Multitasking Doesn't Work
health.clevelandclinic.org
health.clevelandclinic.org
> Other studies suggest that people who frequently “media multitask” (like listening to music while checking email or scrolling through social media while watching a movie) are more distracted and less able to focus their attention even when they’re performing only one task.
This was novel to me. Listening to music negatively affects your ability to focus? I guess it's time to throw out my headphones...
Well, but that article links to "Multicosts of Multitasking"[1], which references studies made on "heavy multitaskers", defined as having high "MMI" scores. What's MMI?
"A high MMI score means an individual engages in a lot of media multitasking (e.g., checking email while also perusing Facebook and watching Netflix), and a low score means he or she does not (e.g., checking email without any secondary media)."
Whoa there - checking Netflix and Facebook are completely different than listening to music, and it should be obvious that the former two are mutually exclusive with performing other tasks (simultaneously) - but that doesn't imply the same for music.
I think that the HN submission page might be making some unsupported claims...
I think I'm convinced that the difference is how well I know the music - listening to the radio is all songs that I've heard hundreds of times, so I don't think I'm subconsciously spending effort listening to the lyrics since I know them already?
Where do you find them?
that's the key. You are possibly using 'the right song' to isolate yourself, so that helps you concentrate. The fact that it's on repeat also makes you don't pay much attention to it anymore. But try to listen to 'the wrong song', or worse 'the wrong new song', and it's probably going to have some effect.
I mostly listen to 'concentration music' and lo-fi without vocals. Put something more complex and with lyrics...
First, the point of a study is to be generalizeable - that is, to apply to as many people as possible, usually other than oneself. If a particular technique applies to you, that's great! However, most people aren't you. Starting with a known general principle/technique (most people can't multitask) will, on average, put you in a better position than randomly picking someone's anecdotal evidence.
Second, there are many documented cases where self-assessment of one's own traits or capabilities is inaccurate. Performance while cognitively impaired (alcohol & sleep deprivation), assessment of skill as a novice (Dunning-Kruger), and the ability to multitask are all well-known cases of it, with many more lesser known (e.g. whether one is a "visual" or "verbal" learner[1]) - so, you should be wary about having too much self-confidence in your own capabilities, especially for something as poorly-understood as software engineering.
That's why I specified myself. I completely understand the difference between generalizable results and individual results.
> many documented cases where self-assessment of one's own traits or capabilities is inaccurate
I completely understand that self-assessment is flawed, but there is also something to understanding oneself and trusting a self-assessment that is born out over decades of experience. I have never even come close to being as productive as I can be with the right musical piece and sharing this insight with others may prompt further research that sheds new nuance on the topic.
Trance is distracting for me, but melodic progression works well for me, and sometimes vocal chillout trance. I still have to process random stuff with background noise from an urban setting, even on one of those noise generators.
Back when I was living in a more rural setting, the song birds, the wind between the trees, the rain makes for something that is easier for me to focus.
That is what works for me. My wife has difficulty concentrating without Netflix in the background.
Music helps me focus more than say, a noisy coffee shop.
For activities that require more intellectual attention, the articles assertion would be accurate to me personally.
I know lots of people though who must have music on, or they simply cannot concentrate. For them, it might be that the music gives the mind something to hold on to, kind of a rail to help stay on track, and in the absence of that, their mind would wander farther afield.
When I really get focused on a bit of work, whatever music is playing literally disappears - if you came to me 3min later and offered me $millions to tell you what song just played, I'd get nothing.
Cats are even more amazing - I can't find it right now, but I remember in reading studies in some of my college neuroscience classes where electrodes were put on cats to track their audio system with a clock/metronome sound in the background. You can trace the signal processing up through various nodes of the system at successively higher levels. When a mouse was let into the area, the metronome click trace disappeared from the data of almost every level of processing - it was entirely filtered out almost at the raw audio stage when the mouse showed up. That's focus.
The way it feels to me is it gives me a consistent background to work. For example, once I have decided on a particular way to solve a problem and its time to code - I put a long DnB track.
