Why are magnets not bad?

August 24, 2007 4:51 AM

Why does holding a strong magnet up to your brain not do anything bad?

I understand that the brain does not operate completely off chemistry and uses electricity, especially within the neurons, so it seems that there would be a fair assumption that holding a good powerful magnet up to your head would have some deleterious effects and at least be "felt" in some way (distorted thought patterns or seizures). Is there a biochemical reason why you can scramble tapes and screw up a lot of circuit boards by goofing with a magnet, but the brain is impervious?

I'm aware that magnets are a staple of alternative medicine, but since this is mostly voodoo (or empirical at best) it's not really the answer I'm looking for.

posted by zek to health & fitness (26 comments total)

Your magnet just isn't strong and varying enough. See transcranial magnetic stimulation.
posted by DarkForest at 5:00 AM on August 24, 2007 [3 favorites has favorites]

It's not strong enough - as DarkForest says, see TMS. You'd have to have a really frickin' strong magnetic field to affect the brain in a noticeable way.
posted by spaceman_spiff at 5:06 AM on August 24, 2007

Also, the iron in your blood is non ferrous, so it has no effect on that either.

Experiments intended to test magnetic effects on your brain tend to use massively strong electromagnets. Consider the field strength in a MRI machine, it's way beyond anything a regular magnet can provide, yet there are no effects at all.
posted by tomble at 5:10 AM on August 24, 2007

MRI is pretty much based on holding a strong magnet to your brain. So the brain isn't really "impervious", it does do something.

I understand what you're wondering about though, and it's hard to explain. It's not so much biochemistry but physics, and the difference between inorganic and organic molecules. It's been too long for me to be able to explain.
posted by easternblot at 5:13 AM on August 24, 2007

Magnets aren't that strong. Your nervous system works at minuscule voltages and currents, so the magnetic field they generate is also minuscule.

Weak magnets + very weak magnetic fields = nothing happens.

Even very strong magnets + very weak magnetic fields = nothing much happens - which is just as well, otherwise MRI's would give you a bad case of Asploding Head Syndrome.

As for the iron in blood: well, it is still ferrous (by definition; if it's iron, it's ferrous...), but it's molecular, so very very tiny, and there's bugger all of it (what - 4 haem groups in a haemoglobin molecule?). Compared to the mass of a red blood cell, the amount of iron is also miniscule.

(IIRC it is possible to separate blood magnetically given a strong enough magnetic field, enough time, and low/no flow.)
posted by Pinback at 5:22 AM on August 24, 2007

The simple answer is that brains don't operate in the same fashion as electronics. Importantly, neurons don't transmit signals in the same way as electronics or magnetic storage (like old floppy disks or hard drives). And brains are simply massive networks of neurons that transmit and process signals.

But let's talk about how a neuron transmits a signal inside itself and out to other neurons...

Think of a neuron as having an input (dendrite) and output (axon terminal). A signal is received at the input, gets transmitted along the cell "electrically", and finally causes something to happen at the output. I say "electrically" because unlike electronics, where electrons are moving in metal wires, neurons move charged atoms (ions) in and out of different parts of the cell. Those charged atoms are mostly unaffected by a magnetic field.

Also, neurons talk to other neurons in two ways:

Chemical synapse
Neurons release and sense neurotransmitter chemicals in the tiny gaps (known as synapses) between neighboring neurons. This is how signals are sent (and added and subtracted) between neurons. And these chemicals don't really respond to magnets.

"Electrical" synapse
Sometimes neurons are connected to one another through little pipes (gap junctions), and the movement of charged atoms I talked about can go directly from one neuron to the next. Thus again, mostly not affected by magnets.
posted by Mercaptan at 6:23 AM on August 24, 2007 [1 favorite has favorites]

ok, to make up for that i'll try again. magnetic fields affect (moving) charge - if something is charged and moving, a magnetic field pushes it sideways, that's just a fact of life (you really don't want to know why). so the question is - what charges are moving inside a brain, and how does this affect them?

- some people are mentioning mri, but that's mainly based on the effect of magnetic fields on bound electrons. electrons are charged particles that form atoms and are responsible for binding atoms together to form molecules, which are the building blocks for the brain (and everything else). now these electrons are moving (much like planets around the sun, they can be thought of as "orbiting" the atom, although that's only an approximation to the truth) and so are affected by magnetic fields, and in some cases we can detect this. but the effect is small - it doesn't stop the electrons from doing their job as atomic "glue". so the brain doesn't fall apart.

