Does your MBTI type predict your attachment style? What the research says

Posts like “attachment styles by MBTI type” and “the most avoidant MBTI types” are all over social media and blogs. Some give every type a style (“INFPs are anxious, INTJs are avoidant”), and others sort by a single letter (“I means avoidant, F means anxious”). It is easy to see the appeal: when someone you have been seeing starts texting less, their type is the first thing you want to look up. But if you are an INFP who came out secure on an attachment style test, or you and a friend share a type and got opposite results, you may wonder whether these pairings hold up.

The short answer: the MBTI does not measure attachment, and we could not find a journal article that gave the same people both the MBTI and an adult attachment questionnaire. So there is no research-backed ranking of the 16 types by attachment style. Attachment and personality traits have been studied together a good deal, though, and the clearest link runs between attachment anxiety and neuroticism, a trait none of the four MBTI letters measures. On average, avoidance might lean slightly toward Introversion (I) and Thinking (T), and anxiety toward Introversion (I) and Perceiving (P), but these are indirect guesses about small differences, and every type includes secure, anxious, avoidant and fearful-avoidant people.

This guide covers what the MBTI and attachment styles each measure, what research on attachment and personality has found, and what those findings look like in MBTI letters. The last sections show how to read a type-by-style list with a critical eye and how to line up your own two results.

The MBTI turns four preferences into letters. Extraversion-Introversion (E–I) is about whether your attention and energy turn mostly outward or inward, Sensing-Intuition (S–N) about whether concrete facts or the big picture catch your eye first, Thinking-Feeling (T–F) about whether logic or people come first when you decide, and Judging-Perceiving (J–P) about whether you prefer plans settled or left open. The four letters together name one of 16 types. The questions are about everyday life in general; none asks about one particular relationship.

Adult attachment is the pattern that shows up with the people you care about most, such as a partner, family or close friends. Researchers usually describe it on two dimensions. Anxiety is how much you worry that someone will leave or lose interest; avoidance is how uncomfortable it feels to rely on someone or get close. Low on both is called Secure, high on anxiety alone Anxious, high on avoidance alone Avoidant, and high on both Fearful-avoidant.

Attachment can look different from one relationship to the next. In a study that asked the same questions separately about mother, father, partner and best friend, the authors concluded that people who are secure with their parents may be insecure with their partners (Fraley et al., 2011). It can also shift as new relationships come along. So an attachment style is closer to a current tendency in close relationships than to a fixed personality type. Our guide to reading attachment results covers how to read yours, and our guide to attachment styles in dating covers how each style tends to show up with a partner.

The two tests don’t even ask about the same situations. The MBTI asks what feels natural in daily life; an attachment questionnaire asks what worries come up around the people closest to you, and how much you want to pull back from them. Guessing one result from the other would take research that links the two.

We started by looking for studies that compared the MBTI and attachment directly. In November 2026 we searched the OpenAlex and Crossref research databases and the Korean database RISS for the two together, and found no journal article that gave the same people the MBTI and an adult attachment questionnaire. The closest thing was a 2007 undergraduate senior thesis at a US university. It used a short MBTI-style questionnaire (the Keirsey Temperament Sorter II) alongside attachment questionnaires, and its summary says more of the people the sorter classified as extraverted or as sensing types were also secure, but it gives no comparison figures or statistical tests (Rosswurm, 2007). An undergraduate thesis that never went through peer review cannot settle the question on its own.

Attachment and the Big Five personality traits, on the other hand, have been studied together for more than 30 years. In 1992, Phillip Shaver and Kelly Brennan measured adult attachment styles alongside the Big Five, using the NEO Personality Inventory. The styles were related to the Big Five in the ways theory predicted, but they were not simply the Big Five under another name. And when it came to predicting how romantic relationships turned out, the Big Five scales were generally less powerful than the attachment measures (Shaver & Brennan, 1992). Later, a study that gave 8,318 US university students a romantic attachment questionnaire (the ECR) and a Big Five measure also found that anxiety correlated most strongly with neuroticism, at 0.42, and that both anxiety and avoidance were linked to extraversion, agreeableness and conscientiousness by about the same amount, from −0.15 to −0.23. Attachment also predicted relationship quality better than the Big Five did (Noftle & Shaver, 2006).

