Talk:Action potential/Archive 2

Action potentials in dendrites

it is very rare for an action potential to occur in the dendrites.

Might it be worth expanding on this point a little? Given that studies have shown back-propogation of action potentials into some dendrites (Stuart et al. (1997) Action potential initiation and backpropagation in neurons of the mammalian CNS. Trends In Neurosciences, 20(3)) and this idea has already been used in models of neuronal activity (Markram et al. (1997). Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs. Science, 275, 213 – 215).

I don't think I know enough about the subject to know if it's worth putting in, or to be capable of doing a good enough job, but figured it might be worth bringing up Jasonisme (talk) 14:33, 30 May 2008 (UTC)

K+ I-V Curve

The K+ I-V curve is misleading, if not entirely incorrect. The K+ channels that are primarily responsible for the repolarizing phase of the action potential (the delayed rectifier K+ channels) open in a voltage-dependent manner, similar to Na+ channels. Thus, the depiction of an entirely linear K+ I-V (which asserts that G=Gmax at all potentials) is inaccurate in this context. There are several examples in widely available reference texts (Hille, Kandel) which could be used as a template. -Mark 67.176.225.196 (talk) 04:27, 2 January 2008 (UTC)

Article Contradicts Itself

Compare these three examples...

Stimulation

A local membrane depolarization caused by an excitatory stimulus causes some voltage-gated sodium channels in the neuron cell surface membrane to open, causing sodium ions to rush in at high speed through the channels along their electrochemical gradient.

Depolarization ("Rising phase")

As sodium ions enter and the membrane potential becomes less negative, more sodium channels open, causing an even greater influx of sodium ions.

Peak See also: Goldman-Hodgkin-Katz voltage equation

By the time the membrane potential has reached a peak value of around +50 mV, time-dependent inactivation gates on the sodium channels have already started to close, reducing and finally preventing further influx of sodium ions.

With this assertion...

It is important to appreciate that very few ions actually cross the membrane at any stage in the action potential.

--RadioElectric (talk) 17:42, 13 January 2008 (UTC)

I may be misreading, but I'm not seeing a distinct contradiction here. Can you point out the problem more explicitly? --David Iberri (talk) 18:09, 13 January 2008 (UTC)

The first three emphasise the role of many ions crossing the membrane through ion channels whilst the last bit I quoted says that actually very few ions cross the membrane.--RadioElectric (talk) 18:57, 13 January 2008 (UTC)

It's really just a question of magnitude. Ions do traverse the membrane, just not in large enough numbers to cause a notable change in concentration (thus the "very few" bit). There are exceptions to this, such as potassium accumulation in the T-tubule system of skeletal muscle, but generally there's no huge concentration change. Those most important aspect in an action potential is the change in permeability, but people have a tough time understanding how permeability changes lead to action potentials...which is why it's typically presented in the manner of the article (my neuro 101 class in college went about things in the same way). --Dpryan (talk) 03:23, 14 January 2008 (UTC)

I don't see a contradiction. Axl (talk) 15:14, 30 January 2008 (UTC)

Poor referencing

The referencing in this article is extremely poor and inconsistent. There are only 5 in text references within the article. In general the differing methods of referencing is not a good sign of an article. For scientific articles it should contain only in text citations and there should be no need to clarify if something is a primary source as it would become clear from the citations as to which journals were used more often.

This is a side point and there is nothing actually wrong with it. It is stated that the bulk of the information is obtained from Hodgkin and Huxley 1952. Surely a featured article would use sources more up to date than 1952. --Medos(talk) 22:24, 29 January 2008 (UTC)

Why should we use a more up-to-date reference that just says the same thing? Newer doesn't mean better. --Dpryan (talk) 07:51, 31 January 2008 (UTC)

That is true but new aspects are discovered all the time. The physiology will not change but that does not mean that we will not discover anything new through investigation. It's generally good practice to use more recent scientific research than older. Merely looking at old literature and deciding that is sufficient makes no attempt at improving the standard. Also there can be changes in terminology. An example that springs to mind is Coronary Artery disease is generally now referred to as Coronary Heart Disease.

Also the scope of the literature search is very limited. The primary sources are only from 3 authors. And by the nature of Wikipedia it should be even easier to generate a greater of sources.--Medos (talk) 15:29, 31 January 2008 (UTC)

Action potential

This article has been nominated for Featured article review because it does not fulfill criteria 2(c) of featured article criteria. It referencing system is inconsistent and has only 5 inline citations. —Preceding unsigned comment added by Medos2 (talk • contribs) 10:25, 30 January 2008

technical

I have some ideas about possibly clarifying the content of the article. 69.140.152.55 (talk) 11:28, 22 March 2008 (UTC)

Hi, thank you for your excellent suggestions! :) I'll be taking this article under my wing next week and it's good to have any and all ideas before beginning. :) I'll make a few indented comments here, and I would appreciate any other ideas you have for improving the article. Willow (talk) 13:02, 22 March 2008 (UTC)

First, in the Goldman equation, I believe the square brackets refer to concentration; I think that should be explained that directly after (or before) the formula, except that I need a source.

Yes, you're right. Strictly speaking, I believe that the square brackets should represent activity, but here I think they're synonymous. Willow (talk) 13:02, 22 March 2008 (UTC)

Also it should be explained that R is the ideal gas constant and T is temperature, but the source that I have for that is the article on Goldman's equation, and as I understand it Wikipedia is not supposed to cite itself.

Yes, you're right again! :) The RT derives from the fact that Goldman's equation, like the Nernst potential, derives from the Boltzmann factor for ions distributed at equilibrium. Willow (talk) 13:02, 22 March 2008 (UTC)

Second, I also request further clarification (accessible to lay readers) of how the inside of the cell becomes negatively-charged with respect to the outside of the cell, despite the process being described as an opening of channels causing a facilitated diffusion of ions across the membrane, which might be expected to result in the inside and outside of the cell having no charge, rather than reversal of charge. This is already explained but only briefly.

As I understand it, the potassium permeability is dominant, making the resting membrane potential close to that of the equilibrium Nernst potential for potassium. That potential depends on the difference in concentrations between the inside and outside of the neuron, which is established and maintained by the potassium/sodium ion pump, which in turn is driven by ATP. If I recall correctly, a large fraction of your total energy output goes simply to pumping such ions across membranes! At equilibrium, the flux of potassium ions is determined by two factors: (1) the much higher concentration of potassium within the cell, which wants to drive the potassium out; and (2) the negative potential of potassium within the cell, which draws the potassium in. At equilibrium, these two fluxes must be equal and opposite, by definition; hence the potential must be negative. A positive potential would drive the potassium out; does that make sense? Willow (talk) 13:02, 22 March 2008 (UTC)

I thought I understood before, but now I know that I do not. Perhaps a non-technical explanation of Nernst potential may be in order. 69.140.152.55 (talk) 13:25, 22 March 2008 (UTC)

I'll be happy to do that! :) You're right, the idea is so fundamental to the whole process, we should definitely make sure that people get that first. Let me try a brief explanation now? Imagine that, initially, there's no potential across the membrane (the charges are all balanced), but there's a much higher concentration of potassium cations inside the axon than outside. The potassium ions will diffuse around, and some of them will randomly cross the membrane. Feeling no other force, the net flux will be outwards; if each potassium ion has a small chance of crossing the membrane in either direction (initially equal in both directions) and there are many more ions within, then on average more ions will wander out than will wander in — does that make sense? If the potassium ions had no charge, the outward flux would continue until the concentrations had become equal on both sides of the membrane. But as the potassium cations leave the interior of the neuron, the interior is left slightly more negative; that interior negative potential will gradually build up as more and more potassium ions leave, and there is more positive charge on the outside of the membrane. That negative potential will serve as an inducement for the ions to stay on the inside, inhibiting the outward flux (uphill, against the potential) and encouraging the inward flux (downhill, with the potential). The potential makes it less likely for a given potassium ion to leave and more likely that another will enter. Eventually, the potential will become sufficiently negative that the two fluxes (into the cell and out of it) will become equal, giving equilibrium. If some demon were to make the potential even more negative, the inward flux would then predominate, causing the interior potential to become more positive, restoring it to equilibrium. You see, it's a stable equilibrium; there's negative feedback to keep the membrane potential at its resting value, the equilibrium potential. The Nernst equation gives the value of that equilibrium potential for a given difference in the concentrations; if I remember correctly, it's roughly 25 mV per e-fold difference in ionic concentration. Willow (talk) 15:18, 22 March 2008 (UTC)

Third, I might add in regard to the "article contradicts itself" comment above, even though the article does not contradict itself in the strict sense, it might benefit from some cleanup or clarification because, for example "the misconception that sodium 'floods' the cell to cause the action potential" is, in a sense, created within the article itself (and yet removing whatever text that tends to cause the misconception may make the article harder to read).

You're right again! :) We should express that more positively, as I've learned from my friend Awadewit; "say what something is, before saying what it is not." Instead, we should say something like, "Very few ions are required to move across the membrane in order to change its electrical potential drastically."

Fourth, please add more explanation of how reduction in capacitance facilitates the "jumping" of action potential along myelinated neurons. I think this is covered in saltatory conduction, but the Action Potential article is featured and this information will make it more comprehensive.

If I understand correctly, the myelin speeds conduction by limiting the points at which ionic flow can occur. The capacitance of the intervening stretches of axon affects that by affecting the rate at which the electrical potential "diffuses" from one node of Ranvier to the next, as described roughly by the electrical cable equation. Willow (talk) 13:02, 22 March 2008 (UTC)

Finally, the sentence "[t]he model of electrical signal propagation in neurons employing voltage-gated ion channels described above is accepted by almost all scientists working in the field" in the concluding section sounds a little bit weasel-ish, but because omitting it might give undue weight to competing theories, that is not a serious problem in my opinion. 69.140.152.55 (talk) 11:28, 22 March 2008 (UTC)

It may take a while to tidy everything up and give everything its due weight, but we'll certainly try. Thank you for your insightful comments, and please send us any more! :) Willow (talk) 13:02, 22 March 2008 (UTC)

Tags

I'm going to remove the two tags, the one asking for inline citations and the other suggesting that the article is too long. For the former, I intend to add more inline citations, and for the latter, I think the article still has room to grow (it's only 39 kb), provided that it's written well and captivatingly so that the reader is drawn forward inexorably. Hopefully, we can make the article a spellbinder! ;) Willow (talk) 04:42, 24 March 2008 (UTC)

Lacking material

This article lacks a few points such as what determines the equilibrium potentials of potassium and sodium, i.e. why are they different when both are monovalent cations and a mention on how astrocytes are involved in maintaining the ion homeostasis required for action potentials. Dendritic action potentials or spikes should also be mentioned. How different types of neurons and different states of the same types of neurons can exhibit different firing patterns of action potentials would also be good to mention, that it happens is currently mentioned in the overview section but not mechanisms (perhaps that could go in another article). —Preceding unsigned comment added by 129.241.172.206 (talk) 17:47, 26 March 2008 (UTC)

I agree with all of your points except the proposal to discuss the details of equilibrium potentials in this article. Those details are best discussed at equilibrium potential and resting membrane potential. --David Iberri (talk) 21:56, 26 March 2008 (UTC)

Copy editing

I noticed at the FARC that there are several of us copy editing this article. What does everyone think about dividing up the work? Each taking several sections? We could at least start that way. I have to admit that copy editing the entire article is daunting! Awadewit (talk) 17:16, 28 March 2008 (UTC)

Good idea. I took a whack at the lead, but everyone should feel free to review that. How about we break it into the following chunks?

A - 1. Cellular and biophysical context and 2. Sequence of events

B - 3. Phases and 4. Threshold and initiation

C - 5. Propagation

D - 6. Refractory period through 14. Circuit model (all short)

Do we have 4 reviewers? Is this too many chunks? If this works, I'll take Chunk A. – Scartol • Tok 19:11, 28 March 2008 (UTC)

I wouldn't want you all to work in vain, either! You really shouldn't copy-edit sections that I haven't had a chance to fix up yet, since everyone seems to agree that they're pretty bad. Today, I managed to draft something in History, Propagation, Mathematics, and the sections through "Ion pumps" in "Cellular and biophysical context". Maybe you could look those over and see if I'm on the right track? The History section is the only one I've referenced decently, so maybe start there? The Propagation and Mathematics sections are second-tier, so those could be next, while the initial sections are still rather poor. I'm curious about the Math section; it may be totally clear, or it may be hopelessly obscure and maddeningly laconic. At least the references are doing OK; I think we have about 43 now (some of them have multiple references for a single number).

I might not be able to do much typing this weekend. With all the knitting I've been doing lately, plus all the typing these past three days, my right wrist really hurts. :P I may rest it up, go to the library, read some more and just brood over the weekend. But I really appreciate you all coming to the article's rescue. :D I might try to dash off something about the ion channels this weekend, and I'll try to give you more "grist for the mill" next week. Con affetto, mille grazie, Willow (talk) 22:51, 28 March 2008 (UTC)

I'll take the "Mathematics" section. Deep breath! :) Awadewit (talk) 16:31, 29 March 2008 (UTC)

I've pre-copyedited it for you. In particular, I've tried to explain the math rather than just state it. Geometry guy 20:36, 29 March 2008 (UTC)

Hang on just a second for the Math sections, would you mind? I need to make a few improvements, but I'm just too tired to finish them right now... Willow (talk) 07:24, 31 March 2008 (UTC)

Holding. Awadewit (talk) 17:30, 31 March 2008 (UTC)

Thank you. ;) I'll try to be quick about it. :) Willow (talk) 19:07, 31 March 2008 (UTC)

Okay, it's a lot better now; the Fitzhugh-Nagumo section needs work, but the others might be at least intelligible? Any and all suggestions would be very welcome, though, no matter how critical! :P Willow (talk) 21:26, 31 March 2008 (UTC)

Yes fantastic work, it is a lot better, but it may be too long: many of these details may be better placed in the linked subarticles: this article has to give the main ideas. It should be copyeditable by Awadawit without her having to get stuck deeply into the sources. I'm also a bit concerned about the introductory paragraph. This could easily be seen as OR right now. My own view on both these issues is that we should make intelligent cuts, not to the weak math section that we had previously but to a math section of a similar length based on what we have learnt. Geometry guy 21:52, 31 March 2008 (UTC)

I've taken a whack at the History section. Comments are in my sandbox. Cheers! – Scartol • Tok 18:12, 30 March 2008 (UTC)

I'm going to wait for clearance on other sections until you've finished content-ing them up; is that okay? Just lemme know. – Scartol • Tok 12:10, 1 April 2008 (UTC)

Unfortunately, I didn't get to this article today — sorry! I ran out of time. I have tons of notes on everything I want to add tomorrow, though — hope springs eternal.... ;) Willow (talk) 19:30, 1 April 2008 (UTC)

I wish I could do more to help.. – Scartol • Tok 19:32, 1 April 2008 (UTC)

Please don't worry about that; I'm sorry for not writing faster, to give you something to review! :P Willow (talk) 11:18, 2 April 2008 (UTC)

At the risk of being obnoxious, I'll note that we advocated at the FAR for a deadline of 4 April to finish all of this. Should we ask for more time? – Scartol • Tok 12:30, 2 April 2008 (UTC)

I don't like to go back on my word. If Sandy et alii choose to delist it on Friday, then I'll bow to their decision. The article had it coming; and, honestly, if we fall, why then, we'll be the Fallen. ;) But I'm hoping for grace and good works. ;) April 4th is the day of the patron saint of the Internet, Isidore of Seville, and one of the first encycopedians. Therefore, like St. Crispian, we'll remember, with advantages, what feats we do this week; and Wikipedians now abed shall hold their sainthoods cheap whiles any speaks that fought with us for Isidore's day. ;) I'll try at least to reward your faith, as long as G-guy doesn't distract me like Atalanta with any more golden apples. ;) Willow (talk) 16:49, 2 April 2008 (UTC)

I had another look at the math section, and my feeling is that neural networks are just a bridge too far for this article. I can sympathise with the desire to mention them in a featured article on neurons, but this one is really about part of the chemistry of an individual neuron, and I don't see that neural networks are relevant, per WP:WIAFA 4. Do others agree, or am I missing the point? Geometry guy 19:54, 1 April 2008 (UTC)

That thought'd also crossed my mind, but I'd hoped to include something about the temporal encoding of information in the time between action potentials. More generally, I wanted to place the action potential into its context in nervous systems, to describe how they're used, what role they can play, does that make sense? I realize that that doesn't really come across in the present section, but I'm hoping to make a sow's purse from a silk ear eventually — wait, is that right? ;) Willow (talk) 11:18, 2 April 2008 (UTC)

I'm kind of lost now - can I go ahead and copy edit the math section? Awadewit (talk) 14:35, 2 April 2008 (UTC)

Yes, please. The section could be improved as is, no doubt, but it's better to know what's unclear, rather than to guess at it. If it's completely unclear, which shouldn't take long to see, then I'll try again. The neural-network blurb is modular and can be deleted at a moment's notice. Willow (talk) 16:49, 2 April 2008 (UTC)

Update on size and exposition

Umm, the article seems to have swelled by 80kb in the last few days; it's now at 102 kb, which seems long even for talkative me. :P I'm going to start cutting and trimming discussions down, since there are several redundancies from before. The "bridge too far" of neural networks may also go the way of all flesh. ;) However, some of the article's bulk comes from its references, which I've been trying to be scrupulous about and wouldn't want to delete. I seem to remember hearing about a tool for measuring the size of an article w/out the references; does anyone know how to do that?

