Above:Electrical brain stimulation is one of several experiments that the author, pictured, subjected himself to.

I am, at this moment, an overgrown lab rat. A blue plastic band secures a pair of electrodes to my head: a negatively charged cathode on my forehead above my right eye, and a positively charged anode on the left side of my scalp over a portion of my brain called the dorsolateral prefrontal cortex. The DLPFC, as it’s known, is an area of the brain involved in episodic memory, the conscious recollection of the events of one’s past. The electrodes are connected by narrow wires to a gadget that, with its panel of dials and gauges, looks a bit like a ham radio. That device is steadily delivering a mild two-milliampere charge straight into my cranium.

As the electricity flows through my brain, David Gallo readies materials for a memory test. A tweedy, excitable cognitive scientist in his mid-40s, Gallo is the chair of the psychology department at the University of Chicago and heads its Memory Research Laboratory, a center dedicated entirely to the study of how and why we remember (or fail to).

“What do you feel?” Gallo asks. “Do you feel like your memory is charging up?”

“I feel a tickle over my right eye,” I say. “And there’s heat through my scalp and forehead.”

“Almost like a low-grade burning sensation,” he says cheerfully.

Apparently the burning means it’s working.

The University of Chicago’s David Gallo presents the author with a picture to remember (or not).
The University of Chicago’s David Gallo presents the author with a picture to remember (or not).

Gallo, three doctoral students, and I are in the lab’s main testing site, on the third floor of the ivy-covered building housing the psychology department. The narrow room is furnished with tables, chairs, and desktop computers loaded with memory tasks that study participants are asked to complete during experiments. The only decoration: an artistic rendering of the hippocampus, the part of the brain thought of as the seat of memory.

I overhear one of the doctoral students saying something to Gallo about shock therapy. The image of Jack Nicholson convulsing on a sanitarium table in One Flew Over the Cuckoo’s Nest bolts to mind. Sensing my concern, Gallo adopts a soothing bedside tone: “With electroconvulsive therapy, the idea is to deliver enough electricity to induce an epileptic seizure. We’re trying to do the opposite. We want just enough electricity so that we’re tickling the brain region that we want to be more active during a cognitive test. Seizure would not be a good outcome.”

As I finally begin to relax into what will be a 20-minute session of transcranial direct current stimulation — tDCS for short — Gallo says he’s going to give me a memory recall test. The format is identical to the test he gave me before hooking me up to the electrodes: Once he finishes reciting a list of words, I am to write down as many as I can remember. It seems simple enough. What I won’t know until later is that the lists are specifically designed to elicit false memories by triggering powerful semantic associations. After hearing, for instance, the words “banner,” “American,” “symbol,” “stars,” “anthem,” and “stripes,” you’ll swear on your mother’s life that you heard the word “flag.” In this sly bit of laboratory trickery (it’s called the Deese-Roediger-McDermott task and was the subject of Gallo’s 2006 book Associative Illusions of Memory), that misremembered word is known as the “critical lure.”

He reads aloud the first list: “Low, clouds, up, tall, tower, jump, white, dark, cat, charred, night, funeral.” (Critical lures: “high” and “black.”)

When he’s finished, I scribble furiously on a small yellow notepad.

“OK, ready for the next one?” Gallo asks. “Hard, light, pillow, plush, loud, cotton, butter, food, eat, sandwich, rye, jam, table, sit, legs, seat, couch, desk.” (Critical lures: “soft,” “bread,” and “chair.”)

Once the tDCS treatment has ended and I’ve completed the third and final memory test, Gallo tallies the results. “You were 60 percent false memories before stimulation, 30 percent false memories during, and 46 percent false memories afterward,” he reports. “So if we tested 50 subjects and averaged all of those responses and we found this same pattern, we might think that getting tDCS helps you monitor your memory and realize what was there and what wasn’t.”