Another way to describe it is that it gives my brain's CPU a consistent clock, so to speak xD
For most tasks, lyric free chill hop is a perfect balance. It replaces the chaos of my environment with a predictable backdrop that can fade out of focus.
But for tasks that require extreme focus, like putting together a novel (to me) algorithm or exploring a new code base, even chillhop is too much. I have to find a place with absolute silence - anything that grabs my attention is devastating to the task at hand.
it's your familiarity with it.
The human voice is highly distracting for most people.
For the latter example, I wonder how much of it is though neaural networks wired together fire together. In essence Ive always listened to music while riding so it feels right and known, what if I had ridden all this time without music, would I be better? I don't know.
I’ve tried it a couple of years ago, and for me it was mostly repeating tunes and nothing particularly good for programming/focus work.
As I understand it, the relevant studies do not support this.
For me it feels more about familiarity. If I already have the lyrics memorized then it does not feel distracting whereas unfamiliar music can draw my focus.
For understanding a new concept or problem the finding here probably applies.
Perhaps the following is too personal of a take, but we're talking about music YMMV ;-)
Here it comes:
Music doesn't make me feel lonely when I'm behind a computer. And feeling lonely is what made me into a mediocre programmer. For years I've not touched a computer because it just feels so isolating when you do it for 8 to 10 hours straight.
But with music, with music I feel a shared experience between me, the artist, the feelings and the concepts. It helps that this form of "musical reasoning" doesn't cost me a lot of brainpower since I've been doing this on autopilot since I was a kid.
I've got about 50 albums that I rotate through while I'm coding and I can focus on the task at hand fairly well with those on, but if I put on a new album by an artist I like, my ability to concentrate on code goes way down.
Listening to music is bad if you're in city traffic and need to use all your attention, but good if you're on a country road and risk just losing focus.
Or how does an open office compare to listening to music?
For me I sometimes use music to mollify ruminative thoughts or feelings that intrude and distract me. Which results in a net increase in focus. Similar for the din of the open office.
When I need deep thinking time both background noise and music are too interruptive or distracting and I need full silence. But I don’t require that level of focus all the time.
I can't. Especially not with repetitive, melodic music like other siblings say helps them. That's an absolute killer as I starts repeating the melody in my head. Potentially even after I shut the music itself off.
My suspicion is that many people have too many other distractions and music helps them because it's less distracting than the other things. Some people say white noise, such as from an AC helps them sleep. It drives me crazy and I can't fall asleep at all. It does beat a busy street with cars zooming by, revving, honking etc.
Where music can help if there's a mundane repetitive task I need to just get done, because automating it does not make sense at that point. Certain music helps. Allegedly farmers use the same technique to make strawberry pickers faster.
Have you never heard of or witnessed people that turn off their car stereo when they need to drive backwards?
Your TV example I think shows some of this very nicely. You are able to drown out the TV in your head but you aren't able to actually multitask. If you were, you'd know what went on in the show and also get your task done productively. Unless the TV itself drowns out other more distracting things, just not having the TV on would probably be better for concentration.
YMMV in case you are so used to all the noise that you need "something there" because otherwise you get say tinnitus.
I do the TV thing too sometimes but inadvertently. I watch TV, am a little bored perhaps and start say reading HN and at some point notice that the show is over. I totally drowned out the show.
Music, to me, seems to be the perfumed room deodorizer of noisy environments - it drowns out something that’s a problem rather than lending concentration. I understand why people use it in open workspaces.
When running or exercising music is great because it distracts the conscious mind and lets you avoid getting lost in thought. I think this is the same task it serves when, for example, coding. But hard tasks require that otherwise distracted mind and that’s where it falls apart.
But obviously YMMV.
Not so fast.
Actually, listening can help you focus IF it helps you block some other distraction AND you choose something that you can stop focusing on.
When I worked at the office I used headphones whenever I wanted to block off chatter around me.