The article is still rather rough-hewn, so I wouldn't want anyone to devote too much time to copy-editing it. Perhaps tomorrow it'll be better! :) But if perchance you did want to glance over it, and had any suggestions about the exposition, that'd be very helpful. I'll try to make the article illuminating and picturesque for everyone. :) For now, I'm going to go off and recharge my own neurons. ;) Willow (talk) 22:05, 2 April 2008 (UTC)

Copy editing questions

Thanks for your help, Awadewit! :) I'm sorry that it's so foreign to things that you love, but it speaks really well for you that you're willing to venture out into these badlands. :) Willow (talk) 20:19, 3 April 2008 (UTC)

• There are numerous types of ion channels, each with several states whose populations depend on the current conditions (such as voltage, temperature, pH, etc.) and past activity. - I'm not really sure how this connects with the previous sentence - the "there are" is tripping me up. Also, I'm not a big fan of "etc." - lay readers such as myself cannot fill in a list like this! :)

I re-phrased the sentence and left it vague as to what "conditions" I meant. The main idea I was trying to get across was that the Hodgkin-Huxley model is great for modeling the giant axon the the squid, but may be grossly inaccurate for other types of excitable membrane that have different ion channels in them. I believe that animals even have several minutely different versions of each ion channel that they can swap in or out, or use in different tissues, to modulate the electrical activity there. Willow (talk) 20:19, 3 April 2008 (UTC)

• The second purpose is to understand qualitatively the neural computation that occurs when the axon hillock generates a new action potential in response to postsynaptic signals on the dendrites and its own past history. - The "its" refers to the new action potential? Is that correct?

The "its" referred to the axon hillock, which is generating the action potential. A given dendritic stimulus can provoke an action potential in a fresh hillock, but fail to do so in an axon hillock that's tired out from having recently made many action potentials. There are several factors in being "tired out", i.e., that affect the hillock's threshold for initiating an action potential, and I didn't want to go into them all, so I finessed it by saying "recent history". Does the present wording make sense? Willow (talk) 20:19, 3 April 2008 (UTC)

I think grammatically the "its" refers back to the "action potential". What do you think? Awadewit (talk) 21:31, 4 April 2008 (UTC)

I think you're entirely right in the original wording; too bad there aren't verbal wikilinks to clarify pronouns! ;) Anyway, I was kind of dissatisfied with that, and I had to re-write the paragraph after the neural-networks section was deleted. The present wording is much simpler: The second type of mathematical model is a simplification of the first type; the goal is not to reproduce the experimental data, but to understand qualitatively the role of action potentials in neural circuits. Is that OK? Willow (talk) 21:54, 4 April 2008 (UTC)

Excellent. Awadewit (talk) 22:06, 4 April 2008 (UTC)

• where g(V) is a cubic function of the voltage V, that has one minimum, one maximum, and diverges to ±infinity as the voltage does likewise. - I assumed the ± is supposed to be there in front of "infinity"?

That's right! :) I replaced that infelicitous wording, though, and added a picture to illustrate it. I also fleshed it out a bit more, talking about other variants, especially one that has had an illustrious history of research in mathematics. Willow (talk) 20:19, 3 April 2008 (UTC)

I'm not sure how much I really helped out in this section! Awadewit (talk) 00:49, 3 April 2008 (UTC)

Well, I think you helped, and my opinion is the only one that matters. ;) Princess Willow (talk) 20:19, 3 April 2008 (UTC)

PS. Oh, would that that were so...I'm beginning to suspect that we're not going to make it by tomorrow, and that AP is doomed. :( But I'll not give up now; it's St. Crispen's day on my calendar. ;) Willow (talk) 20:19, 3 April 2008 (UTC)

Whenever you're ready, give me another section. Awadewit (talk) 21:31, 4 April 2008 (UTC)

Thank you, thank you, thank you! :) Let me scan the article for a halfway passable section... :) Willow (talk) 21:54, 4 April 2008 (UTC)

...the next best section is History, which Scartol is doing, but if you had any passing suggestions, that'd be great as well; two heads and two friends and all that. :) I'm going to try to fix his suggestions right now. The sections that could use some work copy-editing-wise, although they're poorly referenced, are the Propagation and Context sections; but please don't read the final subsection ("Resting potential") of the Context, since it's still pretty ghastly. [Not that the others are much better at the moment :(] The Propagation one is rather technical, but very important and unlikely to change too much. I was thinking of adding a table showing the diversity of conduction velocities in different types of neurons and in different animals, and maybe a graph showing the dependence of conduction velocity on neuronal diameter, but aside from those "ornamentals", it shouldn't change much. Thank you! :) Willow (talk) 22:08, 4 April 2008 (UTC)

PS. I'll try to answer your scholarly-study questions this weekend; as you know, I've been rather busy! :) Willow (talk) 22:08, 4 April 2008 (UTC)

I'll try my hand at "Propagation" later today. Awadewit (talk) 17:27, 5 April 2008 (UTC)

Hi all... duh, should have checked the talk page before I dug in, but I think I restricted myself to minor and uncontroversial edits so far. Although I realize I'm reading an article that is actively being edited and modified, I jotted down some comments here. Hope they are useful. Cheers, AndrewGNF (talk) 19:35, 5 April 2008 (UTC)

• In 1925, Lillie was the first to suggest that myelin served to restrict the action potential to the nodes of Ranvier.[2] The first experimental evidence for saltatory conduction came from Tasaki[3]. and Takeuchi[4] and from Hodgkin and Stämpfli. - If this isn't going to be explained, perhaps it should be removed? Awadewit (talk) 05:09, 7 April 2008 (UTC)

Umm, I'm not sure what you mean about it being explained, but it was rather orphaned and alone there, poor thing. :( I integrated it into the paragraph above, to give the scientific context for saltatory conduction. Willow (talk) 18:18, 7 April 2008 (UTC)

• For example, the time-scale τ increases with both the membrane resistance rm and capacitance cm; as the capacitance increases, more charge must be transferred to produce a given transmembrane voltage (by the equation Q=CV), and as the resistance increases, less charge is transferred per unit time, making the equilibration slower. - This is too long, but I can't fix it. :( Awadewit (talk) 05:09, 7 April 2008 (UTC)

Yes, I was afraid of that. I'll try to make the sentences snappier and more digestible: less linguini and more gnocchi. ;) Willow (talk) 18:18, 7 April 2008 (UTC)

Again, I don't really think I'm doing much here. Sorry! Awadewit (talk) 05:09, 7 April 2008 (UTC)

Concerns about article content and organization

YIKES!! What happened? What used to be a very clear and concise article about action potentials is now a strewn-together mess of anything related to electrophysiology of excitable cells. With everything from mechanism of resting membrane potential, to voltage-clamp methods to neurotoxins. Much of the material is not even directly related to the mechanisms underlying an action potential, and is poorly organized. Nrets 00:59, 18 April 2008 (UTC)

Please see my response here. – Scartol • Tok 02:00, 18 April 2008 (UTC)

My impression is that not everyone shared your good opinion of the clarity and concision of the original article. See, for example, the FAR and this discussion of TimVickers and SandyGeorgia. There was also the matter of its few illustrations and referencing. But my goal is not to criticize the old version, but to argue in favour of the new version. Willow (talk) 10:10, 18 April 2008 (UTC)

I'll gladly concede that the present article is overly long and should be cut down somewhat, although I hope you'll eventually agree with the scope of the article. We need to tune the harp, tightening its strings but not removing them altogether. For example, I would argue that the following topics are appropriate to be covered in a Featured Article about action potentials:

• action potentials in skeletal muscle fibers, the heart, and plant cells;
• quantitative models of the action potential, especially the Hodgkin-Huxley model for which they were awarded the Nobel Prize;
• experimental methods used to gain the knowledge of the action potential, especially pivotal technical advances such as the glass micropipette electrode and patch clamping that allowed the action potential to be understood at a molecular level.
• the historical development of the understanding of the action potential
• a mental picture of the action potential at the molecular level that is intelligible and can be visualized by lay-people without having to follow lots of wiki-links
• the "life-cycle" of an action potential: its creation, propagation and ultimate fate at the synaptic knob
• the effects of action potentials on other systems, e.g., the neuromuscular junction
• how famous poisons affect the action potential at the molecular level. Admittedly, the pufferfish picture was eye-candy, but tetrodotoxin is relevant to the action potential and many people will have heard about fugu poisoning

The old article was perhaps concise and intelligible to someone who knew the subject well already, but such people are not our target audience, right? Experts will neither read nor cite a Wikipedia article about their specialty; we shouldn't write the article for them, or for other people who understand the subject already. Non nobis solum. Rather, let me convince you that a better target audience consists of humanists such as Scartol and Awadewit, or mathematicians such as G-guy, or perhaps better, sincere pre-med students at a small liberal-arts college in the provinces, far from good libraries. I feel that we should write for people who sincerely want to learn about action potentials, but who might never have understood the difference between an ion channel and an ion pump. I'm not sure that I've done that well, I just dashed it all off, but part of the article's length is devoted to setting the stage for newcomers in the Context section. Plus, I scattered little bits of brain- and eye-candy throughout the article, to keep it from becoming too taxing to read; brain candy does cause cancer of the semicolon, ;) but I think it's important to keep a lively tone and engage lay-people's attentions. Anyway, you may disagree, but as we improve the article, I think we should bear our audience in mind while also trying to be as encyclopedic as possible. Willow (talk) 10:48, 18 April 2008 (UTC)

Hi WillowW, I appreciate your effort to make the article more accessible to a wider audience, however the way it is written now I think it is far less accessible. It's just to confusing! Before, it was written at the level of a basic undergraduate neuroscience text or a basic medical student text, which I think is just about right for Wikipedia. Having taught Neuroscience to undergraduates for several years I can definitely tell you that I would not recommend the article to my students in its present state. It seems like a lot of the problems people had with it was the lack of inline citations, which is ridiculous, since this is basic textbook material and I think is fine to cite those for an article at this level. As far as your proposed organization, maybe you could move the "life cycle" of an action potential further up in the article, maybe the second section. That way, if someone wants to simply read an article about what action potentials are and how they are generated, they can get all the information they need from the beginning of the article and people interested in specialized information (mathematicians, science historians, etc.) can read further down. I'm still of the view that in many cases more expansive is not necessarily better and cutting the scope would be much more beneficial. Wikipedia is not a Textbook, thus articles should not necessarily follow the format of book chapters, since articles can be easily hyperlinked together, going from one well organized, focused chunk of information to another. Nrets 14:01, 18 April 2008 (UTC)

Hi Nrets! :)

I in turn appreciate your courteous letter and advice on the presentation! I feel sure that if we Talk through our differences, we'll come to appreciate each other's point of view; the article as a whole can only benefit. I think we need only clarify our thoughts to one another, and then we'll either find a higher, better compromise, or we'll agree to disagree, which reasonable people may reasonably do. :)

It probably doesn't surprise either of us that we each prefer our own versions. ;) I acknowledge that the present version is as stuffed as a Victorian living room, and that it could be made sleeker, faster and more intelligible, and would surely benefit from the pedagogical insights you have to offer. I've really just dashed it off, and there's much left to improve and to cut. On the other hand, I and several others found the original version unworthy to be a Featured Article in several respects, including writing, clarity and completeness. I hope that you'll be gracious enough to concede that professors are not always able to understand how students can misunderstand their specialty, the professors having studied it for decades and never having misunderstood it themselves. A text that may appear perfectly clear and concise to you may not appear so to a newcomer to the field, don't you agree it's possible? On technical accuracy, I would of course defer, especially in the face of cited scientific literature, but on intelligibility to lay-people, I feel that my opinion has at least some standing.

It's true, but after years of teaching I am familiar with how students think and how to explain things to them. I should say that I had nothing to do with the original article, I just thought it was well organized and concise, although I agree that the language could have been a tad clearer. There was a figure in there of a current voltage relationship for Na+ and K+ currents that you removed that is probably one of the most useful pedagogical graphs for understanding an action potential and have been using a version of that in my class for years. Nrets 01:56, 19 April 2008 (UTC)

I like that graph, too, and removed it with some hesitation and planning for its comeback. The reason I didn't keep it then was that it didn't fit exactly with how I was trying to present the material. I was trying to describe the phases of the action potential at the molecular level, rather than the more classical electrophysiological description. I suspect that it's easier for newcomers to visualize little tubes, the ion channels snapping open or shut, than to intuit how nonlinear electronic circuitry behaves. I recognize that we have to include currents and voltages in the article, and I'd like to re-introduce that figure; but I hope you also see the benefits of the molecular description for lay-people. Willow (talk) 02:35, 20 April 2008 (UTC)

Your letter and edits raise several issues, which might be better discussed separately. When working with other editors, I've always find it good to untangle the threads of our disagreement so that we can solve them one-by-one. I hope you agree that's a good approach, but I'm also open to other approaches. Willow (talk) 17:09, 18 April 2008 (UTC)

On the audience: it was written at the level of a basic undergraduate neuroscience text or a basic medical student text - I would suggest that this is not the level the article should aim for. Since most readers are not going to have the vocabulary or the mathematical background that these texts assume, this article is going to have explain far more than such textbooks. As a graduate student in English literature who is fascinated by science and reads popular science of all kinds, I can tell you that this is aiming too high. A good example to follow in this regard is Introduction to general relativity (an even more mind-twisting topic!). In my copy editing comments below, you'll notice that I point out where I think the article begins to lose people like myself. However, overall, I think that it is well on its way to explaining the topic well. After reading it, I was able to explain what I had learned to someone else who knew the topic and isn't that test of whether someone is beginning to learn something? Awadewit (talk) 23:19, 18 April 2008 (UTC)

I think that basic undergraduate texts are abut as basic as one can get. The standard in WP used to be that science articles should be at the level of Scientific American, which is about the same as a basic text. Nrets 01:56, 19 April 2008 (UTC)

Hi everyone, thought I'd chime in my two cents as well. I share Nret's concerns that the scope of this article has grown too broad, and I seem to recall someone invoking WP:SIZE previously. I like to think that for a topic I'm interested in I can read everything in one sitting (say, 15-20 minutes) to get the high points. But this article is pretty intimidating, and I can feel myself hesitate just because of its length. Specifically with regard to Willow's list above, I agree that all topics should be mentioned, but I also think that not all should be discussed to the same depth. Having all the relevant pieces here is a luxury of course (thanks to Willow and others for that!), and now perhaps we should discuss what less-critically-relevant content could be moved to linked articles? I'll make one specific suggestion below, and hopefully we can discuss to reach consensus? AndrewGNF (talk) 00:32, 19 April 2008 (UTC)

Perhaps we should split this article up into sections, such as

• EPSP/IPSP
• biological action potentials (Hillock through to axon)
• chemical synaptic communication
• quantitative models of action potentials

I think it goes without mentioning that I am against including sections on ionic solutions and cell membrane properties =P Paskari (talk) 17:10, 2 July 2008 (UTC)

How much referencing is needed and of what type?

Several editors have argued that action potential does not require inline citations, because its neurophysiology is standard textbook material, having been worked out (mainly) half a century ago in the 1950's and 1960's. Such editors would seem to prefer a bibliography of textbooks at the very end of the article, to which interested readers can turn to learn more. The chosen textbooks would presumably contain the verification for the assertions in the article. For example, the student might turn to the index of one such textbook and find the pages dealing with ion channels to verify the properties of ion channels.

(1) My own feeling is that inline references improve the article and should be used.

