Quick answer
Impett, Park, and Muise report that research does not strongly support three love-language assumptions. Their balanced-diet picture is a proposal, and a news comment is kept separate from the paper.
In 2024, Emily A. Impett, Haeyoung Gideon Park, and Amy Muise published "Popular psychology through a scientific lens: Evaluating love languages from a relationship science perspective" in Current Directions in Psychological Science. The review starts from a very popular book and asks whether relationship science can carry the claims that traveled with it. The careful sentence in the paper is that the research, taken together, does not provide strong empirical support for three assumptions. A comment Impett gave to a University of Toronto news report is sharper, and it has to stay labeled as a comment. A metaphor the team offers at the end is a proposal, not a law.
The book the review is answering
Gary Chapman's The 5 Love Languages was first published in 1992. The SAGE page for the 2024 paper says the book sold over 20 million copies and was translated into 50 languages. Those two figures are publishing facts attached to that page. They are not evidence that the book's claims are true, and they are not evidence that the claims are false. They explain why a scientific review exists at all. A book that many people have read will shape how couples talk about care, even if the studies do not line up behind it.
The five languages named in popular use are physical touch, words of affirmation, acts of service, quality time, and gifts. The review is about the assumptions that grew up around that list. It is not a takedown of touching someone you love, saying something kind, doing a chore, spending time, or giving a present. Ordinary kindness can matter whether or not a five-part theory is right. The question Impett, Park, and Muise ask is narrower: do the studies give strong support to the theory as Chapman framed it?
Assumption one: a preferred language
The first assumption is that each person has a preferred love language. The review says the research, taken together, does not provide strong empirical support for that idea. "Taken together" means they are judging the body of work, not crowning one study as the last word. "Does not provide strong empirical support" is the phrase to keep. It is not the same sentence as "nobody has a preference about anything." People can like a back rub more than a gadget. The unsupported claim is the stronger one: that each person has a preferred love language in the sense the theory needs.
If you have been introducing yourself with a primary language, this assumption is the one you have been using. The 2024 review does not hand you a replacement label. It says the evidence for the preferred-language premise is not strong. You can still notice what you enjoy. You should not treat that notice as confirmation of a theory the review could not strongly support.
Assumption two: that there are five
The second assumption is that there are five love languages. The same careful verdict applies. The research, taken together, does not provide strong empirical support for a five-language structure. The popular names are still the ones people use: physical touch, words of affirmation, acts of service, quality time, and gifts. Familiarity is not the same as a validated partition of love into five channels.
A list can be useful as a reminder and still fail as a model. Five is a memorable number. The review's point is that memorability and sales, including the SAGE page's report of over 20 million copies and 50 languages, do not establish that human care arrives in that count. If a sixth way of showing care matters in your house, the five-language assumption has no strong empirical warrant to exclude it.
Assumption three: matching the preferred language
The third assumption is that couples are more satisfied when partners "speak" each other's preferred language. Again, the paper's claim is the absence of strong empirical support, not a finding that mismatched couples are doomed or that matched couples are fine. Satisfaction, in this assumption, is supposed to rise when you deliver the other person's preferred channel. The review says the studies, taken together, do not strongly support that.
This is the assumption that turns a quiz into housework: find the language, then perform it. If the preferred-language premise is weakly supported, and the five-count premise is weakly supported, the matching premise is standing on both of them. The review evaluates it in its own right and gives it the same verdict. You can still do kind things on purpose. You cannot cite this paper as proof that hitting the right language is what makes a couple satisfied.
A comment to the news office, kept separate
The University of Toronto news report on the project quotes Impett. She said they reviewed existing studies and "None of the 10 studies supported Chapman's claims." That sentence is her comment to the university news office. It is not a line this article will pretend sits in the abstract. The abstract-level claim to use is the paper's: the research does not provide strong empirical support for the three assumptions.
