in generalizing these results to operational environments awaits further valida- tion research in complex scenarios. REFERENCES Adams, M.J., Y.F. Tanney, and R.W. Pew 1991 Strategic Workload and the Cognitive Management of Advanced Multi-Task Sys- tems. Report No. CSERIAC-91-6. Wright-Patterson Air Force Base, Ohio: Crew Systems Ergonomics Analysis Center. Biederman, I. 1973 Mental set and mental arithmetic. Memory and Cognition 1:383-386. Chou, C.D., and K. Funk 1990 Management of multiple tasks: Cockpit task management errors. Pp. 470-474 in Proceedings of the 1990 IEEE International Conference on Systems, Man, and Cybernetics. Los Angeles, California: IEEE. Cumming, R.W., and P.G. Croft 1973 Human information processing under varying task demand. Ergonomics 16(5):581- 586. de Jong, R., M.G.H. Coles, G.D. Logan, and G. Gratton 1990 Searching for the point of no return: The control of response processes in speeded choice reaction time performance. Journal of Experimental Psychology: Human Perception and Performance 16(1):164-182. Easterbrook, J.A. 1959 The effect of emotion on cue utilization and the organization of behavior. Psycho- logical Review 66:183-207. Goettl, B.P. 1991 Gopher, D. 1982 Tracking strategies and cognitive demands. Human Factors 33(2):169-183. A selective attention test as a predictor of success in flight training. Human Factors 24(2):173-183. 1991 The skill of attention control: Acquisition and execution of attention strategies. In D. Meyer and S. Kornblum, eds., Attention and Performance, Volume XIV. Hillsdale, New Jersey: Erlbaum Associates. Gopher, D., and D. Navon 1980 How is performance limited: Testing the notion of central capacity. Acta Psychologica 46: 161-180.
OCR for page 226
226 WORKLOAD TRANSITION Granda, T.M., D.H. McClure, and J.W. Fogerty 1991 The development of an altitude awareness program. Pp. 47-52 in Proceedings of the Human Factors Society 35th Annual Meeting. Santa Monica, California: Hu- man Factors Society. Jersild, A.T. 1927 Mental set and shift. Archives of Psychology 89. Kleinmann, D. 1991 Models of attention control. In D. Damos, ea., Multiple Task Performance. Lon don: Taylor and Francis. Logan, G.D. 1981 Attention, automaticity, and the ability to stop a speeded choice response. In J. Long and A.D. Baddeley, eds., Attention and Performance, Volume IX. Hillsdale, New Jersey: Erlbaum. On the ability to inhibit complex movements: A stop-signal study of typewriting. Journal of Experimental Psychology: Human Perception arid Performance 8:778 792. On the ability to inhibit simple thoughts and actions: I. Stop-signal studies of decision and memory. Journal of Experimental Psychology: Learning, Memory and Cognition 9:585-606. 1985a Executive control of thought and action. Acta Psychologica 60:193-210. 1985b On the ability to inhibit simple thoughts and actions: II. Stop-signals studies of repetition priming. Journal of Experimental Psychology: Learning, Memory and Cognition 11:675-691. Logan, G.D., and J. Burkell 1986 Dependence and independence in responding to double stimulation: A comparison of stop, change, and dual-task paradigms. Journal of Experimental Psychology: Human Perception and Performance 12:549-563. Logan, G.D., and W.B. Cowan 1984 On the ability to inhibit thought and action: A theory of an act of control. Psy- chological Review 91:295-327. Logan, G.D., W.B. Cowan, and K.A. Davis 1984 On the ability to inhibit responses in simple and choice reaction time tasks: A model and a method. Journal of Experimental Psychology: Human Perception and Performance 10:276-291. Logan, G.D., B.H. Kantowitz, and G.L. Riegler 1987 On the Ability to Inhibit Selectively: A Model of Response Interdiction. Unpub- lished manuscript, University of Illinois. Logan' G.D., and N.J. Zbrodoff 1982 Constraints on strategy construction in a speeded discrimination task. Journal of Experimental Psychology. Human Perception and Performance 8 :502-520. Logan, G.D., N.J. Zbrodoff, and A.R Fostey 1983 Costs and benefits of strategy construction in a speeded discrimination task. Memory and Cognition 11:485-493. Logan, G.D., N.J. Zbrodoff, and J. Williamson 1984 Strategies in the color-word Stroop task. Bulletin of the Psychonomic Society 22: 135-138. Meacham, J.A., and B. Leiman 1972 Remembering to perform future actions. In U. Neisser, ea., Memory Observed. San Francisco, California: W.D. Freeman, Inc. 1982 1983 Moray, N. 1986 Monitoring behavior and supervisory control. Pp. 40.1-40.51 in K. Boff, L. Kaufman,
OCR for page 227
