Is tarot real? What research says about tarot readings and why they feel accurate

Tarot is a deck of 78 cards in two parts (Sosteric, 2014). The short answer is that very few studies have tested tarot readings under controlled conditions (Ivtzan, 2007), and the experiments found for this guide did not show that readings describe people accurately. No study that tested whether tarot predicts the future was found. Readings can still feel accurate for psychological reasons, such as the skill of refining a reading from the client’s reactions and statements that fit almost anyone (from the abstract; Dutton, 1988). Why a reading that fits everyone can feel personal is explained in the guide to the Barnum effect.

Many people consult tarot for fun or as a way to sort out a worry. In a report on 9,593 US adults, 30% said they consult astrology, tarot cards or a fortune teller at least once a year, and most said they do so just for fun (Pew Research Center, 2025). This guide sets out where tarot comes from, what tests of it found, why readings feel accurate, how to use it lightly and when to be careful. It is not meant to criticise anyone who enjoys tarot. mind.io offers no tarot readings.

Playing cards came into Europe from the Islamic world in the second half of the 14th century (Dummett, 1993). Tarot was invented at the royal courts of 15th-century Italy (Sosteric, 2014). Sometime in the 1400s, a Metropolitan Museum of Art curator wrote, someone added 21 trump cards. The Visconti-Sforza cards, dated to between 1450 and 1480, are some of the oldest tarot cards in existence. According to the Morgan Library & Museum in New York, tarot in 15th-century Italy was simply a “courtly game of skill” (Smithsonian Magazine, 2026).

The card-game historian Michael Dummett summed up that the tarot pack was not associated with fortune-telling or the occult until the late 18th century (Dummett, 1993). According to a sociology paper citing Dummett’s work, two essays in Court de Gébelin’s nine-volume Le Monde primitif, published in 1781, set off the idea of tarot as a key to life’s mysteries, and cartomancers such as Etteilla took it up (Sosteric, 2014). A history paper adds that fortune-telling with cards became increasingly popular in France in the late 18th century (Pooley, 2023). The Dummett and Pooley papers were read as abstracts.

The story that tarot came from ancient Egypt was put forward by some writers in the late 18th century, and the Victoria & Albert Museum in London regards it as unsubstantiated. By the late 19th century tarot had become firmly associated with the occult and fortune-telling, and the Rider-Waite-Smith deck, conceived by the British mystic Arthur Edward Waite and illustrated by the artist Pamela Colman Smith, was first published in 1909 (Smithsonian Magazine, 2026). The sociology paper, citing Dummett and others, likewise sums up that occultists’ claims about tarot’s significance have no historical basis (Sosteric, 2014).

Few studies have tested tarot readings under controlled conditions. A 2007 review in a parapsychology journal itself says that tarot experiments are few (Ivtzan, 2007). In a 1983 experiment it describes, 10 students on a parapsychology course who were interested in tarot each received a reading made without face-to-face contact with the reader, and they rated their own reading 4.1 on average and the others’ readings 3.2, a statistically significant difference. But the researcher criticised that experiment because the participants knew each other, and in two repeats own and others’ readings scored 3.3 against 3.2 and 3.4 against 3.3, with no significant difference. The researcher concluded that participants were not able to pick out their own description (Blackmore, 1983, as cited in Ivtzan, 2007).

In the same 1983 experiment, participants gave the ordinary face-to-face readings much higher scores than the test readings given without contact (5.7 against 4.1), although they rated the ordinary readings straight afterwards and the test readings a week later, in writing (Blackmore, 1983, as cited in Ivtzan, 2007). Why a reading given face to face feels more accurate connects with the account of cold reading below.

In a 2004 study comparing real readings, in which 30 participants chose their own cards, with control readings, in which the cards were chosen at random, believers in the paranormal gave both readings higher ratings than non-believers did, but neither group rated the real and control readings significantly differently (Ivtzan & French, 2004, as cited in Hofer, 2009). In a study that randomly gave students either a true fortune or a false one made by swapping the meanings traditionally linked to the suits, belief in fortune telling, astrology and clairvoyance went with higher perceived accuracy of the fortune received (Hughes et al., 2001). That paper was read as an abstract, which does not say whether true and false fortunes were rated differently. The abstract also does not say whether tarot cards were used.