- another thing charged particles are used for in the brain is carrying signals. in some sense the brain is an electronic circuit. for example (and i'm not a biologist, so may have the facts a bit wrong here) charged particles (ions, i believe) travel across the gap between neurons. now if the magnetic field were strong enough, it could mess this up, because those particles would be pushed sideways and so not make the correct connection. but this doesn't happen because even the strongest magnetic fields are small relative to the electrostatic forces involved. it's something like a bee trying to push a rock sideways as the rock rolls down a hill.

i think those are the two main mechanisms you might think of.

in short, the problem is not that magnetic fields have no effect, but that their effect is small. you need systems that are particularly designed to use magnetism (tvs, magnetic media, etc) and the brain is not "designed" to use magnetism - it uses the much more powerful electostatic forces (the normal electrical force in batteries - the +/- thing).

and there's a reason magnetism is so small. it's actually related to the more powerful electrostatic force through special relativity. and special relativity is, as you may know, a theory deeply connected with the speed of light. and the speed of light is fast. and fast means it's a big number. and it turns out that the strength of magnetism is related to the strength of electrostatic forces by dividing by that big number. i can't remember the details, but it's not random - magnetic fields are small for a "reason". so, again, i assume brains evolved to use the more powerful electrostatic force because it is stronger.
posted by andrew cooke at 6:26 AM on August 24, 2007 [2 favorites has favorites]

Andrew Cooke - Funny, I was about to say how I'm a biologist, not a physicist, and thus couldn't say how strongly magnets would affect ions or neurotransmitters. Glad you addressed that.
posted by Mercaptan at 6:36 AM on August 24, 2007

'some people are mentioning mri, but that's mainly based on the effect of magnetic fields on bound electrons.'

No, it's a nuclear effect. NMRI, with the N dropped because people got freaked out by it.
posted by edd at 6:52 AM on August 24, 2007

charged particles (ions, i believe) travel across the gap between neurons

Actually, IIRC, the neuron maintains a charge difference between the inside and the outside of its cell membrane. When the neuron 'fires', this charge difference is allowed to collapse. This collapse travels as a wave down dendrites of the neuron. When the signal reaches the end of the neuron, it signals the release of neurotransmitters, exciting the next neuron(s) in the chain.

TMS works, again IIRC, by inducing small currents in the brain, triggering the neurons to fire purely by electrical disruption.
posted by DarkForest at 7:01 AM on August 24, 2007

No, it's a nuclear effect. NMRI, with the N dropped because people got freaked out by it.

it's a bad day. you're right - for some reason i was thinking about zeeman splitting.

ok, so in that case it's the charged particles in the atom's nucleus. and again, while it has some effect on those, they are deep inside the atom and don't affect the chemistry or physical structure of the brain.
posted by andrew cooke at 7:05 AM on August 24, 2007

sorry, dendrites = axon
posted by DarkForest at 7:06 AM on August 24, 2007

Note: I am a neuroscientist, but I am not your neuroscientist, and all that. I also know just enough about physics not to electrocute myself, so some of the physics here might be off-base: but the neuroscience should be solid.

One thing to remember is that the vast majority of neurons in the brain are insulated by myelin, a fatty covering. There are tiny gaps in the myelin coating. Ions only cross the neural membrane at these gaps. The remainder of the charge is transmitted by ion movements within the insulation, inside the nerve axon. Thus, most of the net transmission of action potentials is buffered against interference and signal leakage.

Most importantly, in a neuron, signal transmission is an electrochemical movement, not a strictly electrical movement; even disrupting the electrical gradient across a membrane would still leave a chemical diffusion gradient. The neural membrane is largely impermeable at rest, so ions can't cross. Pump proteins in the membrane actively push ions one way or the other, creating a net negative charge inside the neuron. When a neuron fires, ion channels open in one spot, allowing ions to cross the membrane - positive sodium in, negative potassium out. The chemical diffusion of ions across the membrane is helped by the electrical potential: Many large molecules, such as DNA, inside the cell are negatively charged, so positive ions are attracted inwards.

The potential - the electrochemical difference between the inside and the outside - goes away. This change triggers nearby ion channels to also open, and the signal moves forward along the axon. Physical properties of the ion channels stop channels in a previously-depleted region from opening for a few moments, ensuring that the signal can only move in one direction (typically down the axon to the axon terminal, but there's no physical reason signals can't be sent in the opposite direction).

After firing, pump proteins use energy to grab positive sodium ions and force them out of the neuron, trading them for negative potassium ions to reset the membrane potential.

What would a magnet do? Potentially disrupt the electrical gradient, perhaps, but it wouldn't affect the chemical gradient. Ions would still move. You can kill the neuron's ability to fire by upsetting the chemical gradient, though. That's why injections of potassium chloride can be lethal: Too much potassium outside the cell, and instead of having potassium rush out when ion channels open, they either stay put, or, if enough potassium is present outside, they might even rush in, stopping the signal from being sent. If this happens in, say, nerves running to the heart, well, it won't be contracting, and blood flow will cease.