A meta-analysis has pooled the evidence too. Kate Warnock’s team gathered studies of attachment at work, covering 109 samples and 32,278 people; the links with the Big Five came from 11 to 14 of those samples, about 10,000 people each. Of the Big Five traits, the one most strongly tied to attachment anxiety was neuroticism, with a correlation of 0.49 after correcting for measurement error, the largest of all the attachment–trait links in the analysis. Attachment avoidance correlated −0.28 with extraversion, so more extraverted people tended to score a little lower on avoidance. Its correlation with neuroticism was only 0.18, far smaller than for anxiety. Both anxiety and avoidance also went with somewhat lower agreeableness. Most of the other links fell between about −0.1 and −0.3 (Warnock et al., 2024).

Because those samples came from studies about work, the numbers may differ in studies of romantic relationships alone, though the student study above, which measured romantic attachment, showed much the same pattern. Attachment overlaps with personality traits to a degree but is not the same thing, and the biggest overlap is between anxiety and neuroticism.

Since we found no study measuring the MBTI and attachment together, the Big Five can serve as a bridge for translating those findings into MBTI letters. McCrae and Costa reported that the four MBTI scales partly measure four of the five Big Five dimensions (McCrae & Costa, 1989). When a later team gave both tests to 900 people, Extraversion-Introversion lined up with extraversion, Sensing-Intuition with openness, Thinking-Feeling with agreeableness and Judging-Perceiving with conscientiousness (Furnham et al., 2003). Neuroticism, the tendency to feel negative emotions such as worry and tension easily, has no MBTI scale at all. The mind.io Big Five test calls it Emotional sensitivity.

Translated this way, attachment anxiety is harder to guess from a type than avoidance, because its biggest overlap is with neuroticism, which the MBTI leaves out. Anxiety also had small negative links with extraversion (−0.18) and conscientiousness (−0.29), and one in the same direction with agreeableness; carried over, they would tilt it slightly toward I and P, and toward T, about as much as or a little less than the avoidance tilt below. That points to Thinking rather than Feeling, so it gives no support to the popular claim that F types are the anxious ones.

Attachment avoidance correlated −0.28 with extraversion and, by a similar amount, negatively with agreeableness. Since E–I lines up with extraversion and T–F with agreeableness, I and T types might score slightly higher on avoidance on average. If the guess held, IT types such as ISTP and INTJ would include a few more avoidant people, and EF types such as ENFJ and ESFJ a few fewer. That does not make an INTJ likely to be avoidant; it only nudges an average. Openness and conscientiousness, the traits behind Sensing-Intuition and Judging-Perceiving, had even smaller links with avoidance, so they offer little to go on.

This guess, though, crosses two bridges, and neither is sturdy. The MBTI letters and the Big Five traits only partly overlap, and the attachment links the guess relies on are modest, around −0.28. Chaining two modest links leaves an even smaller gap, and we found no study that checked the guess by giving the same people both tests. Even if it held, it would describe averages: every type still includes secure, anxious, avoidant and fearful-avoidant people.

First, there is no data behind them. It is hard to find a ranking post that says how many people took both tests or how the results split by type. Perhaps a type description such as “needs plenty of alone time” looked a lot like the avoidant description “keeps people at a distance,” and the two got stitched together.

Second, a small gap in averages cannot predict any one person. With a correlation of −0.28 between avoidance and extraversion, introverts’ and extraverts’ avoidance scores overlap for the most part. Plenty of introverts score low on avoidance, and plenty of extraverts score high. A ranking erases that overlap and hands each type a single style.

Third, both results can change. Attachment scores shift with the relationship you have in mind when you answer and with how that relationship is going, and new experiences can move them toward the secure side. The MBTI, too, measures four fairly independent dimensions rather than truly distinct types (McCrae & Costa, 1989), so an axis near 50:50 can flip when you retest. A line like “ENFPs are anxious” ties together two results that can each change. Our guide to attachment test accuracy and guide to MBTI accuracy explain why.

Fourth, introversion is not avoidance. Enjoying time alone is one thing; feeling uneasy about leaning on someone close is another. Someone who needs a quiet weekend at home to recharge can still confide in a partner and lean on them easily. And someone who livens up every party may start texting less the moment a relationship gets serious. The FAQ in our guide to coping with avoidant attachment covers that difference too.