A study that compared birth-date “tarot personality” categories with the Big Five scores of 494 psychology students concluded that none of the five traits differed significantly between the categories (Santosa, 2012). It tested a birth-date system, not readings with drawn cards. The experiments that tested readings directly were small, and no tarot study that tested predictions of the future was found.

There are related tests that did not use tarot. A study that tested several professional mediums with a method designed to avoid the problems of past research did not support the existence of genuine mediumistic ability (O’Keeffe & Wiseman, 2005). Researchers who support a power to sense the future reported a very small effect (g = 0.09) in a meta-analysis of 90 experiments (Bem et al., 2015), but a large replication designed together by supporters and sceptics found that 2,115 participants making 37,836 guesses scored 49.89%, the same as the 50% chance level (Kekecs et al., 2023). The first two papers were read as abstracts. The blind tests in which astrologers had to match birth charts are covered in the zodiac evidence guide.

The American psychologist Ray Hyman described “cold reading”, which makes a stranger feel that the reader knows all about them, and wrote that props such as a crystal ball, tarot cards or palm reading lend an air of novelty and give the reader time to work out the next statement. Hyman also wrote that people come to a character reader because they need someone to listen to their conflicts over love, money and health (Hyman, 1977). Another researcher describes cold reading as making calculated guesses about a client’s background and problems and elaborating a reading from the reactions, until it seems so uniquely fitting that it appears paranormal, and says the process depends initially on the Barnum effect (Dutton, 1988). That paper was read as an abstract.

The listener fits a plausible statement to their own situation and may later recall specifics the reader never said. Hyman held that this works not because people are gullible but because of the ordinary processes we all use to make sense of what we hear (Hyman, 1977). It has also been suggested that readers themselves may use such clues without being aware of it (Dutton, 1988; Ivtzan, 2007). Hyman’s account of clients being satisfied even when palms were read the opposite way is in the palm reading evidence guide.

Confirmation bias, seeking and interpreting evidence in ways that fit what we already expect, also plays a part (Nickerson, 1998). A “love reading that came true” stays in memory, for example, while the misses are easily let go. Illusions of causality, believing that two unrelated events are causally connected, can happen to anyone, like optical illusions (Matute et al., 2015). These two papers were read as abstracts, and neither is about tarot. One analysis adds that divinatory interpretations are often vague enough to be hard to prove wrong, and that successes are reported more often than failures (Hong, 2025).

A reading can also change what people do and so affect what happens. Self-fulfilling prophecies, in which expectations change behaviour and so outcomes, do occur, but in a review of 35 years of research on teachers’ expectations their effects were typically small (Jussim & Harber, 2005). That review was read as an abstract. In an experiment that did not use tarot but showed people randomly assigned “fortunes” said to be based on details such as birth dates, those who got a positive fortune took more financial risks, an effect the authors report mainly among men. The results come from two online experiments (252 and 441 people) and a real online gambling game in a lab (193 people), where, the authors note, the saliency of the payoffs was relatively low (Tan et al., 2022).

In the same US survey, asked about tarot on its own, 11% of adults said they consult tarot cards at least once a year. Of all adults, 20% said they turn to astrology, tarot or fortune tellers mostly “just for fun” and 10% mostly because they believe these practices give them helpful insights. Only those who used them mostly for insight were asked how much they rely on what they learn when making major life decisions: 1% of all adults said they rely a lot and 5% a little (Pew Research Center, 2025).

A Korean survey did not ask about tarot separately. When 1,507 adults aged 19 or over were asked whether they believe in fortune-telling, saju, face reading, name divination and the like, 40% said they did (3% “mostly”, 37% “somewhat”), and 40% had paid at least once to have their fortune or saju read. Of the 607 who had paid, 59% said what they were told matched reality and 71% said they had taken a reader’s words into account in a decision, but the pollster added that these results alone cannot show a causal link between trust in fortune-telling and its matching reality. 26% had used an artificial intelligence (AI) service for a saju reading or a fortune (Gallup Korea, 2026). In an online survey of 1,007 French adults, 23% said they believe in cartomancy (card reading), and 10% had at some point consulted a specialist or publications such as articles, books or videos on it (Ifop, 2020).