(2) For better or worse, inline citations seem to be a sine qua non for Featured Articles, as outlined here (criteria 1c and 2c) and here. If we agree that we want to return this article to FA status, then inline citations become necessary. One compromise would be to pepper the article with inline citations to textbooks, which would spell out to the reader exactly where they could find the verification of a given assertion. I've done this at some places throughout the article and I'll do more in the coming days.

(3) I personally feel that we should do more. As encyclopedians, I feel we have a responsibility not only to give the currently accepted facts, but also to paint for the reader how those facts came to be known, and a feeling for the evidence that supports them. Not only does that approach give the reader a better feeling for the science, it also pays an endearing homage to the scientists who gleaned that knowledge. Therefore, I've tried to cite some of the original papers that demonstrated this or that fact about the action potential. I recognize, however, that not everyone—including me! :)— has access to the original literature, so I favor a hybrid approach, citing both the original work and relevant commonly used textbooks. I don't believe that citing the original literature harms the article.

That's my take on referencing. Do you all agree/disagree? I think we should all discuss it and try to reach consensus. Willow (talk) 17:38, 18 April 2008 (UTC)

• I understand the point of view of scientists who say "but this information is obvious, it is not controversial, therefore it does not need to be cited". However, I think we do our readers a favor when we cite reliable sources because if they want to learn more about a particular topic, we have made it easy for them: go to this chapter in this textbook. (I have tried to source an article from a bibliography: it is not easy. A list of books is not enough.) Moreover, Wikipedia lacks legitimacy in the public sphere and in academia. Demonstrating concretely that what is in our articles can be found in reliable sources will help to show to the world that we have articles that are themselves reliable. Finally, since the editors of the article would like it to reach FA, a laudable goal, it will have to fulfill the criteria for FA, as Willow has explained. I support this goal, especially since so few important scientific articles have reached FA. Awadewit (talk) 23:26, 18 April 2008 (UTC)

Subsections in the External links?

Personally, I liked having subsections in the External links, to help the reader navigate quickly by grouping the links into common types. I don't think it overwhelms the Table of Contents—as a relative fraction, it's relatively minor—and I honestly haven't found that part of the Featured Article criteria that specifies that External links cannot have subsections. Maybe here somewhere?

That said, we have so few external links of each type that we could just extend the explanation of each link to clarify that it's an animation, a set of lecture notes, a link to simulation software, or whatever. That'd be a little redundant, but I'm open to deleting one or both of the subsection levels, if that's the consensus. Willow (talk) 18:11, 18 April 2008 (UTC)

• One option is to limit the TOC, as was done at Joseph Priestley. Awadewit (talk) 23:27, 18 April 2008 (UTC)

A few technical issues

A few technical issues came up that might warrant discussion? Willow (talk) 18:30, 18 April 2008 (UTC)

• Is the chloride anion relevant for the action potential? As far as I know, they have relatively small effects on the action potential and do not have axonal voltage-sensitive conductances. However, chloride contributes to the axon's leakage conductance and was included in the equivalent circuit modeling of Hodgkin and Huxley. Chloride seems to be the major counterion for the various cations (sodium, potassium, calcium), to help preserve electroneutrality and establish the resting potential. If I recall correctly, chloride conductance is also important in inhibitory post-synaptic potentials. Hence, even if the chloride ions don't contribute significantly during the action potential itself, they affect its propagation and the resting potential context. Admittedly, it's a "cough&spit" role next to the prima donna roles of sodium and potassium, but still important enough to mention, I think.

Although Cl- does contribute to the resting membrane potential, there are no Cl- conductances activated during an AP, adding it makes it a lot more confusing. Inhibitory potentials are irrelevant and these should be discussed in the synapse article. Nrets 01:49, 19 April 2008 (UTC)

I think we're mostly in agreement, except on the delicate question of whether the article should discuss the mechanism of neurotransmission. I sympathize with the POV that this article does not need to describe the events between the extinction of the pre-synaptic action potential and the generation of another at the post-synaptic axon hillock, since no action potentials are present. But I also see narrative advantages to describing the full "life cycle" of an action potential, from its initiation at the axon hillock, its propagation to the synaptic knob, its "posthumous" effects and how those contribute to the birth of a new action potential. It's also nice to go from the neuronal axon potential to the skeletal-muscle action potential without having a gap in the narrative.

The Cl- issue is a minor point. I mentioned it as the most significant anion for the action potential (which perhaps we agree on?); chloride also appears in the equivalent circuit diagram of the Hodgkin-Huxley model. We both seem to agree that the role of chloride channels is quite minor (relative to the cations) for the typical action potential in animals, affecting mainly the propagation through their effect on r_m. However, chloride seems to be one of the main ions in the action potential of Acetabularia; see this reference, for example. (I also noticed the CLC gene family of voltage-gated chloride channels, but from what I can tell, they play no role in nervous conduction.) All I'm asking for is that we restore (and possibly improve!) this sentence. You raise a very good point about the danger of confusing our readers, but I'm thinking that chloride should be mentioned (it is mentioned in almost all of the textbooks I consulted) and that we can find a way to forestall their confusion — peppering the article with a clarifying/warning sentence here or there, spelling out chloride's role and how minor it is? Willow (talk) 19:34, 19 April 2008 (UTC)

Willow, I don't understand what you are trying to do here. On one hand you say you are trying to simplify the article to make it accessible to "lay readers" yet on the other hand you seem to want to introduce every obscure detail known about action potentials. If you look at any neuroscience textbook, the chapter on action potentials is separate from that on synaptic transmission. Of course they are all related to the excitable properties of neurons, but they are separate topics. You are also working on the assumptions that the sole role of action potentials is to cause neurotransmitter release, or that the sole role of neurotransmitter release is to evoke action potentials. So maybe it would be sufficient to say something along the lines that action potentials can be triggered when postsynaptic potentials sufficiently depolarize the postsynaptic cell. But really stop short of explaining synaptic transmission.I think forcing this analogy of "the life cycle" of an action potential is what may be getting in the way.

As far as Cl-, I still think it is extremely misleading to say that "Cl- is the principal anion in the action potential". Could you say, "in some algae, anions such as Cl- are also important for AP generation" ?Nrets 19:24, 20 April 2008 (UTC)

• I tried to distinguish between the instantaneous membrane potential V (which might be anything, in principle) and the voltage V_Goldman, which is defined by the instantaneous ionic permeabilities via the Goldman equation. From that perspective, changing the ionic permeabilities changes V_Goldman but not V directly, although V rapidly approaches V_Goldman, tracking it closely. It was clearly my mistake to call V_Goldman the (instantaneous) "resting potential", since that term is usually reserved for the zero-net-current potential in the absence of action potentials, roughly -70 mV; that resting potential does not change during the action potential, whereas my definition does. Since the distinction between the static resting potential and the changing Goldman potential seems to have been lost even on the neurophysiology teacher, it seems overly subtle for the article. We should probably just use some finessed wording that ignores the voltage relaxation, such as "Due to the falling sodium permeability and increasing potassium permeability, the membrane potential returns to a value close to -70 mV, the original resting voltage." Willow (talk) 18:52, 18 April 2008 (UTC)

I think you are getting things confused here a bit. To say the resting potential does not change during an AP is wrong. I think what you meant to say is during the AP the active conductances are added onto the resting conductances to alter the membrane potential. The term "Goldman" potential is not really used either. The Goldman equation uses the relative ionic conductances at any given point in time to predict the membrane potential. So it can be applied when the membrane is at rest or during an action potential. I think that you should just stick to the term "membrane potential" and just use it consistently throughout. Nrets 01:49, 19 April 2008 (UTC)

I can see that you're a very good teacher, always thinking how your students might've misunderstood something. :) I'll try to benefit likewise from your advice. But you should perhaps give me the benefit of the doubt, and I'll do likewise.

The point I was trying to make — sadly hampered by my ineloquence — was that the membrane potential V need not equal the Goldman voltage V_Goldman. (I know that I'm using non-standard nomenclature, but I want to forestall any chance that you misunderstand my meaning; the Goldman voltage is defined by the formula in the article, when the instantaneous ionic permeabilities are substituted.) Please consider that the Goldman voltage is the voltage at which no net current flows across the membrane; yet since current does flow across the membrane, it follows that V is not always equal to V_Goldman. To illustrate my meaning, consider the following thought experiment. Imagine a membrane with voltage-independent potassium and sodium channels, with the potassium conductance being much greater than that of the sodium. Imagine, however, that the potassium channels are initially completely blocked with a covalent photolabile crosslinker, so that the resting voltage equals the sodium Nernst potential. This crosslinker is then destroyed with a picosecond burst of laser light, causing the potassium conductance to increase to its full value in a picosecond. Thus, V_Goldman changes in a picosecond. Do we imagine that the membrane voltage V — the integral of the electric field across the membrane — likewise changes in a picosecond? I think not; it will decay to V_Goldman with a time constant τ given by the local membrane conductance and capacitance. Therefore, I think it wrong to say that V = V_Goldman.

That said, I completely agree that it's a subtle point and would likely be lost on most readers. Therefore, I'd be in favor of a finessed wording that doesn't equate V = V_Goldman (for accuracy) but might suppress V_Goldman in favor of V, as you seem to be suggesting.

The matter of the "resting potential" nomenclature is again relatively minor. I'd assumed that you had deleted my references to V_Goldman as the "resting potential" because you favored the idea of a static resting potential, say, -70 mV. For example, Ted Bullock et al. define it that way in their textbook

“ This stable negative potential, recorded across the cell membrane in the absence of activity, is termed the resting potential. ” — Bullock, Orkand, Grinnell (1977), Introduction to Nevous Systems, W. H. Freeman, p. 131.

However, your explanation above seems to suggest that your conception of the "resting potential" agrees with my conception, so I think we're in agreement? For the sake of clarity for the reader, however, I might suggest that we stick with Dr. Bullock's definition for the resting potential and use another term (such as V_Goldman, or whatever we want to call it) for the instantaneous resting potential defined by the instantaneous ionic permeabilities. As you say in your edit summary, an "action potential" doesn't sound very resting! :) Willow (talk) 20:04, 19 April 2008 (UTC)

So stick to Vm for the actual, instantaneous membrane potential and Vrest for the resting membrane potential. Nrets 19:24, 20 April 2008 (UTC)

• I'm having trouble finding a reference for the observation that a rested axon, stimulated in its middle, can produce two action potentials moving in opposite directions away from the point of stimulation. Any help on referencing would be much appreciated! :) Willow (talk) 19:00, 18 April 2008 (UTC)

• I'd suggest to delete the very last sentence of the section 1.5 'resting potential, first because this discussion of the action potential peak

does not match the titel of the section, and second because it is not quite correct. It should rather tell that the peak (zero slope) marks the time when depolarization by increasing sodium permeabiliy and repolarization by increasing potassium permeability are equal, i.e. the peak is reached before the maximum sodium permeability. Similar misinterpretations of peaks are found frequently in semi-scientific texts (.e.g. of my students). What so ever, this issue might be treated in correct detail (including capacitive currents as well) elsewhere but not at the prominent last pace of this section 'resting potential'. Solfiz (talk) 15:08, 12 September 2008 (UTC)

Copy editing questions on "Initiation, propagation and termination"

Here are my uninformed and silly questions!

Bless you and thank you for you-know-what and you-know-why. :) My magnolia tree burst into bloom today, so I'm in a much better place. ;) Willow (talk) 22:29, 19 April 2008 (UTC)

Initiation

• A typical action potential is initiated at the axon hillock when the membrane there becomes sufficiently depolarized, i.e., when the membrane voltage reaches threshold. - What kind of threshold? Some sort of electrical threshold?

Yes, it's a voltage threshold? When the membrane voltage rises a certain amount (say, 15 mV) above its normal voltage (say, -70 mV), then a runaway condition results in an action potential; if it's raised by a lesser amount, the voltage will gradually decay back to -70 mV. Willow (talk) 22:29, 19 April 2008 (UTC)

I'm not sure these specifics made it clearer. I think they just made the issue more confused for me. Awadewit (talk) 03:10, 22 April 2008 (UTC)

• The action potential then propagates along the axon without diminishing, being created anew at every step. - "being created anew" - I kind of get this, but not really - could we be more precise?

Yes, I would be grateful for a better way to explain it! Like Winnie the Pooh, I sense that there must be a better way, if I could only stop beating my head to think of it. ;) The basic idea I was trying to convey is that an action potential—seen as a sharp rise and fall of voltage—occurs at a single point on the axon, but that its consequences can spark a new action potential elsewhere on the axon. An analogy might be to a forest fire; the action potential at a point is like the incineration of a single tree; it runs its course from beginning to end, and stops. But sparks from that tree may ignite a neighboring tree, which is then inflamed and runs its course, etc. In each case, the energy released in the fire comes from the burning tree itself, not from the spark that ignited it or the tree that made the spark. So too, an action potential at one point causes internal currents within the axon that can depolarize nearby patches of membrane, causing them to fire; the first and second membrane patches may be well-separated, as they are in myelinated fibers: two nodes of Ranvier. But each such patch of membrane provides its own energy for its own action potential, just as the tree provides its own energy for its own incineration. If the trees are all the same, the fire of each burning tree will likewise be the same; so too, if each patch of (unfired) membrane is the same, the amplitude of the action potential fired there is the same; the action potential and the forest fire travel without diminishing.

That was an excellent explanation. Hmm. Perhaps we should assume the readers can follow your explanation and don't need a metaphor like me? Awadewit (talk) 03:10, 22 April 2008 (UTC)

• The axon may branch along its length, and the action potential may fail to propagate along one or both of the branches. - I think this sentence needs to be connected to the one before it. I wasn't sure of the relationship between the two.

I went ahead and re-wrote that lead-in paragraph rather substantially. Umm, I hope it's better now; it's certainly longer. ;) Willow (talk) 23:40, 19 April 2008 (UTC)

It's a lot more specific. Harder to follow, but probably more precise. Awadewit (talk) 03:10, 22 April 2008 (UTC)

• This binding opens various types of ion channels, changing the local permeability of the cell membrane and thereby altering the resting potential, by the Goldman equation. - Since the Goldman equation isn't explained, should it really be mentioned?

You're right! I took a lesson from Nrets and finessed the wording to eliminate the Goldman equation and "resting potential" idea in favour of "membrane potential". It was a technical point that would down most readers, and not really contribute to understanding. Willow (talk) 23:40, 19 April 2008 (UTC)

• Whether the voltage is decreased or increased, the change propagates passively to nearby regions of the membrane, as described by the cable equation and its refinements (see below); typically, it decays exponentially with the distance from the synapse and with time from the binding of the neurotransmitter. - I wasn't totally sure what the "it" was supposed to be after "typically" - I think it should be replaced with "change" or "action potential".

I replaced "it" with "voltage stimulus"? Strictly speaking, it's the amount by which the voltage exceeds (or falls below) the local resting voltage. Willow (talk) 23:40, 19 April 2008 (UTC)

• In sensory neurons, signals from the environment are transduced into action potentials - Can we come up with less sophisticated word than "transduced"? (This would go for subsequent uses of the word as well.)

Ghastly, no? I went back and could scarcely believe how obscurely it was written — blech! :P Hopefully, it's better now. Willow (talk) 23:40, 19 April 2008 (UTC)

I've never really liked the "as described above/below" formulation. I think many Wikipedia readers jump around. They won't know what that is referring to. Any way to remove those phrases from the page? Awadewit (talk) 03:10, 22 April 2008 (UTC)

I think I got rid of them all. Willow (talk) 20:41, 22 April 2008 (UTC)

• However, not all sensory neurons transduce their signals directly into action potentials. - I had to read this paragraph several times before I understood this idea - it's not actually that complicated. I wonder if there is a way to explain it more simply? I got hung up on all of the terms I didn't know, so it was hard for me to see the simple concept.

Another "did I actually write that?" moment. I merged the two paragraphs and tried to clarify the concepts. Willow (talk) 23:40, 19 April 2008 (UTC)

I still think the end of the paragraph is introducing too much jargon instead of just explaining the simple concept that some sensory neurons convey their signals indirectly. Awadewit (talk) 03:10, 22 April 2008 (UTC)

I tried to simplify it; is it better now? The word "directly" was perhaps not the best choice. In my own mind, I picture the difference between pulse-like action potentials and the other more continuous types (graded potentials, neurotransmitter signals) as being like the difference between staccato prose and lyric poetry. The same ideas can be expressed in both, no? I suspect (OR warning, I have no reference) that the sensory neurons use their initial continuous signals (the lyric poetry) to do some preprocessing among themselves on the information, so that they don't need to send all the information up the optic nerve? Once they "compress" the information down, then they encode it into action potentials (staccato news-reporter prose) and send it to the brain. Willow (talk) 20:41, 22 April 2008 (UTC)

It's an analogue to digital signal conversion. Tim Vickers (talk) 21:23, 22 April 2008 (UTC)

• The timing of such pacemaker potentials can vary with external stimuli, just as the heart rate can be altered by pharmaceuticals as well as signals from the sympathetic and parasympathetic nerves. - But I thought the whole point was that these cells were not being regulated by external stimuli? (Sorry for my denseness here.)