I also have tinnitus -- light music helps me mask it so it stops being distracting.
https://www.nature.com/articles/nature.2015.18248
That's not to say other areas are better by default. There is just a lot of junk science out there. "Study says X" is meaningless to me personally. "Study X was reproduced by A, B, C" is something I'd take notice of.
Any time I have vocals / lyrics in music, it drags my attention away - so I only listen to non-vocal (in my case electronic) music.
For me, and one of the reasons I HATE open office plans, I cannot focus if I can almost hear a conversation. Speaking at a normal tone would be far less distracting. Similarly, someone is speaking in a language with too many borrow words in English. Overhearing Germanic languages in particular, my brain keeps trying to treat it like an English speaker with mush-mouth or having a stroke. If I just squeeze my brain harder with my furrowed brow, it will all make sense.
There is no actual definition of what would constitute a “task” here and that alone makes all of it quite dubious in the way you outline and any other things in how it's often extrapolated.
The part about Movies and Social media makes sense. They are both "heavy" tasks, so they each have their demand. But Music? I would say it depends on the music. Some is distracting, some not. There is optimized music for concentration, usually found in video games. But there is also "heavy" music that demands your attention, IMHO usually those with lyrics, especially inspirational lyrics.
I'm curious whether they checked the type of music in those studies.
All that phrase means is that an habit of listening to music while you do something is correlated with lessen focus while not listening to music.
There's no causality there, and there's nothing about how deep is the focus while listening to music.
These are definitely not all same, unless you're literally just spinning the mouse wheel like a fidget spinner or watching a Fast & Furious movie. Even more complex music isn't the same as ambient or generic background music.
It's not uncommon to hear claims from neurodivergent people about achieving optimal stimulation levels to perform a task. That can be increasing stimulation with music or motion, or decreasing it by blocking out sound with ear plugs. I've seen these play out and it seems to work. I think it's also safe to assume that someone wearing noise cancelling headphones is swapping one set of sounds for a quieter or at least less distracting one. If in fact they aren't just tuning people out by wearing headphones and not playing anything.
And even neurotypical people can maintain throughput on a boring task by accompanying it to music. Laborers and militia just being the most obvious in media. I've talked to a couple of neurodivergent people who admit to "singing" a song in their head while working. Just because you don't hear the music, doesn't mean there's no music.
Tragic though, for advertisers.
But what is a task? You emphasize the term and I have a general concept of what a task is in work and other contexts. But that seems too fluid to make broad statements about human behavior - even with studies combining two activities. How do we know one person's doing a task isn't another person's multitasking? It's intuitively plausible that two activities logically will be somewhat harder than doing each in sequence. But are all such combinations identical? A lot of challenging activities do involve the combination of subtasks - firefighters might break into a building looking for survivors while monitoring the situation of the building for safety. Just about any video involves satisfying multiple constraint simulateously, if they didn't, they'd be boring, etc.
My favorite lately is the La Palma volcano live-streams.
There was a time in my life (when I owned a desktop) that I would listen to music, play a livestream, play a video game cooperatively, and voice chat with my cooperative partner.
It was the perfect stimulation that would have me focused deeply on the conversation as the gameplay was the topic du jour. It came second-nature.
Unplug! Take off your head phones! .... Close the door.
Like many artists, I had the habit of listening to classical music or jazz while working. On one of my first visits to Milt Kahl I innocently asked:
"Milt, Do you ever listen to classical music while you're working?"
"[Expletives]...! I'm not smart enough to think of more than one thing at a time!"
Since it came from a genius, this made quite an impression on me. After this I learnt to face the silence and think before swirling my pencil around. My animation improved right away.
This has been the case with many artists when I've passed this wisdom along. .... They were even more surprised at the startling improvement in their work.
[Image of artist with shirt that reads, "animation is concentration"]
But then you're also less likely to get back to work once you are in social media, aren't you?
If your job is to be "on the bottom of things" doing careful calculations and designs, you should not multitask. But if your job is to be "on top of things", then you will have to.
Can't mix the two safely.
