We Should Listen to Our Emotional Pain: An interview with Paul Andrews

Dr. Paul Andrews is an Associate Professor of Evolutionary Psychology in the Department of Psychology, Neuroscience and Behaviour at McMaster University. His research focuses on understanding the evolution of depression, which he argues may be an evolved emotional response for understanding and solving complex problems. Dr. Andrews is also concerned with the evolution of the serotonin system and the effects of antidepressants on mental and physical health. His research has shown an increased risk of cardiovascular events, as well as death among those who use antidepressants.

The transcript below has been edited for length and clarity. Listen to the audio of the interview here.

 

Brooke Siem:  I want to start with a quote from a writer named Maggie Nelson. She wrote a beautiful book called Bluets, and a line stood out to me, which feels very relevant to your work.

The quote is: “It often happens that we treat pain as if it were the only real thing, or at least the most real thing. When it comes around, everything before it, around it, and perhaps in front of it tends to seem fleeting and delusional.”

As an evolutionary psychologist, keeping that quote in mind, why do we feel pain?

Paul Andrews: Wow, that’s a powerful quote. I love it. We feel pain because, as part of a single evolutionary tree of life, organisms—from flies to mice to squid to humans—all face threatening challenges in their lives, things that can kill them or interfere with their reproduction. We need to be able to avoid them and the harm associated with them. Painful feelings seem to have a common property, which is to promote what we call avoidance behavior—essentially avoiding harmful events in our lives. Additionally, they promote avoidance learning, which teaches us how to behave in order to steer clear of such threats.

For example, take a simple case: when you have a sprained ankle, which most of us have experienced at some point, it’s very painful. The pain helps you keep weight off your ankle so you don’t injure yourself further, giving it a chance to heal. The pain prevents further harm and damage to the injured area. If you take pain medication to reduce the discomfort, you’re more likely to put weight on the ankle, which can delay the healing process and may even cause more injury. I can expand on this, but I wanted to start with a physical pain example because it’s something we all intuitively understand.

 

Siem:  I tore my ACL and had surgery a few months ago, so I’ve spent a lot of time reflecting on the difference between healing physically and healing emotionally. Physical pain is straightforward for people to understand, but emotional pain seems more complex—it’s not linear. What is the evolutionary purpose of emotional pain, and is it, in a modern world where most of us aren’t facing life-threatening dangers, an outdated adaptation?

Andrews: Great question. Let’s see if we can extend this concept of avoidance behavior and avoidance learning to emotional pain. I like to start with milder forms of psychological pain before diving into depression or anxiety. For example, think about the malaise you feel when you have an infection. It’s unpleasant and distressing, but it serves a purpose. That feeling is worse when you’re physically active, so it forces you to rest and allows your immune system to allocate energy for healing. It’s a form of avoidance—avoiding physical activity to prioritize recovery.

Another example is hunger. Most of us don’t know what starvation is really like, but we know what it’s like to be hungry. When you’re starving, you have this massive amount of hunger that is almost intolerable. It is designed to help you, to motivate you to go out and find food so you can solve the starvation problem. But in an avoidance language, it is avoiding the death that can be caused by starvation. That’s what’s the painful feeling is motivating.

There’s a whole range of painful emotions that can be framed in this avoidance context. Take jealousy: people often think of jealousy as pathological, but it’s a normal response to perceived threats to a valued relationship. Reactive jealousy is a normal thing — it helps you — it’s a sign that you actually value your partner. If your partner’s actions indicate a potential emotional or sexual threat, jealousy motivates you to pay attention and address the situation to protect the relationship.

We can apply the same thinking to social anxiety, which is when you’re afraid of a negative consequence from engaging in social interaction, like embarrassment. If you fear embarrassment, that anxiety might keep you from social situations where the risk is high. Of course, that comes with a cost—missing out on social opportunities and becoming isolated, but you avoid the cost of being embarrassed. It’s easy to think about social anxiety in terms of avoidance behavior and avoidance language.

When it comes to depression, I’d argue that, like other painful emotions, we should consider it through the lens of avoidance behavior and learning. It’s a relatively unexplored area in science because many researchers don’t frame depression this way. Most researchers don’t even think about it in those terms, and so in my mind, it’s a frontier area of science.

