The Inkblots

by Damion Searls

  An inkblot Rorschach made for his work with Konrad Gehring (1911/1912), with interpretations marked on the page (probably responses from Gehring’s students recorded by Gehring). Left side: “Balkan peninsula” (white space, upside-down caption) surrounded by “Adriatic,” “Aegean,” and “Black Seas,” “tongs,” “horse’s head,” “Island of Rügen.” Right side: “cartoon of a dog,” “boy riding hobby horse,” “mouse,” “glove.”

  Supplemental color test: see this page.

  Rorschach’s painting of his apartment, 1918. The new baby, Lisa, is playing with her toys, including wooden animals Hermann made for her; some of the pictures visible through the door are hung low on the wall for the baby to look at.

  Rorschach’s depiction of life with baby Lisa. Top left: “The thing that always works”; top right: “Going out for a trip”; lower right: Rorschach seems to be taking notes on the baby’s response to a symmetrical puppet, also visible hanging over the bed in top center.

  Painting made for Lisa’s first birthday

  From Manfred Bleuler’s 1935 article on Moroccan Rorschachs. Top: Common European interpretation of Card III: two waiters bowing. Bottom: Common Moroccan interpretation: unconnected parts of skeletons, a cemetery. Credit 21

  Roemer images, ca. 1919, published 1966. More artsy than Rorschach’s, but their artistry was precisely the problem: “Roemer tries hard and does his best, but he likes very imaginative images, rich in fantastic figures, which spoils everything” (letter to Emil Oberholzer, January 14, 1921).

  Jackson Pollock, Autumn Rhythm (Number 30), 1950, which looks nothing like Rorschach’s inkblots but accomplishes a similar feat: prompting the viewer to see and feel movement. Credit 22

  Rorschach’s program for the Carnival celebration at Münsterlingen (1911), including a portrait of himself in a wizard’s costume holding calipers in place of a magic wand. The elegant lady next to him, holding the handbag, is Olga, and in the top left corner is Paul Sokolov, the other assistant doctor at Münsterlingen.

  Rorschach in his apartment, dressed in wig and wizard’s garb and holding calipers

  Today, even as the inkblot test rests on a firmer scientific footing than ever before, both as diagnostic tool and as therapeutic method, it is being given less often. Usage has tumbled from its sixties peak of an estimated million times a year in America to a fraction of that—no more than a tenth, perhaps a twentieth. The Rorschach had stayed the most used personality test in the United States for decades until the emergence of the MMPI, and then was second except for a dip in the eighties. Not anymore.

  Chris Piotrowski, a psychologist who has been tracking Rorschach usage for decades, estimated in 2015 that the Rorschach ranked ninth, perhaps lower, among personality tests used by assessment psychologists. It was behind several self-report tests (the MMPI; the Millon Clinical Multiaxial Inventory, or MCMI; and the Personality Assessment Inventory), short checklists (such as Symptom Checklist-90, the Beck Anxiety Inventory, and the Beck Depression Inventory), structured interview scripts targeting specific psychiatric diagnoses, and other quicker projective methods such as Human Figure Drawing and Sentence Completion. Anecdotal evidence suggests a gradual decline in usage, rather than a bombshell effect from What’s Wrong with the Rorschach?, but there are no studies that reveal exactly when and why the shift happened and whether the introduction of R-PAS in 2011 and Mihura’s article from 2013 have hastened, slowed, or reversed the trend.

  Wood’s book seems to be a plausible cause of the decline, but it is hard to gauge its real impact. Most psychologists and evaluators went right ahead doing what they were already doing. Those who disliked the Rorschach welcomed the takedown; those who knew and used the test largely dismissed the book, or used its criticisms to prompt narrow but real improvements. It is also impossible to disentangle Wood from wider dynamics in the field. The Rorschach had become, after Freud, a symbol for everything people didn’t like about psychotherapy: too much unprovable inference, too much room for bias, not enough hard science. Many critics of the Rorschach were also critics of Freud, making the same kinds of argument against both. And so Rorschach researchers had to defend what they were doing far more than other assessment psychologists, even though most of the same problems affected other kinds of testing as well. Many chose to pick other battles.

  In the popular media, at least, skepticism dominates. Whenever Scientific American or Slate has cause to mention the actual Rorschach test and quote an expert, that expert has almost always been one of the coauthors of What’s Wrong with the Rorschach, who invariably says that the test has been scientifically debunked but is still in use. The criticisms brought up are those that were leveled at Exner’s system in the early 2000s, and no one mentions any of the developments since.

  Information about how often the test is taught, rather than used, is more mixed. Whether due to skepticism or to wider shifts in the field such as increased specialization, accredited graduate schools and internships have reduced their emphasis on projective, or “performance-based,” techniques. The Rorschach was not among the top ten most covered tests in a 2011 survey of clinical psychology programs; Piotrowski called the decline “precipitous,” concluding that the Rorschach would soon “become non-existent in clinical psychology training in the USA.” A more recent study suggests that this prediction was too stark: while course coverage of the Rorschach dropped from 81 percent of programs in 1997 to 42 percent in 2011, it bounced back to 61 percent in 2015 (or perhaps the 2011 figure was too low). And almost all “practitioner-focused” programs, as opposed to “research-focused” programs, continue to teach the Rorschach, even though such training has declined in graduate schools overall.

  Then there is the quality of Rorschach instruction. The APA requires clinical psychologists to be competent in psychological assessment, but doesn’t say what that means: Students used to take five semesters on personality assessment but are now likely to have a one-semester course on theories of personality, which also covers how to establish rapport in testing situations and a wide range of specific tests. In 2015, the whole Rorschach test, history and theory and practice, Exner or R-PAS or both, might get two three-hour class sessions.

  Eugen Bleuler worked to bring Freud’s costly methods to the people who needed them most—the poor, the hospitalized, the psychotic. Rorschach, too, aspired to create a method that could be used with everybody. But the wider forces of inequality and specialization seem to be working against that vision. Assessment and psychotherapy in general are becoming more like pay-out-of-pocket counseling or coaching: exploratory and improvisational, with less of an emphasis on a specific diagnosis. The very ethos of assessment—trying to get a view of the whole person—seems not to fit into the managed-care system that we still have today. Maybe the technocratic Rorschach test will simply be unable to compete in the marketplace, and the more exploratory Rorschach will go the way of Freudian analysis and other open-ended client services, a luxury for those who can afford it. This more artisanal approach is likely to last as long as people want to know more about themselves.

  “Even for sympathizers like me,” in the words of Chris Hopwood, a young psychologist active in the assessment community, the Rorschach “is sort of like vinyl: you only use it if you really want the music to be good.” If the Rorschach were merely one nuanced but inefficient test in a battery of assessments, this would be the end of the story.

  —

  The decline of the Rorschach in clinical psychology should not be exaggerated: R-PAS is gaining ground, and a fraction of a million times a year is still a lot. The inkblots are used as a test all over the world, sometimes to assign a diagnosis, sometimes to less measurably change how a therapist understands a client. If a woman comes to see a psychologist for help with an eating disorder and then has a high Suicide Index score on the Rorschach, her psychologist might talk to her differently: “There are ways that you have your world organized that are a lot like people who go on to kill thems
elves. Should we talk about that?”

  Examples like this will seem suspect to psychologists or laymen who think the Rorschach finds something crazy in everyone. But the test is also used to find mental health. Recently, at one state psychiatry facility in the criminal justice system that houses people declared to be Not Guilty by Reason of Insanity or Incompetent to Stand Trial, a violent man had been undergoing extensive treatment. (This story has to be kept vague for confidentiality.) The treatment seemed to have worked—the man’s psychotic symptoms were gone; to all appearances, he was no longer a danger to himself or others—but the team of doctors on his case was divided over whether he had really improved or was faking health to get out of the facility. So they gave him a Rorschach test, which turned up no sign of thought disorders. The test was trusted enough as a reliable and sensitive indicator of such problems that the negative finding convinced the team, and the man was released.