You do not want the engineer/architect calculating the load on a beam to be checking email, on slack, answering the phone and other "on top of things" tasks.
But some managers think you are a slacker if you are not running around with your hair on fire at all times. They don't understand the difference.
www.coloradocollege.edu
For example, driving involves keeping the car at a reasonable, avoiding obstacles on and off the road, deciding whether you need to turn soon, noticing the condition of the vehicle and related things. These could all be considered tasks and are indeed difficult to keep straight when one begins driving. But they can indeed be done together. And while it might be easier to somehow do each separately, the combined task of driving isn't possible without this "multitasking".
I understand that nearly everyone talking about multitasking here is referencing whether or not tasks at work can be efficiently combined. But I don't think that addresses the larger issues (what's a single task, when two tasks intersect, how can you not multitask, is the difficulty of "multitasking" a limitation of humans or is it that the task itself becomes inherently more difficult).
(2) Frequently people multitask to manage their emotions. It is like having a choice between a room that is too cold and too hot and you can be comfortable moving from one to the other.
Sometimes, when I know starting a Slack thread will lead to lots of discussion, I wait until I have mental space to be ready to switch to them. Once those conversations start, it's very hard to concentrate on the current task.
Other times, I know people will take a long time to respond, so I'll kick off those discussions early and be comfortable in being able to context switch away from it when needed.
For example, while my code is compiling, I might think through the specific steps I'm going to take to try it out when I run it here in a minute. Or try to plan out the next change I'm going to make to the code if this current code works. Or start a list of some unit tests I need to add.
Sometimes it might be worth multitasking and being less efficient at both tasks. If I'm washing the dishes and listening to a podcast I'm completely okay if I'm at 80% efficiency at washing the dishes and paying attention to the podcast. Neither activity generally requires all my attention. The same goes for many other activities that we do.
Playing a video game and listening to a lecture is probably still better than just playing the video game. Just don't expect to absorb the lecture as well as you would if you were to fully concentrate on it.
“I think better when I walk.” is a commonly heard statement; many research scientists when thinking about a problem also take a walk outside and find that thinking while walking improves it.
But in soft science research without any real definitions it's of course quite easy to not incorporate such things because terms have no meaning and then say that “research proves it” while all it did is prove things about their ad hoc definition of “multitasking” which is far removed from the real world.
I also listen to books when doing hobby woodworking, as most of the time spent is on rather menial tasks like sanding or planing. However, not all of it is like that, and you often need to focus, do some calculation or planning. During those times, I need to pause the book, as I can neither focus on it nor on the task at hand.
Human bodies have lots of ability to multitask, it’s just their brains have very little.
> Other studies suggest that people who frequently “media multitask” (like listening to music while checking email or scrolling through social media while watching a movie) are more distracted and less able to focus their attention even when they’re performing only one task.
Compared to what? Music for me is usually a way of escaping the hell that can be open spaces.
I generally keep a text document open all the time. When I have to task switch I write down where I am leaving the current task and what I have to do when I come back. Sometimes it is a couple of lines of text. Other times I have to add more to it.
For me this works very well. It avoids the issue of an interruption or having to deal with multiple things in a day killing your productivity because it takes an hour to get your brain back into the prior task.
Simple. Effective. At the end of the day it looks like you did multiple things simultaneously, just like a multitasking OS. In other words, my system tick is >= 1 hour, not milliseconds.
"You're not really multitasking, you're just screwing up multiple things at the same time"
I do however feel a great sense of accomplishment whenever I have several browser windows open while I work though. :|
This is a subtle point. I can easily have 10 tabs open and swap between them while mono-tasking. E.g — ignoring for a moment the fact I'm writing on HN... — right now I have 4 tabs open on Sentry, where I'm comparing stack traces for a couple of instances of two related errors. I have another tab on the relevant PR, and a few more with docs pages for the relevant APIs/libraries. Swapping between these things isn't multitasking, it's just managing the complexity of the one single task.
