AFTER FINISHING OUR PhDs at Penn in the late 1960s, Steve Maier and I went in opposite directions. I spun outward from learned helplessness, applying it to larger and larger targets: human depression, physical illness, psychotherapy, schools, businesses, the army, and, as you will soon read, even a flourishing human future. Steve went inward. He essentially switched fields and retrained as a serious neuroscientist studying how the brain circuits in rats influence behavior. He did not return to helplessness until the 1990s. When he did, he changed everything.
Steve took an assistant professorship in the University of Illinois’s psychology department in 1968. He had a weird teaching style, but it won the undergraduate teaching award. He did not lecture; he only answered questions. I met Greg Kimble, a venerable learning theorist, and he told me about a new department he was forming at the University of Colorado in Boulder. He asked me if I was interested. I had just been tenured at Penn and was not moving again. Steve, I recalled, loved the outdoors, so I told Greg, “Try Steve Maier.” Greg did, and Steve left the plains of Illinois for the mountains of Colorado, where he stayed for the rest of his career—hiking and bicycling through these mountains on weekends. He molted, no longer a flabby city kid, and even in his mid-seventies, he is, annoyingly, a model of trim, muscular fitness.
Steve went on to make the most important discovery in our field, and his discoveries turned learned helplessness on its head. He showed that the arrow of causality that we had postulated was wrong and that it was not helplessness but control and mastery that were learned. He discovered the brain circuits that produce and prevent helplessness, and their functions are wildly different from what we had expected all those years ago. How Steve discovered this is a mite technical but the prettiest example of neuroscience that I know, so I urge the reader to bear with the telling. This chapter recounts his story; I play the role of admirer. The years I’d spent working on positive psychology gave me a special appreciation for just how profound his findings were.
I first need to remind you what our theory said originally and what we discovered fifty years ago. Our theory essentially claimed that animals learned that nothing they did mattered, and so they expected that nothing they did in the future would matter. This expectation of helplessness undermined trying to escape bad events in the future. Our evidence came from the triadic design: in the first phase, one group (ESC) of animals got escapable shock, a second group got exactly the same pattern of shock but the shock continued regardless of what they did (INESC), and a third group got no shock (Zero). In the second phase all the animals went to a shuttle box, usually twenty-four hours after the first phase. Most of the animals in the INESC group then failed to escape in the shuttle box; they just lay there passively. Almost all the animals in the ESC and Zero groups learned to escape shock well in the shuttle box.
Steve, of course, was eager to know what the brain of the rat was doing, but good tools for looking closely at brain circuitry only got developed in the 1990s. Steve wondered where to start and reasoned backward from the passive behavior and heightened anxiety in the INESC group. Helpless rats showed aborted fight and flight and increased anxiety (panic), and the brain structures for these were known by the 1990s: the dorsal periaqueductal gray controls fight or flight and the amygdala controls panic. It looked as if INESC rats had inhibited dorsal periaqueductal gray function and exaggerated amygdala function.
And there is a structure—the dorsal raphe nucleus (DRN)—that projects to both, inhibiting one and potentiating the other. It does this by releasing serotonin, 5-HT, which in turn leads to the release of 5-HT in the amygdala and dorsal periaqueductal gray down the line. This produces passivity and panic.
To my non-neuroscience reader: I am going to simplify and keep to the bare bones of what you need to know about the brain to appreciate how Steve did it. The dorsal raphe nucleus is one structure you will need to remember and the transmitter, 5-HT (serotonin), is another, because activating the DRN directly and getting its release of 5-HT produces just the same deficits—panic and inhibited fight-or-flight response—as inescapable shock does.
So Steve reasoned that if inescapable shock produces a powerful activation of the DRN 5-HT neurons leading to the release of 5-HT in the amygdala and dorsal periaqueductal gray, then the DRN might be the crucial node in INESC producing learned helplessness. But it would also have to be true that escapable shock does not activate the DRN.
Got it so far? If not, stop scanning, slow down, and reread the previous paragraph.
