Electricity, Brain "Disruption", and Math

Dec 08 2010 Published by under Behavioral Neuro, Neuroscience, Uncategorized

Sci got an email the other day. Ok, I get lots of emails, but this one asked a cool question, which is always nice. All it asked for was an opinion on an article in Scientific American: "Get Better at Math By Disrupting your Brain". Sci looked. Was intrigued. Read the actual paper...and found the SciAm coverage of it somewhat misleading (or at least, really confusing).

Basically, the article at SciAm states that

The goal of the study was to assess whether modifying activity in the parietal lobes affected the acquisition of number competence.

If the brain functions by optimizing behavior, it might be possible to worsen numerical competence by disrupting parietal function, but it should not be possible to enhance it that way. However, that is precisely what Cohen Kadosh's team found. Remarkably, this improvement was still present six months after the training.

The problem isn't so much in the total coverage itself. The problem is in the word "disruption". You keep using that word. I do not think it means what you think it means.

dis·rupt (ds-rpt)
tr.v. dis·rupt·ed, dis·rupt·ing, dis·rupts
1. To throw into confusion or disorder: Protesters disrupted the candidate's speech.
2. To interrupt or impede the progress, movement, or procedure of: Our efforts in the garden were disrupted by an early frost.
3. To break or burst; rupture.

So you'd THINK, based on the SciAm article and the definition of the word "disruption", that this paper used magnets to disrupt me, by which they mean to impede or interrupt, implying a negative effect on brain activity in the parietal lobe, and this paradoxically (as the SciAm article notes) made people better at math.

That would be cool. If that was what the paper actually FOUND. It wasn't.

And so, I'm going to cover the paper for you all. Let's clear this up.

ResearchBlogging.org Kadosh et al. "Modulating Neuronal Activity Produces Specific and Long-Lasting Changes in Numerical Competence" Current Biology, 2010.

So basically, there are a lot of people out there who are bad at math. I know that sounds really silly, but apparently about 20% of people have 'numerical difficulties', wherein math is very difficult for them, making it hard for them to succeed in certain jobs or at certain tasks. Not only that, many people who suffer from strokes will suffer damage to their numerical processing abilities.

(someone who does not suffer any difficulties in numerical processing. Really, this isn't an example, but I love him so much!)

People who have difficulties with numbers often have functional or structural abnormalities in the right parietal lobe. This is an area of your brain that is usually thought of as being involved in sensory processing, but it's also involved in things like your "number sense".

EDIT: The following has been clarified by commenter Superkuh. So this paper involves math, the parietal lobe, and a technique called transcranial direct current stimulation (TDCS). Basically, you just attach conductive wires to the skin and use a current limited circuit to run electricity one direction through a the volume of tissues between the two electrodes. The stimulation, depending on the current polarity, electrode placement on the scalp, and pulse rate, can depolarize different kinds of neurons (depolarization is part of an action potential, and what makes a neuron "fire". For more on that, see my Science 101 post on it). TDCS can also cause long term effects within the neurons, leading to potential long term changes in function.

So the idea here was to see how TDCS affected numerical processing in the brain, using humans. TDCS is not particularly risky (there's a minor risk of seizures). Sci has actually been in a study on this before (they made my fingers JUMP using magnets on my HEAD. There is nothing quite so creepy as watching your finger move entirely of its own volition). They had the participants learn the associations between random symbols, without learning their number value. So, by the rules of the test $ is greater than &, but you don't necessarily know whether $ is 45 or 3, and whether & is 27 or 1. You just know that one is bigger than the other. Each time the participants were trained on the task, they got a weak current applied to their heads. 1/3 of the group got a a cathodal application to the right parietal lobe and an anodal application to the left parietal lobe (RC-LA), 1/3 got anode application to the right and cathode on the left (RA-LC), and one group just got a quick jolt that didn't have any known effects.

After six days of training, they tested their performance in various mathematical tasks.

