To what extent can people successfully multitask

Multitasking in realistic situations: effects on performance, subjective perception and physiological parameters

Summary A number of studies show that performance deteriorates when people “multitask”, i.e. work on two or more tasks at the same time. Often unrealistic reaction time tasks are used in corresponding investigations. In the present study, the effects of multitasking were examined using two realistic scenarios. Changes on three levels were checked: performance values, subjective experience and physiological parameters. The data were also analyzed for possible gender and age differences. Since possible changes in working age were in the foreground, test subjects between 21 and 60 years of age were examined. It was shown that double tasks lead to a loss of performance and increased subjective tension. Changes in physiological parameters seem to depend on the nature of the task. There were neither gender differences nor increasing deterioration with age. Overall, the data suggest that multitasking should be reduced in work activities, at least for certain tasks, in order to avoid errors and possible accidents. Longer-term effects of multitasking on performance and health should be the subject of further research. keywords

· Multitasking

· Realistic situations

· Gender differences

· Age differences

· multitasking

· Real-world scenarios

· Gender differences

· Age differences

The term “multitasking” is currently being discussed in many media. It is an essential feature of the modern world of work1. But is multitasking, i.e. the simultaneous processing of two or more tasks, even possible? Which performance losses and psychological stress can this lead to? Can women “multitask” better than men? Does the ability to multitask with age decrease? All of these are questions that are relevant to occupational safety.

The simultaneous execution of two tasks has often been investigated in basic psychological research with the help of reaction time tasks. It has been shown here that performance usually worsens if you work on two tasks at the same time: The reaction time to a second stimulus is significantly longer if it is presented (very) quickly after the first stimulus2. If such a disturbance occurs, one speaks of “dual task interference”. Research into the causes of this interference is an important topic in basic research. It is believed that there is a bottleneck between the perceptual and motor processing stages. Where exactly this bottleneck is to be located has not yet been conclusively clarified.

Multitasking in the world of work can lead to errors, to overlooking important information as well as to psychological stress.

Presumably, both input processes (perceptual encoding) and output-related processes (reaction selection and initiation) are affected2. Regardless of where there is a bottleneck in the manufacturing process, it should be noted that the performance usually deteriorates with double tasks. However, the delay that arises when working on tasks at the same time can be reduced through practice3, 4. However, a delay effect usually persists even after intensive practice. The strength of an exercise effect depends on the type of task and on the subject himself.

Outside of basic research, which deals with the causes of double-task interference, there are still few systematic studies of realistic multitasking situations5. One of the few everyday situations that has meanwhile been examined more frequently is driving a car while using the phone at the same time (cf. 6). Overall, these studies also show that driving performance deteriorates. Ophir et al.7 examined "media multitaskers", i.e. people who use more than one medium at the same time. For the investigation, it was first necessary to identify people who are very strong media multitaskers and those who are only to a small extent. For this purpose, people were first asked how many media they consume on average at the same time, for example they watch videos, click on graphics, look at texts etc. at the same time. Then those people were selected who use a lot of media at the same time and those who use less media at the same time than average . The subjects were checked to see how good they were in “cognitive control”, i. H. how well you can intentionally pay attention to certain elements while blocking out other unimportant elements. It was shown that the strong “multitaskers” (so-called heavy media multitaskers, HMM) are less able to block out irrelevant stimuli than the below average “multitaskers” (light media multitaskers, LMM). Another experiment showed that the HMM are less able to switch between tasks than the LMM. Those who practice media multitasking less often are better able to voluntarily focus their attention on certain aspects, even if they are distracted by distractors. It has not yet been clarified whether the poorer performance of the HMM is a result of multitasking or whether people who are generally poorly able to concentrate on one thing tend to multitasking in the media. If it turns out that the poorer performance is a result of multitasking, this means for the future that people will find it increasingly difficult to concentrate on important elements.

So far, there is no clear evidence of gender differences in multitasking. For a long time, the question of gender differences with simultaneous task processing has not been investigated, so that no reliable statements on this question have been possible so far. However, initial studies indicate that there are no gender differences (e.g. 8).