In a master’s thesis based on interviews with four people who use tarot regularly and in a self-reflective way, the participants said they use tarot to gain insight into their current situation and possible plans of action. They used the cards most often in difficult times, when the cards offered comfort: confirmation that things were OK and that life was not without order (Hofer, 2009). These are four people’s accounts, not evidence that tarot helps. Other studies have looked at uncertainty and stress. A study of 174 Israeli citizens during the Gulf War found magical thinking more often among people living in areas exposed to missile attacks (Keinan, 1994). In a study of 40 students in introductory courses on astrology, psychology and German, interest in astrology went with a larger number of personal crises (Lillqvist & Lindeman, 1998). Both studies were read as abstracts, neither is about tarot, and as a comparison of areas and a correlation they cannot show that stress causes such beliefs.

Some people now use generative AI to interpret tarot. In a study that interviewed 12 tarot readers who use AI in their own divination practice, the readers used AI to navigate uncertainty and self-doubt and to explore other perspectives. The researchers started from the premise that no causal relationship exists between the question and the cards drawn, and noted that AI chatbots are known to be sycophantic, confirming what users already know or the interpretation they already have. They also named the small sample of only 12 participants as a limitation (Prock et al., 2026). No study testing how accurate free or AI tarot is was found.

If you enjoy tarot, here are some ways to use it. They are this guide’s suggestions, not methods tested by research. Use a reading as a question that helps you sort out your thoughts rather than as an answer, and make the decision yourself. Set how much money and how many readings you will spend in advance, and do not repeat the same question until you get an answer you like. Note down the misses as well, so that only the hits do not stay in memory. If you want to look at your personality as scores from your own answers, you can try the Big Five personality test.

A fortune-telling site that seemed free can keep taking money. According to Japan’s National Consumer Affairs Center, consumer centres across the country received more than 1,000 consultations a year about fortune-telling sites and apps: 1,490 in fiscal 2016, 1,406 in 2017, 1,430 in 2018 and 1,837 in 2019, and 1,016 in fiscal 2020 up to 31 October. In the problems it describes, people who signed up thinking the service was free were told by someone calling themselves a fortune teller or reader that they had wonderful luck with money or would meet a good partner, exchanged messages many times, saw no improvement in their money or love luck, and paid large sums. The centre advises not entering personal details such as your name, date of birth or e-mail address just because a reading is free, and not continuing the exchange on the reader’s word (National Consumer Affairs Center of Japan, 2020). These figures are consultations, not confirmed cases of loss.

Rules elsewhere point the same way. According to advice from the body that writes the UK’s advertising rules, claims that psychic or spiritual readings are truthful or accurate are likely to be considered problematic by the advertising regulator, and marketers should not make promises they cannot keep, such as breaking curses or improving health, wealth or other circumstances (CAP, 2026). The same body’s guidance says readings should not be presented as personalised when the same or a substantially similar report is sent to everyone, and that claims about healing must make clear that they refer to spiritual, not physical, healing. It also notes that many consumers enjoy seeking such advice (CAP, 2016). In France, answering a written question in parliament, the government stated that fortune-telling is a legal activity but is subject to several rules of consumer and criminal law, and that the consumer protection authority (DGCCRF) can check and sanction false advertising and dubious commercial practices (Assemblée nationale, 2023). No Korean official warning dealing specifically with fortune-telling services was found.

Readings can become hard to stop. A case report of one patient who kept seeking fortune-telling services described it as a type of behavioural addiction (Skryabin, 2020), and a review of increased astrology use summarised case studies that conceptualised this as a possible behavioural addiction with symptoms such as distress and cravings, while saying that further research is necessary (Das et al., 2022). Both papers were read as abstracts, and this is not a condition with established diagnostic criteria. Japan’s National Consumer Affairs Center also lists consultations from people who wanted to stop but were held back by the reader (National Consumer Affairs Center of Japan, 2020).

If you keep asking the same question again, if the cost of readings weighs on you, or if you find it hard to decide anything without one, pause and talk with someone you trust or a professional. If your mood has been low for more than two weeks or worry is big enough to get in the way of daily life, look first at the depression check-in (PHQ-9) or the anxiety check-in (GAD-7) rather than the cards, and talk with a doctor or counsellor. A tarot reading does not replace medical care or counselling. No study measuring whether readings make people put off health problems was found.

