The water glass
The young busboy was embarrassed. He’d rushed over to fill the man’s water glass, only to find it already full. It kept happening, and he felt he owed the customer an explanation.
Every time he looked at the man’s glass, he said, he thought it was empty. And he wanted the man to know he wasn’t trying to work him for a tip.
The customer was Milton Erickson, and the confused busboy was part of an experiment he didn’t know he was in.
We know about it because Erickson described it in a letter to his friend André Weitzenhoffer, the hypnosis researcher who had helped develop the Stanford Hypnotic Susceptibility Scales. Written in Erickson’s characteristic purple ink, the letter was donated to the Erickson Foundation’s archives in 2005 by Weitzenhoffer’s widow, Mildred. The Foundation’s director, Jeffrey Zeig, published it in 2022, describing it as an account of the “field experiments” Erickson used to understand people and sharpen his perception.
Erickson called this one “the coffee cup routine”.
The coffee cup routine
He set it out for Weitzenhoffer as a set of instructions, and told him to try it himself.
When your coffee is placed in front of you, move it to a noticeably different spot. Then, every time you take a sip, pick up the cup in a very precise way and put it down in a rigidly precise way, always when the host, waitress or busboy is watching. Repeat it many times, so that they learn a precise pattern of your movements without knowing they have learned it.
Then, when the cup is empty, put it down with a clearly different movement. Unobtrusively move the saucer at least an inch from where it was, and at the same time, quietly shift your own position.
Erickson believed this created a “subliminal recognition”, as he put it, that something had changed, which led the staff to wonder whether the cup was empty.
He said he had tried it many times on hosts, hostesses, waiters, waitresses and busboys, and not only with an empty cup, but also with the cup three-quarters, half and a quarter full. And he didn’t just watch. When staff came over to refill his cup, he asked them what had made them think he wanted more coffee.
They couldn’t really say. Their answers were vague: “Things didn’t look right.” Things looked different. Something seemed wrong. One said she’d looked and tried to work out what she hadn’t done right.
The busboys were his favourites. Their duties were limited, he wrote, to butter pats and iced water, and so they limited their vision. He had seen busboys rush over to fill a glass that was already full, and stand there confused. One apologised for being absentminded, because he kept forgetting he had already filled Erickson’s glass.
The locked drawer
The coffee cup was just practice. The letter ends with something far more remarkable.
Erickson kept a drawer in his office, and only his secretary had the key. Whenever he noticed what he thought was a meaningful tiny clue, he wrote down what it told him, sealed it in an envelope, and gave it to his secretary to date and lock away.
Over the years, he could check his sealed predictions against what actually happened. He wrote that he became more and more expert, and that he also learned from his mistakes. It was, in effect, a private experiment he ran on himself for years, a way of testing his intuition instead of simply trusting it.
He drew one firm rule from it: never use these observations on friends or family, and never get emotionally involved.
He gave a striking example. He could tell when patients were pregnant, he wrote, sometimes before they knew it themselves. But with his own daughter-in-law, he had to be told before he could see it.
The great observer was blind exactly where his feelings were involved.
The man who couldn’t see colour
Erickson’s perceptiveness is all the more remarkable given what he couldn’t perceive.
A former patient, Walter Mills, recalled Erickson describing how, while teaching at Wayne State’s medical school, he had worked around being unable to see colour under the microscope by focusing on shapes instead. Mills also remembered Erickson saying he had no sense of rhythm. He could pick up the rhythm of music only by putting his hand on a piano while it was being played.
With some of the usual channels closed to him, Erickson trained himself to notice what most people overlook: the way someone sits, a slight change in their voice, a cup moved an inch. The coffee cup routine was one of the ways he practised.
The ability he was training turns out to be something all of us use, mostly without realising it. Sometimes it’s astonishingly sharp. Sometimes it fails spectacularly. A century of research has explored both sides.