Steve’s group then demonstrated exactly this: inescapable shock activates the neurons in the DRN that contain 5-HT, and exactly equal escapable shock does not.1
Yet, to withstand scientific scrutiny, these findings alone were not enough. The fact that INESC shock activated DRN 5-HT neurons did not mean that this process is either necessary or sufficient to produce helplessness. So next Steve showed that DRN activation was necessary for helplessness to occur and that it was sufficient. In other words, DRN activation always produced helplessness, and nothing else did. He determined necessity by blocking the DRN activation produced by inescapable shock with lesions or injection of drugs to the DRN.2 These prevented inescapable shock from producing its usual poor escape and panic: rats that got INESC but had the DRN blocked were not helpless.
He determined sufficiency simply by injecting drugs that activate 5-HT neurons into the DRN, producing the same passivity and panic as does INESC shock. Rats that got no shock at all but had an excited DRN were helpless.
So the activation of the DRN is necessary and sufficient to produce passivity and panic.
(Pause. If your hold on this material is shaky, go back and read the last page again.)
To this point, Steve’s experiments had shown only the brain chemistry behind the behaviors we’d witnessed decades before. But now a new question arose, one that we couldn’t have considered in 1964 because neuroscience hadn’t come far enough: Why does the DRN respond only if the shock is inescapable? The obvious option is that the DRN detects that shock is inescapable. However, the DRN is a tiny brainstem structure (perhaps 20,000 to 30,000 neurons in the rat, 150,000 cells in the human), and so it is unlikely to have the processing power to detect no control. The next possibility is that the DRN receives greater excitatory input during inescapable than during escapable shock, thereby leading to more activation with inescapable shock. But this turned out not to be so.
If the rats were not learning helplessness, only one obvious possibility remained: that being helpless was a natural, unlearned, default response to prolonged shock. This, in turn, meant that the presence of control must somehow be learned high up in the cortex, and this learning must turn off the DRN and so abort helplessness. In other words, helplessness wasn’t learned; control was. Steve tested this theory and confirmed it was true. Through further experiments he located the actual pathway—within the medial prefrontal cortex (MPFC)—that was the agent of the change we’d seen all those years before. The MPFC-DRN pathway was the key, and it didn’t teach those rats helplessness; it taught them control.
Steve showed that turning this pathway on or off chemically overrode the external contingencies. He first taught rats to escape by turning a wheel. They did this perfectly well, even when the MPFC-DRN pathway was turned off chemically. But crucially they were later helpless in the shuttle box. Completing the tour de force, he first made the rats passive and helpless, because their wheel turning did not matter. But he turned this pathway on chemically during the helplessness. Crucially these rats were not helpless later in the shuttle box.3
The development of a new circuit that sent the information about control to turn off the DRN required the production of new proteins, and Steve’s group showed that blocking new protein synthesis in the MPFC after escapable shock allowed helplessness to later occur.4 That is, even though the subjects exerted control and the MPFC was activated, the rats were helpless later unless new proteins could be formed. These particular proteins, called plasticity proteins, were indeed induced in the MPFC by escapable shock.5 So Steve concluded that the MPFC-DRN pathway was modified in the several hours after the experience of escapable shock, and this altered pathway was crucial to the expectation that shock would be controllable in the future.
Keep this newly formed MPFC-DRN pathway in mind. What in the world is it?
Steve’s findings turned learned helplessness on its head. Helplessness wasn’t learned; only control was. Did this mean our first great finding was a mistake? What did the original learned helplessness theory get right, and what did it get wrong?
As the original theory claimed, organisms do indeed learn about the dimension of control, and this dimension is critical. However, the dimension detected and learned about was the presence, not the absence, of control. This did not mean that animals were born helpless, but it did mean that they became helpless when faced with sustained negative experiences. The passivity and panic that followed INESC for several days was an unlearned reaction to prolonged aversive stimulation. This meant that helplessness is not learned; it is some kind of mammalian default response to bad things. Steve had showed that learning about control aborted this default process. He’d discovered a pathway for defeating helplessness.
Clearly, this was a substantial revision to the theory we developed fifty years ago. When we found that dogs given inescapable shock later failed to learn to escape in a shuttle box, but that dogs given escapable shock later escaped normally, the ideas we developed were our guess about what is most adaptive for dogs, rats, and people. We reasoned that active coping was best because this would minimize bad events, and so we assumed that organisms would initially expect control to be possible and try to escape. If the stressor proved to be uncontrollable, organisms would then learn this and expect it to be true in the future, with this expectation of uncontrollability undermining trying. However, the neural data do not support this. Instead, the presence of control is the active ingredient, leading to the inhibition of default DRN giving up.
Perhaps this counterintuitive arrangement makes more sense if we consider evolution. For our ancient ancestors, threats engaged defensive reflexes, and these reflexes cost energy. If the reflexes were unsuccessful, it would be adaptive to shut them down and conserve energy for adjustments that promote survival, such as altering the immune system to better fight any infection or wound that might occur. Primitive organisms, such as reptiles, do not have the sensory apparatus to detect threats at a great distance; nor do they have complex responding to deploy for control. Bad events for primitive organisms are generally uncontrollable by any available voluntary behavior, and so primitive organisms do not need mechanisms to detect controllability. Thus, the success of defensive reflexes is likely related to how long the existing threat lasts—if it is prolonged, then passivity and conservation would be adaptive, with energy being shifted to physiological adjustments to threat. It is important that 5-HT is phylogenetically ancient and has, from the beginning, been involved in shifting the balance of the flow of energy.
As organisms became more complex in evolution, they began to detect and identify threats at a distance. And so they developed rich behavioral and cognitive skills to control threats. Control became available even against prolonged threats. So passivity and other energy adjustments set in motion by the continuation of threat should shut down when control works.
Thus, when high mammals encounter a threat, here is what happens. First, defensive reflexes occur, activating the DRN. 5-HT builds if the threat continues long enough, and if 5-HT reaches some threshold, voluntary trying is shut down, and energy flow is shifted. However, detection of control shuts down the DRN so that trying will continue. Further, when control exists, plasticity occurs in the MPFC-DRN circuit so that the system now learns, reacting to bad events as if they are escapable, thereby prolonging trying. This is why expert athletes, soldiers, and pilots are calm under pressure. Their brains detect and expect control when others panic and freeze.
In short, control is a relatively recent adaptation in evolutionary history. Our experiments showed just how powerful it can be, and Steve’s showed exactly where in the brain it lodged.
THERE ARE TWO main takeaways. The first is that our default response to prolonged bad events is helplessness and heightened anxiety. The second is that our higher cortical processes inhibit this default helplessness. These higher cortical processes buffer against, but do not annihilate, the default passivity of the DRN.
This paints an up-from-helplessness neural picture that fits human development well. As a species we are born helpless, and as infants and toddlers we stay immature for a very long time. Indeed, this slow maturation may be crucial to our success as a species. Perhaps during this time we are gradually learning more and more control. Importantly rats that learn mastery early in life resist helplessness when exposed to inescapable shock later in life, and it is exactly the MPFC-DRN pathway that prevents this later helplessness.6 Early experience with control immunizes us against the ravages of uncontrollability later in life, perhaps through this pathway. But I still wondered exactly what psychological process this all-important circuit was.
All this has major implications for therapy. Therapy that tries to undo trauma is not likely to work. Trauma is here to stay. Treatment that dwells on past trauma and present misery without emphasizing planning for a better future may be for naught. Consider the classes of therapy in which the patient reviews a past trauma in order to gain insight into its causes or to have catharsis about it. The circuitry suggests that merely confronting, understanding and reliving the trauma cannot achieve much. Confronting the past is not the province only of psychodynamic therapy; it is also common in popular cognitive-behavioral therapies: discussion of rumination, discussion of memory biases like selective filtering (where the patient only attended to a negative part of something that happened and ignored the positive parts), behavior-chain analysis (where the patient looks at all the steps that led up to an eating or drinking binge and considers how those steps set him up to “fail”), and catastrophizing about the present are all modern techniques that work on the present without necessarily addressing the future. A skilled and experienced therapist can lead a patient through these exercises about the past and the present with the added, explicit purpose of changing future behavior, such as better recognizing triggers for past maladaptive responses in order to avoid those triggers in the future or gaining insight into catastrophizing in order to learn how to be more optimistic. So therapy that explicitly creates end runs around trauma, buffers against it, and uses the past and the present as vehicles for planning a better future is more likely to work.
