# Dating dealbreakers study: why flaws predict choices more than virtue lists

> A 2026 study of 1,074 adults: people agree on good partner traits but not on flaws. Negatives predicted hypothetical swipes better — and red flags matter for gym buddies.

Published 2026-11-21 · Updated 2026-11-21 · revns editorial

**Dating dealbreakers carry more variety — and in hypothetical app choices, more predictive weight — than the short list of virtues most people say they want**, according to a peer-reviewed study published online **6 October 2026** in the *European Journal of Social Psychology*. Across **four studies** and **seven preregistered experiments** with **1,074 participants**, Tabea J. Zorn and Christian Unkelbach at the University of Cologne report that desirable partner traits cluster tightly, while undesirable traits spread across many meanings. Negative traits then helped distinguish who would swipe left or right in lab tasks better than positive traits did. The work is about **online dating psychology**, not the gym, but it sharpens a habit that also matters for [training buddies](/blog/workout-accountability-partner-does-a-training-buddy-help): **listen for one solid red flag**, not only for shared ideals.

**TL;DR**

| Design piece | Main number (Zorn & Unkelbach, 2026) |
| --- | --- |
| Total participants | **1,074** across Studies 1–4 (seven experiments) |
| Trait listing (Studies 1a–1c) | **3.61–4.26** positive traits per person vs **2.44–2.88** negative traits listed |
| Unique English trait words pooled (1a+1b) | **181** negative vs **152** positive (plus four judged both ways) |
| Semantic distance in meaning map (Study 2) | Positives **1.96** vs negatives **2.33** (higher = less similar) |
| Importance ratings (Study 3, 7-point scale) | Avoiding negatives **5.51** vs seeking positives **5.37** |
| Hypothetical swipes (Study 4a) | **95%** right swipes for lone positive word; **15%** still right for lone negative word (varied by flaw) |
| Top rated dealbreaker words (Study 3) | **Abusive** 6.98/7; then unfaithful, dishonest, deceiving, untrustworthy |
| Top dealmaker words | **Trustworthy** 6.73/7; then kind, loyal, genuine, understanding |



## What did the 2026 dating dealbreakers study find?

Zorn and Unkelbach tested whether **valence asymmetries** — ideas from their **Evaluative Information Ecology** framework — show up when people think about partners on dating platforms. In everyday social life, positive information is often **frequent and similar**, while negative information is **rarer and more diverse**. The authors asked whether that pattern also describes the **traits** people list when they think about who to date.

The peer-reviewed paper is **"Valence Asymmetries of Traits in Dating Contexts"** (*European Journal of Social Psychology*, published online **6 October 2026**; [doi:10.1002/ejsp.70121](https://doi.org/10.1002/ejsp.70121)). Every experiment was **preregistered**, and the team reports open data and code.

Across **Studies 1a and 1b**, **200** English-speaking adults on Prolific with prior dating-app experience listed traits to seek and avoid. **Study 1c** added **98** German-speaking adults in a Cologne lab (61 with online dating experience, 37 without). After filtering entries that were not personality traits (height, hair colour, music taste and similar), individuals still wrote **more positive than negative traits**, but the **combined word pool** held **more unique negative than positive labels** in the English samples.

**Study 2** had **175** adults rate **333** trait words on three affective dimensions (pleasantness, intensity, control). **Positive traits clustered closer** in that space (**1.96** average distance) than **negative traits** (**2.33**), meaning flaws share less overlap in meaning. Lead author Tabea J. Zorn has left academia; co-author Christian Unkelbach commented to PsyPost that most people agree on positive ideals while disagreeing more on what counts as negative.

**Study 3** (**200** adults) rated how important each trait was for dating on a **1–7** scale. Avoiding negatives scored slightly higher on average than seeking positives. For both valences, traits **closer in meaning to other traits of the same type** tended to earn higher importance ratings — and that link was **stronger for negative traits**.

**Studies 4a and 4b** (**200** and **201** adults) simulated **left or right swipes** (and a **1–7** likelihood scale in 4b) when a profile showed **only one word**. For positives, **stated importance** strongly predicted swipes; **structural similarity** did not add significant prediction. For negatives, **importance and similarity each predicted choices**, and responses varied far more — some mild flaws still drew yes votes, severe words mostly drew no. Authors note a **progression bias**: **52.9%** of all choices were right swipes even though negative prompts slightly outnumbered positive ones.

