Habitual Multitaskers Do It Badly 386
iandoh writes "According to a group of Stanford researchers, people who frequently multitask don't pay attention, control their memory or switch from one job to another as well as those who prefer to complete one task at a time. In other words, multitaskers are bad at multitasking. The research team is also studying how to design computer voices for cars that result in safer driving."
Reader AliasMarlowe adds "The comparison involved multitasking with a number of attention or context related tests. For the study, multitasking was defined as consuming multiple media sources at once — gaming, TV, IM, email, etc. Interestingly, the habitual multitaskers were much worse at multitasking than the single taskers in these relatively straightforward tests. In self-assessment the multitaskers considered themselves good at it and the single taskers considered themselves bad at it. An extreme case of the Dunning-Kruger effect, perhaps, with consequences for business and society."
When I multitask... (Score:5, Interesting)
Re:Makes sense (Score:4, Interesting)
Humans Can't Multitask (Score:3, Interesting)
Multitasking in humans is a myth. You might be able to rapidly switch between tasks, but processing more than one thing simultaneously can't be done.
Re:Makes sense (Score:5, Interesting)
Conversely, I believe that being forced to multitask by my environment has created attention deficit disorder in me. I can't pay attention to things like I used to, and staying focused is very difficult for me. Even if NOTHING is demanding my attention, I feel like I have a compulsion to switch to a different task every few minutes. It's horrible. I used to be able to focus on a single task for long stretches, sometimes I could read a book for 14 hours or more in a day if I was sufficiently interested in it. Now, every three paragraphs or so, I feel like I want to check my email.
Re:Texting while driving (Score:3, Interesting)
Texting while driving is STUPIDLY UNSAFE and I only do it when I feel the situation has appropriate trade-offs (no cars close or medium-close in front of me, no turns in the road, importance of sending the text, etc.)
According to my wife (Score:2, Interesting)
It's a productivity killer! (Score:2, Interesting)
Re:Humans Can't Multitask (Score:2, Interesting)
Targeted test, towards a targeted response. My wife and I get in this discussion all the time. I prefer to watch TV, play with my cats, and work/play on my computer at the same time. I typically work best like this. The key is, as people have mentioned, multitasking is very much like a computer. Each device/sense/ability are only capable of one process at a time, however, typing, breathing, sitting upright, listening to music, do not conflict with each other. If I were to add in anything that took away from one of the others, I would be unable to maintain.
The above test measured the wrong information. It was looking for cognitive multitasking. I seriously doubt it is possible for a human to cognitively multitask. We can hear and parse multiple conversations (I would have no idea of how many the average person can handle, but I peak out at about 3, and I am not very good at it), but we can only maintain one conscience stream of data at a time. This means, you have a conversation with someone, you may be able to keep track of what someone else is saying, but you are not likely, and definitely not proficiently going to be able to carry out a conversation with another. You may be able to hear what they say (one of the ways we can switch conversations and topics in conversation, the above mentioned ability to parse multiple conversations). But each time you switch between conversation, you have to break your stream of thought, interject the new conversation, and carry on with it. Then you switch back, but it's not multitasking, it's more in line with serial processing.
So, can people multitask, you have to, in order to function. But, to what level, and how do you do it is a different question (or 2).
Personally, I will listen to music, or watch TV while programing because it helps me focus on programming. It gives me a constant stream of data to keep my other senses busy while I focus my train of thought on what I am writing/designing. However, when the wife stops in, and sees me doing this, and deciding I can handle a conversation everything goes south, and a fight will typically break out.
Re:Multitasking just has to be done properly (Score:3, Interesting)
It's been shown that the average attention span runs about 20 minutes. After that, you _will_ lose the ability to concentrate and your mind will naturally wander
It's also been shown that it takes about 40 minutes to enter flow, at which point your productivity increases. Somehow, these seem contradictory...