Witnessing the apparent positive effect of tDCS is eye-opening, yet I spend the drive home from Hyde Park preoccupied with all my false memories. How could I have been so confident in my responses while unconsciously committing so many mental errors? Earlier, Gallo had tried to reassure me, explaining the double-edged nature of memory’s malleability: “These errors serve an adaptive function. They’re being driven by heuristics, or shortcuts, that your brain uses to process information in memory, because you can’t remember everything. You remember selectively and you reconstruct. Sometimes your brain gets it right; sometimes your brain gets it wrong.” And sometimes you’re stuck in traffic on I-94, the faint impression of an electrode on your forehead, questioning whether any of the memories of your life are accurate.


Memory, in essence, is who we are. As Gallo puts it, “Memory shapes everything. Not only our understanding of ourselves, but our understanding of anything in the universe.” This helps explain the tragic lament of Auguste Deter, the first person diagnosed with what would become known as Alzheimer’s disease: “I have lost myself.”

You don’t have to look hard to discern a growing preoccupation with memory and cognitive acuity in the popular imagination. While channel-surfing late one night, I stumble across an infomercial-style program on WTTW in which the controversial psychiatrist Daniel Amen, calling to mind the Ronco ads of yore, ballyhoos his book Memory Rescue: Supercharge Your Brain, Reverse Memory Loss, and Remember What Matters Most. Another time, in line at the grocery store, I spot a memory-themed special edition of Time, its cover blaring “Build a Sharper Mind” in the breezy tone favored by the neighboring fitness magazine promising instant six-pack abs.

You can even buy cheap home versions of tDCS devices. Not long after my brain-zapping session with Gallo, I decide to try one out — in the name of research and in the vague hope of offsetting the memory-eroding effects of stress, insomnia, and what researchers have dubbed “cognitive offloading.” Basically, that’s what we do when we let the internet, our smartphones, and Alexa do our remembering for us.

I have reason to be hopeful. Upon arriving home after my tDCS session with Gallo, I flipped on Jeopardy! A longtime fan of the show, I watch nearly every day, shouting out answers in the form of questions, usually on the too-late side. That afternoon something felt different. It wasn’t as if my head were suddenly filled with trivia previously unknown to me. But correct responses seemed to leap to mind more rapidly: What is The Birds? Who is Cardi B? What is a glockenspiel, Alex? I was focused, sharp. When I mentioned the experience to Gallo, he said he couldn’t definitively ascribe my success to the tDCS but noted that some research has shown the presence of an “aftereffect period,” when “enhanced cortical readiness lasts even after the electrical current is turned off, maybe upward of an hour.”

The Brain Driver, the home tDCS kit I chose, is made by a local entrepreneur named Alex Heredia, who has said he was inspired to create his own gizmo after stimulation helped him overcome bouts of depression and foggy-headedness in the wake of a bicycling accident. The iPod-ish gadget sells online for around $150 — considerably less than Gallo’s Soterix tDCS unit, which a doctoral student told me costs thousands of dollars and is available only to physicians and institutions. The cover of the Brain Driver’s user manual exhorts: “Power Up Your Brain. 20 Minutes Per Day.” That’s what I begin to do, as part of my morning routine. Coffee, cereal, and Brain Driver on the table in front of me, I slip the electrodes under the headband and slowly crank up the unit to two milliamperes. Afterward, I have a bit more energy, as if I’ve downed an extra shot of espresso, though I don’t experience any particular cognitive breakthroughs. I continue the morning sessions for a few more days, until I come across an unsettling post on the tDCS Reddit thread — 11,500 subscribers and counting — in which a panicked-sounding user describes disturbing lapses in concentration, growing lethargy, and uncomfortable cranial pressure after just a few days of home stimulation sessions.

I ask Gallo if he thinks DIY biohackers are crazy to be electrifying their brains without the supervision of a professional. “I wouldn’t say ‘crazy,’ ” he says, “because in psychology that’s a loaded term.” But he emphasizes that while one-time use in a laboratory setting is relatively safe, the great unknown is the potential side effects of “repeated administration over the course of days.” A group of neurologists echoed that uncertainty in an open letter about do-it-yourself tDCS published in the Annals of Neurology in 2016. “Whereas some risks, such as burns to the skin and complications resulting from electrical equipment failures, are well recognized,” they wrote, “other problematic issues may not be immediately apparent.”

