Mia is a teenage girl from Mexico who can read her mother’s mind. Akhil is a young man in New Jersey who is seeing things through other people’s eyes. And Lily parleys brain-to-brain with friends who are far away from where she lives in Georgia. According to The Telepathy Tapes, a 10-part audio series that is now one of the most popular podcasts in America, Mia, Akhil, and Lily are nonspeaking people with autism who have a special skill: They’re savants for ESP.
Some nonspeaking children with autism end up learning to use language when they get a little older. Others may use sets of cards with pictures to communicate. But Mia, Akhil, and Lily, whose last names aren’t given in the podcast, are “spellers,” which means that they express themselves, with other people’s help, by pointing to a letterboard or typing on a screen. Proponents of this method say that it unlocks its users’ hidden gifts and proves their competence. An abiding faith in spelling has been expressed in documentary films, TV news reports, and best-selling books. But The Telepathy Tapes takes that faith one step further. Spelling is not just inspirational, the show suggests; it redefines the limits of perception and reality. 
However grandiose this mission, to call the podcast “scrappy” would be to oversell its polish: Hosted and created by the documentary filmmaker and self-described “science nerd” Ky Dickens, The Telepathy Tapes’ editing is choppy and its music sounds homemade. The scientific tests that it describes, which aim to prove that Mia and the others really do spell out their parents’ thoughts, are of dubious design. Yet the show has been a fixture since December on the Apple charts for audio, and was nominated for an iHeartMedia Podcast of the Year award; production just began for an associated documentary film. Dickens declined interview requests and did not answer emailed questions for this story, but last week she was a guest on The Joe Rogan Experience, where she spoke for two and a half hours about her podcast. “I’m sure you’ve opened up a lot of people’s minds to this,” Rogan said to her with admiration. “This could be a profound change in, just, society as a whole.”
But for people who may be less inclined than he to accept the podcast’s story at face value—those who have what Dickens and her subjects call a “materialist mindset”—another sort of message might be coming through. The Telepathy Tapes is more interested in showing off the extraordinary talents of its subjects than in engaging with skeptics. The show responds with moral outrage to any evidence that might cast doubt on ESP, or indeed on spelling itself, as if disbelief could only ever be a form of ableism. But in assuming a reality where reading minds is possible and astral planes can be traversed, it misses something more compelling: Humans really do possess a superpower for connection, with unpredictable effects.
Okay, so Mia is supposedly a telepath: Show her mother a number or a picture, and Mia will tell you what it is. On The Telepathy Tapes, Dickens tests this skill with the help of a psychiatrist, paranormal investigator, and anti-vaccine campaigner named Diane Hennacy Powell: “The crew, myself, and Dr. Powell were meticulous about making sure this test was bulletproof.” They set up at a rented house in Glendale, California. At Powell’s instruction, a wall mirror is removed and a TV screen is covered up, so that Mia, who is also wearing a “really intense and very dark” blindfold, can’t catch sight of any cue in their reflection. Before each trial of the number test, Powell opens up an app that generates random digits and taps it not just once but many times, to make sure that “it was super random.”
The podcast’s long, initial description of this testing process leaves out one important factor: how Mia’s spelling works. Video clips of the experiments (which are posted to the podcast’s website behind a $9.99 paywall) show that every time she reaches to the letterboard, her mother grasps her chin or puts a finger on her forehead, and also holds the board herself. The contact is nonstop. I’ve seen the same dynamic many times before in my own interviews with spellers and have learned that even gentle touch, or minor movements of the letterboard, can be a source of subtle and subconscious cues. In this way, Mia’s mother could be guiding Mia’s spelling and nudging it toward certain answers, whether they’re aware of it or not.
I first learned about this method of communication in 2014 while reporting for a story, published in The New York Times Magazine, on the case of Anna Stubblefield, then a philosopher of race and disability at Rutgers University, who had taught a student’s brother—a nonspeaking man diagnosed with severe physical and cognitive disabilities—to type into a keyboard as she held his hand. In the months that followed, the man’s inner life, and his prodigious talents, appeared to be suddenly let loose. He started taking college classes. He published a peer-reviewed, academic paper. He made plans to move into his own apartment.
The man also fell in love with Stubblefield, at least according to the typed-out messages. And she fell in love with him. Eventually they started having sex. His consent was given via spelling, which he could do only with her support (and that of certain others). A criminal trial followed, on the premise that the messages were fake—or rather, that Stubblefield herself had been their author, perhaps unwittingly. If that were true, it meant their love affair had been a story of her own creation—that she’d somehow managed to seduce herself, through the pressure she was placing on his hand. Stubblefield wound up pleading guilty to aggravated criminal sexual contact and serving close to two years in jail. (Nick August-Perna’s documentary film about this case, Tell Them You Love Me, was a Netflix hit last year.)