 

Siem: Can you give an example? I love that your work takes these broad concepts that are often ignored at the individual level and makes them applicable. How can an individual use this knowledge to get out of a depressive episode or anxiety? What might avoidance behavior look like in depression?

Andrews: I’ll share an example from the literature. It focuses on physicians and psychiatrists and their emotional responses to making medical errors. Let’s take physicians—surgeons, for example. When they make a significant error that harms a patient, they often experience intense emotional distress, which can sometimes meet clinical criteria for depression, acute traumatic stress, or generalized anxiety. The same goes for psychiatrists who lose a patient to suicide. Hopefully, when they make a mistake, they learn from their mistake, right?

The literature shows when most physicians make a serious medical mistake, they feel a lot of emotional agony over having hurt a patient. We also know that the emotional agony can reach clinical levels. It’s not uncommon for physicians who have made these kinds of mistakes to meet clinical criteria for depression, for acute traumatic stress, for generalized anxiety and so on. The whole purpose of avoidance behavior and avoidance learning is to figure out what you could have done differently to change the outcome.

But if the physician concludes that it was the stupid nurse’s fault, or the stupid lab technician rather than reflecting on their own role, they won’t learn and they won’t reduce the chance of it happening again in the future.

Rumination over mistakes is inherently self-blaming because the goal is to figure out what you could have done to prevent the outcome. It has to be self-blaming otherwise it’s not going to be effective in producing avoidance behavior. One of the puzzling things for a clinician is, why, if rumination is supposed to help you solve problems, why does it have to be so self-blaming and self-deprecating? The answer to that is it’s hidden in the fact that it’s designed to promote avoidance behavior and avoidance learning which is all about what you could or can do. You have to go searching back through that causal chain of events for things you could have done differently.

 

Siem:  How do we differentiate between normal avoidant behavior and rumination that produces positive change versus pathological rumination where people spiral over inconsequential things?

Andrews: Psychiatry itself has changed. In the 1970s, there was a strong influence of Freudian thinking, which understood that painful feelings are pretty normal. The psychiatrist would try to evaluate what’s going on in your life socially and situationally and evaluate your pain in relation to that. If it was proportional they’d say, “It seems like it’s a natural normal response.”  It was only if the pain was disproportionate to the stressors that it would be viewed as a mental health problem. After the 1980 DSM revision, this perspective was lost.

Now, the field of psychiatry has essentially produced a checklist of feelings and symptoms, and if you check off enough of them then, boom, you’ve got the disorder.

 

Siem: Even using the word “pathological” bothers me in the context of mental health.

Andrews:  But as an evolutionist, here’s the thing: you and I are both wearing glasses, right? That means our eyes can malfunction, and they’re clearly evolved organs. Even if there’s some adaptation in our brains for producing depression, it has to malfunction at some rate. I just don’t think it’s as high a rate as the diagnostic criteria suggest. That’s my point: the real problem, as far as I can see, is how do we accurately distinguish between normal and pathological?

 

Siem: I don’t want to move past this point too quickly because it’s really important. One of the many a-ha moments I’ve had with your work is the observation that, in general, organs don’t fail that often.

Andrews: Especially during the reproductive years.

 

Siem: And yet, we’re creating this narrative that so many children have malfunctioning brains and that all these young, healthy people are struggling with malfunctioning brains. You don’t see one in four 14-year-olds with heart disease or liver failure. It doesn’t make sense.

Andrews: Agreed. Let’s consider this anecdote from a colleague of mine, Joanne Cacciatore, a Professor of Social Work at Arizona State. She shared a story about a young mother who was moving heavy furniture in her living room. She thought her one-year-old daughter was in the kitchen playing. The mother tried to move a heavy TV set, and as she did, it started slipping out of her hands. It turned out the daughter wasn’t in the kitchen after all—the mother accidentally dropped the TV on the child, and the child died.

 

Siem: Horrible.

Andrews: Incredibly horrible. But you don’t need to be an evolutionist to understand a few points here—just go with your gut instinct. If this woman’s brain is normal and she’s not a psychopath, is she going to have some agony over killing her child? The answer is, yeah, massively. Is she going to meet the DSM criteria for depression, which includes five out of nine symptoms for at least two weeks? Absolutely. Is she going to be depressed for two weeks? Yes. Two months? Definitely. She may never fully recover.