It's a little bit of a semantic issue. The cells would fire on their own, but the rate at which they fire can be tuned by other inputs, just as you can adjust an old-fashioned windup metronome by fiddling with the weight. The internal spring drives the oscillation, but the weight adjusts its rate? Hoping I'm being clear, Willow (talk) 23:40, 19 April 2008 (UTC)

Clearer now. Thanks. Awadewit (talk) 03:10, 22 April 2008 (UTC)

Propagation will follow in a moment! Awadewit (talk) 22:12, 18 April 2008 (UTC)

Thank you so much; as you see, your suggestions are really helpful! :) Willow (talk) 23:40, 19 April 2008 (UTC)

Propagation

• The currents flowing inwards at a point on the axon during an action potential spread out along the axon - Sorry, what currents?

Those are the sodium ion currents that flow inwards when the sodium channels are opened during the action potential. The currents are responsible for changing the membrane voltage. Should I specify that they're (usually) sodium ion currents? In some cases, other ions might contribute, such as calcium. Willow (talk) 03:12, 20 April 2008 (UTC)

Best not add that confusion - this is probably just an artifact of me having the read the article is pieces over several days. Awadewit (talk) 03:16, 22 April 2008 (UTC)

• The currents flowing inwards at a point on the axon during an action potential spread out along the axon, and depolarize the adjacent sections of its membrane - of the axon's membrane?

Yes, exactly! :) Willow (talk) 03:12, 20 April 2008 (UTC)

• I noticed that sometimes the article refers to the "synaptic buttons" and sometimes to the "synaptic knobs", but they seem to be the same thing. It would be easier for readers like myself if the article stuck with a consistent terminology since we are trying to learn lots of new terminology.

Yes indeed, they're the same. I believe that the more-or-less official English term is "synaptic knob"; I started using "synaptic button" because I liked the corresponding French term, bouton, which sounds better to my ears and is more picturesque for the reader. Hardly a good reason, I know, and I tried to switch them before, but I guess I missed a few. One second...okay, I think I got them all. Thanks for catching that! :) Willow (talk) 03:12, 20 April 2008 (UTC)

• If a neuron is myelinated, an action potential at one node of Ranvier provokes another action potential at the next node, although no action potential occurs on the intervening segments of membrane. What do you think about explaining myelination and the nodes of Ranvier? These seem crucial to understanding this kind of conduction. How much clicking should the reader have to do? (Currently, myelin is described a bit more in the "Cable theory" section, but earlier sections depend upon knowing what it is.)

That's a great idea! I mentioned them earlier in the article, but it'd be good to refresh our readers' memory here, just before the information is needed. I might move some of the "Cable thory" material upwards...Willow (talk) 03:27, 20 April 2008 (UTC)

This is very helpful - thanks. Awadewit (talk) 03:16, 22 April 2008 (UTC)

• Although the mechanism of saltatory conduction was suggested in 1925 by Lillie,[1] the first experimental evidence for saltatory conduction came from Tasaki[2] and Takeuchi[3] and from Hodgkin and Stämpfli. - Can we get first names?

Yes, it's better to be more personal, and avoids confusion with other neurophysiologists of the same name, who do exist. :) Willow (talk) 02:56, 20 April 2008 (UTC)

• Myelination confers several important advantages. - On the conduction process?

Myelinations confers faster conduction velocity for a given axonal diameter or, conversely, a smaller diameter for a given velocity, thereby saving space. By preventing ions from leaking out between the nodes of Ranvier, not as much energy is wasted, too; myelination gives a metabolic savings. Willow (talk) 03:27, 20 April 2008 (UTC)

That ever-helpful colon. :) Awadewit (talk) 03:16, 22 April 2008 (UTC)

• This saving is a significant selective advantage, since the human nervous system uses approximately 20% of the body's metabolic energy. - We are talking about frogs and squid and suddenly humans - if the logic could be explained a little painstakingly, it would help those of us struggling with the topic!

OK, I'll try my best! :) Willow (talk) 03:27, 20 April 2008 (UTC)

Last section coming right up barring any unforeseen events! Awadewit (talk) 22:50, 18 April 2008 (UTC)

Termination

• Action potentials on an axon almost always move in the same direction: "downstream" from the axon hillock to the axonal termini, which are called the synaptic knobs or buttons. - I feel like this sentence is repeating information we have already been given.

Yes, you're right; I eliminated that whole paragraph. That particular sentence was merely meant as a recap to frame the question and refresh the reader's memory; but the whole topic was better covered under "Propagation". Willow (talk) 11:55, 20 April 2008 (UTC)

• The absolute refractory period of the action potential allows this one-way propagation - I'm not really sure what "absolute refractory period" means.

I merged this stuff into the Propagation, and explained the absolute refractory period in the opening sentence. It was explained earlier in the article, and I'd been relying on the reader's memory; but for an article of this length and variety, it's good to refresh their memory. Willow (talk) 11:55, 20 April 2008 (UTC)

• However, certain arrythmias of the heart result from an action potential stimulating an 'echo" of itself. - Interesting, but crucial?

Yes, you're right. I wanted a counterexample to unidirectional propagation, a bad situation where the neuron (or a neural system) constantly restimulated itself, instead of moving linearly from beginning to end. I deleted this, although I might re-introduce it someday into the Cardiac action potential section, where arrhythmias are discussed. Willow (talk) 11:55, 20 April 2008 (UTC)

• The action potentials that do reach the synaptic knobs generally cause a neurotransmitter to be released into the synaptic cleft. What is the "synaptic cleft"?

It's the gap between the pre- and post-synaptic cells; the neurotransmitter is released from the former, diffuses across the cleft, and binds to receptors on the latter. I moved the image up to clarify that; it's better earlier, I think. Is the present version OK, do you think? Willow (talk) 11:55, 20 April 2008 (UTC)

I still don't really know what the cleft is from the article's description. Awadewit (talk) 03:21, 22 April 2008 (UTC)

• The post-synaptic membrane on the dendrite is often raised up to form a spine. - Is this part of the description of what goes wrong? I wasn't sure what its relationship to the rest of the paragraph was.

Ummm, its relationship was "random factoid that occurred to Willow as she was editing." It's gone, gone, gone. ;) Willow (talk) 11:55, 20 April 2008 (UTC)

• In some cases, the cytosols of two excitable cells (such as neurons) are connected directly by relatively large pores known as connexins. - What are cytosols and connexins, exactly?

They weren't essential terms, so I kind of finessed the wording? For completeness, the cytosol is the "juice" inside a cell, the aqueous solution that holds the ions whose currents are so important here, whereas a connexin is the building block protein of the pore (the gap junction) connecting the two cells. The idea is that the ionic current flowing in the presynaptic neuron can crossover into the second cell through the gap junction, stimulating the postsynaptic cell without the intermediary of a neurotransmitter. Such electrical synapses are faster and have more precise timing than do the more customary chemical synapses. Willow (talk) 11:55, 20 April 2008 (UTC)

If the terms aren't essential, I wouldn't introduce them. This article is full of new jargon for the reader. It is overwhelming. Any way to cut down on that should be seized upon! Awadewit (talk) 03:21, 22 April 2008 (UTC)

I hope this helps. By the way, I was rather overwhelmed by images in this section. You might try cutting some out (gasp, I know). Awadewit (talk) 23:04, 18 April 2008 (UTC)

I tried to space out the images a little better and got rid of one as well. Is it better now? I suppose there is such a thing as too much eye-candy. ;) Thank you again, from the bottom of my heart, Willow (talk) 11:55, 20 April 2008 (UTC)

Much better! Let me know if you need any more help. Awadewit (talk) 03:21, 22 April 2008 (UTC)

Proposal to move "Mathematical Models" section to sub-page

How do people feel about moving essentially the entire Mathematical Models section to linked pages? I think the fact that there *are* mathematical models to describe action potentials is clearly relevant, and perhaps we can summarize that fact in a few sentences instead of discussing each model in detail. (In a couple of cases, the discussion of these models here seems to duplicate or even exceed the depth of the pages themselves.) How about leaving a stub like this:

"Several mathematical models have been developed which analytically describe experimental measurements of current and voltage. The most popular of these are the Hodgkin-Huxley model, for which Alan Lloyd Hodgkin and Andrew Huxley were awarded the Nobel prize in 1963, and the Fitzhugh-Nagumo model."

And then we could move the bulk of the content here into the specific articles on those two models. Oh, and of course my summary "stub" could be vastly improved and made a bit beefier, but how do people feel about the general idea of "delegating" some of this discussion to other pages (with the goal of reducing length and improving readability)? AndrewGNF (talk) 00:48, 19 April 2008 (UTC)

• This makes me nervous. As I have been repeatedly told by more informed people than myself, without the math, there is no science. And when I mentioned this proposed move to my roommate, who is a physics major interested in biophysics (a fascinating field!), I received a shocked "What?!". :) Perhaps someone could explain to me how we can understand action potentials without these mathematical models? That would go a long way to assuaging my nervousness. How similar is this to removing Schrodinger's equation from an article on quantum mechanics or Newton's laws from the force article? Awadewit (talk) 19:42, 19 April 2008 (UTC)

I think we agree that some of the MM section should remain; the question is how much? I can see both sides. On the one hand, maybe all the casual reader wants is a qualitative understanding, rather than a quantitative understanding. For example, if we were to describe the flooding of New Orleans, it might be fine to say that the levies gave way and the cite was mostly flooded; we might not need to describe how quickly it flooded or where exactly it flooded? I feel it's generally better to be more specific and more complete, within the limits of the reader's powers of concentration; but that's probably why my articles tend to as stuffed as a Victorian living room until my friends help me redecorate. ;)

On the other hand, mathematical/computational models of action potentials (and more generally, excitable membranes) are a major area of research and have been so for fifty years. Hodgkin and Huxley were arguably awarded the Nobel Prize mostly for their mathematical model, since they didn't invent the apparatus. New studies are being done all the time, to find enlightening simplifications, to model new types of channels or more complex systems of neurons. The van der Pol model seems to be a standard model system for the study of dynamical systems, although I'm guessing most mathematicians don't know that it's a model of the cardiac action potential. ;)

Maybe I'll just try to shorten it as Andrew suggests, and then we can experiment, if we felt that too much or too little had been cut? But before that happens, I'd appreciate other people's inputs! Willow (talk) 20:51, 19 April 2008 (UTC)

For what it's worth, I think the MM section is now a *fantastic* summary the topic for the action potential article. I think of it as a teaser -- enough so that every AP reader knows that there's a much deeper discussion available, but without being overwhelming the "main topic". Bravo... At the risk of becoming that guy who's full of suggestions but doesn't actually do anything, might I suggest that this treatment could also be applied to the "Experimental methods" section? This one might be harder because of the diversity of topics under this broad heading, but again, I think the argument could be made that these important related topics should be in separate article(s). But nice job! AndrewGNF (talk) 20:10, 21 April 2008 (UTC)

Proposal to Rename Article

Based on all of the new information added to this article, and the arguments of whether to add or retain certain sections, I would like to propose to rename the article something like "Properties of Excitable Membranes" or "Membrane Excitability", of which the action potential is only one manifestation. We could have action potential redirect here. In this way everyone is happy, the article can include all the subsections without any of them being beyond the scope of an action potential, and all of Willow's hard work goes to good use. Any thoughts?? Nrets 19:30, 20 April 2008 (UTC)

A rose by any other name... :)

The re-naming is fine with me, and I would happily go along with what everyone else wants. But I'm a little uncertain about the scope of the new title, "Properties of Excitable Membranes". If it's meant to cover only membranes that can support an action potential, then why not stick with the more familiar title, "Action potential"? If, on the other hand, it's meant to cover all electrically active membranes, such as those of photoreceptor cells, then the present article seems inadequate, even at its current size. If, God forbid, I should find myself wanting to write an article on chemical synapses (or electrical synapses, or the neuromuscular junction), I would feel compelled to make it encyclopedic, e.g., talking about every major variety of channel, with maybe a sentence or phrase each about its discovery, electrical properties, physiological function, tissue expression, subcellular localization, posttranslational modifications, structure, regulation of activity, connections to diseases, and so on. I'm sure, for some of you, that confession only illustrates the madness that led the present article. ;) And I do recognize the possibility of summary style. My only point is that, if the second scope is intended, then I suspect we would need to re-engineer the article to focus much less on the action potential and much more on other electrophysiological phenomena. That would require a lot of work, and time.

I'll confess that I don't see why it's so undesirable to discuss the events that produce action potentials and the effects that action potentials have once they've run their race. For comparison, if we were discussing proteins, wouldn't we want to discuss their synthesis and degradation? I think we all would want to have sections about ribosomal synthesis and proteasomal degradation, no? Similarly, when we write biographies, we often mention the historical context into which the subject was born, and the legacy that they left behind. Describing a little bit about the chemical/electrical events at synapses and the neuromuscular junction—our coverage of them is pretty minor, right?—also sets a nice contrast against which the action potential might be better understood. Of course, we can cut them down to their bare minimum, if that's desired.

I truly am sensitive to the length of the article and will be trying to cut it down, beginning with the unloved "Mathematical models" section. I also appreciate everyone's input and corrections, even if I offer questions and concerns; we're in it together and pulling together! :) I may also trim the "Experimental methods" section, since I recognize that most of the methods are used to study both action potentials and other potentials. I'm of two minds, though, since I also feel we should tell our readers how scientists arrived at the present understanding, and the evidence that supports it. Just because an experimental method works for two systems, doesn't mean that we can't mention it at both places, right? I think the methods help people to realize the significance of an advance like patch clamping or the glass micropipette electrode,; they see how huge fields of study were opened up for research because scientists overcame a few apparently minor technical hurdles.