Re:From the people who brought us clippy (Score:2, Interesting)
Your referenced intrigued me, so I looked into Cliff Nass. There's a very interesting interview [adlininc.com] with him where he talks about both Microsoft Bob and Clippy. While he defends Bob (and I do see his point), he freely admits to the problems with Clippy--and better--explains why it failed. He seems to be well-respected in the industry for his contributions to the social-aspects of software design.
I don't know if your intention was to dismiss the research because of Cliff Nass, or if it was just to poke fun, but one might not want to dismiss this research just because Cliff Nass is involved.
Re:But it looks so good on my résum (Score:3, Interesting)
I don't attribute this to malice.
Ask someone: Which would you prefer: that you candidate be good at doing only one thing at a time, or that he be good at doing many things at a time? What do you think their answer is likely to be? More is more, isn't it?
The problem is, the requirement going out doesn't get compared against the capabilities of actual working people.
How it should go is like this:
PHB: Find me someone who's good at doing lots of things all at once.
HR: Can't; Tried, Looked, none exist. People are better at doing one thing at a time.
PHB: OK, let's set up our processes so that people only do one thing.
How it actually goes:
PHB: Find me someone who's good at doing lots of things all at once.
HR: These candidates who applied all say that they can do this. Interview them and pick the best one we can afford.
PHB: OK. [Hires one]
[Weeks, months, or years pass]
PHB: Find me someone who's good at doing lots of things all at once. The last guy wasn't as good as he said he was, and burned out.. Now there's even more things that need to be done.
HR: OK. Here's another pool of applicants.
And the thing is, I am good at doing a lot of different things. This makes me valuable, because I'm flexible and also because I can put my various skills together to better effect than I could if I only knew how to do one thing. That doesn't mean that I should be actively involved in several ongoing projects all at the same time.
I can't work well if I'm expected to work on several development projects concurrently, while at the same time supporting the production environment, all while thinking about our existing processes and policies and trying to think of better ones, and keep up with the constant changes with the existing ones that come from my supervisor or higher-ups on a day to day basis. And so, I don't work well. But I do manage to meet expectations. It's just that the expectations are managed downward to where we don't really expect all that much.
Re:Makes sense (Score:3, Interesting)
I just plain don't have the time to focus on any one thing for that long.. but anyway I digress:
The same way a computer multi-tasks is exactly why I find this study is flawed. The preface the study by saying that they were trying to find out why multi-taskers could do so so well. They then threw out this premise by saying that these multi-taskers single threaded performance was low while forgetting that their group in question was known to be good at multi-tasking.
I'll use, for example, a box I'm currently beating up for performance testing. Nice spanking new Nehalem based dual-quad w/ 48GB ram. When in Hyper-threaded mode the per-logical-core performance goes down by a significant amount (say 25%) but you have double the logical core to work with SO if you have a single threaded application to run you will do better on the non-HT mode but an application that can multi-thread well will do better on the HT mode.
I see similar situations here: Aside from questioning their test group (there is a big difference between someone with ADHD and say your average /.er who multi-tasks like most people breathe {(c)The Core ;)} but anyway.. I agree that when I multi-task my per-task focus goes down a measurable amount BUT as long as I add in some protection routines to make sure that reduced performance != reduced accuracy I am able to accomplish more / unit time than someone who can only do one thing at a time even if they can maybe do a single task a bit faster than I can.
In my current job I find it impossible to NOT multi-task and, given the large amount of distraction coming in, someone who Can't multi-task will suffer because they are not allowed the single threaded environment they need.
Re:One look at our faces tells us a lot (Score:3, Interesting)
Tho observationally, prey animals are even worse at multitasking (ie. quick task switching). Most literally can't chew and scan for predators at the same time, and tend to panic each and every time they're forced to task-switch.
Re:One look at our faces tells us a lot (Score:3, Interesting)
Why do you think the human mind isn't a computer?