Myth 1: tarot reflects the unconscious. This is tarot literature’s own account. The 2007 review explains that, in this view, the deepest understandings of the moment exist in the individual’s unconscious, and the cards chosen in a reading are not random but archetypes that express that inner knowledge (Ivtzan, 2007). But when the experiment testing whether people could tell their own reading from others’ was repeated, no significant difference appeared (Blackmore, 1983, as cited in Ivtzan, 2007), and readings with self-chosen cards were not rated significantly differently from readings with cards chosen at random (Ivtzan & French, 2004, as cited in Hofer, 2009).

Myth 2: tarot feels accurate because of the cards’ power. In one experiment, face-to-face readings got much higher scores than test readings received in writing a week later, 5.7 against 4.1 (Blackmore, 1983, as cited in Ivtzan, 2007). The feeling of accuracy can also come from cold reading, which refines a reading from the client’s reactions (Hyman, 1977), and from statements that fit almost anyone, and it has been suggested that readers themselves may use these clues without being aware of it (Dutton, 1988; Ivtzan, 2007).

Myth 3: free tarot and AI tarot are accurate too. No study testing the accuracy of free or AI tarot was found. Researchers who interviewed readers using AI for tarot noted that AI chatbots tend to confirm the interpretation a user already has (Prock et al., 2026). The survey finding that 26% of Korean adults have used AI for saju or a fortune (Gallup Korea, 2026) is about use, not accuracy. There is also official advice not to enter personal details just because a reading is free (National Consumer Affairs Center of Japan, 2020).

Myth 4: my love reading came true, so tarot works. Hits stay in memory while misses are easily forgotten. Confirmation bias, seeking and interpreting evidence in ways that fit what we expect, can produce such memories (Nickerson, 1998). One analysis adds that divinatory interpretations are often vague enough to be hard to prove wrong, and that successes are reported more often than failures (Hong, 2025). One reading that came true is not a test of whether tarot works.

Myth 5: psychologists use tarot too. A researcher who argues for using tarot in counselling presented tarot readings from the perspective of the projective hypothesis, with one person’s reading as a case (Semetsky, 2006), but that does not make tarot a recognised projective test. A 2023 review of 28 Korean papers on tarot found that Jung’s synchronicity, Maslow’s hierarchy of needs and psychological testing tools were the main frameworks, and said that psychological research on tarot is very scarce, partly because tarot is seen as unscholarly (Choi, 2023). Both were read as abstracts. No study comparing the effects of tarot in counselling with other methods, and no survey of how many mental health professionals use tarot, was found.

Q. Can tarot predict the future? A. No study testing whether tarot predicts future events was found. In a related test that did not use tarot, a large replication designed together by supporters and sceptics of a power to sense the future found a hit rate of 49.89%, the same as the 50% chance level (Kekecs et al., 2023).

Q. Do tarot cards have fixed meanings? A. Tarot was made as a deck for a card game in 15th-century Italy, and after two essays published in Court de Gébelin’s Le Monde primitif in 1781, cartomancers such as Etteilla took it up for fortune-telling. A paper citing Dummett and other historians concludes that there is no historical basis to the occultists’ claims about the significance of the tarot (Sosteric, 2014). This guide does not give card meanings.

Q. I have readings often. Is that OK? A. Many people consult tarot for fun. In a US survey, 1% of adults said they rely a lot on such practices when making major life decisions (Pew Research Center, 2025). If you keep repeating the same question, if the cost weighs on you, or if you find it hard to decide without a reading, take a pause. A case report of one patient described repeatedly seeking fortune-telling services as a type of behavioural addiction (Skryabin, 2020), but this is not a condition with established diagnostic criteria.

Q. How is a tarot consultation different from counselling? A. Using tarot in counselling has been proposed (Semetsky, 2006), but no study comparing its effects with other methods was found. UK advertising guidance also keeps readings from claiming physical healing (CAP, 2016). If emotional difficulties last, please see a doctor or a counsellor. What a first visit involves is covered in seeing a psychiatrist for the first time.