We also need more referencing; that bi-directional propagation reference isn't the only one! ;) If you know of good references, textbook or otherwise, or better ones than I've used, it'd be a very nice gesture to add them to the article. Thanks muchly! :) Willow (talk) 21:30, 20 April 2008 (UTC)

Wording

"Membrane Excitability" is fine, as long as "action potential" searchers are directed to this site. I'm puzzled about 'potentials' in electrophysiology anyway when voltages (= differences of electrical potentials between inside and outside) are meant. Many authors do avoid this slang and use "reversal voltage", "membrane voltage", "clamp voltage" etc. instead. Only the "action potential" is so well established that it may hardly ever be extinguished. My personal approach is: spell it "electrical excitation" and pronounce it "action potential". Solfiz (talk) 15:48, 27 August 2008 (UTC)

I don't know if it would be the wisest idea to rename it Membrane Excitability. Most people search for action potential. And quite frankly, this would be like taking the Ford article and renaming it the automobile article with the justification that Ford is just one manifestation of Automobiles. But it might be a good idea to create another section called Membrane Excitability. Another problem is that this article is already 100KB, at this stage we should consider splitting it up. If we make the article any bigger, it might be unmanageable. Paskari (talk) 22:30, 27 August 2008 (UTC)

In my view, the rationale behind the renaming to "membrane excitability" would be that this article does waaaay more than just talk about action potentials. It's not quite the same as the "Ford vs. Automobile" argument because the Ford article talks specifically about Fords and not about automobiles in general. --David Iberri (talk) 01:09, 28 August 2008 (UTC)

Darn it, and I spend so long coming up with that analogy. I suppose you are right in that this article does so much more than just describe the action potential. Furthermore, it's almost impossible to take most of the material out, because one needs an understanding of ion flows to understand how action potentials are brought about. It's just that Membrane Excitability is no where near as sexy as action potential.Paskari (talk) 12:13, 29 August 2008 (UTC)

Oh, I agree entirely. This article should definitely remain at Action potential (although there's nothing saying we can't have a separate Membrane excitability article). The problem is with its scope. All the topics mentioned on this page are relevant -- membrane potential, propagation, synaptic transmission, mathematical models, etc. -- but this article doesn't need treatises on each of those topics. That's the point of wiki links: discuss only what's necessary to empower your readers so they can understand the article in some context; everything else belongs in a wiki linked article. --David Iberri (talk) 15:38, 29 August 2008 (UTC)

This is going to be a tricky one.Paskari (talk) 17:50, 29 August 2008 (UTC)

We could start by merging the material in sections 1.1,1.2.1 and 1.5 with the membrane potential article (it's not as well written an article as this one). Also, we could rename this article something like action potential (in neurons) and move section 6 into different articles. Also, I'm not too sure why neurotoxins has its own section.Paskari (talk) 16:50, 31 August 2008 (UTC)

Trimmed article

The article has been trimmed by roughly 23kb, and focused more on the action potential itself. We now have roughly 50.5 kb of readable prose, which falls within WP:SIZE (barely). We also now have 130 scientific references — but well-meaning editors can always add more, where there are gaps. :) Willow (talk) 11:00, 21 April 2008 (UTC)

Goldman Equation

Looking at the way it is written, it looks like the superscripted plus (+) signs of the ions next to the P's are actual plus signs, meaning that one should calculate P+[ion] rather than P.[ion]. Would it be clearer if the equation could just be re-stated using specific ions (K, Na and Cl), rather than the more complicated al-purpose version with generic ions? See here for an example: http://www.math.pitt.edu/~bard/classes/passive2/node3.html. (No that's not my website). Nrets 14:30, 21 April 2008 (UTC)

Sure, that's easy to do! If you're OK with it, I'll also add Ca, since it comes up a few places within the article. As an aside, I've never believed that you were acting out of anything but a sincere desire to improve the article for Wikipedia's readers, and I'm very sorry if I gave you another impression. :( It's possible for two well-meaning people to respect one another, even when they differ, don't you agree? :) Willow (talk) 15:03, 21 April 2008 (UTC)

I hate to keep disagreeing with you, and being a pain, BUT... typically Ca++ is not included in the equation because, (i) the concentration of Ca++ relative to other ions is so low that it will have only a small effect on Vm (except maybe during a Ca++ spike) and more importantly, (ii) as far as I'm aware, there is pretty much no resting conductance to Ca++, and you are using in this case the Goldman equation for Vrest. If you insist on adding Ca++ just make sure you account for valence (z) in the RT/zF part of the equation. Nrets 01:14, 22 April 2008 (UTC)

Please don't worry about disagreeing (nicely) with me — I like it! :) I'm keenly conscious of my blind spots, and appreciate people opening my eyes to what I've missed. I think you raise some very good points about calcium. I totally agree about its negligibility most of the time, but I'd wanted to include calcium because it shows up in a few key places, like the cardiac action potential, where the Goldman voltage matters? But I ended up not including it in the Goldman equation, because it really complicates the formula. :( When the valencies z differ in the G. eq. derivation, you can't factor them out, unlike the Nernst equation. :( I included a reference that explains it better than I can, but I might update the Nernst and Goldman equation articles for completeness. Hoping that you're liking the article's recent improvements, Willow (talk) 11:21, 22 April 2008 (UTC)

I included a referenced (monovalent) derivation at the Goldman equation article; I hope you like it! :) Do you think I should write out the divalent derivation, too, or maybe just give the final result? Thanks again for your help and thought-provoking suggestions! :) Willow (talk) 15:54, 22 April 2008 (UTC)

Trimmed Experimental methods section

I trimmed and re-organized the Experimental methods section. Comments, suggestions and critiques are always welcome! :) Willow (talk) 20:16, 22 April 2008 (UTC)

PS. I notice that Medos2 has revoked his Remove vote at the FAR. We must be doing something good! :) Willow (talk) 20:16, 22 April 2008 (UTC)

Willow without a doubt you were the main force in me doing that. You are have been a star with your edits. Medos ^{(talk • contribs)} 14:13, 25 April 2008 (UTC)

Awadewit is imploring that more editors help Willow with Action potential. What kind of help are you looking for, Awadewit? I see lots of talent around here! –Outriggr § 02:13, 23 April 2008 (UTC)

Am I allowed to answer? ;) The main objection remaining at FAR seems to be the several sections that are as yet unreferenced. If you have access to a modern science textbook that covers this material, or to a database of scientific papers, you might be able to track down the references, say, for a paragraph or two? if we all did that, it would go a lot faster.

But there are many other ways of helping, too. For example, it would be great if you could just read through the whole article from beginning to end, and try to find the "glitches", the parts that are confusing or unclear or inconsistent or (Heaven forfend) incorrect. Perhaps the whole article is all four, in which case I should move to Outermost Thule and raise merinos; I know I can spin that yarn. ;) Willow (talk) 15:44, 23 April 2008 (UTC)

Sorry Willow, I realized later that I may have appeared to be cutting you out of the question—not my intention at all. It was my intention not to be presumptuous about the "help required", since you're the driving force here, and have improved it so much already.

I like neuroscience topics, but am not qualified to reference-hunt. I would also take the top-down approach to referencing—much the opposite of the prevailing winds these days—meaning, what needs to be referenced? Again, will leave it to the experts.

I will print the article, and see if I can provide any [bad pun alert] potent actionables! –Outriggr § 23:53, 23 April 2008 (UTC)

Thanks for your help, Outriggr! :) It was really good to hear from you again — it's been too long, no? As I think you know, I was just pulling your leg about being allowed to answer. ;) Anything you'd like to help with, whether copy-editing or smoothing bumps in the exposition or whatever, would be very welcome. :) Oh, and you could vote Keep at the FAR. ;) Willow (talk) 18:34, 24 April 2008 (UTC)

It has been a while! I have fond memories of the time we imagined ordered animation in the molecular twittering image (don't ask me what FA that was), and the time I wrote to you in all math markup.

Onto business, and with the caveat that I still haven't printed the article, can I ask that you reconsider all the "Main article" templates? The article is chock full of intermingling concepts, and I find that most of the "see alsos" after the headings are linked in the text, often in the first sentence afterward. I find the templates add repetition and density to an already dense article. If you don't see this message for a while, I might make an edit to show you what I mean, and you can amend as you see fit. For now, –Outriggr § 00:17, 25 April 2008 (UTC)

Na/K pump figure

May I suggest removing the Figure of the Na/K exchanger? The function of the pump is somewhat beside the point for this article and it just increases the amount of preliminary information that needs to be "digested" before the reader finally gets to the relevant bits. Nrets

Sure! :) When I saw your edit summary, I thought you wanted me to make a more realistic animation of the pump's action, maybe based on the crystal coordinates. What a relief! :) OK, consider it gone. Willow (talk) 18:14, 23 April 2008 (UTC)

Here to help.

I think I can help. I'm a biology student, and my University gives me access to a lot of journals, I also have a reasonable number of sources here at home. What needs done first? Shoemaker's Holiday (talk) 23:51, 24 April 2008 (UTC)

• Thanks Shoe! We appreciate the help! Awadewit (talk) 08:55, 25 April 2008 (UTC)

Thank you also from me, Shoe! :D I'm really glad that you'd like to help out. I think the main thing now is to smooth out the exposition, so that the reader has the feeling of inexorable flow; there shouldn't be any glitches in style or logic. However, if you saw sentences that needed a more specific citation, then please feel free to add them. For example, we still need a reference for the observation that stimulating a fresh axon in its middle produces two action potentials, one moving orthodromically and the other antidromically. If you could find the original reference for that, that'd be great! :) Also, I don't have access to the Kandel reference, which I feel sure should be a rich source of supporting references. It's probably mostly redundant with the 4-5 textbooks we've used already such as Purves et al. (2008); but still Kandel seems to carry such weight that if you could add those references to the others, I think it would strengthen the article's credibility.

Alternatively, you could tackle another topic in biology and bring it to FA status. You shouldn't be daunted; it's not that hard and you'll have plenty of help, methinks. There are so many articles that need help desperately, that I'm sure you could find one that inspires you. You might make more of a quantum-leap contribution that way, too, rather than refining this article. Maybe you'd like Acetabularia? That was the organism in which it was first shown that genes reside in the cell nucleus, that DNA was likely the carrier of genetic information, not proteins. Or how about methylglyoxalase, an extremely cool enzyme that I was working on with Tim? You'll find him fun to work with. :) Willow (talk) 17:41, 25 April 2008 (UTC)

Alright! Sorry I was a bit slow to leap in and thus missed out in a lot of it - I'm a little ill of late. Think I might try and tackle evolution of parasitism in insects and related taxa. Though it needs a better name. Shoemaker's Holiday (talk) 00:23, 26 April 2008 (UTC)

Congratulations on a fantastic save!

I read the article last night, and was amazed by the wealth of information: I learnt a lot about the amazing way that nerves work; cell biology is awesome. I found a few nitpicks and was planning to help out a bit this evening, but I see, as ever, that I'm too slow! Congratulations to all, especially Willow, on producing a work of startling depth and information, and hence saving another featured article for the encyclopedia and its readers. Geometry guy 18:55, 25 April 2008 (UTC)

Well done Willow! –Outriggr § 23:56, 25 April 2008 (UTC)

Information overload

First, I want to acknowledge the time and effort that a few people have dedicated to revise and improve this entry. There are several good ideas in this article, and their work is to be commended. However, I am seriously concerned with how it is presented.

This entry, as written, seemingly attempts to summarize an entire field of neuroscience, often called "cellular neurophysiology." For example, concepts like resting membrane potential, dendritic integration, synaptic transmission, ion channel structure/function, neuropharmacology, and more are all covered. This is done at the expense of fulfilling the fundamental obligation of any encyclopedia: to provide a layperson with a clear, concise, and accurate source of information.

Indeed, when one attempts to condense so much information, there are bound to be factual errors. Since I would advocate cutting this article in half (or more), it doesn't make sense to go through and fix them point by point at this time.

I would ask that the person/people most actively responsible for editing this entry ask their parents, grandparents, siblings, friends, children, etc. to read through the current entry and gauge their level of comprehension. The beauty of Wikipedia is that it can jump start investigation into unfamiliar intellectual territory, but it can't cover everything. Unfortunately, I have the feeling this article will confuse more than it will clarify. Mark 67.176.225.196 (talk) 08:13, 25 June 2008 (UTC)

Admittedly, it is a lot of information to absorb, especially terminology. But it is intelligible and enlightening to lay-people, as experience shows. The article is also broad in its coverage, but we're supposed to be encyclopedic; the article needs to cover most aspects of the action potential, and I think you'll find that everything in the article pertains to AP's.

My own feeling is that I don't have to write technical articles for my grandparents, who have never been online and likely have never heard of an ion, but for serious students who wish to be informed. In writing this article, I was actually inspired by friends' comments about Wikipedia, who complained about (1) having to jump back and forth between linked articles and never being able to read a self-contained article, and (2) articles being written for the expert (which makes them useless) by assuming too much prior knowledge ("review article wannabees"). It's a delicate balance, I realize, and every editor has to follow their best judgment. I feel that the amount of space devoted to giving background knowledge here is minor compared to the space devote to action potentials proper. Perhaps my mind will change with further reflection, but I feel that you could not remove half of the material here without damaging the article profoundly.

I do appreciate where you're coming from, though, and see two solutions. First, we can work on making the lead of this article more intelligible. Or perhaps we could start Introduction to action potentials? Willow (talk) 11:26, 26 June 2008 (UTC)

I would be content with a revision of the lead paragraph(s) such that it served those who were interested in as brief, clear, and comprehensive explanation as possible. In my opinion, it would start with striking the phrase "pulse-like wave" from the first sentence, as it simply isn't very clear, and I'm not convinced it applies in all cases. The definition from Dictionary.com is much more straight-forward, and could be used as a starting point: "A momentary change in electrical potential on the surface of a cell, especially of a nerve or muscle cell, that occurs when it is stimulated, resulting in the transmission of an electrical impulse." I will then attempt to identify and provide suggestions/corrections (with appropriate citation) for what I perceive are confusing/inaccurate passages in the body. Thank you for your willingness to continue to revise this article. Mark 67.176.225.196 (talk) 22:08, 26 June 2008 (UTC)

I have to dart off to work, but here's a quick note. I don't like the dictionary.com definition, since it omits the crucial point of self-propagation? For example, a transient stimulation of the membrane that decayed exponentially in space and time would also fulfill their definition, it seems. Also, I'm leaving in a few days for my sister's wedding, and won't be back until late July; please don't imagine that I don't want to work with you, it's just that I'll be away. Thanks for being so polite and understanding, and I appreciate your advice on how to improve the article very much, Willow (talk) 22:39, 26 June 2008 (UTC)

Ions and the forces driving their motion

Ions and the forces driving their motion

Ions cross the cell membrane under two influences: diffusion and electric fields.[6] Diffusion allows net flow of ions from regions where the ions are highly concentrated into regions of low concentration. Ions also move in response to an electric field. By definition, the integral of the electric field across a patch of membrane equals the voltage Vm across that patch.[note 2] Likewise by definition, the flows of different ions through that patch are the ionic currents at that patch; the total current is the sum of all the individual ionic currents. Using these definitions of voltage and current, such a membrane patch can be modeled by an equivalent electronic circuit.[7] In particular, for each type of ion the patch will have a capacitance C and a conductance g; according to Ohm's law, the current I of each ion type is related to the transmembrane voltage Vm by the equation I = g Vm. For a given set of ionic conductances, there is an equilibrium voltage E at which the total current across the membrane is zero; the natural flow of ions generally causes the membrane voltage Vm to approach E.[8]

Is there any SPICE file that may be used elsewhere?

How does an electronic circuit take account of moving ions?

How does an electronic circuit take account of ionic surface density?

Do ions move at the same speed? Do ions interact equally at every distance?--Somasimple (talk) 05:17, 28 June 2008 (UTC)

I don't know if there's a SPICE file, there are enough differences between a stereotypical RC circuit and a membrane that it might be difficult to make one (though I've never played with SPICE types of modela). As for how electronic circuits take into account moving ions, the answer is that they don't normally need to. The parameter of interest is current, as with a traditional circuit. Ion movement (i.e. changes in concentration) is important in some locations, such as T-tubules. In those instances, changes in ion concentration is a function of specific current (e.g. K⁺ current) and diffusion. These concentration changes alter the ionic driving force (remember that the voltage across each resistor is not the membrane voltage but rather the driving force of the relevant ion(s)). By ionic surface density I assume you mean channel density. This is conveyed by the resistance. If you mean the concentration of each ion, then this is what sets the driving force. As for atoms moving at the same speed, they don't. Questions regarding atom speeds and atom-atom interactions would be better addressed on a physics related page (or reference any intro physics text). --Dpryan (talk) 06:27, 30 June 2008 (UTC)

I have some difficulty to follow your scientific reasoning, here :

It is said :

In particular, for each type of ion the patch will have a capacitance C and a conductance g; according to Ohm's law

The Ohm's law fits perfectly my present preoccupation. I tried to simulate the propagation of the HH model within a SPICE simulator and it doesn't work, as previewed. A simple RC circuit (even with many cells) is unable to create a delay.

Secondly, I'm afraid to think that a moving ion is also an atom. Physics rules Biology. And it has been proved that ions cross the membrane in both directions. If they did, perhaps they interacted with something since they are charged particles?

The ionic surface density tells us that two ions with the same charge may have distinct electrostatic interactions: Na+ is 1.0 where K+ is only 0.42. When Na+ ions enter the cell, it works in a better manner than k+.

Of course, this fact, contradicts the principles you listed earlier. --Somasimple (talk) 06:03, 1 July 2008 (UTC)

N.B. since this discussion isn't directly relevant to the Action Potential article followups should be directed to my talk page.

An RC model that follows HH equations is capable of reproducing propagation and the various delays properly. You can see a working model's output here and here which shows and action potential propagating down a stretch of skeletal muscle. If you'd like, I can supply you with the fortran code used to generate the traces in the paramyotonia congenita article so you can more closely inspect how a simple model works.

Regarding electrostatic interactions, yes different ions will behave differently in this regard. Unless you're dealing with a molecular dynamics type simulation then such interactions are not something with which we ought concern ourselves. The exception to this is often calcium, which is heavily buffered and acts as an intracellular signal.

As for ionic surface densities, you'd be advised to talk to a physicist but I don't think your understanding of how that might be relevant to action potentials is correct. It will certainly affect interactions, but not in the way you seem to be implying.--Dpryan (talk) 20:22, 1 July 2008 (UTC)

CAUTION : This discussion is relevant to the subject : Action potential and its propagation.