The form of communication that Stubblefield was teaching had been invented in Australia in the 1970s by a woman named Rosemary Crossley to help a nonspeaking girl with cerebral palsy who was languishing inside a state-run home. Crossley, who died in 2023 (and with whom I spoke for many hours), claimed to have discovered that the girl and others in the home could spell out words on a letterboard if their arms were being held a certain way. “I was acting as a responsive item of furniture,” Crossley wrote in a 1980 memoir, “not moving her arm but simply facilitating her own movement.” This method would be called “facilitated communication,” or FC.
By the early 1990s, FC had been imported to America and declared a miracle. News outlets covered its astonishing success, as kids who had been diagnosed with low IQs and never spoken in their lives appeared to have their minds unlocked by Crossley’s method. The movement had the spirit of an uprising. Every child whose inner life FC revealed was a challenge to the system that had failed them; their messages were taken as a source of liberation, whether from actual imprisonment in institutions or from the softer tyranny of meager expectations.
But from the very start, this story lent itself to other readings too. Children with autism made a spate of sexual-abuse allegations via keyboard that could not be corroborated, and seemed to be untrue. And certain aspects of FC just came off as fishy. Kids who had barely gotten any education were able, all at once, to communicate through text. Some went on to college. How could they have learned so much, so quickly? When similar doubts had been raised in Australia, Crossley offered some unlikely explanations—for instance, that the kids she’d worked with learned addition and subtraction by watching Sesame Street, as her book explains, then figured out multiplication and division, as well as how to work with fractions, on their own.
As FC spread throughout U.S. education, skeptics mobilized to run double-blind experiments and determine whether spelling really works. In one often-used test, researchers showed a picture to a speller while his or her supporter or facilitator wasn’t in the room; then they’d have the speller write out, with the facilitator’s help, what they’d seen. Few, if any, such experiments succeeded. Under these controlled conditions, the same kids and young adults who were writing poetry in school appeared unable to declare that they’d just seen a picture of a sandwich. This suggested that they were not actually—or not entirely—the authors of the words they’d been typing. Rather, their messages had been channeled from the people who were touching them, or holding up the letterboards, in the same way that people using a Ouija board can spell out messages without realizing they’re doing so.
As of 1994, it’s fair to say, the method was debunked. The American Psychological Association came out against its use—“studies have repeatedly demonstrated that facilitated communication is not a scientifically valid technique,” it said. The American Academy of Child & Adolescent Psychiatry and lots of other professional groups agreed. But FC never really went away. Instead, its users floated off into the fringe, where any criticism could be tossed into the sea. The message-passing tests had been unfair, adherents told me, because they made spellers angry or confused. Why should the spellers have to acquiesce to “scientific” inquiries, their families and facilitators asked, when they’ve already shown they have a voice?
Certain users of FC—it’s hard to know how many—really can communicate, to some extent, without another person’s touch. I’ve seen this for myself: I met a few typers who could spell out words completely on their own, though the messages they produced with others’ help had much greater fluency. But such cases appear to be exceptional. For most spellers—nearly all of them, perhaps—message-passing is a challenge even when they aren’t being made uncomfortable. At a conference held in Syracuse, New York, in 2014, I observed a friendly class on how to pass this sort of test. Every speller in the room was told to pull an object from a bag—a toy car, a crayon, a sock—while their facilitators looked away. Could they then type out what they’d seen? They could not. This mattered little to the parents at the conference, who told me that they simply didn’t worry much about the so-called science. You could puzzle over tables in someone’s published meta-analysis, one kid’s mom explained, or you could keep on talking to your son.
The Telepathy Tapes is well aware of all this history; it just rejects it. The scandals of the 1990s arose from the errant work of “untrained, or barely trained facilitators,” the podcast says, rather than from FC itself. Since then, the old mechanical keyboards have mostly been swapped out for iPad screens, plastic boards, and letter stencils, as part of newer versions of the method such as Rapid Prompting and Spelling to Communicate. Whatever it might now be called, FC remains as unreliable as ever, according to a 2018 statement from the American Speech-Language-Hearing Association. But Dickens claims the opposite, and argues that the research purporting to discredit spelling has fallen out of date. “The accusation that most spellers are not actually communicating, that someone is pushing their hand around or moving around the letterboard, is unabashedly, unequivocally false,” she says on the show. As one parent tells her: “The thing is, Ky, we can’t all be lying.”
Indeed, the podcast goes so far as to invert the question that skeptics started asking in the ’90s: whether spellers were capable of sharing information that their facilitators didn’t know. The Telepathy Tapes asks instead whether spellers might have access to information that only their facilitators know. Or put another way: Can they read their facilitators’ minds?
If anything could prove that facilitators are at risk of substituting their own ideas for those of the people they intend to help, this should be it. And yet Dickens and her subjects have drawn the opposite conclusion. The parents seem delighted by their children’s ability to read their minds, not suspicious of it. Some do end up feeling a bit unnerved, Dickens says, by the fact that they can no longer hide birthday presents from their kid, or protect their private thoughts. But instead of taking that supernatural knowledge as a hint that spelling is unsound, they conclude that ESP is real.