Q. If a reading fits well, does the reader have a special ability? A. Not necessarily. A fitting reading can be produced by cold reading (Hyman, 1977), and it has been suggested that readers themselves may use such clues without being aware of it (Dutton, 1988; Ivtzan, 2007). After three tests without face-to-face contact, the researcher concluded that participants were not able to pick out their own reading (Blackmore, 1983, as cited in Ivtzan, 2007). Writing down both what felt right and what missed can help you judge.

References: Sosteric, M. (2014), A sociology of tarot, Canadian Journal of Sociology 39(3), 357-392, https://doi.org/10.29173/cjs20000; Ivtzan, I. (2007), Tarot cards: A literature review and evaluation of psychic versus psychological explanations, Journal of Parapsychology 71(1), 139-149, https://www.parapsychologypress.org/jparticle/jp-71-1-139-149; Dutton, D. L. (1988), The cold reading technique, Experientia 44(4), 326-332, https://doi.org/10.1007/BF01961271; Pew Research Center (2025), 30% of Americans consult astrology, tarot cards or fortune tellers, 21 May 2025, https://www.pewresearch.org/religion/2025/05/21/3-in-10-americans-consult-astrology-tarot-cards-or-fortune-tellers/; Dummett, M. (1993), The history of card games, European Review 1(2), 125-135, https://doi.org/10.1017/S1062798700000478; Anderson, S. (2026), See visions of the past and future in this New York City exhibition on the Renaissance roots of tarot cards, Smithsonian Magazine, 11 June 2026, https://www.smithsonianmag.com/smart-news/see-visions-of-the-past-and-future-in-this-new-york-city-exhibition-on-the-renaissance-roots-of-tarot-cards-180988940/; Pooley, W. G. (2023), Paper tools for broken hearts: Fortune-telling with cards in France, c. 1803-1937, French History 37(4), 379-400, https://doi.org/10.1093/fh/crad043; Blackmore, S. (1983), Divination with tarot cards: An empirical study, Journal of the Society for Psychical Research 52, 97-101 (cited in Ivtzan, 2007); Ivtzan, I., & French, C. C. (2004), Testing the validity of tarot cards: Can we distinguish between a real and a control reading?, The Parapsychological Association Convention, 437-439 (cited in Hofer, 2009); Hofer, G. M. (2009), Tarot cards: An investigation of their benefit as a tool for self reflection, MA thesis, University of Victoria, https://dspace.library.uvic.ca/handle/1828/1553; Hughes, M., Behanna, R., & Signorella, M. L. (2001), Perceived accuracy of fortune telling and belief in the paranormal, The Journal of Social Psychology 141(1), 159-160, https://doi.org/10.1080/00224540109600543; Santosa, D. A. (2012), The validity of tarot as a personality inventory, ANIMA Indonesian Psychological Journal 28(1), 56-61, https://doi.org/10.24123/aipj.v28i1.4183; O’Keeffe, C., & Wiseman, R. (2005), Testing alleged mediumship: Methods and results, British Journal of Psychology 96(2), 165-179, https://doi.org/10.1348/000712605X36361; Bem, D., Tressoldi, P., Rabeyron, T., & Duggan, M. (2015), Feeling the future: A meta-analysis of 90 experiments on the anomalous anticipation of random future events, F1000Research 4, 1188, https://doi.org/10.12688/f1000research.7177.2; Kekecs, Z., Palfi, B., Szaszi, B., Szecsi, P., Zrubka, M., Kovacs, M., et al. (2023), Raising the value of research studies in psychological science by increasing the credibility of research reports: The transparent Psi project, Royal Society Open Science 10(2), 191375, https://doi.org/10.1098/rsos.191375; Hyman, R. (1977), “Cold reading”: how to convince strangers that you know all about them, The Zetetic 1(2), 18-37, https://skepticalinquirer.org/1977/04/cold-reading-how-to-convince-strangers-that-you-know-all-about-them/; Nickerson, R. S. (1998), Confirmation bias: A ubiquitous phenomenon in many guises, Review of General Psychology 2(2), 