If you're interested at least in consciousness and brain waves, you could do worse than to look into reading some of the references in "States of Consciousness: Experimental Insights into Meditation, Waking, Sleep and Dreams" Dean Cvetkovic & Irena Cosic Editors
Pretty much the exact same way as modern hardware works.
Alarms and lists serve as my task manager, since my brain fails at it by default.
I know two people who can, and they both have ADD. (But when I asked around, many others who have ADD can’t, though they are capable of fast context switching).
A friend knows how to fork and merge parallel streams at will, but can get bored. The mental intensity he can achieve when all of his streams merge is much greater than typical. He tells me he’d run podcasts at 4x or something while he works on other stuff.
> Other studies suggest that people who frequently “media multitask” (like listening to music while checking email or scrolling through social media while watching a movie) are more distracted and less able to focus their attention even when they’re performing only one task
The causality could just as well run the other way: people with poor focus adherence presumnly exhibit it regardless of the amount of distraction.
Of course, if I ever had the opportunity to work on a single task, I can be very productive at that. Easiest way to get into flow state. However, that isn't common for my work. Often, instead, I am working on something that crosses many boundaries and has unavoidable downtime. For instance, right now I am working on visualizing erlang processes using Phoenix and d3js to see if I can detect unusual memory use growth. I'm going back and forth between erlang docs and Phoenix views and elixir code and javascript (with, of course, lots of reading and searching in between). I have no idea if that's multitasking, but when I have to move from one layer to another I experience the same kind of disruption and need to refocus as I do when I'm interrupted.
Another quality of multi-tasking that I frequently experience is the difference between needing to respond to external stimulus v.s. being able to orchestrate multiple processes to run in parallel. They both involve context switching and loss of focus, but there are huge differences in the order of magnitude.
How many studies never saw the light of day because the result was uninteresting, or ran counter to the narrative the study's author wanted to put out?
Just say "there is evidence to suggest" instead.
If I have laundry, dishwashing, etc going. I am multitasking but I don't need to expend attention.
So there's certainly a threshold where multitasking becomes a problem.
In MSP world, I didn't control my own schedule. If the schedulers thought I only needed 30 minutes to do a 3 hour task. That's all you got. Some coworkers would hard limit. Put the 30 minutes in and leave the job unfinished.
The stated rule was double/triple billing; which is a felony but you know they'll throw you to the wolves if you actually break the law and get caught. However, the rule was that if you needed more time. You had to work on multiple things at once. If you did not accomplish a task, you had to set the ticket back with justification why you didn't get it done. About 25% of the time if you didnt get it done and put in a justification. Someone would come talk with you.
Not to help, but usually chastise you.
.. and a queue of anxious white collar parents forms to get their child tested to determine if they are in the 2.5% /s
I have asked around of other people who have ADD, and they don’t report parallel thoughts, just very fast context switching.
Though typically when I do put on music to work, it's old Amiga soundtracks or something. I wonder how much the difference in levels of distraction for lyrics vs. no lyrics.
So I only pretended to multitask and my error rate plummeted. Does each generation have to figure this out on their own?
Best I can do is reading hacker news while compiling or training a convolution network.
But anyone can hire multiple people to do multiple tasks.
For the known task related to low cognitive focus I use for background classical music (Corelli, Telemann) with low volume.
When I want full cognitive load (learning, coding), silence and isolation are working for me the best.
No it does
Maybe by being provocative and choosing an absolute position on the matter, the author was hoping to attract readers.
For those interested in the explanation given by Huberman, skip to the section titled “Multi-Tasking Is Real” found in the description of the video [0].
Right brained people, those with the right brain dominant over the left, have an easier time multitasking because the right brain is image oriented. The right brain processes information as images are processed, In a visual simultaneous manner.
That being said, brains are weird and if you find you can multitask then go for it. I've gathered data on my own attempts to multitask and established that I cannot.
[1] https://www.sciencedaily.com/releases/2013/08/130814190513.h... [2] https://www.health.harvard.edu/blog/right-brainleft-brain-ri...