Indeed, much of the billion-dollar-plus memory and cognitive improvement industry has the whiff of snake oil. Among the abundant resources out there, you can find everything from memory boot camps to Brain Gym International’s “learning through movement” program. In 2016, the maker of Lumosity, a popular online brain-­training program, settled a Federal Trade Commission lawsuit alleging that the company deceived consumers with unfounded claims about boosting users’ mental performance and reducing or delaying cognitive impairment.

So what are Gallo and his colleagues hoping to achieve with tDCS? Among other things, carefully controlled electrical stimulation experiments are a cheap, noninvasive method of manipulating brain regions to determine which are causally involved in certain memory processes. They can also reveal interesting things about how mental acuity changes with age or even time of day. (A couple of weeks after my own tDCS session, I watched as a sleepy U. of C. student, her DLPFC freshly tickled, completed a memory task while undergoing a brain MRI; it was part of the lab’s inquiry into whether tDCS has a greater effect if it’s administered during a subject’s nonoptimal cognitive processing time — mornings for college students, evenings for senior citizens.)

Does tDCS work? At least when it comes to certain kinds of memory, Gallo’s research suggests it does. A paper he coauthored for the journal Cortex in 2015 concluded that electrical stimulation had a “significant” effect, allowing subjects “to more accurately recollect specific details of their experiences, demonstrating a causal role of DLPFC in the retrieval of episodic memories.” My own tDCS test — and maybe, just maybe, my Jeopardy! performance — seemed to bear this out.


Electricity is not the only stimulus Gallo and his fellow memory scientists are using to study the brain. On the fourth floor of the University of Chicago Medical Center, home to the Human Behavioral Pharmacology Laboratory, he and Harriet de Wit, a professor of psychiatry and behavioral neuroscience at the university, have been dosing people with various psychoactive drugs — THC, MDMA, amphetamines — in order to observe the effects on episodic memory. It’s a line of study that few had pursued before Gallo and de Wit began their research a decade ago, and now it’s taken on a more urgent relevance, especially in light of the progressive legalization — and growing potency — of marijuana.

“Why is it,” asks UIC’s Karl Szpunar, “that three years from now I can think back to this interview and imagine that someone else was hear with us the whole time?”
“Why is it,” asks UIC’s Karl Szpunar, “that three years from now I can think back to this interview and imagine that someone else was here with us the whole time?”

A few weeks after my tDCS session, I show up at the lab and am led by a research assistant to a small testing room that more closely resembles a grandmother’s parlor than a lab: a plush lounge chair, framed images of serene landscapes, a tray of scented candles, an array of mass-market magazines, a TV, an adult coloring book, a pack of colored pencils — amenities designed to keep drugged subjects on an even keel over the course of their four-hour stay. There will, however, be no psychoactive substances coursing through my system during the tests I’m about to undergo. Researchers can legally administer the drugs, which are obtained from pharmaceutical companies, only to full-fledged study participants, so I’ll have to settle for imagining how I might handle the various queries if I were high as a kite.

For today’s experiment, I’m asked to sit at a computer and perform a series of tasks designed to determine the effect of certain drugs on emotional memory. A software program prompts me to rate photographs based on my positive or negative emotional reaction to them, as well as the degree to which I find them mentally “arousing” — that is, stimulating or boring.

A photo of heavenly clouds appears on the screen, then quickly disappears. I rate it with two clicks of the mouse: highly positive on the emotional scale, neutral on the mental arousal scale. A celebratory scene in front of Cinderella Castle at Disney World? Very positive, quite arousing. A patterned china plate? Emotionally neutral, totally boring. A filthy, clogged toilet? Extremely negative, extremely arousing. Were I an actual subject in the study, I would return to the lab 48 hours later to have my memory of the images assessed.