175-220, https://doi.org/10.1037/1089-2680.2.2.175; Matute, H., Blanco, F., Yarritu, I., Díaz-Lago, M., Vadillo, M. A., & Barberia, I. (2015), Illusions of causality: How they bias our everyday thinking and how they could be reduced, Frontiers in Psychology 6, 888, https://doi.org/10.3389/fpsyg.2015.00888; Hong, Z. (2025), Divination, Cambridge Elements, Cambridge University Press, https://doi.org/10.1017/9781009541961; Jussim, L., & Harber, K. D. (2005), Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies, Personality and Social Psychology Review 9(2), 131-155, https://doi.org/10.1207/s15327957pspr0902_3; Tan, X., van Prooijen, J.-W., & Van Lange, P. A. M. (2022), Positive fortune telling enhances men’s financial risk taking, PLoS ONE 17(9), e0273233, https://doi.org/10.1371/journal.pone.0273233; Gallup Korea (2026), Do you believe in fortune-telling and saju? 1991-2026 [in Korean], Gallup Report, 30 April 2026, https://www.gallup.co.kr/gallupdb/reportContent.asp?seqNo=1638; Ifop (2020), Les Français et les parasciences, étude Ifop pour Femme actuelle, 10-12 November 2020, https://www.ifop.com/wp-content/uploads/2020/12/117725_Ifop_FA_2020.11.27.pdf; Ifop (2020), La vérité est ailleurs ? Voyance, sorcellerie, astrologie, Focus no. 211, December 2020, https://www.ifop.com/wp-content/uploads/2020/12/Focus-211-Voyance-sorcellerie-astrologie.pdf; Keinan, G. (1994), Effects of stress and tolerance of ambiguity on magical thinking, Journal of Personality and Social Psychology 67(1), 48-55, https://doi.org/10.1037/0022-3514.67.1.48; Lillqvist, O., & Lindeman, M. (1998), Belief in astrology as a strategy for self-verification and coping with negative life-events, European Psychologist 3(3), 202-208, https://doi.org/10.1027/1016-9040.3.3.202; Prock, M. K., Epstein, Z., Schroeder, H., Smith, A., Lee, C., Goblot, V., & Jahanbakhsh, F. (2026), Interpretive cultures: Resonance, randomness, and negotiated meaning for AI-assisted tarot divination, Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, 1-15, https://doi.org/10.1145/3772318.3791571; National Consumer Affairs Center of Japan (2020), Is that fortune-telling?! Messages keep arriving from people calling themselves fortune tellers or readers and you cannot stop: Beware of trouble with fortune-telling sites [in Japanese], 26 November 2020, https://www.kokusen.go.jp/news/data/n-20201126_1.html; Committee of Advertising Practice (CAP) (2026), Psychic and spiritualist services and lucky charms, AdviceOnline, 31 March 2026, https://www.asa.org.uk/advice-online/psychics-spiritualists-fortune-tellers-astrologers-and-clairvoyants.html; Committee of Advertising Practice (CAP) (2016), Guidance on making responsible ads for spiritual and psychic services, astrologers and lucky charms, 4 January 2016, https://www.asa.org.uk/static/uploaded/39684a29-6499-4de6-be34c05b41a99a17.pdf; Assemblée nationale (2023), Question écrite n° 6608, Encadrement de l’activité de voyance, and the ministry’s answer, Journal officiel, 27 June 2023, https://www.assemblee-nationale.fr/dyn/16/questions/QANR5L16QE6608; Skryabin, V. Y. (2020), An addiction to seeking fortune-telling services: A case report, Journal of Addictive Diseases 38(2), 223-228, https://doi.org/10.1080/10550887.2020.1740637; Das, A., Sharma, M. K., Kashyap, H., & Gupta, S. (2022), Fixating on the future: An overview of increased astrology use, International Journal of Social Psychiatry 68(5), 925-932, https://doi.org/10.1177/00207640221094155; Semetsky, I. R. (2006), Tarot as a projective technique, Spirituality and Health International 7(4), 187-197, https://doi.org/10.1002/shi.252; Choi, K. (2023), A proposal for the psychological application of tarot cards [in Korean], The Journal of Humanities and Social Sciences 21 14(1), 3837-3848, https://doi.org/10.22143/hss21.14.1.270