Fifth, Feeling (F) is not anxious attachment. In the MBTI, Feeling describes how much weight you give to people and values when you decide. Worrying that a slow reply means someone has cooled on you is a different matter. The Big Five bridge does not point toward F either, but the link is so small that neither F nor T tells you much about anxiety.

Sixth, a line like “you want to get close but step back for fear of getting hurt” describes something almost everyone has felt, so it reads as personal. The pull of descriptions that fit nearly anyone is called the Barnum effect. The more accurate a ranking’s description feels, the more it is worth reading the one for another type and checking whether it fits you just as well.

Rather than looking up your type in a ranking, take both tests and see your own results. On mind.io, the attachment style test (28 statements, about 6 minutes) and the MBTI test (32 statements, about 4 minutes) are both free.

The attachment style test asks about anxiety and avoidance with 14 statements each, 28 in all, written for mind.io. On each axis, 9 statements are scored as written and 5 are worded the other way around and scored in reverse. Each axis’s average becomes a score from 0 to 100, and 50 is the dividing line. Anxiety alone above 50 gives Anxious, avoidance alone gives Avoidant, both give Fearful-avoidant, and neither gives Secure; a score of exactly 50 counts as not above the line. A score of 40–60 on either axis is “Near the midline,” so the result page also shows the neighboring style you would get if that axis crossed to the other side.

The MBTI result shows how far you lean on each of the four axes, in percent. No combo card puts these two results on one screen, so open both result pages side by side and compare them yourself. If your Extraversion-Introversion or Thinking-Feeling percentage is close to 50%, that letter could change next time, and if your anxiety or avoidance score is near the midline, so could your attachment style. See whether you lean introverted and thinking and also score high on avoidance, as the guess above would suggest, or land on a different mix entirely. Either is fine: what holds on average need not hold for you.

Your attachment result can depend on which relationship you picture. Answer once with your current or most recent partner in mind and once with family or a close friend, and you can see how your attachment differs between relationships. Comparing with a friend who shares your type works well too. If you open a friend’s shared attachment result and then finish the same test, your result page shows a reading of your two styles together. Two people with the same type can come out with quite different attachment styles.

If you are curious about neuroticism, the trait most tied to attachment anxiety, check your Emotional sensitivity score in the Big Five test as well. It gives you a score for a trait your MBTI result cannot tell you about.

All the tests in this guide are tools for self-reflection, not diagnoses. The mind.io MBTI test is not the official MBTI (Myers-Briggs Type Indicator); it is a free MBTI-style test built around the same four preference axes. The attachment style test is a self-check built from mind.io’s own statements around the two axes used in research, anxiety and avoidance; it is not a validated clinical scale.

Please do not use an attachment style as a label for judging someone else. Deciding “they’re an INTJ, so they must be avoidant” makes it easy to miss the ways that person actually reaches out. And checking someone’s phone or location out of jealousy, keeping them from friends, threatening to hurt them if they leave, or causing physical harm is not an attachment style. It is abuse, and in that situation your safety comes before any communication tips. Where to get help is listed in the “When it is not about attachment” part of our guide to talking with an avoidant partner.

References: Shaver & Brennan (1992), Personality and Social Psychology Bulletin 18(5), 536-545, https://doi.org/10.1177/0146167292185003; Noftle & Shaver (2006), Journal of Research in Personality 40(2), 179-208, https://doi.org/10.1016/j.jrp.2004.11.003, https://adultattachment.faculty.ucdavis.edu/wp-content/uploads/sites/66/2015/09/Noftle_2006_Attachment-dimensions-and-the-big-five-personality-traits.pdf; Warnock, Ju & Katz (2024), Journal of Business and Psychology 39(6), 1239-1257, https://doi.org/10.1007/s10869-024-09960-9 (figures from the authors’ posted manuscript, https://osf.io/c83ke); McCrae & Costa (1989), Journal of Personality 57(1); Furnham, Moutafi & Crump (2003), Social Behavior and Personality 31(6); Fraley, Heffernan, Vicary & Brumbaugh (2011), Psychological Assessment 23(3), 615-625, https://doi.org/10.1037/a0022898; Rosswurm (2007), Personality and relational style, undergraduate senior thesis, Ball State University, https://cardinalscholar.bsu.edu/handle/20.500.14291/188716 (accessed November 2026).