Tests like this one have revealed to Gallo and de Wit something that none of their predecessors had uncovered: When subjects are under the influence of THC during memory creation, the drug’s well-known amnesiac effect is disproportionately stronger for emotional stimuli. Put another way, you’re more likely to forget an event that happened when you were stoned, but that likelihood increases markedly if the event carried an emotional charge (Cinderella’s castle, say, or a dirty toilet).

That finding came as a surprise. Emotion has been shown, in nonstoned people at least, to make any event more memorable. It’s a big reason we vividly recollect a first kiss or a car accident, even decades later, but struggle to remember what we ate for lunch yesterday. Gallo and de Wit suspect that what they observed is a symptom of THC dampening the processing in regions of the brain, such as the amygdala and hippocampus, that help encode emotional memories, therefore taking away the boost that you typically see in relation to positively or negatively charged information.

Gallo and de Wit have also observed how people on drugs retrieve memories. When participants studied images during a computer task while sober, then took a recall test two days later while under the influence of THC, they displayed fairly accurate memories of the images they had seen, as compared with the placebo group. But they were more likely to produce false memories, claiming to remember stimuli — whether positive, negative, or neutral — that they had in fact not studied.

“You can think of different consequences of that,” Gallo says. “If you study for an essay exam under the influence of THC, you’re less likely to remember the facts when you take that test. Now let’s say you study sober, and you think, I got this, and you go and take the exam under the influence. What’s going to happen is you’re going to falsely remember all sorts of stuff. You’re going to fill up the page with all the stuff you studied, as well as all sorts of stuff that’s nonsense.”

Given such conclusions, you might expect Gallo to be firmly against pot, but his take is more nuanced. “There’s this received wisdom that cannabis distorts your brain and your mind — this Reefer Madness rhetoric,” he says. “[But] we put someone who’s on THC in a memory test, and it’s not like they can’t remember anything, like they’re blacked out. They remember a lot. … If you are living in chronic pain and cannabis is the only thing that seems to help, maybe you would feel better about taking the drug if the literature tells you, ‘The memory impairments are not as large as you might think.’ ”


Memory is more than a window into the misty, water-colored moments of our past. It is also, I come to find out from Karl Szpunar, the key to imagining the future. Szpunar is the 39-year-old head of the University of Illinois at Chicago Memory Lab, and he offers the reassuring observation that the malleability of memory — its unsettling tendency to be inaccurate — is as much a blessing as a curse.

“Why is it that three years from now I can think back to this interview and imagine that someone else was sitting here with us the whole time?” he asks when we meet in his office on campus. “If you have a memory system that’s malleable, the payoff is that we can imagine what has not happened yet. It allows us to imagine the future.”

Our recollective mind is traditionally thought of as a mechanism for one-way time travel, a tool for retrieving information from the past to help guide us in the present. Szpunar and his colleagues have helped make it clearer that we also reach back to our past in order to mentally jump forward, simulating events that haven’t yet happened in order to envision what may lie ahead. With the aid of functional MRI scans, Szpunar has observed that subjects’ brains exhibit a similar pattern of neural activation whether they are remembering past events or imagining future ones.

“If I’m remembering this morning’s breakfast or imagining tomorrow’s breakfast,” Szpunar says, “I’m going to activate the same memory.” He demonstrates with a simple exercise in which he asks me to monitor my own mental response to a hypothetical situation. “Sometime this summer,” he says, “you and I are going to be at a Cubs game and it’s going to start raining.” My brain instantly extracts salient details from the past — games I’ve attended at Wrigley Field, times I’ve been rained on — in order to conceive of a totally novel future event. “That’s an important adaptive feature of our memory.”

Early notions of a link between memory and future thinking arose from the study of people with profound amnesia. Those with little or no access to their episodic memory, typically as a result of disease or injury, frequently describe their past as a void — and they tend to say the same of the future. “They don’t have the building blocks of the past to construct future scenarios,” Szpunar explains.