Joanne’s other work shows that it’s not uncommon for parents who’ve lost a child to still meet the DSM criteria for depression four years after the loss. In this case, because the mother can point to her own behavior as the cause, she may never recover. From my perspective, given the avoidant learning part of this, I’m not sure it would be a good idea for her to completely recover. You don’t want people to live in misery forever, but if something like this happened to me, I wouldn’t want to forget it entirely. I’d want a part of me to always remember to be careful with my kids.

The question is, would something like this happen again? If she’s got a normal brain and experiences agony for a long period, she’ll probably never let anything remotely similar happen again. More importantly, she might become ultra-protective of her children—maybe even too protective. But while that isn’t ideal, it’s understandable in terms of avoidance behavior and learning. This anecdote illustrates that, when contextualized, even severe and prolonged depression can be viewed as a normal response.

Siem: Let’s shift to your latest work. You mentioned earlier that you had just published a paper on successfully inducing depression in the lab. This ties into our discussion: if you can effectively induce it at will, it suggests that depression might not be a brain disorder. After all, you can’t snap your fingers and give someone fatty liver disease. Tell us more about this research and its implications for psychiatry.

Andrews: I’ll provide some context for why we developed this paradigm. There are various methods to induce sadness experimentally, but no one had developed a paradigm for triggering depression, which is a syndrome involving multiple symptoms—behavioral, cognitive, and emotional—that align with what we think of as DSM depression. The debates around depression—whether it’s a disorder or a normal emotional response—don’t progress because we lack experimental control over what we’re studying. We often rely on correlational studies or case-control studies that can’t address causality.

Experimental paradigms with randomized controls are the classic way of testing causal hypotheses. So, we created a paradigm that involves putting people through social rejection in the lab. It’s deceptive because they’re not actually being rejected by others, but they don’t know that and they think they’re being rejected by other people. The other component is that we give them a reason to analyze the rejection through root cause analysis and ask, “Is there something I could have done to avoid this?” That’s when you start seeing rumination.

Siem: Interesting.

Andrews: That’s why we’re not just inducing sadness. We published a paper using this paradigm, and we found that we can increase sadness, but we can also induce symptoms of anhedonia, which is the loss of interest in normally pleasurable activities like sex, social companionship, or humor. We can induce feelings of guilt or self-blame and low self-esteem, which manifest as a sense of worthlessness.

We also have some evidence that this paradigm triggers difficulty concentrating, another symptom, though we want to test that more in the future. I’m not saying we’ve met all the formal criteria for depression because that needs five out of nine symptoms and we have no evidence of inducing any suicidal ideation. But we have four out of the nine symptoms, including sadness, anhedonia, guilt, and worthlessness, which are core features of depression. Other symptoms like lethargy, restlessness, or appetite changes can occur in many other conditions as well. We know we are inducing a syndrome because it’s multiple symptoms with cognitive and affective or emotional features.

Our sample consists of university students, averaging around 18 years old, essentially at the beginning of their reproductive years. To get into university, you typically need to be fairly healthy with a well-functioning mind. From an evolutionary standpoint, natural selection doesn’t tolerate high levels of dysfunction during reproductive years. As we get older and past reproductive age, the rates of true disorders go up. So if we can trigger depressive symptomology in young healthy people, that is an argument against it being a disorder.

Some of the participants—15% to 20%—have psychiatric diagnoses. I find that problematic, given what I just told you about natural selection not tolerating high levels of this. But nevertheless, under the idea that they ostensibly have something wrong with their brains, we can exclude them from the analysis and see if the ones who don’t have the diagnosis still show these symptoms. We find no difference in the effect, which means, again, we are triggering this in now the most ostensibly healthy individuals in our population.

Our argument is that this is a normal response. Even the feelings of guilt and worthlessness can be useful, but how could it be a useful normal thing to feel like you’re a piece of crap? In a sense, worthlessness is actually motivational. For example, consider that physician who makes a serious medical error. If he believes he’s God’s gift to medical science and doesn’t experience a dip in self-esteem, he won’t be motivated to change his behavior. A little bit of self-criticism can drive corrective actions. That is something that the researchers in social psychology explicitly recognize, that loss of self-esteem can actually help young training physicians to learn from their mistakes and go on and be better physicians.