Neuroscience rocks.
posted by caution live frogs at 7:45 AM on August 24, 2007 [6 favorites has favorites]

The earth has a magnetic field stron enough to bend solar wids and cause the polar lights. Your brain is subject to this magnetic field every day, and yet there is no effect. Why would your puny electro magnet do anything?
posted by Pollomacho at 8:51 AM on August 24, 2007

Adding another perspective, the forces of the electro-magnetic interactions are well below the forces of the dynamic interactions (running, jumping, etc) that the brain is capable of compensating for with no effect on performance. Until you can generate an EM force that is the equivalent to a boot to the head, the brain will be able to filter it out.
posted by cardboard at 8:59 AM on August 24, 2007 [1 favorite has favorites]

You can scramble tapes with a magnet because the information on tapes is magnetic, not electrical. Magnetism and electricity are related, but not the same.

You can induce an electrical current in a conductor by subjecting it to a varying magnetic field. The stronger the magnetic field and the faster it varies, the stronger the current induced.

Moving a permanent magnet fast enough and close enough to your skull to induce TMS-like currents in the brain would involve enough acceleration to shatter it. You wouldn't have to worry about the magnetism of it so much as its resemblance to buckshot.
posted by flabdablet at 9:15 AM on August 24, 2007

The earth has a magnetic field stron enough to bend solar wids and cause the polar lights. Your brain is subject to this magnetic field every day, and yet there is no effect. Why would your puny electro magnet do anything?

it's not just the strength of the magnet that matters, but also the distance from the magnet. you can think of the earth's magnetic field as being a very powerful magnet at the "centre" of the earth, but it's also far away, while a magnet in your hand is very close. you can see this with a compass - it will point to even a weak magnet if nearby, rather than follow the earth's magnetic field.
posted by andrew cooke at 9:24 AM on August 24, 2007

I say "electrically" because unlike electronics, where electrons are moving in metal wires, neurons move charged atoms (ions) in and out of different parts of the cell. Those charged atoms are mostly unaffected by a magnetic field.

To expand on that: a stationary charged particle in a magnetic field experiences no force from the field. A moving charged particle in a magnetic field does, but here's the catch: the force it experiences is proportional to its velocity (to be more precise, to the component of its velocity which is perpundicular to the magnetic field.)

An electron zipping through a copper wire typically moves at something like 1/10 the speed of light. The ions which carry charge in your nervous system move--well, I don't know exactly, but you can bet it's orders of magnitude slower than 1/10 the speed of light, and the force it experiences from the magnetic field is proportionately smaller.

Second, the ions carrying charge in your nervous system are much more massive than the electrons that carry charge in non-biological electrical systems. A potassium ion is about 70000 times more massive than an electron. So even if that potassium ion were zipping alongside an electron, both going at 1/10 the speed of light, through a magnetic field, they'd both experience the same force, but since the potassium ion is 70000 times as massive, it's only deflected 1/70000 as much as the electron (to be more precise, it experiences 1/70000 the acceleration the electron does).

Bottom line: the charged ions which carry electrical charge in your nervous system are both much slower and much more massive than electrons flowing through a copper wire, and both contribute to those ions being relatively unaffected by magnetic fields.
posted by DevilsAdvocate at 9:37 AM on August 24, 2007

On MRI..

It is important to understand that MRI is an active, not a passive, sensor. You first excite nuclei (the active part), and then monitor the relaxation. It is not unlike testing shock absorbers by shoving on the hood of a car, or testing the stability of a shelf by pushing on it. By watching how the energy you put into the system dissipates, you can learn things about the system.

When I say you can learn things by watching the reaction, it is actually a lot simpler than it sounds. All you are looking for is differences, not any particular reaction. Given that we know a lot about what a body is made and how it works via other mechanisms, any difference at all is enough to tell you a great deal. If you've ever watched Time Team, think of the geophysical surveys they like to show. They measure something about the ground in a grid pattern. If differences show up in a pattern, they know they've found something.

Add to that very sensitive instruments, and very tiny differences can be detected. Like the difference between organ, muscle, and fatty tissue..

MRI uses a property of 'body stuff' that, with very strong fields, can be excited just a little. And, with very sensitive instruments, that kind of excitation can be measured.

The fact that MRI uses such intense fields, much more intense than any other magnetic field you are going to encounter, might actually answer the question in itself. Given a strong enough magnetic field, it alone might do something bad to the body, but you simply aren't ever going to experience one that strong.



the forces of the electro-magnetic interactions are well below the forces of the dynamic interactions (running, jumping, etc) that the brain is capable of compensating for with no effect on performance.