## How does this compare with earlier dealbreaker research?

PsyPost and prior work link the pattern to **relationship dealbreakers** research. Jonason and colleagues surveyed common flaws people avoid in mates ([doi:10.1177/0146167215609064](https://doi.org/10.1177/0146167215609064); *Personality and Social Psychology Bulletin*, **2015**). Across their studies, people weighted **negative dealbreakers more heavily** than positive dealmakers when judging described partners, especially in long-term mating contexts.

The 2026 paper adds a **structural** layer: it is not only that a flaw feels bad, but that **different people name different flaws**, so a single negative cue can separate choices when positives sound alike ("kind," "loyal," "genuine") on every profile.

## What are common dating dealbreakers in this study?

Content rankings and structure rankings **did not always match**. By stated importance in Study 3, **trustworthy** led positives and **abusive** led negatives. Structural "centre" traits — those most similar to other traits of the same sign — included **clever, natural, cultured, helpful, open** on the positive side and **crude, stuck-up, unhumble, disloyal, shallow** on the negative side. Only **fake** appeared among both the most important dealbreakers and the most structurally central negative traits.

That gap matters for searchers asking **"what are some dating dealbreakers?"** There is no single universal list. Severe harm categories (abuse, dishonesty, unfaithfulness) rated high for many, but mild negatives such as **unfashionable** still left a minority willing to swipe right in the hypothetical task.



## What should you not conclude from hypothetical swipes?

The authors state plainly that results are **self-reports in imagined scenarios**. Real apps add **photos**, chat, location, and bundles of traits at once. Attractiveness and status cues can dominate early screens. Cultural samples were mainly **UK, US and German** adults; patterns may differ elsewhere.

Zorn and Unkelbach also asked whether admitting a small flaw might signal honesty. In their one-word tasks, **even mild negatives mostly failed** as a strategy — but they call that an open question for fuller profiles.

Nothing here diagnoses personality disorder, predicts violence, or tells you to end a real relationship from one trait word. It is social psychology on **information structure**, not clinical guidance.

## What might training partners learn from dealbreaker psychology?

Revns readers often pair up for walks, lifts or team [challenges](/features/teams). The 2026 paper **did not study gym buddies**. Use it only as a loose mirror:

1. **Virtue lists collide.** Almost everyone wants a "motivated, reliable, supportive" partner on paper — in dating and in sport. Sameness on the good side makes positives **less informative** in Zorn and Unkelbach's swipe models.
2. **Red flags are personal but decisive.** You and a friend may disagree whether "always late" is minor or a dealbreaker; one behaviour that breaks trust for you — ridicule, ignoring injury, competitive shaming — may matter more than five shared goals.
3. **Hypothetical generosity is not consent.** The study's progression bias (many right swipes) fits giving strangers benefit of the doubt online. In person, **do not** extend that benefit to pain, coercion or humiliation in training.

For setup steps and evidence on dyadic exercise, see [workout accountability partner](/blog/workout-accountability-partner-does-a-training-buddy-help) and [training with a partner](/blog/training-with-a-partner-why-partners-stick-with-it). For mood when a partnership ends, see [missing a season through injury](/blog/missing-a-season-through-injury-loneliness-mood-and-what-helps-2026) and [loneliness and exercise](/blog/loneliness-and-exercise-how-training-with-others-helps).

## What are training partner red flags worth acting on?

These are **general training habits**, not definitions from the paper:

| Red flag | Why it matters for health and consistency |
| --- | --- |
| Pressure to train through sharp pain or illness | Raises injury risk and hides overreaching |
| Public shame or comparisons on body or pace | Tied to poorer adherence in many group studies |
| Unpredictable scheduling or guilt trips | Undermines the simple check-in structure that helps dyads |
| Dismissing sleep, rest days or clinician advice | Conflicts with [recovery](/blog/recovery-in-training-what-it-means-and-how-to-tell-you-have-recovered) basics |
| Boundary crossing outside agreed metrics | Oversharing weight, food or private health data |

One honest conversation or ending a pact beats years of dread before sessions. Teams in revns can limit visibility to scores and streaks you both choose — see [couples workout tracker](/blog/couples-workout-tracker-train-together-without-oversharing).



## What to do this week if you date and train

If you use dating apps and also want a gym buddy:

1. **Name one non-negotiable** behaviour (honesty about plans, no mockery, respect for rest days) before you name five ideals.
2. **Meet once for a low-stakes walk or mobility session** before you share a heavy programme or sign up for a paid challenge.
3. **Keep health data private by default.** Use [mood tracking](/blog/how-to-track-mood-and-energy-next-to-training) for yourself; do not treat a new match as a therapist.
4. **Protect sleep** the night before early partner sessions — short sleep blunts positive mood in lab work, including the 2026 [sleep and enjoyment study](/blog/mild-sleep-loss-positive-mood-aesthetic-study-2026).