  The Rorschach also continues to be used in a research context. It is often hard to distinguish between Alzheimer’s-type dementia and other effects of age and mental illness—could the inkblots tell them apart? In a 2015 conference, a Finnish scientist presented his analysis of Rorschach tests given to sixty patients in a Paris geriatrics unit, ages fifty-one to ninety-three (average age: seventy-nine). Twenty of the patients had mild or moderate Alzheimer’s and forty had a range of other mood disorders, anxiety, psychosis, and neurological problems. The test found many common elements between the two groups but also a range of distinguishing features. Half a dozen Rorschach scores showed that Alzheimer’s patients were less psychologically resourceful, with less cognitive sophistication, creativity, empathy, and problem-solving ability; they distorted information and did not integrate ideas and perceptions. Most intriguingly, despite putting a normal amount of effort into processing complex and emotional stimuli, Alzheimer’s patients gave fewer Human responses—a kind of content response still generally accepted as an indication of interest in other people. The Alzheimer’s patients, more than their peers, had checked out of the social world. This finding was new in Alzheimer’s research, with implications for treatment and care.

  Outside of clinical psychology, the fact that there is so much data about how the inkblots are perceived makes them useful in a range of applications. In 2008, when a team of Japanese neuroscientists wanted to study what happens when people see things in original ways, they needed recognized, standardized criteria for whether something a person sees is common, uncommon, or unique. So they took what they called “ten ambiguous figures that have been used in previous studies” and projected them inside an MRI tube equipped with a voice scanner, tracking brain activity in real time as subjects gave typical or atypical answers to the inkblots.

  The study demonstrated that seeing something in a standard way uses more instinctive, precognitive brain regions, while original vision, requiring a more creative integration of perception and emotion, uses other parts of the brain. As the Japanese scientists pointed out, Rorschachers had long argued precisely that Original responses “are produced from the interference of emotion or personal psychological conflicts…on perceptual activities.” The MRI study confirmed Rorschach tradition, just as the inkblots had made the MRI experiment possible.

  Another conclusion from this research was that people who see forms less well have larger amygdalas, a sign that this region of the brain, which processes emotions, has been activated more often. “This suggests that emotional activation greatly influences the extent to which one distorts reality,” just as Rorschach had posited a century ago with his correlation of Color and poor Form (F–) responses.

  Other recent studies of perception have used new technologies to investigate the test-taking process itself. Since typical test takers give two or three responses per card on average but can give nine or ten when asked, a team of research psychologists at the University of Detroit argued in 2012 that people must be filtering or censoring their responses. Maybe getting around this censorship would make a performance-based test more revealing. If only there were an involuntary reaction to an image, or at least a reaction “relatively more difficult to censor.” There was: our eye movements as we scan an inkblot before we speak.

  So, building on eye-movement Rorschach studies going back to 1948, the researchers put a head-mounted EyeLink tracker on thirteen students, showed them the inkblots, and asked, “What might this be?,” then showed each blot again and asked, “What else might this be?” They quantified and analyzed the number of times each subject stopped and looked at one place on the image, how long they looked, how long it took to disengage from the whole image and start looking around, and how far the gaze jumped. They drew general conclusions, too, such as that we hold our gaze longer during second viewings, since reinterpreting an image is an “attempt to acquire conceptually difficult information.” This is paying attention to how we see, not what we say, with a vengeance. Eye movements will never reveal as much about the mind as what we see in the inkblots, but researchers are exploring what they do reveal about how we see—and returning to Rorschach’s original vision of the test as a way to understand perception.