The faction of facilitators who have landed on the same belief dates back to the 1990s too. One important, early advocate for FC suggested that users might have access to “vestigial psychic abilities.” Arthur Golden, who appears on the podcast and has said that he began communicating telepathically with his nonspeaking, autistic son in April 1994, told me that many who were getting trained in the method at that time reported having similar experiences. But Crossley and the other prominent advocates for FC shut down this sort of talk, perhaps because they worried it could only undermine their method’s credibility. To this day, many in the spelling world rebuff supernatural claims: On the podcast, parents say that they’ve been “shushed” in online forums, or “shut down” from sharing their experiences at conferences. On Rogan’s show, Dickens called it all “a big cover-up.”
Soma Mukhopadhyay, who created the Rapid Prompting method, told me that a perception of “telepathy” may arise from the fact that facilitators will at times predict what a speller wants to say, just like your cellphone does while you’re typing in a text message. Mukhopadhyay said that attributing this effect to psychic powers won’t help autistic children in the long run, but she shied away from telling people what to do or think. “I wouldn’t ask the parents not to believe in telepathy,” she said. “Everyone is a grown-up. Everyone can do what they want.” Diane Hennacy Powell, the psychiatrist who consulted on the podcast’s experiments and who has published a book about ESP, was more direct. “I am more conservative than Ky about the scientific conclusions and don’t believe telepathy has been sufficiently proven,” she told me via email. “She is a filmmaker and not a scientist, so it’s important to acknowledge our different perspectives.” (Powell also said she could not discuss the matter in more detail because of a nondisclosure agreement.)
The telepathic spellers, then, are on the fringe’s fringe, shunned by both spelling’s critics and its true believers. And as The Telepathy Tapes moves along, Dickens and her subjects’ claims become even further unmoored from any dock posts of common sense. Spellers whom Dickens first describes as merely having ESP are soon credited with even more amazing powers and “uncanny knowings,” such as detecting cancer and finding ways to cure it. By the end of the series, Dickens—who repeatedly declares herself a “skeptic”—has explained that spellers may project themselves into alternative dimensions; that they can speak with angels; that they visit friends in dreams and prophesy disaster. “Could the government utilize these individuals?” she said to Rogan last week. “It’s a fair worry.”
Any suspicions one might have about spelling’s authenticity are submerged into the same phantasmagoric narrative. What explains these kids’ stunning breadth of knowledge? Naturally, they did their coursework in the great beyond. One of the podcast’s telepaths is said to have “instantly learned everything there was to know about The Great Gatsby” during a nightly visit to a “school in Heaven.” And Akhil, the young man in New Jersey, is said to be sharing messages from long-dead relatives—“a very common gift” among the “spiritual savant” spellers, according to the podcast.
Someone else might take the latter as another sign that spelling works by leveraging a Ouija-board effect—given that here it’s being used, just like a Ouija board, to communicate with ghosts. But Akhil does appear to be doing something unique among the spellers whose tests have been filmed and posted on the podcast’s website. After many years of practice spelling with his mother, he now appears to do so independently, typing on an iPad keyboard without ever being touched at all, and with no one’s hand on his device.
But a closer look suggests that his mother is still guiding him, even though she may not realize it. She does not touch him, but she is highly active all the same: She issues words of encouragement and makes other sounds; she flutters her hands in concert with his typing; and she adjusts her posture while he’s reaching for the screen, as if her body were a joystick. One of the video clips shows an experiment in which he spells out mariposa. You can see his mother in the background, leaning to the right as Akhil’s finger moves in that direction to type out i–p–o; and then moving leftward, as his finger does the same, for s–a.
If such cueing, rather than the spirit realm, is the source of Akhil’s messages, then one must still acknowledge that he has extraordinary talents. For this mode of spelling to “work” as well as it does, Akhil would need to possess an exquisite sensitivity to his mother’s subtle cues; he’d have to be so attuned to her every gesture and expression, that even chirps and leans and flutters could serve as radar signals, directing him to specific letters on the screen. In other words, he’d need to have special access to his mother’s mind.
This is not telepathy, but it is connection—a connection so intense that I don’t think it would be far off to call it love. A different, reckless version of this bond was at the center of the Stubblefield trial. When my story on that case was published, I quickly came to understand that readers were responding to the romance, real or not, that formed its core. A flock of artists’ emails quickly landed in my inbox, pitching plans for heartfelt adaptations: movies, operas, poems. The story has “everything we are looking for,” one French film producer said: “love, desire, manipulation and mystery.”
“We have a deep drive to connect and know each other,” Dickens says at one point in her series. But the podcast shows this instinct is a complicated gift. Humans are virtuosic at creating bonds; we’re savants for filling space with meaning. But we’re also zealous at defending them. Pointing at a letterboard, aided by a gentle touch, can seem to open up a window on another person’s mind. That effort at communion is a holy, humanistic act. But its magic can be isolating, too, by conjuring relationships that don’t exist, and marooning us within a universe where anything is real.