## When to see a professional

See a clinician or licensed mental health provider if sadness, anxiety, panic attacks or loss of interest interfere with work, sleep or safety for **two weeks or more**. Seek **urgent** help for threats, stalking, coercive control or any assault — online or in person.

**Crisis lines (examples):** US — call or text **988**; UK and Ireland — **Samaritans 116 123** (free, 24/7). If you are elsewhere, use your local emergency number or health service crisis line.

Relationship distress alone does not require a label; still, a counsellor can help when the same argument loop blocks sleep and training for weeks.

## Related

- [Workout accountability partner: does a training buddy help?](/blog/workout-accountability-partner-does-a-training-buddy-help)
- [Training with a partner: why friends and couples stick with it](/blog/training-with-a-partner-why-partners-stick-with-it)
- [Loneliness and exercise: how training with others helps](/blog/loneliness-and-exercise-how-training-with-others-helps)
- [How to track mood and energy next to training](/blog/how-to-track-mood-and-energy-next-to-training)
- [Teams and challenges](/features/teams)

This is general training and wellbeing guidance, not medical advice, psychotherapy or legal advice about relationships.

## Sources

- [Zorn TJ, Unkelbach C. Valence Asymmetries of Traits in Dating Contexts. European Journal of Social Psychology. Published online 6 October 2026](https://doi.org/10.1002/ejsp.70121)
- [Jonason PK, et al. Relationship dealbreakers: traits people avoid in potential mates. Personality and Social Psychology Bulletin. 2015](https://doi.org/10.1177/0146167215609064)
- [Dolan EW. Psychologists find a hidden asymmetry in how people judge potential partners. PsyPost. 21 November 2026](https://www.psypost.org/psychologists-find-a-hidden-asymmetry-in-how-people-judge-potential-partners/) (news reporting; scientific claims traced to the journal article above)

## Questions

### What are dating dealbreakers?

Dating dealbreakers are partner traits or behaviours that make someone unwilling to pursue a relationship. In a 2015 Personality and Social Psychology Bulletin survey, Jonason and colleagues grouped common dealbreakers into categories such as apathy and poor hygiene for long-term mating. A 2026 European Journal of Social Psychology paper by Zorn and Unkelbach found abusive, unfaithful and dishonest near the top of negative ratings in their word lists, while trustworthy, kind and loyal ranked highest among positives.

### Why do people disagree more on negative traits than positive ones?

In four studies with 1,074 participants total, Zorn and Unkelbach report that daters name more positive traits than negative ones individually, but the pooled vocabulary contained more unique negative words. Negative traits also sat farther apart in a three-dimensional meaning map (average distance 2.33 versus 1.96 for positives). The authors tie that pattern to an evaluative information ecology in which positive social information is frequent and similar, while negative information is rarer and more varied.

### Does one bad trait end a match in the 2026 study?

Not always. In hypothetical swipe tasks, about 15 percent of participants still chose to swipe right when a profile showed only a single negative trait, depending on whether the flaw was mild (such as unfashionable) or severe (such as abusive). Stated importance of the flaw was the strongest predictor; structural similarity to other negative traits added extra predictive power on the negative side only.

### Did the study use real dating apps?

No. Participants imagined partners described by one personality word at a time on Prolific or in a German university lab. There were no photos and no messaging. Authors caution that hypothetical self-reports may not match real swiping, where attractiveness photos and bundled profiles matter.

### What does this have to do with gym buddies or training partners?

The study tested romantic dating, not sport. The transferable idea is narrower: people often list the same good qualities they want (reliable, kind, motivated) but reject different flaws. For a training pact, one clear red flag — pressure to train through pain, public shaming, or unpredictability — may tell you more than a long list of shared goals. That analogy is practical judgment, not a finding from the paper.

### When should relationship or mood problems get professional help?

Contact a clinician or crisis service if low mood, anxiety, panic or thoughts of self-harm last most days for two weeks or more, or if a partner's behaviour feels unsafe. In the US, call or text 988. In the UK and Ireland, Samaritans are on 116 123. Revns does not provide therapy or crisis care.

---

Canonical: https://revns.com/blog/dating-dealbreakers-trait-asymmetry-study-2026
Index of everything: https://revns.com/llms.txt
revns gives training guidance, not medical advice.