Siem: If we can induce depression in a lab, it makes the idea of prescribing antidepressants in response to depression seem questionable. Am I right or wrong about that?

Andrews: The results of our studies suggest that depression can be a normal emotional response to social rejection, even in healthy young people. This makes it harder to justify the use of antidepressants because the underlying philosophy of these medications is based on the assumption that something is going wrong in the brain—that depression is not a normal response. Antidepressants are meant to correct this supposed malfunction. But if depression is a normal emotional response, it makes more sense to turn to psychotherapy rather than medication.

Siem: From your perspective, what happens when we introduce antidepressants into perfectly normal brains? I know you’ve done research on cardiovascular issues connected to antidepressants and early mortality. I’m curious about your take on how medicating “normal” brains plays out.

Andrews: To answer that, I need to explain a bit about how antidepressants, especially SSRIs (Selective Serotonin Reuptake Inhibitors), work. SSRIs bind to the serotonin transporter, which is found on neurons that release serotonin in various regions of the brain. But the serotonin transporter isn’t just found in the brain—it’s present throughout the body, in organs like the lungs, heart, muscles, and even the intestines and reproductive organs. The serotonin transporter allows serotonin to be absorbed by cells that express it. For instance, a lung cell can take up serotonin from the bloodstream through this transporter.

When we call these drugs “selective,” it means they primarily bind to the serotonin transporter and not to others, like the dopamine transporter. But they aren’t selective enough to only target the brain. The idea behind using these drugs to treat depression is that they target serotonin in the brain, but because serotonin transporters are found in so many other parts of the body, SSRIs affect other systems too.

Siem: Wasn’t that also a bit of a marketing trick? The idea that it’s “selective” made it sound like we knew exactly what we were targeting.

Andrews: Yes, that’s a good point. I do believe that SSRIs can reduce depressive symptoms, but they aren’t as effective as people might think. Much of the symptom reduction can be attributed to the placebo effect. That said, I wouldn’t go so far as to say they have no effect at all.

Siem: This leads us to the issue of trade-offs. What are the pros and cons, and what kind of informed consent are people missing out on? Based on what you’re about to share, it seems like people might not fully understand the trade-off they’re making when taking these medications.

Andrews: There are a number of unintended side effects that patients and physicians don’t want. These include cognitive effects, which can be undesirable, but let’s focus on the physical side effects for now.  The Selective Serotonin Reuptake Inhibitor, because it’s not selective enough, will bind to the serotonin transporter wherever it is in the body. If you’ve ever seen an antidepressant commercial in a country where direct-to-consumer advertising is allowed, like the U.S., you’ll hear a long list of side effects: digestive problems, sexual dysfunction, cardiovascular issues, and more. These side effects may seem random, but if you actually understand that the drugs bind to the serotonin transporter wherever it is in the body and prevent serotonin from going into all sorts of organ systems,  you can start connecting the dots.The side effects may be mild individually, but when they occur across multiple systems, the cumulative effect can be significant.

 

Siem: Exactly, and people seem to experience heightened emotional responses during withdrawal. That makes me think: why do people start complaining about all these symptoms when they come off the drugs?

Andrews: In my mind, this is still a frontier area of research, but there are many individual cases—people who’ve gone through discontinuing an antidepressant like an SSRI—that show what you’ve said is true. Understanding the mechanisms behind it is the understudied aspect. I could easily imagine that what’s happening is somewhat the reverse of the side effects. Not everyone is going to experience all the side effects you hear about at the end of an SSRI commercial, but collectively they are there, and we know this from the variety of individual responses that can occur. This is because the serotonin transporter is expressed everywhere. People’s individual side effect profiles, whether they’re starting or discontinuing the drug, are likely to be highly variable and broad-ranging.

Siem: Well, Paul, let’s end this on a positive note. With all your research and knowledge in this field, how can this give people a little bit of hope as they move forward in their recovery?