  Not what you see, but how: Eye movements while looking at Card I. The blot is shown blurred by the researchers so as not to reveal the actual image. Lines are scan paths, and circles are pauses or fixations; this subject paid most attention to the central part of the blot. Credit 19

  The most fundamental question about the test that Rorschach left unanswered at his death was how these ten cards could produce such rich responses in the first place. The mainstream trend in psychology, from Beck to the content analysts to both Exner and his critics, has been to leave this question of theoretical underpinning aside. Empiricists thought of the test as eliciting responses and spent decades fine-tuning how those responses should be tabulated. For Rorschach—and for only a few who came later—the inkblots elicited something deeper. Ernest Schachtel had argued that test results are not words that are spoken but ways of seeing. “It has to be emphasized that this is a test of formal things,” Rorschach wrote in 1921, “how a person perceives and absorbs.”

  Today, we know more about the science and psychology of perception than ever before. As the test breaks free of the clinical psychology culture wars, it may at last be possible to integrate the inkblots into a full-fledged theory of perception, as Rorschach so wanted, or at least to sketch out what it is about the nature of seeing that gives the inkblots their power.

  Look closely at this picture. There will be a quiz.

  Imagine you are given all the time you want to study this picture, and then it’s taken away and you are led to a dark room. Now imagine two different scenarios: In one, with your eyes closed, you have to answer a simple perceptual question about the image—Is the tree wider than it is tall? In the other scenario, you have to answer the same question, but your eyes are open and the picture itself is faintly shown on a screen so that you can look at it while you’re being asked the question.

  This experiment was tried on twenty people, with various images and analogous questions. The subjects’ brain activity was measured during each scenario—that dark room was an MRI scanner. It turned out that the overlap in brain activity between the two scenarios was 92 percent, suggesting that almost all of what your brain does when you see something is the same—or is at least in the same area of the brain—as what it does when you visualize something. The retina actually picking up light or not is only 8 percent of what happens, so to speak. Perception is a mainly psychological, not physical, process.

  When you look at something, you are directing your attention to parts of the visual field and ignoring others. You see the book in your hand or the baseball hurtling toward you, and choose to disregard all the other information that’s reaching your eye: the color of your desk, the shapes of clouds in the sky. You are constantly cross-checking what’s out there against objects and ideas you recognize and remember. Information and instructions are traveling along nerves
from the eye to the brain, and also from the brain to the eye. In another experiment, Stephen Kosslyn, coauthor of the tree-visualization study and one of today’s leading researchers into visual perception, monitored this two-way neural activity moving “upstream” and “downstream” during an act of seeing, and found that the ratio is 50-50. To see is to act as much as react, put out as much as take in.

  Even what seem like completely straightforward optical tasks turn out not to be merely passive or mechanical. Our eyes may register wavelengths, but a lump of charcoal looks equally black whether it is in the bottom of a bag or out in the summer sun at your barbeque—the light it reflects is different, but we see it as black because we recognize it as black. In the same way, a sheet of white paper looks white regardless of the lighting in the room. Painters have to unlearn this way of seeing, so they can paint “black” or “white” things with different colors. As Kenya Hara, a Japanese designer, writes in a beautiful book called White: “Things like the rich golden yellow of the yolk from a broken egg, or the color of tea brimming in a teacup, are not merely colors; rather they are perceived at a deeper level through their texture and their taste, attributes inherent in their material nature….In this regard, color is not understood through our visual sense alone, but through all our senses.” In other words, the most perfect Yolk Yellow swatch in the biggest sample book in the world isn’t pooled in its soft-boiled shell, or shimmering on a clear layer just hardening in a frying pan to the smell of olive oil starting to heat up, so it can’t be the yolk-yellow color we actually see. Colors exist in connection with colored things that awaken our memories and desires. No objective system—no Pantone chart, no color wheel, no grid of pixels of “every” color—can truly represent any color. Even seeing a color is an act of the self, not just of the eye.