The means and methods we use to compute and calculate numbers are all derived from what the mind does. The way we sort and associate data, link information, make comparisons and all the things that computers do were essentially designed after the human mind. We can't say that a computer is a human mind, but we can say that we are continuing to develop technologies "in our own image" and that our own means and methods have a great deal in common with computers as a result.
But I will stop applying computing methodology to the human mind when the computers are not designed after human methodologies.
Re:Yes and no (Score:3, Interesting)
Analysis depends on where you parse your task units. Walking miles while focusing intently on the scenery and composing a complex poem and committing it to memory can be two separate tasks, competing for attention, or they can be one single task. Making them one task was the poetic practice of Wordsworth as well as of Wallace Stevens, Basho, and Gary Snyder. Similarly listening to the melody line of a single instrument may be one task, and listening to each of the other instruments in an ensemble separate tasks, competing. Or they can all be aspects of one task, focusing on the whole of the performance while aware of each instrumental line.
Similarly driving and listening to music can be two tasks, competing for attention. Or they can be one task, where the music and road melds to a single experience. Driving, while listening to music, while holding a conversation, while composing a poem can be separate tasks, dividing the attention, or one task, unifying it. The analyses of people driving poorly when on a cell phone suggest the problem is that people create in their minds two spaces, one of the conversation, one of the road. We might all tend to do that. But is it the only way? Can we blend multiple strands into the same space such that, rather than competing and conflicting, they create a harmonious whole? Because it seems it was precisely from such a creation that Wordsworth's poetry flowed.
To what degree can we take a variety of activities and make a single task of them? Is it multitasking any more when we succeed?
Where's the multi-tasking? (Score:2, Interesting)
So the tests the experiments used were:
the groups were shown sets of two red rectangles alone or surrounded by two, four or six blue rectangles. Each configuration was flashed twice, and the participants had to determine whether the two red rectangles in the second frame were in a different position than in the first frame.
After being shown sequences of alphabetical letters, the high multitaskers did a lousy job at remembering when a letter was making a repeat appearance.
The test subjects were shown images of letters and numbers at the same time and instructed what to focus on. When they were told to pay attention to numbers, they had to determine if the digits were even or odd. When told to concentrate on letters, they had to say whether they were vowels or consonants.
Given three single tasks, they found that "light multitaskers" performed better than "heavy multitaskers." Why is this surprising?
Re:Texting while driving (Score:4, Interesting)
I hate you all, you fucking phone drivers. Get off your fucking phones and out of my damn lane. YOU are the reason that it is such hell to drive now. YOU are the reason there are so many wrecks and red light running. YOU are the reason that so many lives are lost and everyone's insurance is so high. Hang the fuck up.
Yes, because as you can see, the annual number of accidents per vehicle miles traveled has gone up in direct proportion to saturation of the cell phone market.
Except... not. No, accident statistics have stayed pretty darned flat with respect to VMT (which continually goes up, year after year). The severity of injuries and incidence of fatalities goes down as new innovations in passenger safety come out and are implemented in the fleet.
BTW, one thing that has stayed VERY constant, for the last 30+ years: half of road fatalities are caused by a drunk driver. Even though road fatalities have gone down (even as the total population and per capita VMT have risen), 50% are from accidents caused by drunk drivers. You'd think we would have learned better by now, but nooooo.
So it'd be dead easy to determine, once and for all, what effect cell phone use has on driving: run a multiple-regression analysis on accident rates, taking into account VMT per capita, total population, and other such stats that we know influence accident rates, plus add cell phone market penetration over time. Look at data for the last, say, 20 years. That will show you how much cell phones (remember when they were all "car phones"?) impact accident rates. Until someone does this (law enforcement has all the accident data, and I'm sure the cell companies would cough up the subscriber numbers if it meant the possibility of getting all these laws against using their product repealed), everyone needs to calm the f*** down and realize that there are good drivers, and bad drivers, and mediocre drivers, and if it wasn't the phone, it'd be something else distracting the a**hole in front of you.