Research on age differences usually shows that multitasking performance deteriorates with age. However, the extent of the deterioration depends on the nature of the tasks. For example, the deterioration in automatic processes is less than, for example, in difficult motor task components9. However, many studies that deal with age effects only compare two groups with each other, with one group mainly comprising 20 to 30 year-olds and the other group mainly over 70-year-olds (e.g. 10, 11). But there are also studies that compare several age groups with one another. Crook et al.12 conclude that there is a 5–12% deterioration in performance per decade when performing tasks at the same time. However, this deterioration is not always present in the middle age range, i.e. between 40 and 60 years, but seems to depend on the type of performance.

A distinction that is seldom made, but is of great relevance when it comes to multitasking, is that between tasks that require conscious attention and tasks that are automated. Automated activities, i.e. activities that have been learned and are then carried out very frequently, such as walking, require little attention or a low allocation of cognitive resources13. With this type of task, other activities, such as speaking, can be carried out at the same time, as cognitive resources are available for this. However, the less activities are automated and the more conscious attention they require, the less they can be carried out at the same time. This applies particularly to activities that are critical to errors or to activities where the consequences of possible errors are particularly negative.

Overall, it can be said that the simultaneous processing of two or more tasks generally leads to a reduction in performance or an increase in the number of errors, which can only be compensated to a limited extent through practice. However, the degree of deterioration depends on the type of tasks and is influenced by age and possibly gender. However, clear age effects only appear to be evident from an older age of around 70 years.

A corporate culture should be promoted that does not recognize permanent multitasking as a special achievement.

The effects of multitasking on health protection have not been considered until recently. For this reason, two realistic scenarios were used in the present study, in which performance data, subjective tension and heart rate were collected. In addition, possible gender and age differences were checked. The age range in which people are generally employed, i.e. 20 to 60-year-olds, was of interest here.

To study the effects of multitasking, two tasks were used. This involved a driving simulation on the one hand and an office job on the other.

A so-called "lane change task" was used as the driving task (Lane Change Task14). This is a driving simulation that is presented on the PC and in which the lane has to be changed on a three-lane road after appropriate cues (Figure 1). Before starting the actual task, the test subjects could practice the task until they had mastered the operation. After that, the person had to drive once for 3 minutes without any further tasks and once for three minutes with additional tasks. At the same time as the driving task, the person had to enter a telephone number into a cell phone one after the other, take a paper handkerchief out of the pack, take the counted money out of a wallet and read out directions.

In the “office work”, the test subject was shown one word on a screen and had to decide at the push of a button whether the word contained no, one or two spelling errors (Figure 2). The test subject was initially able to practice the task using 10 examples. Then she had to work through 70 words in two blocks, one after the other. In one of the blocks, the person had to perform a second task at the same time. A spoken text was played to her through headphones. She was instructed to pay attention to this text as she was asked questions about it after the block. It was a text on a shopping situation that lasted 70 seconds. Nine questions were asked about this text. A preliminary study showed that an average of 7 questions could be answered correctly if only the text was played to one.

The heart rate was recorded while the task was being processed. The subjective perception was recorded with the help of a scale from “relaxed” to “tense” according to the individual task blocks. The sequence of tasks and the conditions within a task were balanced across the test subjects.

32 female and 32 male people between the ages of 21 and 60 took part in the study. The mean age was 39.8 years. There were four age groups (21-30 years, 31-40 years, 41-50 years and 51-60 years) each of which consisted of eight subjects, i.e. four women and four men.

Lane change task
The lane deviation increased significantly from 0.76 meters without additional tasks to 1.2 meters with additional tasks (t (63) = –7.4, p<.001). auch="" die="" subjektive="" anspannung="" stieg="" auf="" einer="" 10-stufigen="" skala="" signifikant="" von="" 2,9="" auf="" 4,8="" (t(63)="-8,8"><.001). abbildung="" 3="" zeigt="" die="" mittelwerte="" für="" die="" zwei="" bedingungen.="" auch="" die="" herzrate="" unterschied="" sich="" signifikant="" zwischen="" den="" beiden="" bedingungen.="" sie="" stieg="" von="" 80,5="" schlägen="" pro="" minute="" ohne="" zweitaufgabe="" auf="" 86,1="" schläge="" pro="" minute="" mit="" zweitaufgaben,="" (t(51)="–3,9,"><.001). varianzanalysen="" zeigten,="" dass="" es="" weder="" unterschiede="" zwischen="" den="" geschlechtern="" noch="" zwischen="" den="" altersgruppen="" gab.="">Office task
The number of correct, incorrect and omitted reactions did not change between the two conditions in the spelling task. Without or with an additional task, there was an average of 55.5 versus 55.3 correct reactions, 12.2 versus 12.3 incorrect reactions and 2.4 versus 2.3 omitted reactions. The reaction time also did not change between the two conditions. It was 2,645 ms for the individual condition and 2,666 ms for the additional task. With regard to the additional task, however, there was a clear difference. Of the 9 questions that related to the text heard, only 2.27 could be answered correctly on average. In a preliminary study in which only the text was presented, the number of correct answers was 6.98 (n = 40). The difference is significant (t (102) = 13.55, p<.001, abbildung="" 4).="" die="" subjektive="" anspannung="" stieg="" auf="" der="" 10-stufigen="" skala="" signifikant="" von="" 3,4="" ohne="" zusatzaufgabe="" auf="" 4,3="" mit="" zusatzaufgabe="" (t(63)="-5,6"><.001. abbildung="" 4="" zeigt="" die="" mittelwerte="" für="" die="" zwei="" bedingungen.="" die="" herzrate="" unterschied="" sich="" nicht="" zwischen="" den="" beiden="" bedingungen="" (82,5="" schläge="" pro="" minute="" ohne="" zusatzaufgabe="" vs.="" 82,2="" schläge="" pro="" minute="" mit="" zusatzaufgabe).="" varianzanalysen="" zeigten="" keine="" unterschiede="" zwischen="" den="" geschlechtern="" und="" lediglich="" einen="" signifikanten="" leistungsunterschied="" bei="" der="" anzahl="" richtiger="" reaktionen="" abhängig="" vom="" alter.="" post="" hoc="" tests="" zeigten,="" dass="" sich="" die="" 41-="" bis="" 50-jährigen="" signifikant="" von="" den="" 31-="" bis="" 40-jährigen="" und="" den="" 51-="" bis="" 60-jährigen="" unterschieden.="" es="" handelte="" sich="" also="" nicht="" um="" eine="" stetige="" veränderung="" mit="" zunehmendem="" alter.="">discussion
The aim of the present study was to investigate to what extent the topic of “multitasking” is also relevant for occupational safety. For this purpose, two realistic scenarios were used in which the performance, the subjective feeling and the heart rate were measured once without and once with secondary tasks. The data were checked for age and gender differences. It was shown that multitasking in the driving task led to significantly poorer performance, higher subjective tension and a higher heart rate. There were no age or gender differences between the conditions. In the office task, there was a significant deterioration in performance as well as an increase in subjective tension in the additional task. The heart rate did not change. There were no gender and almost no age differences in this task either.

With regard to possible gender differences, the data show that one cannot generally assume that men and women differ in their ability to multitask. This can of course also depend on the type of task and should be checked in further research. The study also shows no age differences for the 20 to 60 year olds. An increasing deterioration in performance does not seem to set in until around the age of 70. So there does not seem to be any difference in the age at which people normally work.

Overall, it can be said that multitasking can lead to errors and psychological stress in realistic situations and thus presumably also in the real world of work. This not only entails economic damage, but also has implications for the safety and health of employees.

However, the available data show that work situations must be viewed in a differentiated manner. The heart rate during the office task remained constant under the different conditions. It can be assumed that the test persons ignored the second task despite instructions to the contrary, i.e. ignored the spoken text. The test persons did not experience the multitasking situation as such and therefore there was no greater psychophysiological stress, which was measured during the entire task processing. However, at the moment when the people realized that they could not answer the questions about the text, the subjective tension increased, which was only measured at the end of the task.

Errors that arise through multitasking can also lead to accidents and corresponding consequences for health in the real world of work. Mental stress that can result from multitasking can also have a negative impact on health. Multitasking can also have long-term effects, as the study by Ophir et al.7 suggests: If the ability to concentrate decreases with increased media multitasking, this generally leads to poorer concentration performance and, accordingly, to more errors. On the other hand, there is also research showing that multitasking performance improves with practice.