What they found was that the RA-LC group had increased mathematical task performance, while the opposite group (RC-LA) underperformed.

And here's where the Sci Am article, in my opinion, got misleading. Because ANODE stimulation ENHANCES neuronal activity in TDCS. It doesn't "disrupt" it or decrease the activity. It ENHANCES it. Cathode stimulation does the opposite.

This means that if you got Anode stimulation in the RIGHT Parietal cortex, you actually got ENHANCEMENT of the neuronal activity there. This is not a "disruption". It means they got what they were expecting to get. Enhance neuronal activity in the right parietal lobe during training, enhance mathematical performance. Decrease neuronal activity there, and decrease mathematical performance.

The best part, though, is actually that the effects persisted. The RA-LC group (the enhanced group) STILL had improvements over six months later (apparently the decrease in performance for the other condition didn't persist). This has some great implications for helping people with severe numerical issues, as well as for helping people who have damage to this area due to stroke (though the stroke condition might make it less effective due to tissue damage in the area, that study will have to be done).

So this is an interesting study. Not sure how it works, as no one is sure how TDCS works, but the effects are there and could be helpful.

But it's not a "paradoxical behavioral improvement". This isn't "disruption" of your neuronal activity proving that your brain works best when it's NOT working the most efficiently. On the contrary, this study shows that enhancement of neuronal activity in a region led to an increase in a numerical processing outcome. So I'm not really sure how this study is like "swimming and guarding the clothes at the same time" (which seems a lot more like a study in attentional shifting). To my mind, this study has a lot more to do with identifying neuronal networks and areas that change specific behaviors, and affect learning and memory. Not about whether your math gets better with disruption. I'm not sure what story the article at Sci Am was trying to tell, but I feel like it may have confused the issue.

But then again, maybe it just had to do with the use of the word "disruption". Am I misinterpreting the article? Am I mis-defining the word? Anyone?

Cohen Kadosh R, Soskic S, Iuculano T, Kanai R, & Walsh V (2010). Modulating neuronal activity produces specific and long-lasting changes in numerical competence. Current biology : CB, 20 (22), 2016-20 PMID: 21055945

14 responses so far

  • Snarkyxanf says:

    Please tell me you've heard the "censored" version of the Count's song.

  • dreg says:

    TDCS is NOT the same as TMS.

  • marius says:

    TDCS is actually different from TMS in that the magnetic field doesn't change, but is constant. (You can actually get little burn marks.) Well, that's what I gathered from all the buzz around it.

    Also, what does enhanced mean? The neurons fire more. Does this mean they function better? Most likely not. Imagine I put a hell of a lot of random cars on an already functioning street. Does the street work better? If you interpret a traffic jam as the pinnacle of street functionality, then yes. Of course, we have no idea what TDCS actually does. So if there are only a few extra cars and street-throughput is simply increased, that could be an enhancement. Well, if they were on the right lanes and with the right speed and stuck to the rules. However, random cars tend not to do that.

    As I always say: never change a winning brain! Of course, if you take losing brains, then shaking things up a bit is like regression towards the mean. Anyway, it would all come down to how you interpret TDCS to interfere with a real brain. I personally would find it rather amazing if such a crude and largely random manipulation enhances functionality (as opposed to activity) which seems to be what they found here.

    • Snarkyxanf says:

      Localized TDCS is arguably less crude and random than, e.g. caffeine, which is usually thought to have some function-enhancing effects.

      • marius says:

        True, of course. Still, if you want to enhance functionality, then simply jolting millions of functionally slightly related neurons does not seem like the best approach to me - at first glance.

        It's amazing how many people think that more activity on the neuronal level is equal to better performance, even in neuroscience.

    • scicurious says:

      Ah, thanks for letting me know. It's a technique I'm not particularly familiar with. That said, this paper seems to imply that enhanced firing during training in this paper did increase function in testing. And they deliberately picked participants with no numerical abnormalities.