STRATEGIC TASK MANAGEMENT 227 and J. Thomas, eds., Handbook of Perception and Human Performance. New York: John Wiley. Moray, N., M.I. Dessouky, B.A. Kijowski, R. Adapathya 1991 Strategic behavior, workload, and performance in task scheduling. Human Fac- tors 33:607-629. North, R.A., and D. Gopher 1976 Measures of attention as predictors of flight performance. Human Factors 18:1- 13. Osman, A., S. Kornblum, and D.E. Meyer 1990 Does motor programming necessitate response execution? Journal of Experimen- tal Psychology: Human Perception and Performance 16(1):183-198. Pattipati, K.R., D.L. Kleinman, and A.R. Ephrath 1983 A dynamic decision model of human task selection performance. IEEE Transac tions on Systems, Man, and Cybernetics SMC-13(2):145-166. Puffer, S.M. 1989 Task-completion schedules: Determinants and consequences for performance. Human Relations 42:937-955. Rabbitt, P. 1989 Sequential reactions. In D. Holding, ea., Human Skills, Second Edition. New York: John Wiley and Sons. Raby, M., C.D. Wickens, and R. Marsh 1989 Investigation of Factors Comprising a Model of Pilot Decision Making: Part 1. Cognitive Biases in Workload Management Strategy. Report No. ARL-90-7/SCEEE- 90-1. Savoy, Illinois: University of Illinois, Aviation Research Laboratory. Ruffle-Smith, H.P. 1979 A Simulator Study of the Interaction of Pilot Workload Errors, Vigilance, and Decisions. Report No. NASA-TM-78472. Moffett Field, California: NASA Ames Research Center. Rubinstein, T., and A.F. Mason 1979 The accident that shouldn't have happened: An analysis of Three Mile Island. IEEE Spectrum 16:33-57. Sanderson, P. 1990 Knowledge acquisition and fault diagnosis: Experiments with PLAULT. IEEE Transactions on Systems, Man, and Cybernetics SMC-20(1):225-242. Sarter, N.' and D.D. Woods 1991 Situation awareness: A critical, but ill-defined phenomenon. International Jour- nal of Aviation Psychology 1 :45-57. Schachar, R.J., and G.D. Logan 1990 Impulsivity and inhibitory control in normal development and childhood psycho- pathology. Developmental Psychology 26(5):710-720. Segal, L.D., and C.D. Wickens 1990 TASKILLAN II: Pilot strategies for workload management. Proceedings of the Human Factors Society 34th Annual Meeting. Santa Monica, California: Human Factors Society. Shanker, R.C. 1991 Z-Scheduler: Integrating theories of scheduling behavior into a computational model. Pp. 1367- 1371 in Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics. Los Angeles, California: Institute of Electronics and Electrical Engineers. Spector, A., and I. Biederman 1976 Mental set and shift revisited. American Journal of Psychology 89:669-679.
OCR for page 228
228 WORKLOAD TRANSITION Sperling, G., and B.A. Dosher 1986 Strategy and optimization in human information processing. Pp. 2.1-2.65 in K. Boff, L. Kaufman, and J. Thomas, eds., Handbook of Perception and Human Performance, Volume I. New York: John Wiley. Stroop, J.R. 1935 Journal of Experimental Psychology 18:643. Tannock, R., R.J. Schachar, R.P. Carr, D. Chajczyk, and G.D. Logan 1989 Effects of methylphenidate on inhibitory control in hyperactive children. Journal of Abnormal Child Psychology 17 :473-491. Tsang, P.S., and C.D. Wickens 1988 The structural constraints and strategic control of resource allocation. Human Performance 1:45-72. Tulga, M.K., and T.B. Sheridan 1980 Dynamic decisions and workload in multitask supervisory control. IEEE Transac- t~ons on Systems, Man, and Cybernetics. SMC-10:217-232. Vallacher, R.R., and D.M. Wegner 1987 What do people think they're doing? Action identification and human behavior. Psychological Review 94(1):3-15. Vidulich, M.A. 1988 Speech responses and dual-task performance: Better timesharing or asymmetric transfer. Human Factors 30:517-5 29. Wickens, C.D., and D. Gopher 1977 Control theory measures of tracking as indices of attention allocation strategies. Human Factors 19:249-366. Wickens, C.D., and Y. Liu 1988 Codes and modalities in multiple resources: A success and a qualification. Hu- man Factors 30:599-616. Wickens, C.D., D. Pizarro, and B. Bell 1991 Overconfidence, preview and probability in strategic planning. Proceedings of the Human Factors Society 35th Annual Meeting. Santa Monica, California: Human Factors Society. Wickens, C.D., P. Tsang, and 13. Pierce 1985 The dynamics of resource allocation. In W.B. Rouse, ea., Advances in Man- Mach~ne Systems Research, Volume 2. Greenwich, Connecticut: JAI Press. Wiener, E.L. 1977 Controlled flight into te~Tain accidents: System-induced errors. Human Factors 19:171. Williams, H., M.P. Tham, and C.D. Wickens 1992 Resource Management in Aviat~on Incidents: A Review of the ASRS Data Base. Report No. ARL-91-5/NASA-91-3. Savoy, Illinois: University of Illinois, Avia- tion Research Laboratory. Zbrodoff, N.J., and G.D. Logan 1986 On the autonomy of mental processes: A case study of arithmetic. Journal of Experimental Psychology: General 1 15: 1 18-130.
Representative terms from entire chapter: