- History of withdrawal
- Dependence and drugs
- The benzodiazepines
- Legacy effects: Tardive dyskinesia
- Treating and stopping
- Protracted withdrawal syndrome
- Drugs and peripheral neuropathies
- Peripheral neuropathies & C-fibers
- Research goals
History of withdrawal
By the mid-1960s, the existence of withdrawal syndromes after antipsychotics were stopped was well documented.1 Within two years of the release of the first antidepressant, imipramine, a withdrawal syndrome had been described.2, 3
Despite this high profile, the recognition that antipsychotics and antidepressants could cause withdrawal then vanished from professional awareness.
There are two good reasons for this disappearance.
The first was social. This all happened in the 1960s, especially 1968, when the powers-that-be were very scared by the democratic revolutions happened all over the Western World – it was a Western Spring. Drugs like LSD seemed part of this revolution. The image above is of a Tokyo University occupied by students, who have sacked the Department of Psychiatry.
The social response was not one of trying to understand what was going on but to ban it. The chosen weapon was to say bad drugs that disrupt the social order cause addiction and madness – but good drugs that put you back in your place in the social order don’t come with problems like this.
On this basis LSD was banned even though it doesn’t cause dependence or addiction and it was assumed the antidepressants, benzodiazepines, and antipsychotic drugs would not be linked to dependence or addiction in any way.
The second reason for the disappearance is biological. We don’t understand the biology of dependence or addiction.
Dependence and drugs
In the 1950s, no one was certain if dependence and addiction were two different things. It was only in the 1950s when the work of Wikler and Isbell in Kentucky demonstrated that the problems after stopping alcohol actually did stem from alcohol withdrawal and not from the effects of other toxic processes – and ditto for barbiturate withdrawal – that firm distinctions were drawn between physical dependence and addiction (see ref 5).
Physical dependence referred to changes induced in the body by a drug that leads to problems on withdrawal from the drug. This was to be contrasted with a personality based drug seeking and criminal behaviors linked to addiction. A dependent individual would not necessarily be a junkie in other words.
No sooner was this made clearer than another concept – abuse liability – emerged in the mid-1960s to add to the confusion. The abuse liability of a drug is its ability to induce pleasure, craving, and a tolerance that led to escalating doses. Abuse liability gives rise it is said to drug dependence – drug dependence is distinct from physical dependence.6, 7
Drug dependence does lead to addiction, making even a previously normal person a junkie, whereas physical dependence doesn’t make you a junkie. Clear?
Enter the benzodiazepines. In the 1970s and early 1980s, these drugs caused the first major split between the public on one side and the therapeutic establishment – medicine and the pharmaceutical industry – on the other.
The benzodiazepines produce a clear physical dependence that develops on low-doses, in individuals taking the agents for therapeutic purposes, who do not become junkies, and who sometimes function better on the drug than off it.
Doctors and companies were reluctant to recognize that there was or could be a serious dependence problem with the therapeutic use of a drug. But the “victims” in this case received public support and sympathy in a way that traditional “addicts” never do.8
The victims were you and me. And the establishment was making an ass of itself. In 1988 the British regulator, the MHRA, claimed it only had reports of 28 people in Britain addicted as it put it to benzodiazepines.
The American Psychiatric Association didn’t do much better:
“Historically, long-term, high dose, physiological dependence was called addiction, a term that applies to recreational use. In recent years, however, it has been apparent that physiological adaptation develops and discontinuance syndromes can appear after regular therapeutic dose administration…. in some cases after a few days or weeks of administration. Since therapeutic prescribing is clearly not recreational use, the term dependence is preferred to addiction, and the abstinence syndrome is called a discontinuance syndrome” (APA 1990).
This view suggests physical dependence on a legitimate treatment is acceptable whereas becoming a junkie is not.
This is not how we the public see it. Most of us have no fears we might become junkies. We want to be able to stop a drug when we want to stop. Many people now talk about being hooked to antidepressants just as they did to benzodiazepines. None of us distinguish between being hooked because a drug causes pleasure and being hooked because difficulties on withdrawal make it difficult to stop.
Part of the reason we don’t buy the professional line is because withdrawal can be so severe and can endure for so long. If the discontinuation syndrome were quick and tolerable – as with rebound to beta-blockers or anticholinergics – there would be much less argument, but when ex-heroin users can say it’s harder to get off SSRIs than off heroin, the professionals have lost the argument.
The problem that was the benzodiazepines deepened with the SSRIs, which were marketed as non-dependence producing antidepressants. One group of people were saying SSRIs don’t do to you what the benzodiazepines do, while another group were still refusing to accept the benzodiazepines caused any problems.
The truth is that after only a few weeks exposure some healthy volunteers had significant difficulties on stopping SSRIs.
The truth is the problems stopping antidepressants look very like the problems stopping benzodiazepines. But what causes symptoms on stopping?
Drug companies spent a decade between 1995 and 2005 frantically denying antidepressants cause withdrawal – discontinuation symptoms is what they insisted on calling them. They completely refused to accept antidepressants cause addiction – but refusing to accept that whatever we call them, the problems are almost identical to the benzodiazepine problems.
Companies and many experts refuse to accept that withdrawal can go for as long as people who have difficulties getting off SSRIs claim.
So what is going on?
Legacy effects: Tardive dyskinesia
Withdrawal from antipsychotics can lead to tardive dyskinesia. No one argues with this. Tardive dyskinesia can last for years.
Tardive dyskinesia is rarely thought of in terms of drug dependence even though it appears on withdrawal and when it shows on treatment it can be managed by raising the dose of treatment just as you would to overcome tolerance for morphine.
But rather than call it withdrawal, tardive dyskinesia is usually seen as a legacy effect of treatment.
The legacy effects of treatment arise in the course of treatment and endure after a treatment has stopped. This means they can be confused with withdrawal making it difficult to be certain exactly what’s legacy and what’s withdrawal and how long withdrawal endures.
In the case of the SSRIs, Post-SSRI Sexual Dysfunction (PSSD) is just like tardive dyskinesia. It often starts in the course of treatment, or it may only appear on withdrawal. It can endure indefinitely after treatment stops, and at present there is no known cure. The symptoms can involve genital numbing, which suggests an element of neuropathy.9
SSRIs can also induce visual problems such as nyctalopia and a bunch of other strange problems, like causalgia, allodynia and all kinds of other things that the average doctor may have dim recollections of having heard about somewhere.
Poorly understood symptoms like these make it difficult once individuals are on psychotropic drugs for some time to know where the treatment ends and the disease begins. These complaints are often dismissed as “neurotic” and certainly dismissed as incompatible with what we know of withdrawal.
Treating and stopping
Treating and stopping a drug is never the same as not treating. If a person has been on a statin or an antidepressant, simply stopping does not restore the person to the way they were before. There may be residual problems such as muscle pain. There may be benefits such as feeling more alive which can cause problems in its own right if the treatment has dulled sensitivities for some years.
Despite this caveat, many people taking drugs do not have a problem on withdrawal. This is true for benzodiazepines and antidepressants as well as PPIs for gut problems and antihypertensives. Some people have no apparent difficulties, even after being on treatment for some years. Some people seem to be able to stop even without tapering.
A second group of people have short term difficulties. These acute withdrawal syndromes can be managed to some extent by tapering and sometimes by the use of supportive medication. Most classic withdrawal syndromes for opiate and opioid dependence, and alcohol dependence fall under this heading. They are usually understood to last from 1-4 weeks depending on dose and duration of exposure.
A proportion of those taking benzodiazepines, antidepressants, antipsychotics, dopamine agonists and other drugs will have problems like this, although the difficulties may last from 1-6 months rather than 1-4 weeks.
The complicating factor is the existence of protracted withdrawal symptoms. The validity of protracted problems have been denied by most of the establishment, despite at the same accepting the existence of tardive dyskinesia. Part of the difficulty in accepting the existence of these issues lies in explaining problems that may only appear some time after the drug has been stopped. Another difficulty lies in not being able to do much to alleviate the problem and a worry that if too many people hear about this kind of problem they may be reluctant to start treatment in the first instance.
Protracted withdrawal syndrome
Antidepressant prescribing took off 25 years ago. From early on, there was a slowly increasing number of people each year unable to stop. There is now one antidepressant prescription per year for everyone in the Western World. Roughly 10% of the population is taking them and 9 out of 10 of these are taking them because they can’t stop.
They try to stop and have difficulties, and are told by their doctors that this is the original problem that has come back – and they believe their doctor. Or they try harder to stop and can’t. Some get off treatment but have enduring problems.
This has led to talk of protracted withdrawal syndrome, with doctors denying that the complaints people have fit any known model of dependence.
The two sides are talking past each other. This could be solved by agreeing that withdrawal is something that last weeks or maybe months but not years. If there is something lasting months or years, it must be something else.
Before abandoning a simple withdrawal model, the first point to make is that withdrawal syndromes from antidepressants are real and are often more severe than many concede.
The initial healthy volunteer studies on paroxetine in the mid-1980s found that up to 66% of subjects had withdrawal problems after two weeks of exposure to treatment. This happened in both young and old, in male and in female subjects.
The withdrawal syndrome included anxiety, depression, malaise, dizziness, agitation, insomnia, nightmares, essentially most of the features of major depressive disorder even though these individuals had not begun with any affective disorder.
It is now clear that all SSRIs cause significant dependence and withdrawal problems with many SSRI users who have also used opiates, cocaine or other drugs of abuse saying that it can be more difficult to get off an SSRI than these other drugs.
The company line remains that any problems are mild and transient. But the studies that companies cite have looked at symptoms which emerge shortly after the discontinuation of treatment. They have not taken legacy effects into account, or attempted to recruit from the hundreds of thousands of subjects who are unable to stop treatment.
In most Western countries, there is now one antidepressant prescription per year for every member of the population. Of these upwards of 90% are for people on treatment chronically.10 This should not happen if there is no problem stopping. There have in fact been more reports to regulators about dependence on SSRIs than there have been for any other psychotropic drugs (see table).
|Benzo – Z Hypnotics
|Gastric – H2 Blockers
|Gastric – PPIs
|HRT & Contraception
|SSRI & SNRIs
There is a failure at social, professional and scientific levels to acknowledge difficulties or more importantly grapple with the nature of the difficulties. Given that the SSRIs are linked to a doubling of the rate of congenital malformations and of miscarriages, and linked to alcoholism, as well as suicide and homicide the issue of just what antidepressants do is an important public health issue.
Drugs and peripheral neuropathies
This section focuses on the Serotonin Reuptake Inhibiting (SSRI) antidepressants but something similar likely plays a role in protracted problems following the use of benzodiazepines, antipsychotics, stimulants, dopamine agonists as well as biphosphonates for osteoporosis and other drugs.
The basic proposal is that SSRIs have effects on peripheral sensory systems, especially on C-fibers. As with tardive dyskinesia, these effects can appear within a week of starting treatment but are more likely to arise later in the course of treatment, and will endure after treatment stops.
The fact that SSRIs do cause peripheral neuropathies and maybe other problems helps explain certain things.
- It would fit the standard understanding that withdrawal is time-limited. We would expect many to have a straightforward withdrawal process, lasting a number of weeks.
- Withdrawal will often be more severe than is conceded. There is substantial evidence that withdrawal doubles the rates of suicidal and aggressive acts. These effects may be so severe than in some instances the process of withdrawing becomes protracted and may be impossible.
- Withdrawal will have certain standard features such as electric zaps and dizziness which will clear after weeks or months when other often strange problems become more prominent.
- We would expect a number of clinically disturbing features to emerge on treatment or its discontinuation that persist relatively indefinitely after treatment stops. The evidence from PSSD makes it clear neuropathic disturbances arise on treatment or in withdrawal.
- Enduring post-withdrawal problems need to be distinguished from protracted withdrawal process because they are two phenomena with different causes and different cures.
- Peripheral sensory neuropathies will create instability of affective tone and a variety of dysphoric conditions that convince clinicians this is the original disorder in need of treatment leaving them deaf to the hunch of the person suffering.
- The need for treatment of a peripheral neuropathy will create a situation where there is no clinical option from increasing drug dose to decreasing dosages that brings relief.
- Adding the effects of withdrawal to the effects of a peripheral neuropathy will make the discomfort of each worse.
- Continuing with a treatment that causes withdrawal is not a huge problem. Continuing with a treatment that has provoked a peripheral neuropathy seems like a poor option.
- All the signs are that this neuropathy has reversible elements to it – hence the need for research.
- In the face of the kinds of difficulties people have on SSRIs, the history of tardive dyskinesia shows that many clinicians are likely to prefer a model that denies the possibility of any such effects and attributes any difficulties to the disorder being treated. In the case of the antidepressants, this will lead to recommendations for treatment options such as the addition of mood-stabilizers to an antidepressant regimen.
The classic symptoms linked to protracted withdrawal are:
- Heart palpitations and other heart problems
- Increased anxiety
- Pains – strange pains around the body
- Muscle tightness and stiffness and twitching
- Food and drug sensitivities or intolerance
- Feeling very flat
- Taste or smell changes
- Sensitivity to sound and light
- Heat sensitivities
- Visual disturbances
- Auditory disturbances
- Memory or concentration problems – Brain Fog – Chemo Brain
These problems overlap with but are different from the electric zaps, dizziness, nausea, vomiting, and nightmares that are prominent features of acute withdrawal but which do not appear to endure for more than several weeks or a few months at most.
Many of these protracted withdrawal features map well onto complaints which when brought to neurological clinics lead to a diagnosis of peripheral sensory neuropathy.
Case A — paroxetine induced penile anesthesia
Waldinger et al (2014) have recently reported a case that might point a way ahead.11 Their patient was referred for paroxetine induced penile anesthesia.
The gentlemen lost his sense of smell and taste while taking an SSRI in addition to having diminished skin sensitivity over large parts of his body including the genital area. Smell and taste returned while on treatment and after he stopped paroxetine his skin sensitivity returned to normal but his genital numbness persisted. He could apply Tiger Balm to his genital area without any sensation.
This led to treatment with LPLI (low-power laser irradiation) to the penile area, which produced a degree of restoration to the ability to distinguish between warm and cold stimulation and about a 20-40% recovery in touch sensation.
Waldinger and colleagues speculate that the action of LPLI treatment is mediated through Transient Receptor Potential (TRP) channels. TRP channels are a series of channels that mediate calcium and sodium ion transport across membranes. They are present on all types of nerves but are particularly prominent in the slow conduction C-fibers that mediate tactile sensitivity, pain, itch, aspects of smell and taste.
Case B — parathyroid gland removal
One of us has recently been consulted on the case of a 60 year old woman who had 20 years successful treatment with dosulepin until following a parathyroidectomy the benefit of treatment failed. Having previously been unable to stop treatment, she was now able to stop successfully.
A year later her mood slipped. She was given a series of serotonin reuptake inhibiting drugs but found herself intolerant to all of them, with disturbances of vision, smell, touch, heart rhythms and an aggravation in terms of intensity and extent of a previously mild degree of numbness and muscle stiffness in her feet. She developed food and drug intolerances she did not have before.
Her current situation is that most antidepressants seem to cause problems in ever lower doses while stopping them also causes problem. There appears to be an instability of some system.
In the face of these developments, she has been advised that her clinical picture suggests either a refractory unipolar disorder or a bipolar disorder and she has been recommended to start lithium or another mood stabilizer. In the absence of any alternate diagnostic option, advice like this is likely to be common but misguided.
Case C — hearing, vertigo, balance
My hearing is a lot worse, I have vertigo and balance problems. I feel unsteady on my feet. My ears are ringing. They are also supersensitive to sounds. As a result, I can’t wear a hearing aid in one ear any more.
I feel like I am only 50% here — kind of like a bad head cold feeling, or living in a dream state. I feel shaky and out of sorts and panicky. I feel weird and feel like I am going to pass out. I can be fine one minute, then BAM — all of a sudden I feel this odd feeling coming on as if my hearing gets very quiet. I feel as if I am chilled. I get a tingly feeling in my head, and then I feel a sort of darkness and closed-in feeling about to happen. I start to shake and sweat, and I just feel as if I am drifting away.
This is indistinguishable from what people on SSRIs say but it in fact comes from someone who had been taking a benzodiazepine.
Case D — anxiety, anger, fog, tinnitus
I’m now two years off Prozac and I can’t believe how much Prozac’s affecting me still. I’m so used to having numb emotions and poor cognitive abilities now that I sometimes forget that this isn’t the real me. Now and then I have days when underneath my “drugged state” I can feel strong anxiety and anger “bubbling” inside of me because I know that my emotions and my view on life isn’t my own.
I have a constant annoying fog in my head, I have constant tinnitus and I have head pressure and headaches still that come and go.
I meet no friends any longer, I have no interest in boys and little interest in sex. All this is very unlike me. Even if I had bad days as well before Prozac happened, I then snapped out of my low mood pretty quickly if I met a friend or took a shower or something.
I can function normally in conversations, but I know that I’m not “in” the discussion. My mom asks me often how I can seem so content with the very uninteresting life that I have. I tell her that I know that in reality I’m not content with my life but that my emotions aren’t “real”. I also tell her I’m sort of “stuck in the moment” and unable to think of the future.
I’m starting to worry that I’ll have to live with this foggy, empty head for the rest of my life. This total lack of motivation and interest in anything (but food really, I still crave carbs in a insane way, and I’m still forty pounds heavier than pre-Prozac), is making life feel really pointless. It also feels strange being 26 and having very little desire and motivation for getting a career and a partner. (Amongst other things).
Before Prozac I sometimes felt low sometimes felt happy, but I always got energy boosts now and then during the day when I did what I needed to get done. Nowadays I never feel really happy and never really sad but just content, and I’m always tired and I can’t think or feel clearly if my life depended on it.
WHEN will I start to care about things again? Is there a way to force forward motivation? Will I be completely recovered and free from my cognitive issues again?
Peripheral neuropathies & C-fibers
Forty years ago when standard medical textbooks covered the peripheral sensory neuropathies they outlined a picture where little was known. People presented, they stated, with problems for which the patient often could not find words.
The complaints often seemed bizarre with pain for instance triggered by heat or touch (tactile allodynia), paraesthesia, or complaints of tight band like sensations that did not map onto standard dermatomes.
People complained about hot or numb sensations in feet or hands that were present in stocking or glove distributions – this does not fit classic nerve distributions, but we now know it maps onto C-fibers.
These pain syndromes responded poorly to standard treatments for pain, and the other sensory problems responded to nothing.
Nothing was known about the basis of these complaints but it was expected that there might be a linkage to disturbances of C-fibers rather than the larger fibers mediating discriminative touch.
The burning sensations some people have, fall under the heading of causalgia. The speculation was that there might be some linkage to the sympathetic system – a reflex sympathetic dystrophy, or dysautonomia.
There are links to drug treatment. Causalgia has long been linked to alcohol use, even though it is reported more often in women than men, and forty years ago women did not drink alcohol as much as men – so it really should have been men who had it.
Triptan use for migraine is now linked to tactile allodynia.
Persistent genital arousal disorder (PGAD), which causes vulvodynia, has been linked to SSRIs, while PSSD and similar conditions have been linked to various drugs including SSRIs, isotretinoin, finasteride and tetracyclic antibiotics.
C-fibers are now at the center of a huge research effort. There is a great deal of research on new receptors – Transient Receptor Potential (TRP) channels, with distinctions now made between 8 superfamilies of TRP channels and efforts to map the roles of each.
There are new distinctions between C-fibers mediating touch, pain, itch and fibers mediating affective tactile sensitivity – a caress system.12
There is a growing body of evidence that maternal touch operating through the caress system in early life is an important determinant of the development of the social brain in the first instance and our later abilities to regulate mood states.
The implications could not be greater. At stake is no less a question than who or what we are.
A century ago the James-Lange Theory of the Emotions shockingly to many argued that our emotions and much of our being might lie primarily in our body rather than in our minds. What we really think happens in our bodies and is secondarily interpreted centrally – the mind guesses as to what is going on. The heart has its reasons of which the mind knows nothing.
The emerging understanding of the C-fiber system fits this model; in this case, nurturing touch, caresses, set our mood and if the system through which this happens is disturbed, then our mood state is likely to be also.
In just the same way, while there is a mental input to libido and sexuality also, peripheral genital sensations and our bodies – sense of smell and other sensory experiences – are likely to do something to create libido so that when genital numbness happens as in PSSD, or disturbances of smell and touch, libido necessarily falls off as a consequence.
There is little information on C-fiber input to memory processes but there are a great number of patients taking drugs from statins to fluoroquinolones to antidepressants who all complain about “Chemo Brain” or “Brain Fog” who have normal results on cognitive function tests. Complaints like this, against a background of normal cognitive tests are consistent with a disturbance in C-fiber function.
On the question of food and drug intolerances it is worth noting that there are smell and taste receptors in the gut, where 90% of the body’s serotonin lies, and a substantial proportion of the nerves in the gut wall are C-fibers.