His lab is now extending its research to study memory’s role in imagining not just personal but also collective future scenarios. To demonstrate, a graduate student seats me in a testing room to complete something called a “future fluency task.” A computer asks me to list things that I’m personally excited about and worried about in the next week, the next year, and the next five to 10 years. I’m then asked to write down what I believe Americans are excited and worried about over the same time periods.

I quickly jot down a bunch of things that have me excited for the future: a Colorado ski trip on the near horizon, my wife and I buying a home somewhere down the line. I have a harder time coming up with things I’m worried about. By contrast, when I turn to the collective future, I find it easy to reach into the country’s seemingly bottomless bag of woe to pull out worries: the 2020 presidential election, global warming, police brutality, and so on. The personal excitement and collective worry lists end up with far more items than the personal worry and collective excitement lists. (Straining to come up with something — anything — the country might be excited about in the next week, I actually type “The Bachelor.”)

My responses fit the pattern Szpunar sees in the lab’s findings, some of which appeared in a paper titled “Staying Positive in a Dystopian Future: A Novel Dissociation Between Personal and Collective Cognition,” published last year in The Journal of Experimental Psychology: General. “People see their own future as positive,” he tells me. “The future of the world around them? Not so much. And we think this has a lot to do with how we project forward into our personal future using our own experiences [which are influenced by a personal positivity bias]. When we try to imagine the collective future, we’re also using memories — but much of it is memories of information that we got from the news media, which is inundated with negative information.” Asked, for instance, about the future consequences of social media or artificial intelligence, subjects tend to give mostly positive answers about how the technology will affect them personally and largely negative responses about how the same technology will effect the country.

Szpunar surmises that people could be persuaded to think about topical issues in very different ways if they were primed to favor a personal perspective over a collective one. At the very least, he says, if we can’t make people feel less negative about the country’s future, perhaps the negativity can be harnessed to inspire them to take action to improve what lies ahead.


As I begin to think about some of my most closely held memories in a new light, I wonder: Can the malleability of memory that makes simulating the future possible also be used to simulate the past in order to patch tragic mental gaps?

My father died of a heart attack at the age of 43. I was just 4 years old — too young, it seems, to have retained a single vivid memory of him. Growing up, I was sometimes envious of my two older brothers, who had years’ worth of rich memories of their experiences with Dad. Other times, I counted myself lucky: By not holding on to recollections of my father, I reasoned, I may have been less personally affected by his loss. While I ached for a father, I couldn’t truly miss my dad as a person.

And yet, like others who lose a parent at a very young age, I absorbed a lot of secondhand information about my father — from stories, photographs, home movies, and so on. The strange thing is, I began to semiconsciously reconstruct those details into my own false memories of Dad. It was an attempt, I suppose, to copy and paste him as a character into the story of my life, to feel more intimately connected to a man I’ll never truly know.

I ask Gallo if my experience fits with his understanding of memory distortion. He considers my question and answers in a comforting tone that sounds more doctor-to-patient than source-to-journalist: “Your way of adapting to the loss was constructing who your dad was, in a sense to try to have that influence on your life that lasted beyond him. Those memories can be very real, and they can be accessible. There was a researcher once who didn’t want to call it episodic memory; he wanted to call it repisodic memory, because you remember just what you repeat over and over in your head, whether it’s true or not. In many ways, that’s very much true of all our memories — they’re all reconstructed in some way. They might be more or less accurate, based on how much of the original information is intact.”

“That’s what’s so frustrating,” I say. “Little to none of my original information about my dad is intact.”

“You might not have very rich memories of your dad,” Gallo says, “but you also have to think of this: He had amazing memories of you. At one point in time, memories of your experiences together did exist — just maybe not in your brain in the same way as in his. Those memories existed in his mind, and even though he’s gone now, that doesn’t undo the fact that those memories were there.”

The insight falls over me like a magic spell. Suddenly I feel lighter, freed of some lifelong filial burden. The false memories I’d constructed of my dad? I realized they could finally be forgotten.