The horse who read minds, or didn’t
At the turn of the twentieth century, a horse in Berlin became a sensation. Clever Hans, owned by a retired mathematics teacher named Wilhelm von Osten, seemed able to solve arithmetic problems, read a clock and identify coins, tapping out his answers with a hoof. When sceptics took von Osten away, Hans could still answer questions put to him by strangers.
In 1904, a thirteen-member commission, including a veterinarian, a circus manager, a cavalry officer and the director of the Berlin zoo, concluded that no trickery was involved.
Then the psychologist Oskar Pfungst investigated further, and solved the mystery. When Hans could see his questioner, he answered correctly 89 percent of the time. When he couldn’t, he managed just 6 percent. Hans could only answer when the person asking knew the answer.
Pfungst worked out what was happening. When a questioner set Hans a problem, they would involuntarily lean very slightly forward, and Hans would start tapping. When he reached the right number, they would make a tiny, involuntary upward movement of the head, and Hans would stop. The questioners had no idea they were giving the answers. Pfungst found he produced the cues himself, even when he tried to suppress them.
Hans wasn’t doing maths. He was doing what Erickson’s busboys were doing: reading a pattern in someone’s body that the person had no idea they were giving away. Today, the “Clever Hans effect” is one of the reasons scientific experiments are run “double-blind”, with neither participants nor researchers knowing who is in which group.
Missing the obvious
If people can pick up on such tiny cues, you’d expect them to notice a big change. Often, they don’t.
In 1998, the psychologists Daniel Simons and Daniel Levin ran one of the most startling experiments in modern psychology. A researcher stopped people on a university campus to ask for directions. Partway through the conversation, two workmen carried a large door between them. Hidden behind it, the researcher swapped places with a different person, who carried on the conversation.
Only around half of the people giving directions noticed that they were now talking to someone else.
It’s called change blindness, and it’s a humbling reminder that we see far less of the world than we think. Yet Erickson’s waiters seem to have done the opposite: they noticed a change too small to name. How can both be true?
Sensing without seeing
In 2004, the psychologist Ronald Rensink at the University of British Columbia tested something very like the waiters’ experience. He used two versions of the same photograph, identical except for one small detail, and showed them one after the other in the same spot on a screen: the first photo, then a brief blank grey screen, then the altered photo, then the grey screen again, then the first photo, over and over. Each image flashed up for only a fraction of a second. The brief grey screen was the key: it hid the flicker that would normally make the change jump out, so people had to search for the difference.
Viewers are told that the flickering display may contain a change, and their job is to find it. Viewers pressed one button when they felt something was changing, and another when they could actually see what it was. Some people, on some trials, felt the change a second or more before they could see it. Rensink called this “mindsight”, and argued it was a distinct way of perceiving: an awareness that something has changed, without yet seeing what.
Not everyone agreed. A year later, Daniel Simons and his colleagues replicated the finding, but offered a simpler explanation: those people had seen the change, but weren’t confident yet, and kept looking to make sure before saying so.
Either way, the feeling Rensink’s participants described is the same one Erickson’s waiters gave him: something isn’t right, but I can’t say what.
The body knows first
Another famous experiment suggests our bodies can sometimes pick up a pattern before our minds can explain it.
In 1997, Antoine Bechara, Antonio Damasio and their colleagues at the University of Iowa asked people to play a card game with four decks. Two of the decks were secretly rigged to lose money over time. Participants began choosing the better decks before they could say which strategy worked best. Even more strikingly, their skin began to show small stress responses when they reached for a risky deck, before they knew consciously that it was risky.
The researchers concluded that unconscious biases can guide behaviour before conscious knowledge does. But this finding has been challenged too. Other researchers, using more detailed questioning, found that participants could often describe the better strategy as early as they began using it. The question of exactly how far ahead of conscious knowledge the body gets is still debated.
Half a minute
We also read other people astonishingly fast.
In 1993, the psychologists Nalini Ambady and Robert Rosenthal showed strangers silent video clips, under thirty seconds long, of college teachers at work. The strangers’ judgements significantly predicted the students’ end-of-semester ratings of those teachers. Even six-second clips were strongly related to the final ratings.