I don't think that is a good analogy.. Just try to extend it to X-rays, for example :P

X-rays are different from MRIs, by the way, because they are making use of an effect that causes damage to the body at strengths only a little higher than the strength needed to get useful readings.
posted by Chuckles at 9:52 AM on August 24, 2007

FWIW, caution live frogs, with answers like that, you will always be my neuroscientist.

/sorry for the derail
posted by misha at 9:56 AM on August 24, 2007

If it helps, I have noticed that I can feel strong magnets when my fingers have been very close to them for many seconds. It's a very subtle feeling, but there have been several occasions when I've been engrossed in something I'm working on, then throught "that's weird... my [left index, or whatever] finger has that weird magnet ache feeling", then look down and see that that finger is resting right next to a strong magnet moved there while rummaging through stuff previously. It's anecdotal, but there have been no false positives as yet.

Try it yourself. Buy some strong magnets (strong enough that you would hurt or injure yourself if you got any skin caught between two when attracting each other), hold you hand at rest so that one finger is only a few mm from a magnet pole, leave it there for 20 seconds, and see if you can sense any difference in how that fingertip feels, compared to the other fingers. If you can, remember the sensation, and have a friend put the magnet under a sheet of card, so you don't know which part of the card has the magnet underneath, and see if you can locate it. (I haven't tried that yet).
posted by -harlequin- at 10:00 AM on August 24, 2007

MRI uses a property of 'body stuff' that, with very strong fields, can be excited just a little. And, with very sensitive instruments, that kind of excitation can be measured.

Actually, that is wrong in important ways because MRI uses two types of excitation in combination. But, we aren't trying to build an MRI here, so..
posted by Chuckles at 10:10 AM on August 24, 2007

I just want to amplify what flabdablet said: magnets won't affect the majority of electronics, either. They'll mess with magnetic storage, of course, and they'll mess with precise particle-beam stuff like CRTs. But other than that, a strong static magnetic field won't disrupt circuitry. (It would have to be a really strong field before the Hall effect started messing things up in a typical circuit.) Maybe you could saturate a transformer core or something, but even that seems unlikely to me. (Disclaimer: I am an EE, but I am not your EE, etc.)

(A strong magnet can disrupt mechanical stuff like wristwatches, by magnetizing the parts and making them stick to each other.)
posted by hattifattener at 2:02 PM on August 24, 2007

Think of a neuron as a tube, or a pipe (thus making the brain/nervous system a series of tubes). When a neuron is triggered, it either depolarizes or hyperpolarizes. This is done by pumping charged ions into and out of the neuron. So if you took a tube (like a paper towel roll or a PVC pipe) and poked a lot of holes in it, the ions would be flowing directly in or out through the holes in the side of the tube. In a wire, the current flows along the length of the wire (basically) whereas in a neuron the flow is not along the length but in and out of the holes on the side. So there is no current, per se, to interact with the magnet. You have some very-slightly charged particles moving a very small distance independent of (i.e., at right angles to) the direction of the signal. Electro-magnetic interaction is based on a linear propagation of fields; in the brain it's more like a diffusion. Yes, it's caused by electrochemical interaction, but the signal itself is not strictly electrical in nature.

This is only a part of the picture, and meant to supplement the quality answers already provided in the thread. As others have said, you really need to get to the level where the magnetic field is so strong it strips the electrons from the atoms for there to be any noticeable effect (e.g. explosion).
posted by Eideteker at 4:46 PM on August 24, 2007

After firing, pump proteins use energy to grab positive sodium ions and force them out of the neuron, trading them for negative potassium ions to reset the membrane potential.

Sodium and potassium ions have the same charge: +1. My understanding was that the Na⁺/K⁺ pump creates a net charge by transporting 3Na⁺ out and 2K⁺ in, which also creates a concentration gradient which allows K⁺ to diffuse out of the cell through potassium specific ion channels, further increasing membrane potential.

I'm sure you were just simplifying things for the sake of brevity, but I'd like to know if my understanding is accurate.
posted by [expletive deleted] at 10:44 PM on August 24, 2007

-harlequin- - you're probably thinking you can feel it just because you know it's there. You'd need to properly double-blind it to be sure you can do it - noone in any contact with you between the magnet being placed and you declaring where it is should know where the magnet is - and I'd lay money on you not being able to tell in those conditions.

I have been subjected to both MRIs and TMS for research purposes. Only the TMS generates noticeable sensations. One reason is that it is generating signals in your brain, and that does do odd stuff. The other is that it can generate what you might call a very minor case of the aforementioned head asplodey syndrome, in that the currents it generates produce their own magnetic field which interacts with the TMS's field, and this generates a noticeable force between your brain and the TMS device. Not enough to asplode your head, but it can be felt as a slight tug on your head.

Even strong permanent magnetics are very weak in comparison to either device.
posted by edd at 9:56 AM on August 25, 2007