This circuit also implies something that—skeptical about the overselling of neuroscience—I have never countenanced before: that there may come a day when we can treat depression by turning on the MPFC-DRN pathway directly or by turning off the DRN directly. Steve’s discoveries were about rats, of course. But human beings have quite similar structures. Right now we do not have anything like the tools to identify these circuits in human beings. Nor do we have the tools to turn on or turn off these structures by drugs or transmagnetic stimulation or optogenetics (the use of light to stimulate neurons). But that day will soon come, and depression and panic in humans—I say this for the very first time—might become curable.
It is gratifying to have lived long enough and to be in a vibrant enough science to find out you were wrong. Helplessness is not learned as we first thought; it is the default reaction of mammals to prolonged bad events. What is vouchsafed to humans, however, is a magnificent cortex, a top-down structure that learns mastery over bad events. We can learn—and teach—that future bad events will be controllable, and this will buffer us against helplessness and anxiety.
THIS DISCOVERY WAS profound. But I still didn’t fully grasp what this circuit really was until the middle of a long, strange night.
I had been mulling over Steve’s data for more than a year. We were writing a Psychological Review article together,7 but I was still bewildered. I felt the MPFC-DRN circuit was tremendously important, that it made evolutionary sense, and that it was more important even than Steve thought, but I couldn’t put my finger on why.
My friend and patron Jack Templeton died of brain cancer on May 22, 2015. Jack and I still had so much unfinished business to do together: to increase the creativity of the next generation, to raise positive psychology to the next level, and to understand the biology of hope and how it relates to spiritual flourishing. He had been a rock of support for all of these.
His memorial service was scheduled for June 11, 2015, at 10:30 a.m. at an evangelical Presbyterian church in nearby Bryn Mawr. A thousand people from all over the world would come—representatives of all the good works that Jack supported—and I planned to be there as well.
But on June 10, Mandy took me to hospital. There was a cyst on my spine, and I was having trouble walking. I’d had a similar episode a year before, and surgery to remove the cyst had kept me in bed for almost a month. I recovered completely, and the cyst was benign. The surgeons now operated while visualizing my spinal cord in real time—an amazing new technique—and the new cyst was aspirated. I came out of general anesthesia after midnight. All was well—the cyst was benign again—but I was advised to stay in bed for the next few days. Mandy put me to bed at 1 a.m.
Three hours later I awoke suddenly from a deep sleep, completely alert. My very first thought was that I knew what the MPFC-DRN circuit was. I just knew it. It was the hope circuit. All this fell suddenly into place. I didn’t know the details yet, but I knew I was right and rushed downstairs to write an email to Steve. I started the email at 5:35 a.m. and finished at 5:57 a.m. At the moment I put my finger on the send button, all the electricity in Lower Merion Township died.
A scene from the movie Field of Dreams crossed my mind. The protagonist, Ray, tells his wife Annie, “When primal forces of nature tell you to do something, the prudent thing is not to quibble over details.”
I want to tell Jack Templeton about a discovery he would have loved. Jack and I talked often about a science of hope and its relation to faith. Jack wanted there to be a “hope circuit” in the brain, and I can tell him now that there is. Photo courtesy of the John Templeton Foundation.
The power came back on at 6:20 a.m., and I hit send. The email was delivered to Steve.
I next opened my snail mail and read a note of thanks from Mathew White, sent with a copy of his new book on positive education at Saint Peter’s College in Adelaide.8 I glanced at the chapter on positive Christianity and read, “Religion and science are opposed but only in the same sense in which my thumb and forefinger are opposed—and between the two, one can grasp everything.” This was the focal issue for Jack and his father, Sir John Templeton. I burst into tears and had a command hallucination to put on a suit and go to the memorial service. Which I did.
I arrived at the church shortly before the service began and saw Jack’s widow, Pina. I told her about my midnight epiphany. She told me that I should testify. There were two eulogies for Jack scheduled, and I should give a third. So after Stephen Post’s eulogy, I was announced.
“I want to tell Jack what happened to me this morning and to tell Jack about a discovery he would have loved. Jack and I talked often about a science of hope and its relation to faith.”
I narrated the surgery, the sudden awakening, the MPFC-DRN mystery, and the blackout. Finally I added, “Between science and religion one can grasp everything.”
“Jack wanted there to be a ‘hope circuit’ in the brain, and I can tell him now that there is,” I concluded.
There was a stunned silence. I left the church and went home to bed. I stayed there for days.