Both can be true in different registers. A researcher can tell a news office that none of the 10 studies she and her colleagues reviewed supported the claims, and the paper can state a more measured summary for the scientific record. If you quote only the news line, you should say Impett said it to the university. If you quote only the paper, you should not inflate "does not provide strong empirical support" into a slogan she did not put in the abstract. This article keeps both, with the attribution attached.
Their metaphor, offered as a proposal
The team proposes a different metaphor: love as a balanced diet, needing a range of nutrients, including factors outside those five, such as companionship and emotional support. Read the verb. They propose it. A proposal is not a result you can treat as settled law. The balanced diet is their suggestion for how to think once the three assumptions lack strong support. It is not a new five-item test, and it is not a finding that a couple who "eats a balanced diet of love" will be happier by a measured amount. No such amount is in these notes.
The content of the proposal still matters. A range of nutrients is a picture of variety. Companionship and emotional support are named as factors outside the five popular languages. That is an explicit widening. Physical touch, words of affirmation, acts of service, quality time, and gifts do not exhaust the list the team wants you to consider. Sitting with someone, and offering emotional support, count in their metaphor even when they do not fit neatly into a gift or a chore.
Companionship is also a platonic word. The advisory literature on friends and neighbors is a different document, but the diet proposal's mention of companionship and emotional support points past a romance-only menu. Friends, colleagues, and neighbors are part of how connection gets defined outside a couple.
What you can do with a popular quiz anyway
You can treat the five names as conversation prompts without granting them the status of a finding. Ask what you appreciated this month, and let the answer include touch, words, help, time, gifts, or something else, including company and support. You can decline to rank yourself with a primary language if the first assumption lacks strong support. You can decline to judge the relationship by whether you matched a preferred channel if the third assumption lacks strong support too.
You can also keep the scale of the book's reach in view without bowing to it. Over 20 million copies and 50 languages, as the SAGE page states, mean you will meet the vocabulary everywhere. Meeting a vocabulary is not the same as meeting a well-supported model. When someone says "that is not my love language," you can hear a preference in ordinary speech and still remember that Impett, Park, and Muise did not find strong empirical support for the theory wrapped around that sentence.
The balanced-diet metaphor is available if the five-box quiz feels too small. Use it as their proposal. Ask whether the relationship is getting a range of care, including companionship and emotional support, rather than whether you have diagnosed a single channel. Do not turn the metaphor into a rule with a score. They did not publish it that way.
This reading is a summary of a review and a news comment. It is not a verdict on your relationship. If the way you give and receive care has become a source of ongoing pain, a couples clinician can help you talk about that without a five-language script.
Which of the three assumptions have you been treating as fact: that you have one preferred language, that five languages cover love, or that matching that language is what makes a couple satisfied?
Review and news comment
Impett, E. A., Park, H. G., and Muise, A. (2024). Popular psychology through a scientific lens: Evaluating love languages from a relationship science perspective. Current Directions in Psychological Science, 33(2), 87–92. https://doi.org/10.1177/09637214231217663
University of Toronto. Little evidence linking the five love languages to healthy relationships, researchers say. https://www.utoronto.ca/news/little-evidence-linking-five-love-languages-healthy-relationships-researchers-say
Frequently Asked Questions
- Which three ideas lacked strong support?
- That each person has a favorite channel of care, that the true count of those channels is five, and that couples feel better when they perform the favorite. The paper's wording is that the studies, together, do not strongly back those ideas.
- What did Impett tell the university news office?
- She said the ten studies reviewed did not back Chapman's claims. Keep that sentence with the news interview. Do not treat it as a line from the abstract.
- Is the balanced-diet picture a proven law?
- No. It is the team's suggestion: love needs a mix of nutrients, including company and emotional support outside the popular five.
- What reach does the SAGE page state for the 1992 book?
- Over twenty million copies sold, and translation into fifty languages. A wide audience is not the same thing as strong evidence.
- Are the five popular behaviors useless?
- The review does not say that touch, kind words, help, shared time, or presents are worthless. It says the theory wrapped around them lacks strong empirical backing.