I do not constest that a model works in a dedicated environment. But you're saying it follows the Ohm's rules. So it must work in a general environment dedicated to electronics laws.

Such graph, as you said it, without any context neither a single description is not a valid argument.

So, here is my exemple that comes from this page. Here is the model that propagates. I took this model and put it in a SPICE software and I was able to produce the same graph. That's fine ! But this graph does not validate the concept of any propagation. You must analyze the falling phase and it does not create any delay. No Delay, No propagation.

Your last paragraph lets assume that I certainly unable to grasp the concepts I'm speaking about. I'm certainly stubborn but if I apply strictly what Physics tell us (acceleration, speed, electrostatics...), I find that my model seems to follow the facts. It raises some good questions: What happens when the local concentration level falls down? What consequence it brings to refractory periods? (concentration is at top right corner).

Of course, it contests, one more time, your point of view but you're free to bring any argument that may destroy it: That is the fun with Sciences! --Somasimple (talk) 09:07, 2 July 2008 (UTC)

Just because a discussion has to do with the topic of a particular article does not mean it has to do with the article itself.

That a set of equations conforms to Ohm's law does not imply that it will be accurately represented in a general purpose environment. That the relevant equations and constants are available in published literature (the graphs I posted are from a reconstruction of Wallinga et al., Eur. Biophys. J. 1999), that working code has been offered, and that others (including myself) can and have used such to reconstruct action potentials in silico implies that the problem is at your end. In short, PEBKAC. --Dpryan (talk) 21:23, 2 July 2008 (UTC)

Just because a discussion has to do with the topic of a particular article does not mean it has to do with the article itself.

Since A contains B then B belongs to A. That's Mathematics and it's logical! Your sentence doesn't follow the Logics' rules. As a scientist you must answer to all the questions that contradict your point of view and carefully examine the arguments proposed by your opponent. I have clearly understood that virtual reality may be created in silico: That's what I see when I play on my computer. It mimics a world where real facts may be ignored.--Somasimple (talk) 05:33, 3 July 2008 (UTC)

I have a problem with this paragraph. It starts off fine, but then after this line

By definition, the integral of the electric field across a patch of membrane equals the voltage Vm across that patch.

There might as well be a note stating "Non physicists need not read on". Is there anyway we could make this section more understandable? Thank You. Paskari (talk) 15:38, 30 July 2008 (UTC)

Now that I read it again, it sounds more like I copied it from a textbook and pasted it into wikipedia. Directly preceding it is a very basic description of how diffusion affects the ionic concentric. However, once we get into electric fields, for some reason, the integral of the electric field is described. Paskari (talk) 18:09, 30 July 2008 (UTC)

Myeline Sheeth

in the section "Initiation, propagation and termination", I have a problem with this line "Once started, the action potential propagates down the axon without diminishing; the inwards current of an action potential at one patch of membrane depolarizes nearby membrane patches, sparking another action potential there." I am familiar with the domino theory of action potential propagation, but I was under the impression that the entire axon is insulated with myeline sheeth, therefore, ion channels can only "spark another action potential" at the nodes of ranvier (which is so overly dramatic it might as well be a character in the Lord of the Rings). What exactly does "nearby membrane patches" mean here? Paskari (talk) 12:35, 2 July 2008 (UTC)

This is another problem that remains unsolved (in my humble opinion) actually.

Why the ratio myelin/axon is quite constant all over the neurons of the planet?

When the tight junctions disappear the action potential slows but insulation is unchanged, why?--Somasimple (talk) 14:11, 2 July 2008 (UTC)

If you go down further to the section "Myelin and saltatory conduction" it talks about how the action potential propagates smoothly between ranvier nodes, and at the ranvier nodes it triggers another action potential. So basically, "nearby membrane patches" should be changed to read "at every ranvier node". After having read From Neuron to Brain (12 times), I got the impression that upon reaching threshold, that point would trigger its immediately adjacent point to fire an action potential, and so on. So it would take a thousand, or a million action potentials before reaching the axon terminals. Well since action potentials only really occur at the axon hillock, and the axon is generally myelinated, it's really just an ionic signal which gets repeated at each ranvier node, correct?Paskari (talk) 16:19, 2 July 2008 (UTC)

You're correct in your understanding. Since there are no active conductances under the myelin sheath the action potential diminishes between nodes of Ranvier (it follows the cable equation here). In myelinated neurons "nearby membrane patches" would then mean subsequent nodes of Ranvier. Perhaps the wording in the article could be tweaked a bit here along the lines of saying "nearby exposed membrane patches"...I'm sure someone else can come up with a more eloquent wording.--Dpryan (talk) 20:08, 2 July 2008 (UTC)

Sorry, but I'm not convinced by the cable theory or this section.--Somasimple (talk) 17:10, 2 July 2008 (UTC)

What do you find unconvincing? Xargque (talk) 23:01, 2 July 2008 (UTC)

A real cable has an inductive component that do not exist here. No inductance, no delay.

See the above discussion. The given model is unable to create such a miracle.--Somasimple (talk) 05:12, 3 July 2008 (UTC)

Try again, the delays in HH-type models are derived from the differential equations that specify gating parameters (these along with channel density are what end up determining conduction velocity). As for inductance, have you considered that perhaps others have already discussed models incorporating such parameters (such as Lieberstein in 1967). Of course, others further analyzed such models and did some calculations of their own (see Kaplan & Trujillo 1970 and Scott 1971 in Math. Biosci., which should probably be gone through as a pair) and found that inductance is insignificant in neurons. It's important to remember that while we may think of the axon as a cable and even find electronic cable related equations useful when describing axons (or other excitable tissue) that axons are not cables. To quote Rall (Handbook of Physiology, 1977), who wrote at length about the cable equation and its solutions, "Many have recognized that nerve impulse propagation has much more in common with the traveling chain reaction of a lighted fuse than with electromagnetic wave propagation". So, next time you feel like dismissing something you can't figure out as a "miracle" you should try doing some reading first. --Dpryan (talk) 08:36, 8 July 2008 (UTC)

You summed it up perfectly by saying "Perhaps the wording in the article could be tweaked a bit here along the lines of saying 'nearby exposed membrane patches' ". Paskari (talk) 12:35, 3 July 2008 (UTC)

Axon Terminals and Myeline Sheeth

Are the axon terminals also myelinated? There is a lot of space devoted to the axon, but then it seems as if the ionic charges magically appear at the axon terminal buttons. Are there voltage gated channels on the axon terminals, or are thy insulated with myeline? Paskari (talk) 16:22, 2 July 2008 (UTC)

No, the terminals are not myelinated. Just as an action potential in one node causes sufficient depolarization in the next node to open voltage-gated sodium channels and lead to AP propagation, AP invasion of the final node (just before the terminal) causes sufficient depolarization to open voltage-gated calcium channels and trigger the release of neurotransmitter. This page correctly refers the reader to the page about the chemical synapse for more information on this matter. Cheers, Xargque (talk) 23:03, 2 July 2008 (UTC)

Well it appears as if the page chemical synapse uses the terms chemical synapse and axon terminal button interchangeably (in all honesty, it is often times hard to tell the difference between a synapse, and an axon terminal button, and a neurite...). Nevertheless, it has little information on the arrival of the action potential. The axon and the axon terminal buttons are pretty clear. There is a point when the axon branches into numerous paths (lets call them axon terminals). My question is whether the myeline sheeth branches as well, or if the axon terminals are uninsulated. Basically does the myeline sheeth insulate the axon from hillock all the way to the begining of the axon terminal button? Thanks Paskari (talk) 12:43, 3 July 2008 (UTC)

Does this picture from the article clarify? http://en.wikipedia.org/wiki/Image:Neuron-no_labels.png

Regarding the terms: "axon terminal" is the end of the axon. Usually the end has a chemical synapse, although occasionally there is an electrical synapse. The term "neurite" is a generic term that means any process (or protrusion) coming off of a neuron. Axons and dendrites are the two most common types of neurites. So a neurite is the whole thing from the cell body to the terminal, and the terminal is the end. Many texts do use "chemical synapse" and "axon terminal" somewhat interchangeably. Strictly speaking, an "axon terminal" is an anatomical term (it describes a physical structure) and a chemical synapses is a physiological term which describes a functional structure. It so happens that essentially all chemical synapses happen to be physically located inside axon terminals and most axon terminals house chemical (as opposed to electrical) synapses. I haven't read the chemical synapse article very closely, but it is possible that the terms are used somewhat sloppily there.

One other point: these unmyelinated axon terminal branches that you are describing occur at the neuromuscular junction, however, at many synapses in the mammalian central nervous system, you simply just have a single, unbranched (and unmyelinated) terminal. These central nervous system axons start with the axon hillock and then have alternating nodes and myelin. After the last section of myelin, the axon simply ends in an unmyelinated terminal which has many voltage-sensitive calcium channels that allow calcium in, triggering neurotransmitter release as described in this article and in the article on chemical synapses. Xargque (talk) 20:27, 3 July 2008 (UTC)

Perhaps I should rephrase my question. What happens to the action potential between the point where the myeline ends and the chemical synapse begins? Does the signal attenuate or do voltage gated sodium channels boost the signal to make sure that every chemical synapse gets triggered. If the signal attenuates, then those synapses situated fartherst away are at a disadvantage since they always receive weaker signals. Regarding the picture [1], I've seen it before, but it doesn't look reliable, and it directly contradicts a picture in Neuroscience (Purves et. al.) which shows the myeline extending the entire way to the chemical synapse. Paskari (talk) 12:51, 4 July 2008 (UTC)

I (unfortunately) don't have my copy of Purves et al any more, but I suspect that what they are showing is a CNS synapse (as opposed to a motorneuron synapse onto a muscle (which is depicted in the picture I linked to above). I have always suspected that there are sodium channels along the terminal branches of motorneurons, but I must confess that now I can't seem to find a reference for that.

Regarding the question of whether every synapse gets activated: I don't see any reference to this fact in Wikipedia, but at most central synapses, even when the action potential invades the synapse, the release of neurotransmitter is just a probabilistic event. The probability of release varies from synapse to synapse, but ranges from ~10% to ~90%. For a good review, check out the section entitled "Ca2+ TRIGGERING OF NEUROTRANSMITTER RELEASE" in Sudhof TC (2004) Annu Rev Neurosci 27:509-547. Note, you need a subscription to Annual Reviews of Neuroscience to view this (most Universities should have this or else be able to get the bound version of the journal through inter-library loan). Best, Xargque (talk) 17:30, 4 July 2008 (UTC)

I just think this should be addressed once we can establish whether the myeline continues to the end of the axon terminals, whether there are sodium channels, or whether it depends on the neuron type. I have read Sudhof's SYNAPTIC VESICLE CYCLE, so I am familiar with the concepts of probability. In fact I'm looking at plasticity models for the hippocampus (p,N,q), which is partly why I'm asking such detailed questions. The other reason I am so picky about being as clear yet concise as possible is because I don't believe that anyone who comes after me should have to go through so much trouble to understand plasticity (most notably LTP and LTD). Thanks for the reference though, I'll check out that paper. Paskari (talk) 14:55, 6 July 2008 (UTC)

Thanks for asking all the questions and putting in the time to improve the sites. I am sure that myelin does not extend to the terminals (or to terminal branches in motorneurons). I have asked people I know if there are voltage-gated sodium channels in the terminal branches, and everyone seems to believe that there are, but no one can tell me any reference to show it. The best evidence I can give you is based on calcium imaging, with which people have shown that the distal terminal branches are not "disadvantaged" as you had suggested. In other words, the same amount of calcium comes in to distal terminal branches (those far away from the last bit of myelin) as to proximal terminal branches (those synapses closer to the last bit of myelin). The suspicion, which I believe has never been confirmed, that most neurobiologists have is that there are voltage-gated sodium channels that aid signal propagation. However, it cannot be ruled out that the signal might be propagating via depolarization as a direct result of the calcium flowing in (calcium is also positively charged) or that there might be more calcium channels at the distal terminals, allowing them to compensate for a lesser depolarization.

So, in summary, the terminals are not myelinated. Conventional microscopy techniques have firmly established that. However, as to whether there are sodium channels along the terminal branches so as to ensure that all the vesicular release sites get equal depolarization: indirect evidence suggests "yes", but direct evidence does not appear available. Xargque (talk) 16:36, 6 July 2008 (UTC)

I see. I guess it's safe to assume that there are channels on the terminal branches. Now the tricky part is somehow including this into the article. There's nothing I hate more than presenting a partial answer, which even then only applies half the time. Thank you for the reply. Paskari (talk) 11:29, 7 July 2008 (UTC)

The curses of a finite data set, my friend. Xargque (talk) 17:21, 7 July 2008 (UTC)

From some quick searching on pubmed it seems that there's some variability by neuron type and study as to the density of presynaptic sodium channels. In the Calyx of Held (Leão et al. J. Neurosci. 2005), which is one of the few (if not only) recordable presynapse, sodium channels are isolated to a small strip in the heminode. It seems that this is sufficient to carry the action potential into the presynapse (though with a shortened half-width). However, there do seem to be higher densities of sodium channels in pinceau-type nerve terminals of cerebellum and mossy fiber boutons. --Dpryan (talk) 07:40, 8 July 2008 (UTC)

Well this is quite a pickle. Intuitively, I would think that the axon terminals would be myelinated or at the very least have sodium channels. If a pyramidal cell makes 10,000 connections, and the action potentials branches off to reach the different axon terminal buttons (and the propogated signal decays exponentially with distance), I doubt any voltage gated calcium ions will be activated by the time the signal reaches the synapse. Paskari (talk) 18:02, 25 July 2008 (UTC)

Well, as the papers I referenced above indicate, there tends to either be sufficient densities of sodium channels at the synapse or just outside of the synapse to propagate the action potential. I don't know if anyone has been able to do recordings from pyramidal cell axons, it's a technically very difficult procedure.--Dpryan (talk) 19:13, 28 July 2008 (UTC)

Image of distribution of currents for propagating action potential in unmyelinated neuron

Anyone that tries to make such a drawing will fall in a dead end. That's normal if you try to integer an electrotonic sequence in a process that does not really need it.--Somasimple (talk) 05:03, 23 July 2008 (UTC)

you need to understand this diagram [2] --Somasimple (talk) 10:06, 24 July 2008 (UTC)

Dendritic contribution to AP

I'm notorious for being very picky, so I appologize in advanced. I am having a problem with this line from the article Whether the voltage is decreased or increased, the change propagates passively to nearby regions of the membrane...typically, the voltage stimulus decays exponentially with the distance from the synapse and with time from the binding of the neurotransmitter. Some fraction of an excitatory voltage may reach the axon hillock. What exactly reaches the axon hillock? Lets assume decay is negligeable. Does a 1 mV depolarisation of the spine depolarise neighbouring membrane potentials until it finally depolarises the axon hillock membrane potential. Or do the sodium ions diffuse downstream, depolarise the neighbouring membrane potentials and so on. Or are the two just different sides of the same coin? Thank You Paskari (talk) 17:47, 25 July 2008 (UTC)

It's the voltage change that propagates, not the sodium ions. --Dpryan (talk) 19:31, 28 July 2008 (UTC)

I'm just trying to picture how a depolarisation caused at the dendrites could propagate all the way to the axon hillock without having reached threshold. Once a region reaches threshold, I can see how it would depolarise the neighboruing membrane patch, which would trigger an AP and so on. But this is generally not the case with the dendrites. If threshold is not achieved, then the depolarisation would decay exponentially with distance. The only thing I can think of is that surface area of the neuron converges dramatically at the axon hillock. Is there a near constant ratio of number of spines/dendrites and the average distance to the axon hillock (much like how the nodes of ranvier are spaced apart about 1/3 the way before the signal begins to attenuate). Paskari (talk) 19:59, 30 July 2008 (UTC)

Regarding your first question, this article once contained an excellent explanation from User:Synaptidude about the propagation of passive potentials. He likened this passive propagation of voltage to the action of a Newton's cradle device. I rather liked that description myself, and it goes a long way towards conceptualizing how graded potentials (ie, the ones that occur between dendrite and axon hillock) work.