      I would like to see better testing as well, it seems like a very crude manipulation, as you said.

  • Jason Dick says:

    The Count is even more fun to listen to censored 🙂

  • JJ says:


    First, as mentioned in the comments above, tDCS is different than TMS. TMS relies on a magnetic field to elicit current in the neurons and depolarize them whereas tDCS (transcranial DIRECT current stimulation) relies on the current itself. It does not depolarize the cells but it brings them closer to (anodal stimulation) or further from (cathodal stimulation) depolarization. tDCS is known to influence brain plasticity and has been shown to enhance learning and/or retention in several motor and perceptual tasks.

    This Kadosh paper studies how tDCS can affect mathematical skills... However, in the current biology study, I have the impression that a better retention of the trained skill (as demonstrated for other tasks) would lead to better mathematical skill during the test session because training and test are too similar. I bet that the same procedure would not improve the calculus abilities of the participants 🙂

  • Mark says:

    The state of this kind of research is pretty undeveloped. All of the cogneuro people are extited about TMS and similar methods, but don't have a firm grasp of exactly what is going on. Paricually the disruption or enhancement or whatever it is we are doing to your neurons issue. Just look at how fMRI shapped up from the early days. A little too much excitement about the method, some poorly designed studies and over eager data interpretation. That is probably what is going on here. Though I haven't had time to reread the article.

  • superkuh says:

    What the others comments say is mostly correct. TMS has nothing to do with tDCS. Absolutely nothing. rTMS is voltage based, where voltage across an axon is proportional to the rate in change in the external magnetic field generated by pulses of thousands of Amperes in metal wire placed adjacent to the head. tDCS is simply running very low currents, thousandths of a single Ampere, through the head by direct ohmic contact.

    You should just remove all mentions of TMS. It only confuses an otherwise clear and interesting summary.

    If you chose to leave the TMS information in, then at least don't called it honkin' magnets. It is big honkin' *electromagnets*, which are an entirely different animal.

    • scicurious says:

      Good point, I changed the relevant bits.

      • superkuh says:

        Uhm, I'm glad you changed the bit on TMS to electromagnets but tDCS doesn't use electromagnets; that was TMS. tDCS just uses electrical current and the tiny magnetic field generated as a byproduct is insignificant. It is literally just running electricity through the skin, around the skull, through the dura matter, and within the brain.

        I'm sorry to be such a pedant but may I suggest,

        "Electrical current, Brain “Disruption”, and Math"
        "So this paper involves math, the parietal lobe, and a technique called transcranial direct current stimulation (TDCS). Basically, you just attach conductive wires to the skin and use a current limited circuit to run electricity one direction through a the volume of tissues between the two electrodes. The stimulation, depending on the current polarity, electrode placement on the scalp, and pulse rate, can depolarize different kinds of neurons (depolarization is part of an action potential, and what makes a neuron “fire”. For more on that, see my Science 101 post on it). TDCS can also cause long term effects within the neurons, leading to potential long term changes in function."

        • scicurious says:

          AHhhhhh. Excellent. Thank you for the clarification. My other sources said it was magnetic, not electrical. Can I ask which source you are using so I can take a look? Obviously the SciAm article was confused as well, it's not well clarified in the paper.

  • Saif says:

    I have a question unrelated to this article but related to brain activity. I will some of the more knowledgeable people can answer.

    My mom just went through Chemotherapy and one of the conditions she developed while recovering completely from Lymphoma, was an occurrence of Dementia which is raising a question in my mind.

    She would completely go back to different times in her life and be locked into it for that period, at times lasting at times for about 5 hours.

    For example, she would be telling me regarding the medications that my dad has to take in evening, but who has actually passed away many years ago in 1994. But fortunately she has completely recovered.

    My question is.... Does human brain, much like computers, has a a time index pointer , which is what starts to malfunction and in my mom's case was pointing at a time in past when my dad was alive, rather than staying correctly pointed at current time zone?

Leave a Reply