More generally C-fibers and their associated TRP channels both in skin and in internal organs have a role in the recognition of irritants, inflammatory products and xenobiotics. While little is understood about how they might mediate the effects of drugs, this makes them, until proven otherwise, the system in place to mediate whatever physiological disruptions might stem from taking foreign agents chronically.
TRP channels play a significant role in cardiac functioning. Through mechanisms not clearly understood, the antidepressants and antipsychotics have significantly more effects on aspects of cardiac functioning expressed in QT intervals and rhythm abnormalities than almost any other groups of drugs.
We know the fact that known channelopathies in the heart – disturbances of sodium, potassium or calcium channels – mediate shorter or longer QT intervals, increasing or decreasing the risk of arrhythmia from treatment induced QT changes.
Similar channelopathies elsewhere in the body might explain why some of us are susceptible to the development of treatment related peripheral sensory neuropathies.
The pharmacology of channelopathies, and TRP channels, remains rudimentary, there are likely many treatments out there that have potentially helpful effects. Among these are the calcium channel blocking drugs. There are scattered reports of benefits from calcium channel blocking drugs on “withdrawal” syndromes linked to antipsychotics.
Other options lie in drugs with effects on sodium or potassium currents.
What is needed is a map of the calcium, potassium and sodium blocking actions that SSRI and related antidepressants in fact have.
This will require a lot of research because there are fast and slow sodium currents and the drugs in the calcium channel blocking group block many different types of calcium channels, so one of them might help some the features we seek to influence but others not. One channel blocking drug seems unlikely to help all problems.
Note: Details of serious side effects are increasingly being removed from Wikipedia’s pages. We therefore recommend against using Wikipedia as a reliable source of information on medications.
- To find out everything we can about peripheral neuropathies and get the material in understandable language so that everyone can participate in solving the problems.
- To find everything possible about C-fibers.
- To find out everything possible about sodium, potassium, calcium and TRP ion channels.
- To keep in focus the fact that a range of people from hairdressers and complementary therapists to laboratory scientists may have important contributions to offer. Remember – the Ark was built by amateurs, the Titanic was built by experts.
The elements of the model are almost all poorly understood, so we are actively calling on anyone with an interest in these issues to help us move some of these areas forward.
- Healy D. The Creation of Psychopharmacology. Harvard University Press (2002). ↩︎
- Mann AM, MacPherson AS. Clinical experience with imipramine (G22355) in the treatment of depression. Can Psychiatric Assoc J 1959;4:38–47 ↩︎
- Kramer JC, Kline DF, Fink M. Withdrawal symptoms following discontinuation of imipramine therapy. Am J Psychiatry 1961;118:549–50. ↩︎
- Battegay R. Drug dependence as a criterion for differentiation of psychotropic drugs. Compr Psychiat 1966;7:501–9. ↩︎
- Hollister L. From hypertension to psychopharmacology: a serendipitous career. In: Healy D, editor. The psychopharmacologists, vol. 2. London: Chapman and Hall; 1998. p. 215–35. ↩︎
- World Health Organisation. Expert committee on dependence producing drugs, 14th report, WHO Techn Rep Cir 312; 1965. ↩︎
- Nutt DJ. Addiction, brain mechanisms and their treatment implications. Lancet 1996;346:31–6. ↩︎
- Bury M, Gabe J. Tranquillisers and health care in crisis. Soc Sci Med 1991;32:449–54. ↩︎
- Hogan C, Le Noury J, Healy D, Mangin D. One hundred and twenty cases of enduring sexual dysfunction following treatment. Int J Risk Saf Med. 2014;26(2)109–16. ↩︎
- Healy D, Aldred G. Antidepressant drug use & the risk of suicide. International Review of Psychiatry, June 2005; 17(3): 163–172 ↩︎
- Waldinger MD, van Coevorden RS, Schweitzer DH, Georgiadis J. Penile anesthesia in Post SSRI Sexual Dysfunction (PSSD) responds to low-power laser irradiation: a case study and hypothesis about the role of transient receptor potential (TRP) ion channels. Eur J Pharmacol. 2015 Apr 15;753:263-8. ↩︎
- McGlone F, Wessberg J, Olausson H. Discriminative and affective touch: sensing and feeling. Neuron. 2014 May 21;82(4):737-55. ↩︎