As for your other question, I can't speak to literature on ratios of spines/dendrites and distance to axon hillocks, etc., but I may be able to help your understanding of graded potentials and their journey to the axon hillock. As you know, graded (passive) potentials propagate according to cable theory and therefore their amplitude decays with distance and time. A postsynaptic potential at the dendrite will reach the axon hillock so long as its initial amplitude is sufficient to withstand complete decay as it propagates from dendrite to axon hillock. Regardless of initial amplitude, all graded potentials will experience some decay as they propagate, as determined by the familiar λ and τ constants. In fact, EPSPs decay so markedly with distance and time that it takes many hundreds (thousands?) of EPSPs converging on a single neuron nearly simultaneously in order to initiate an action potential at the axon hillock. Herein lies the computational power of the neuron. --David Iberri (talk) 20:33, 30 July 2008 (UTC)

Thank you, that makes a lot of sense. One thing that surprises me, is that the axon hillock is considered to be the "spike initation zone" because it has the highest concentration of voltage activated sodium ions. But, I have never read any mention of the fact that such a huge surface area rapidly converges onto the incredibly small surface area of the axon hillock. Why is that? Also what happened to User:Synaptidude's explanation? Paskari (talk) 22:10, 2 August 2008 (UTC)

Action potential branching

In the introduction it says The axons of neurons generally branch, and an action potential often travels along both forks from a branch point. This implies the axon always branches off into exactly two branches, is this actually the case? Paskari (talk) 19:24, 29 July 2008 (UTC)

grammatical error

The first line in the section "Ions and the forces driving their motion" reads Electrical signals within biological organisms are generally by ions. I know this is getting picky, but I wouldn't say a signal is by something. Electrical signals are by ions doesn't really roll off the tongue. Paskari (talk) 15:30, 30 July 2008 (UTC)

WP:SOFIXIT. ;-) --David Iberri (talk) 20:59, 30 July 2008 (UTC)

Changed it to Electrical signals within biological organisms are generally driven by ions. 78.86.221.4 (talk) 16:26, 1 August 2008 (UTC)

Cell Membrane

The following few lines could be better worded

These systems can be divided into two classes: pores ("channels") that allow passive transport of ions, and ion pumps...for active transport of ions. The ion pumps tend to work continuously...By contrast, the ion channels open and close in response to signals...

I think, the distinction of continuous and discontinous transport is void here. The difference is only that the channel activity can be observed on the level of individual molecules because of their high turnover (MHz range), and these observations tell that the channels are either open (active) ore closed (inactive) with unresolved fast transitions in between. The activity of the slower pumps (<1kHz) are only observed in ensembles; but there is no doubt, that pumps like channels and other enzymes change the activity of the individual molecule by discontinuous transitions between active and inactive, e.g. by binding/debinding of an inhibiting ligand. Only the observation of a population causes the impression of an intermediate activity, whereas the underlaying mechanism consists of a certain probability of molecules to be fully active or fully inactive. These circumstances do not need to be discussed here, and the pseudo-distinction between continuous and discontinuous action may readily be skipped here without loss of comprehensibility of the whole article. If, however, a classification of ion transporters in biomembranes is wanted, one should distinguish (i) uniporters, i.e. channels and carriers which simply catalyse the transition of ionic and non-ionic particles with higher or lower turnover, respectively, (ii) cotransporters, where the translocation of one substrate is coupled to the translocation of one or several different substrate(s) in the same (symporter) or the opposite (antiporter) direction, and (iii) pumps, where transport is coupled to a non-transport reaction of e.g. metabolic (ATPases) or photochemic (electron transport in photosynthesis) nature. Solfiz (talk) 20:12, 7 September 2008 (UTC)

Notice how the two classes initially presented are the pores/channels and the ion pumps. But when being described, the former is renamed "ion channels". Paskari (talk) 15:46, 30 July 2008 (UTC)

Also in this section, I am having trouble visualizing this

The two classes play complementary roles; the ion pumps generate the differences in ion concentrations across the membrane, which the ion channels exploit to carry out electrical signaling.

First it can be misleading because some readers might get the impression that ion channels don't affect the ion concentrations (NMDARs can drastically increase the Ca2+ concentration). Furthermore, how do ligand activated receptors fit into this? They have channels, yet they do not exploit any ion concentration to allow transport of ions. Sorry about being so critical, I just feel these articles should be as clear as possible =) Paskari (talk) 16:20, 30 July 2008 (UTC)

I've gone ahead and updated the cell membrane section. I don't like using words like 'active' or 'passive' transport. I find they complicate the sentence without really explaining anything. Also, page 28 of Neuroscience (Purves et. al. 2008) has a much better comparison of ion channels and ion pumps. I'm going to go ahead and include that Paskari (talk) 13:44, 21 August 2008 (UTC)

Electrochemical Gradient

Can someone break down this whole issue of ions moving into and out of the cell based on electrochemical gradients. With regards to diffusion I'm assuming that the sodium pottassium leak channels are a good way to visualize it. Basically pumps move ions from highly concentrated to low concentrated sides. But I don't quite grasp the concept of electric fields moving ions around. Are they also transported through the passive leak channels (pumps), or do they operate through the active ligand/voltage activated ion channels? From the below sentance

Ions cross the cell membrane under two influences: diffusion and electric fields.

I get the feeling that these are the only two ways (which I know is not the case b/c neurotransmitters activated influx of ions does not fall under either category). Perhaps the problem could be solved by changing the sentance to

Ions cross the ion pumps under two influences: diffusion and electric fields.

Thanks Paskari (talk) 19:43, 30 July 2008 (UTC)

From this and your question above I can see that you have a misunderstanding of how channels and pumps work. Pumps work by utilizing some energy source (be it ATP or a concentration gradient) to drive the movement of an ion or molecule either with or against its electrochemical gradient (usually against, but this is manipulable). If you pharmacologically block pumps, membranes will slowly depolarize. Regarding channels, all channels, regardless of their gating mechanism (i.e. NMDA receptors are included in this grouping), pass ions according to their electrochemical gradient (i.e., diffusion and electric fields). In this, it's implied that the channel is in an open state...otherwise there would be no movement to begin with. You might want to refresh your understanding of channels and pumps/transporters.

I'll also take this opportunity to address something you said above about how channels affect ion concentrations. In general, channel openings result in insignificantly small changes in both intra- and extracellular ion concentrations. As you noted, an exception would be intracellular Ca²⁺ concentrations, though these changes don't immensely affect the ionic driving force (depending on your definition of immense, of course). The best exception to the "channels don't affect ionic concentrations" rule would be K⁺ accumulation in T-tubules, though this is more of a special case scenario. Even here, though, there's <1mM change per action potential. Regardless, such changes are transient. --Dpryan (talk) 20:20, 31 July 2008 (UTC)

Yes I have been looking at the Ions and the forces driving their motion section all day (and all last night) and I'm trying to get my head around it. Here's how I see it so far:

The channels, when open, let ions in or out. Which way they travel, or how fast they travel, is based on the concentration of ions on either side, as well as the membrane potential. I think this is what is commonly referred to as electrochemical gradient. This has a marginal effect on either of the ion concentrations. It, however, has an appreciable effect on the membrane potential. The pumps, on the other hand, use the cell's energy to counteract the concentration imbalance (however minute it might be) caused by the channels. The role of the pumps is to maintain stable concentrations. These concentrations need not be at equilibrium.

I also think we should state what it means when we use concepts like 'against concentration gradient' or 'with concentration gradient'. It would help the reader if we meant from high to low concentrations, or vice versa. I make the same mistake when I use words like depolarize/hyperpolarise, and my audience struggles to keep up if they are not familiar with this material. Paskari (talk) 16:54, 1 August 2008 (UTC)

I hope I won't confuse things any further, but I wrote much of that section and would be happy to answer any questions. You seem to know the basics, but I'll just reiterate a few basic facts.

• Ions can cross the membrane through channels and pumps. They can also diffuse through the phospholipid bilayer, but that diffusion is very slow.

• Pumps require energy, and they usually transport ions against their concentration gradient. In other words, they carry ions from the low concentration side to the high concentration side. If you poison the pumps, i.e., stop them from working, the concentration gradients gradually die out, as ions diffuse from high to low concentrations. Transport by pumps is slow compared to the motion of ions across the membrane through channels, so that a millisecond action potential could not be engineered using only pumps.

• Ion channels don't require energy, since they're just holes in the membrane that allow specific ions to pass through. Their power is to open and close in response to environmental stimuli, such as a voltage applied across the membrane, or the binding of a neurotransmitter or neurotoxin.

• I mentioned the electric fields because I wanted the reader to understand how a voltage across the membrane could act on an ion to affect its motion. The voltage difference ΔV corresponds to an electric field E within the membrane; that field creates a force F on the ion by the defining equation F = qE, where q is the electric charge of the ion; that force affects the ion's motion by Newton's second law of motion, F = m a, where m and a are the ion's mass and acceleration, respectively. I didn't want to get into the daunting nitty-gritty of statistical mechanics and electrophoretic mobility, however, so I left out all the more quantitative stuff.

Hoping that this is helpful, and open to more questions, Willow (talk) 17:57, 1 August 2008 (UTC)

Thank you, that does help a lot. One question, when the channels open, the ions move with their electric fields, and with, not against their concentration gradients, correct (unless the electric field interferes I suppose)?

I understand why you choose not to include the detailed equations, but is there any way to present this information in a simplified manner? For example, one of your lines reads

• The voltage difference ΔV corresponds to an electric field E within the membrane

However, I have also seen this phrased as

• At rest the intracellular fluid is more negative than the extracellular fluid. To maintain this, the membrane would have to maintain a potential, this is called the membrane potential.

Your explanation is a little more understandable, but the problem I have with both is that they require me to accept something, not understand it. So I must accept that "the voltage difference corresponds to an electric field within the membrane" and I must accept that "the membrane would have to maintain a potential". I'm complaining about these things as I read them because once I understand them, I'll probably forget which parts gave me the most difficulty. Cheers Paskari (talk) 11:46, 2 August 2008 (UTC)

Thanks, Paskari, I'm happy I could be helpful. :)

To answer your first question, during a typical action potential, ions do usually move on average with the electric field and with their concentration gradient (usually called "down" their concentration gradient) — but not always. When the sodium channels open early in the action potential, the net motion of the sodium ions is inwards, into the neuron, towards the region of lower sodium concentration and towards the region of more negative voltage (for example, V = -45 mV). As soon as the voltage becomes positive, however, the electric field opposes further inward motion of sodium ions, while their concentration gradient still drives them inwards. Similarly, when the potassium channels open up more fully in the later part of the action potential, their net motion is outwards, out of the neuron, towards the region of lower potassium concentration and towards the region of more negative voltage; since V might be, say, +25 mV, the outside is more negative relative to the inside of the neuron. As soon as the voltage becomes negative again, however, the electric field opposes further outward motion of the potassium ions, while their concentration gradient still encourages it.

It's important to remember that the electric field and the concentration gradient are two independent forces; if they're opposed to one another, the ions will on average follow the force that's stronger. Equilibrium is reached when the two forces are exactly balanced. For example, at the resting potential for sodium, roughly the peak of the action potential, the net motion of sodium ions is zero because the concentration gradient is driving the sodium ions inwards as usual, but now the electric field is driving them outwards.

It's also important to note that we're talking about the average motion of the ions. Any single ion, if you could watch its motion, might cross the membrane in either direction, due to random diffusion at room temperature. But the concentration gradients and electric fields impose biases on the statistical behaviour of ions. These forces act like cheaters loading the dice at a gambling game; any single roll of the dice (or movement of an ion) might seem random, but the biases becomes apparent when you roll the dice many times (or watch a mole of ions).

The last two statements of your message, the ones about membrane voltage, leave me a little puzzled. My statement is a definition. A voltage difference is by definition a net electric field along a path. Imagine if you said to me, "A triangle corresponds to having three sides." and I said, "why is that?" You wouldn't be hiding your reasoning from me, or making me accept something; you're just telling me the definition. Similarly, I'm just giving the definition of a voltage difference, as a verbal transition from talking about voltages to talking about electric fields, for the purpose of making it easier to understand how the voltage across the membrane can affect the net motion of an ion.

The second statement seems to me just a repetition, since "more negative" means "negative voltage difference"; it's like saying, "4 is less than 6, therefore 6 is greater than 4." Repetition in different words is sometimes helpful for a reader to pick up on something, but it's not like it introduces a new thought.

I hope these remarks are just as helpful. I'll be reading a new book for most of the weekend, so please be patient if you'd like to Talk some more. Thanks! :) Willow (talk) 12:38, 2 August 2008 (UTC)

OK now that was extremely helpful. I think a good statement to make in an article describing the flow of ions is that the concentrations inside and outside can be considered to never change. This is often mentioned in passing, usually when describing the role of ion pumps. I, however, think that it would be more useful right before discussing ions moving based on electric fields. So basically if you had a membrane potential of zero, then the ions would move from the side with high concentration, to the side with low concentration. In this process, the charge of the ion would affect the membrane (correct?). This then changes the membrane potential, which begins to affect the flow of ions.

Here is an example you gave which I think should be used in the body of the article

When the sodium channels open early in the action potential, the net motion of the sodium ions is inwards, into the neuron, towards the region of lower sodium concentration and towards the region of more negative voltage (for example, V = -45 mV).

This gives a perfect example of motion based on concentration gradients and on electric fields. But for somebody who's illiterate in physics, even something as simple as region of more negative voltage has thrown me off. I know that an ionic solution, like the intracellular fluid, can have a voltage. Further that the extracellular voltage is more positive than the intracellular voltage. But there's also this issue of membrane potential in the middle. My question is, are, for example, the sodium ions attracted to the negative membrane potential, or the negative intercelular voltage? Is there even a difference between the two? Is membrane potential just an easy way of saying "the outsde voltage - the inside voltage"? Basically I don't understand what a membrane potential is. I know it can be measured as a voltage. I also know that it arises because of the voltage difference of the intercelular and extracelular voltage, but if I write "the outside voltage is postive, and the inside voltage is negative, therefore, there must be a membrane potential", and then at my viva someone will, rightly, ask me to explain why, and I won't be able to answer this.

I hope you don't think I'm here gathering answers for my thesis. I just believe that if I don't understand something on here, then a high school student, or and undergrad won't either.

I think it would be good if we could update the section on ion forces. Maybe put in a few examples. Also your explanation of the equilibrium potential with regards to the competing forces of electric fields and concentration gradients I thought were more clear than what is currently in the article. It really comes down to what I said in my previous post: some of the material in this article requires me accept something, as opposed to understand it. Thanks Again Paskari (talk) 21:57, 2 August 2008 (UTC)

Goldman Equation

There is one minor problem with the Goldman Equation, if you look closely, on the top the chloride permeability is P_cl- whereas on the bottom it's just P_cl. Paskari (talk) 21:42, 30 July 2008 (UTC)

Actually, the convention is to leave the ion charges off of P_Na, P_Cl, etc. So I've removed them. --David Iberri (talk) 20:29, 2 August 2008 (UTC)

animation

I'm working on some animations that might be useful for this article; comments/suggestions are welcome.

Action potential video. A larger version (720x480) is available. QuickTime or VLC players are better for video play than the Java-based player.

The animated gif (below, right) illustrates action potential propagation in an axon. Three types of ion channel are shown: potassium "leak" channels (blue), voltage-gate sodium channels (red) and voltage-gated potassium channels (green). The movement of positively-charged sodium and potassium ions through these ion channels controls the membrane potential of the axon. The negative-inside resting potential is mostly determined by potassium ions leaving the cell through leak channels. Action potentials are initiated in the axon initial segment after neurotransmitter activates excitatory receptors in the neuron's dendrites and cell body. This depolarizes the axon initial segment to the threshold voltage for opening of voltage-gated sodium channels. Sodium ions entering through the sodium channels shift the membrane potential to positive-inside, approaching the sodium equilibrium potential. The positive-inside voltage during the action potential in the initial segment causes the adjacent part of the axon membrane to reach threshold. When positive-inside membrane potentials are reached, voltage-gated potassium channels open and voltage-gate sodium channels close. Potassium ions leaving the axon through voltage-gated potassium channels return the membrane potential to negative-inside values near the potassium equilibrium potential. When the voltage-gated potassium channels gate shut, the membrane potential returns to the resting potential.

Animated GIF illustrating action potential propagation.

--JWSchmidt (talk) 16:45, 19 August 2008 (UTC)

First off, great job with both of those. I'm a great supporter of using visual mediums to relay information. One main problem that I have is that it is information overload. So lets take a look at the image: there are three channels with three different colors, sodium and pottasium have their own colors, and finally they are moving all over the place. I think what you're trying to show is one membrane patch stimulating a neirghbouring one, right? Well I can't begin to imagine how much time you spent on this (I suck at graphics), but it would be beneficial if you could break this down. So you might want to take 10 frames to demonstrate the signals from the dendrites propagating to the axon hillock. Then pause the simulation for a few seconds, write "spike initiation zone", then start the action potential. It's not a bad idea to have two channels side by side, but it's too much information to have them working at the same time. So lets call them channels A and B. First, at the hillock, A is activated, so show the sodium rushing in. Make sure to pause the simulation, then write "rising phase". Then when the channels close, pause and write "peake phase", then switch to channel B and do the same, pause, and write "rising phase", then switch back to channel A, show the potassium going out, and write "falling phase".

Visual information can only get so much across. It's too info overload if you should 20 ions bouncing around. Hope that helps

As for the movie you made, try not to use words like "transient", "synapse", "transmembrane", "down concentration gradient", "depolarize". Instead use words like "quick", "connection", "across membrane", "from high to low", "make more positive". Us scientists seem to think that the rest of the world thinks the same way we do. This is a really neat idea. Let me know if I can help further. Paskari (talk) 13:38, 21 August 2008 (UTC)

Thanks for the feedback. Exactly what I was looking for! --JWSchmidt (talk) 21:48, 22 August 2008 (UTC)

The animation is simply false for an squid unmyelinated axon (a reference). Thus, the space constant is too short to enable a polarization at such a distance. a squid axon AP is 40 mm wide where space constant is 5 mm.--Somasimple (talk) 06:55, 22 October 2008 (UTC)

Here is a movie that respects facts Action Potential Propagation--Somasimple (talk) 14:31, 22 October 2008 (UTC)

It may respect facts, but it conveys hardly any information. I'm skeptical that any sort of short movie is going to be very helpful here. The best hope would be to have a movie that shows only voltage-gated sodium channels, sodium ions, and positive charge inside the membrane. Opening of a sodium channel would be followed by inflow of sodium, leading to a spread of positive ions to neighboring channels, thereby triggering them to open, etc. Looie496 (talk) 16:48, 22 October 2008 (UTC)

It was just a short trial and it shows exactly what you are describing. THis is an old one Action potential Cutoff --Somasimple (talk) 17:07, 22 October 2008 (UTC)

To-Do list items

I see you put a To-Do list, excellent. Lets clarify it a little

• Distribution of currents for propagating action potential in unmyelinated neuron I have no idea what this means
• typical branching ratios of neurons; typical number of synapses This one might be a little tricky since there are so many different types of neurons.
• Plant Action Potentials You're on your own with this one, it's not something I'm familiar with.

I'm more concerned with the section Ions and the forces driving their motion it seems like it was written by some Physics guru. I'm in biology so I'm struggling to get my head around it. When we use words like "across membrane", this means next to nothing to some high school student reading it. And it's not just the people who wrote this article, it's in every textbook. Take From Neuron to Brain (Nicholls et. al. 2001), when describing ionic Basis of the Resting Potential it reads At rest, a neuron has a steady electric potential across its plasma membrane. Now if I was studying for a first year biology exam, right there the text book would have lost me. Neuroscience (Purves et. al. 2007) dumbs it down a little, neurons have a means of generating a constant voltage across their membranes, but still I think that it's not good enough. Since the membrane potential is so important in explaining how an action potential works, I think a good deal of emphasis should be put on exactly what it is. Paskari (talk) 10:49, 23 August 2008 (UTC)

Ion Forces edit

I'm sorry, but the explanation in the Ions and the forces driving their motion is just impossible to follow.

Ions also move in response to an electric field. By definition, the integral of the electric field across a patch of membrane equals the voltage Vm across that patch.[note 2] Likewise by definition, the flows of different ions through that patch are the ionic currents at that patch; the total current is the sum of all the individual ionic currents. Using these definitions of voltage and current, such a membrane patch can be modeled by an equivalent electronic circuit.[7] In particular, for each type of ion the patch will have a capacitance C and a conductance g; according to Ohm's law, the current I of each ion type is related to the transmembrane voltage Vm by the equation I = g Vm. For a given set of ionic conductances, there is an equilibrium voltage E at which the total current across the membrane is zero; the natural flow of ions generally causes the membrane voltage Vm to approach E.[8]

I'm going to go ahead and rewrite it following the structure in Neuroscience, Purves et. al. 2007 p 28 - 30. I would appreciate it if you would discuss it here as opposed to reverting it. Thank you. Paskari (talk) 10:58, 23 August 2008 (UTC) I still have to include a section on membrane potentials, since I can't seem to find an easy way of getting the concept of membrane potential across. Here's an example whereby I use the loose term point b/c I don't know what else to use.

• the point at which the electric field completely counteracts the force due to diffusion is called the equilibrium potential.Paskari (talk) 12:31, 23 August 2008 (UTC)

Membrane voltage vs. membrane potential

I completely support these changes, but I wanted to note that "membrane potential" is technically incorrect. In neuroscience, when we talk about "potentials", we're talking about membrane "potentials" (and equilibrium "potentials"). To measure a membrane "potential", two electrodes are placed, one on either side of the membrane. Now the tricky part: each electrode records a single potential. The difference between these two values is what we in neuroscience call the "membrane potential"; but technically this is a "membrane potential difference". In physics, there is a term for a difference in electrical potential: it's called a voltage. So technically, our reference to "membrane potential" is incorrect; "membrane voltage" is actually the correct term. That said, I'm not proposing we use voltage here. I prefer potential because it's so widely (mis)used. --David Iberri (talk) 13:20, 23 August 2008 (UTC)

I see what you mean. My intentions in purging the term membrane voltage was to get a little consistency. I figured since this article made no effort to include a Membrane Potential section, it was unfair to further confuse the reader by using voltage and potential interchangeably. I created a section for membrane potentials, could you please go over it and correct any mistakes I may have made. My background is in computer science and biology, so the term 'potential' means next to nothing to me. It took me a while, and several textbooks, to finally catch on that the membrane potential is just a measure of the difference between the outside voltage and the inside voltage. This is outrageous. One would think that the single most important concept underlying action potentials would be the one most clearly explained :-P Paskari (talk) 13:40, 23 August 2008 (UTC)

Why'd you go to all those textbooks when you've got Wikipedia at your disposal? ;-) Like I said, I totally agree that we should be using "membrane potential" and not "membrane voltage" here, even though the latter is technically correct. So I think you've done a good job of avoiding unnecessary reader confusion. I'm happy to take a look at your membrane potential section when I get a chance. Cheers, David Iberri (talk) 11:56, 24 August 2008 (UTC)

(Moved from #Proposal to Rename Article, above)
With respect to Solfiz's comments about membrane potential as opposed to membrane voltage, I'm starting to question the whole concept of having a membrane potential in the first place. Perhaps it helps with descriptions amongst the academic elite (much like presynaptic, EPSP, concentration gradient), but it's a terrible way to describe it in a Wikipedia article, or a text book. It doesn't do it justice to come up with this complicated term called the membrane potential, then never really describe it, and use it all throughout in the context of "depolarize the membrane potential". I think it would be advantageous to describe it based on what brings about the membrane potential, namely the highly localized concentrations of ions that just traveled through the channels. It's just that the term membrane potential is so abstract that you can't have any faith in it, it absolutely requires a leap of faith on the part first time readers. Paskari (talk) 22:38, 27 August 2008 (UTC)

Just to pick a nit, "depolarize the membrane potential" is incorrect use of the term "depolarize". Membranes are depolarized, not potentials, and not even voltages.

Also, what do you mean by "I question the concept of having a membrane potential in the first place"? Do you mean the article shouldn't discuss it? Or that the current discussion doesn't do the term justice (ie, doesn't help the reader)? I think we agree that this article needs a better description of membrane potentials/voltages, right? --David Iberri (talk) 01:08, 28 August 2008 (UTC)

The use of the term 'membrane potential' requires a leap of faith. Picture yourself giving a presentation to a bunch of high school students (or real estate agents, or accountants...) and you said "all an action potential is, is a sudden increase in the membrane potential from negative to positive and then back to negative again". If there is a section to properly describe the membrane potential that's fine, but I get the feeling that the people that wrote this article (myself included), don't sufficiently understand how the membrane potential works. I just read pages 25-60 of Neuroscience (Purves et. al. 2007), and it does a good job of pointing out that membrane potentials arise because of the selectivity of the membrane to ions, which leads to a buildup of positive or negative ions. But it doesn't explain how the 'signal' decays with time and distance. On p 54 it makes an attempt to warn the reader against going down the wrong road ("Note that this passive current flow does not require the movement of Na⁺ along the axon, but instead by shuttling of charge"), but it doesn't explain to me how the charge 'shuttles'. You get this idea that sodium ions come in and depolarize the membrane. And that is called the action potential, but then the action potential travels down the axon, so you get the sense that the sodium ions are racing down the axon. Is it the case that sodium ions rush in, attract negative chloride ions, and push away positive sodium/potassium ions which do the same thing all the way down? Are electrons travelling through the myeline? Are electrons traveling through the intercellular fluid? I think if we lay the foundations for the membrane potential, and signal propagation through the axon (not necessarily here, but in the membrane potential and axon articles), then explaining action potentials will be much easier Paskari (talk) 17:41, 31 August 2008 (UTC)

Membrane Potential

I created a section for membrane potentials. I was hoping someone else would do it, since it is the one topic I am a little unsure about, but it's been a few weeks, and I decided to do it myself. I tried my best, but if there are problems, please correct them. But try to keep that general structure, I don't want it to get lost in the fog of academia as someone tries to describe it using capacitance and integrations...Paskari (talk) 13:32, 23 August 2008 (UTC)

Thanks for the job. However, there is already a decent wikipedia site membrane potential. So this paragraph can be deleted here and membrane potential should rather be listed in the 'see also' section Nr. 12. Solfiz (talk) 16:21, 7 September 2008 (UTC)

After reading this chapter more carefully, I'd like to urge Paskari to withraw this text for the following problem: the present text only seduces innocent readers to the frequent and wrong idea, a surplus of, let's say K⁺ inside, would cause a positive internal voltage across a K⁺ selective membrane. Correct is, K⁺ tends to diffuse to the low K⁺ external side, leaving some negative charges behind, until outward diffusion and inward electrostatic attraction of K⁺ balance each other (s. Nernst equation). This role of electrodiffusion in the membrane potential is fairly well described by the membrane potential site.Solfiz (talk) 12:46, 10 September 2008 (UTC)

Saltatory conduction

I removed this from the Anatomy of a neuron section, but didn't have the heart to delete it, if anyone can find a suitable place to insert it, be my guest.

• Therefore, the action potential "hops" from one node of Ranvier to the next (the process of saltatory conduction); it does not move continuously down the axon, as it does in unmyelinated axons. —Preceding unsigned comment added by Paskari (talk • contribs) 19:34, 23 August 2008 (UTC)

Read this myelin capacitance and this one Saltatory conduction --Somasimple (talk) 07:04, 22 October 2008 (UTC)

Order of Action Potential

The article dexcribes action potentials in the following order

1. Action Potential Phases
2. Action Potential Initiation
3. Action Potential Propogation

I think a more appropriate order would be

1. Action Potential Initiation
2. Action Potential Phases
3. Action Potential Propogation —Preceding unsigned comment added by Paskari (talk • contribs) 20:36, 23 August 2008 (UTC)

So I went through and I reorganized the sections action potential initiation, action potential phases and action potential propagation into one large section Action Potential. But the Action Potential Propagation section has major flaws. Whoever wrote it was not aware that the myeline insulation makes it impossible for the action potential to be propagated (as it does at the axon hillock, nodes of ranvier and the axon terminals). Also, the Myelin and saltatory conduction section goes into a lot of detail about myeline (ie. fatty, lipid rich, oligodendrocytes, metabolic energy). Seeing as how this article is approaching 110 KB, maybe we should remove some of this. Paskari (talk) 15:56, 24 August 2008 (UTC)

I disagree with this reorganization. The phases of the action potential are entirely separate from the initiation, propagation, and termination of the action potential. The AP is classically separated into phases such as rising phase, falling phase, aftershoot, etc. They are used to discuss the molecular basis of the AP. How an AP is triggered, how it moves along an axon, and how it stops, are different phenomena altogether. Also, using the article name in the section headings (ie, calling the section "action potential") is redundant and generally discouraged. I'm in favor of the previous organization of the article, with sections entitled "phases" and "initiation, propagation, and termination". --David Iberri (talk) 16:16, 24 August 2008 (UTC)

Yes I too realized that calling a section "action potential" is a little wierd. Are you proposing that we split up the 'initiation', 'propogation' and 'termination' into different section? If so, that's not a big bother for me. But I strongly disagree with reorganizing it such that first the propogation is presented, then the initiation and finally the termination...it puts the cart before the horse. Paskari (talk) 17:48, 24 August 2008 (UTC)

I broke the Action Potential section apart such that the new sections now go:

• Action Potential Initiation
• Action Potential Phases
• Action Potential Propagation
• Action Potential Termination

However, in doing so I had to move some of the material into the introductory paragraph. Paskari (talk) 16:41, 31 August 2008 (UTC)

Signal across soma

The Action Potential Initiation section states that "...the voltage stimulus decays exponentially with the distance...Some fraction of an excitatory voltage may reach the axon hillock" but it doesn't explicitly state what happens across the soma. Does the soma contribute at all to the signal propagation? I know axon terminal buttons synapse onto the CA3 pyramidal cell bodies. Perhaps we should directly address this. Paskari (talk) 18:59, 31 August 2008 (UTC)

My understanding was that the soma generally lacks Hodgkin-Huxley-type voltage-gated sodium and potassium channels and is therefore unable to participate in the propagation of action potentials. As a result, EPSPs travel as graded potentials to reach the axon hillock where they are summed. However, I'm having a difficult time finding a source for that. --David Iberri (talk) 21:09, 31 August 2008 (UTC)

Yes that's the impression I had, but if that's the case, maybe we should reword the axon hillock from "having the highest concentration of voltage gated sodium channels" to "being the first region with, and highest concentration of, voltage gated sodium channels". Perhaps we could include this into the axon hillock as well. Paskari (talk) 08:50, 1 September 2008 (UTC)

Right. Have you come across a source for this idea though? I know I've encountered it in my neuro classes, but couldn't find a paper referring to it yesterday. --David Iberri (talk) 10:49, 1 September 2008 (UTC)

From Neuron to Brain (Nichols et. al. 2001) has about 10 pages (p 60-71) describing the crucial role of 4 ion pumps in maintaining stable ion concentrations, but it doesn't explicitly say they occur on the soma, just that they occur in all membranes. I'm taking this to mean there are ion pumps all over the soma. As for channels, here's what I could find, Neuroscience (p 26) states that "ion channels [are] protein molecules that span the membrane". Now that might mean that they span the whole membrane. I did a google scholar title search for "soma ion channel" and I only got one hit, which is a bad sign.

The intro in this article makes reference to sodium ion channels in the soma, but it's misleading, so it could just as well be referring to the axon hillock [3]. Also, this article claims that "Cell-attached recordings demonstrated that TC neuron dendrites contain a nonuniform distribution of sodium but a roughly uniform density of potassium channels across the somatodendritic area"[4] Paskari (talk) 12:12, 1 September 2008 (UTC)

There are certainly voltage-gated ion channels (Na+, K+, Ca++) located in the soma (cell body) of several different types of cells. For example, dissociated cerebellar Purkinje cells are a standard preparation for studying channel function. After dissociation, there is little or no axon/dendrite left, yet dissociated Purkinje cells will fire action potentials spontaneously. (See Raman and Bean, 1997). Florescent antibody staining, which can directly address where different types of channels are expressed, adds additional support.

In regards to initiation and propagation of action potentials, it is widely believed that Na+ channels are expressed in the highest density in the axon hillock, and that is the site of AP initiation. This is likely due to the lower capacitance. However, after the hillock spikes, a spike can be recorded soon after in the soma (and in some cases, the dendrites).

Finally, a comment about "spanning the membrane." This phrase is almost certainly intended to mean that there is a permeation pore from the outside to the inside of the cell, allowing ions to pass. In an earlier post, it seems as if Paskari has incorrectly interpreted it to mean the lateral "span" of channels, in other words, what areas of a neuron have channels.Dieselpower81 (talk) 08:14, 10 November 2008 (UTC)

References to Purves' Neuroscience

There are four separate references to this work, one of which refers to the 2nd edition (it's now in the 4th). These should be consolidated so that it appears as one entry in the reference list. Fuzzform (talk) 20:22, 1 October 2008 (UTC)

Also, this talk page is way too long. A new archive page is needed. Fuzzform (talk) 20:23, 1 October 2008 (UTC)