Remarkably, the strangers’ ratings of specific small behaviours were less predictive than their overall impressions. Whatever they were picking up, it was the overall pattern, not any single detail they could point to.
That’s exactly the kind of information Erickson trained himself to catch, and then, crucially, to check.
The drawer, again
It’s tempting to see Erickson as a man with a mysterious gift. His letter suggests something more ordinary, and more impressive. He practised. He set himself small experiments in cafés and restaurants. He wrote down his hunches, sealed them, and had them dated and locked away, so he couldn’t fool himself afterwards about what he’d predicted. He learned from his errors, as well as his successes.
And he was honest about his limits. The man who could spot a pregnancy before the patient knew it couldn’t see it in his own family. Emotion, he concluded, clouds perception, which is why he told his students never to practise their observations on the people they loved.
The waitresses and busboys who refilled his cup never knew why they’d come over. Things just didn’t look right. Their eyes had noticed what their minds couldn’t name.
Erickson’s lesson was that this kind of noticing isn’t magic. It’s something we all do, all the time. With patience, practice and honesty about our mistakes, we can learn to do it better.
Further reading
The Erickson letter
- Zeig JK. “Unseen Erickson: From the Erickson Archives.” Letter from Milton H. Erickson to André Weitzenhoffer. The Milton H. Erickson Foundation Newsletter. Vol. 42, No. 3 (December 2022), p. 7. https://www.erickson-foundation.org/s/EricksonNews-42-3-WebFINAL2.pdf
- Mills W. Interview with Jeffrey Zeig, 1992. Excerpt from An Epic Life: Personal Perspectives. The Milton H. Erickson Foundation Newsletter. Vol. 42, No. 3 (December 2022), p. 14.
Clever Hans
- Pfungst O. Clever Hans (The Horse of Mr. von Osten): A Contribution to Experimental Animal and Human Psychology. Henry Holt, 1911 (German original 1907).
- Luebering JE. “What Clever Hans teaches us about being human.” Britannica. https://explore.britannica.com/explore/savingearth/what-clever-hans-teaches-us-about-being-human
- “Clever Hans.” Wikipedia. https://en.wikipedia.org/wiki/Clever_Hans
Change blindness
- Simons DJ, Levin DT. “Failure to detect changes to people during a real-world interaction.” Psychonomic Bulletin & Review. 1998;5(4):644–649.
- “Eyes wide open: why we don’t see what’s right in front of us.” APA Monitor on Psychology, October 2006. https://www.apa.org/monitor/oct06/eyes
Sensing without seeing
- Rensink RA. “Visual sensing without seeing.” Psychological Science. 2004;15(1):27–32. https://www.cs.ubc.ca/~rensink/flicker/abs.04.0a.html
- “Mapping the sixth sense.” UBC Reports, 8 January 2004. https://archive.news.ubc.ca/ubcreports/2004/04jan08/mindsight.html
- Simons DJ, Nevarez G, Boot WR. “Visual sensing is seeing: why ‘mindsight,’ in hindsight, is blind.” Psychological Science. 2005;16(7):520–524.
The Iowa Gambling Task
- Bechara A, Damasio H, Tranel D, Damasio AR. “Deciding advantageously before knowing the advantageous strategy.” Science. 1997;275(5304):1293–1295. https://doi.org/10.1126/science.275.5304.1293
- Maia TV, McClelland JL. “A reexamination of the evidence for the somatic marker hypothesis: what participants really know in the Iowa gambling task.” PNAS. 2004;101(45):16075–16080. https://pmc.ncbi.nlm.nih.gov/articles/PMC528759
Thin slices
- Ambady N, Rosenthal R. “Half a minute: predicting teacher evaluations from thin slices of nonverbal behavior and physical attractiveness.” Journal of Personality and Social Psychology. 1993;64(3):431–441.
Leave a Reply