Andrews:  I think there’s a lot of hope. One starting point is to rethink depression as a normal response and think about how a diagnostic criteria that pathologizes normal reactions may be doing a lot of harm. We’re taking people with brains that are working normally, and for the most part, pushing them to use antidepressant drugs which actually interfere with the normal working of the brain. The whole purpose of these drugs is to alter normal brain function. With a better understanding of depression as a normal emotional response to the problems that life throws at you, you can use that to seek out better treatments.

I want to point out the difference philosophically, again, between antidepressant medications and psychotherapies. The disorder narrative common in our culture is a good justification for the use of medications. But if the disorder narrative is right, in principle, the medications could potentially fix what’s going on if we’ve got the mechanism correct. On the other hand, psychotherapies are basically just as effective as antidepressants and may have more useful effects in terms of long-term use than drugs. This doesn’t align well under a disorder narrative because psychotherapy requires your brain to be working pretty well.

If you’ve ever experienced CBT, the therapist doesn’t give you answers. They encourage you to develop your own hypotheses about what’s going wrong in your life and test them. That process requires your brain to be working reasonably well. If the brain is truly malfunctioning, you wouldn’t expect it to be capable of engaging in scientific thinking, which is essentially what CBT demands. This is true for all psychotherapies—they depend on the brain’s ability to reason and understand, which isn’t consistent with the idea of a malfunctioning brain.

The alternative view—that depression is a normal response—doesn’t mean you don’t need therapy. Just like you might need help with a sprained ankle or an infection, you might need help dealing with the complexities of life. But none of this means your brain isn’t working well.

This is important for clinicians, especially clinical psychologists, to understand. One of the reasons why antidepressant medications have skyrocketed in prevalence since the 1980s while psychotherapies have declined is because of the way the narrative about depression has changed. It’s now viewed as a disorder, and that disorder narrative is naturally inclined to push the antidepressants. I think the clinical audience ought to care about this disorder narrative, and they ought to think carefully about the justification that they have in their own minds for why they’re using psychotherapy. If you really think it’s a disorder, your psychotherapy justification doesn’t make that much sense, I’m sorry to say.

Another thing that a psychotherapist can do is to recognize that most serious, even chronic, depressions are normal emotional responses. As a therapist, what is a useful way of approaching your patient? With the disorder narrative, under the idea that it reduces stigma, you say, “You don’t have to worry about this; this is not your fault. You just have something going wrong in your brain that you can’t control.” That’s where the medication often comes in.

But a different way of tackling this would be to say, “You know what? This is a normal emotional response to serious life events that are going on in your life. I would be worried if you didn’t feel depressed over this circumstance. This is normal, and this is your brain telling you that you need to spend time thinking about these crucial things in your life and as a psychotherapist, I can help you with some of that.”

Think about the difference there. First of all, if you’re a patient facing serious problems in your life, like divorce, financial issues, or custody battles—really complicated stuff—and then you’re told that your brain isn’t working well, you’re not only going to become depressed over the problems themselves, but you may also get more depressed over the idea that your brain isn’t functioning properly. How are you supposed to solve all your problems if your brain isn’t working well? It just becomes even more overwhelming. But if you tell your patient that they have a normal response, they avoid that second layer of depression on top of everything else.

The other point here is that this approach naturally induces what is called, in the literature, the “therapeutic alliance.” It turns out that psychotherapies are generally effective in treating depression. It’s not due to the differences between the various psychotherapeutic styles, but rather something called the “common factors.” One of these common factors across psychotherapies is how well you connect, get along, and interact with your therapist. By telling your patient that they have a normal emotional response to serious life events, that is one natural way to build that alliance.

As a therapist, sometimes patients don’t know why they’re depressed. When you ask them, they might not fully understand or be able to identify a cause. But you could ask them a slightly different question: “What are you ruminating about?” They should be able to answer that. From a therapeutic perspective, that’s a useful clue that you can use to help your patient gain greater awareness of why they’re depressed.

Siem:  It’s so obvious—what are you thinking about all day? Thank you so much, Paul. Where can people find you if they want to learn more about your work?

Andrews: I’m an Associate Professor at McMaster University. You can Google my name, Paul Andrews, at McMaster University. That’s probably the best way to find me.

 

Siem:  Thank you so much, Paul. Keep up the good work.

***

MIA Reports are supported by a grant from Open Excellence, and by donations from MIA readers. To donate, visit: https://www.madinamerica.com/donate/