When dealing with the question of how to handle multitasking, there are two points to keep in mind:

1. Automated vs. non-automated, i.e. controlled activities

2. possible consequences of mistakes

Since automatic processes, such as walking, require little or no attention, cognitive resources are available for other activities such as speaking. However, this only applies as long as it is a level, easily visible route. At the moment when the path leads through rough terrain, walking requires more attention and less is available for other activities. Driving is also automated to a certain extent. It could therefore be argued that while driving, cognitive resources are available for other activities, such as making calls. This is true to some extent, but only until a situation arises that requires more attention. When talking to a passenger in such a situation, the driver falls silent in order to focus his entire attention on the traffic. However, the person on the phone cannot observe the situation and tries to continue the conversation. Such situations cannot be foreseen in traffic and can have fatal consequences. This is an essential difference to the situation of walking. An accident can happen here too; however, it can be assumed that the possible consequences of an accident are not fatal.

The conditions under which multitasking is permitted and under which conditions it should be avoided in any case depend on the tasks and the possible consequences. The more cognitively demanding tasks are, the less they can be carried out at the same time.There are already first approaches attempting to determine the extent of cognitive strain for certain types of tasks15. Further research would be desirable here.

Companies should determine which activities should definitely not be multitasked. However, simple appeals to employees are probably not enough. In a study, Lehle et al.16 showed that people experienced less subjective exertion when processing tasks in parallel than when processing them in series. However, the performance data clearly showed poorer parallel processing performance. The subjects may have had the illusion that they were better at parallel processing. In the case of critical activities, multitasking should therefore be made more difficult by appropriate work organization. If this is not possible, measures must be agreed with the employees. "Medial multitasking" in particular, i.e. reading e-mails, making phone calls and following a presentation at the same time, is not always actually necessary. However, it can make the individual feel more important. A corporate culture should also be promoted that does not recognize permanent multitasking as a special achievement.

Overall, it has been shown that multitasking in the world of work can lead to poor performance and poor psychological stress. For this reason, it should be observed in occupational safety. It should definitely be stopped, especially in activities where multitasking can have a negative effect on performance and health.


1. Freude G, Ullsperger P. Interruptions at work and multitasking in the modern world of work - concepts, effects and implications for work design and research. Zbl Arbeitsmed 2010: 60: 120-127

2. Koch I. Mechanisms of interference in double tasks. Psychological Review 2008; 59: 24-32

3. van Selst M, Jolicoeur P. Decision and response. Cognitive Psychology 1987; 33: 266-307

4. van Selst M, Ruthruff E, Johnston JC. Can practice eliminate the psychological refractory period effect? JEP: HPP 1999; 25: 1268-1283

5. Kiefer J, Schulz M, Schulze-Kissing D, Urbas L. Multitasking strategies in human-machine interaction. MMI interactive 2006; 11: 26-42

6. McCartt AT, Hellinga LA, Braitman KA. Cell phones and driving: Review of research. Traffic Injury Prevention 2006; 7: 89-106

7. Ophir E, Nass C, Wagner AD. Cognitive control in media multitaskers. PNAS 2009; 106: 15583-15587

8. Charron S, Koechlin E. Divided Representation of Concurrent Goals in the Human Frontal Lobes. Science 2010; 328: 360-363

9. Riby LM, Perfect TJ, Stollery BT. The effects of age and task domain on dual task performance: Ameta analysis. European Journal of Cognitive Psychology 2004; 16: 863-891

10. Voelcker-Rehage C, Alberts JL. Effect of motor practice on dual-task performance in older adults. J Gerontol B Psychol Sci Soc Sci 2007; 62: 141-148

11. Priest AW, Salamon KB, Hollman JH. Age-related differences in dual task walking: a cross sectional study. Journal of NeuroEngineering and Rehabilitation 2008; 14:29

12. Crook TH, West RL, Larrabee GL. The driving-reaction time test: assessing age declines in dual-task performance. Developmental Neuropsychology 1993; 9: 31-39

13. Anderson JR. Cognitive psychology. Spectrum, Heidelberg, 1996

14. Mattes S. The Lane Change Task as a tool for driver distraction evaluation. In Strasser H, Kluth K, Rausch H, Bubb H (Eds.). Quality of Work and Products in Enterprises of the Future 57-60. Ergonomia Verlag, Stuttgart, 2003

15. Strater O, Bubb H. Design of Systems in Settings with Remote Access to Cognitive Performance. In Hollnagel E. Suparamaniam N (Eds.). Handbook of Cognitive Task Design 333-366. Lawrence Erlbaum, Hillsdale, 2003

16. Lehle C, Steinhauser M., Huebner R. Serial or parallel processing in dual tasks: What is more effortful? Psychophysiology 2009; 46: 502-509

October 13, 2010