Knowledge Library
Sofia relationship knowledge base with behavioral and evolutionary biology · reviewed 2026-11-15. Targeted narrative research review, not a registered systematic review or new meta-analysis. Primary studies, reviews, official statistics and official product descriptions were prioritized. No paid in-app or clinical safety audit was performed.
Initial product recommendation: adults 18+. Many source samples are Western, heterosexual, self-selected or mixed in representativeness. Do not assume results transfer to every culture, language or relationship structure. Custom 0-10 needs and gap measures are unvalidated reflection tools. They do not measure love probability, clinical diagnosis, marriage suitability or divorce risk. Not every piece of advice is biologically proven; hypotheses are not prescriptions.
68 results
- S01Measured human associationMachine learning and relationship quality across 43 longitudinal couple datasets (opens in new tab)
Joel et al. (2020)
Design: Longitudinal prediction
Sample & caveat: 11,196 couples; 29 laboratories; largely self-report; predicts levels better than change.
- S02Measured human associationPredictive validity of ideal partner preferences: review and meta-analysis (opens in new tab)
Eastwick et al. (2014)
Design: Meta-analysis
Sample & caveat: 97 studies; distinguish hypothetical preferences from evaluations of actual people.
- S03Measured human associationActual and perceived similarity and attraction (opens in new tab)
Montoya et al. (2008)
Design: Meta-analysis
Sample & caveat: 313 laboratory/field investigations; interpersonal attraction is broader than romance.
- S04Measured human associationThe friends-to-lovers pathway to romance (opens in new tab)
Stinson, Cameron & Hoplock (2022; online 2021)
Design: Descriptive multi-sample study
Sample & caveat: 1,897 Canadian/US students and crowdsourced adults; retrospective convenience samples.
- S05Experimental human resultExperimental generation of interpersonal closeness (opens in new tab)
Aron et al. (1997)
Design: Laboratory experiments
Sample & caveat: Escalating reciprocal disclosure produced closeness; not evidence of guaranteed romantic love.
- S06Measured human associationUpdated meta-analysis of the Investment Model (opens in new tab)
Tran, Judge & Kashima (2019)
Design: Meta-analysis
Sample & caveat: 202 independent samples; 50,427 participants; correlations with commitment.
- S07Measured human associationInsecure attachment and relationship satisfaction (opens in new tab)
Candel & Turliuc (2019)
Design: Meta-analysis
Sample & caveat: 132 papers; 148 independent samples; 71,011 people; actor and partner associations differ.
- S08Measured human associationDoes communication predict later relationship quality and dissolution? (opens in new tab)
Kanter et al. (2022; online 2021)
Design: Longitudinal meta-analysis
Sample & caveat: 64 dyadic studies; 1,784 estimates; small prospective associations.
- S09Measured human associationSexual communication, relationship and sexual satisfaction (opens in new tab)
Mallory (2022)
Design: Meta-analysis
Sample & caveat: 93 studies; 38,499 unique people; primarily associations, not causal prescriptions.
- S10Experimental human resultDoes marriage and relationship education work? (opens in new tab)
Hawkins et al. (2008)
Design: Intervention meta-analysis
Sample & caveat: 117 studies from 86 reports; limited demographic diversity.
- S11Experimental human resultDo premarital education programs really work? (opens in new tab)
Fawcett et al. (2010)
Design: Intervention meta-analysis
Sample & caveat: 47 studies including unpublished work; mixed satisfaction results, communication benefits.
- S12Measured human associationTransition to parenthood and marital satisfaction (opens in new tab)
Bogdan, Turliuc & Candel (2022)
Design: Longitudinal meta-analysis
Sample & caveat: 49 studies; substantial heterogeneity; effects are standardized changes, not percentages.
- S13Measured human associationRelationship satisfaction across the life span (opens in new tab)
Bühler, Krauss & Orth (2021)
Design: Meta-analysis
Sample & caveat: 165 samples; 165,039 people; group trajectories do not prescribe an individual life course.
- S14Measured human associationFinancial issues and divorce (opens in new tab)
Dew, Britt & Huston (2012)
Design: Longitudinal observational
Sample & caveat: 4,574 couples; financial conflict is not equivalent to low income.
- S15Measured human associationQuality, commitment and stability in long-distance relationships (opens in new tab)
Kelmer et al. (2013)
Design: Longitudinal observational
Sample & caveat: 870 individuals; similar breakup likelihood at follow-up; no distance-based personal forecast.
- S16Measured human associationDepressive symptoms before and after young-adult breakup (opens in new tab)
Acolin et al. (2024; online 2023)
Design: Longitudinal observational
Sample & caveat: 156 young adults with one breakup during a two-year study; average symptom trajectory.
- S17Measured human associationSerial infidelity across subsequent relationships (opens in new tab)
Knopp et al. (2017)
Design: Longitudinal observational
Sample & caveat: 484 adults followed through two mixed-gender relationships; self-reported extradyadic sex.
- S18Experimental human resultRandomized controlled trial of OurRelationship (opens in new tab)
Doss et al. (2016)
Design: Randomized controlled trial
Sample & caveat: 300 heterosexual couples; eight-hour program plus brief staff contact; waitlist comparison.
- S19Measured human associationPaired app: mixed methods evaluation (opens in new tab)
Aicken et al. (2025)
Design: Nonrandomized evaluation
Sample & caveat: 440 complete longitudinal participants; web survey 745; interviews 20; self-selection limits causal claims.
- S20Measured human associationA blueprint for connection: interconnected patterns of relationship change (opens in new tab)
Agapé study (2026)
Design: Single-arm longitudinal evaluation
Sample & caveat: 405 couples; one month; no randomized control; some authors had company shareholdings.
- S21Model or heuristicEvaluating love languages from a relationship science perspective (opens in new tab)
Impett, Park & Muise (2024)
Design: Scientific review
Sample & caveat: Core claims about one primary language, five types and matching have weak support.
- S22Measured human associationReward and motivation systems in early romantic love (opens in new tab)
Aron et al. (2005)
Design: Small fMRI study
Sample & caveat: 17 intensely in-love volunteers; neural associations do not measure a partner's love percentage.
- S23Measured human associationNeural correlates of long-term intense romantic love (opens in new tab)
Acevedo et al. (2012; online 2011)
Design: Small selected-sample fMRI study
Sample & caveat: 17 participants; average marriage duration 21.4 years; not a population prevalence estimate.
- S24Measured human associationSex differences in mate preferences across 45 countries (opens in new tab)
Walter et al. (2020)
Design: Cross-cultural self-report
Sample & caveat: 14,399 participants; group preferences do not determine an individual's needs or conduct.
- S25Animal studyOxytocin receptor is not required for social attachment in prairie voles (opens in new tab)
Berendzen et al. (2023)
Design: Animal experiment
Sample & caveat: Receptor knockout study; do not translate vole findings into human hormone treatment advice.
- S26Population statisticsUS marriage and divorce totals and rates for 2024 (opens in new tab)
CDC/NCHS (Nov. 6, 2026)
Design: Official vital statistics
Sample & caveat: Marriage: 2,128,446, 6.3/1,000; divorce: 662,950, 2.3/1,000; five states excluded from divorce data.
- S27Population statisticsFirst marriages: 2006-2010 National Survey of Family Growth (opens in new tab)
CDC/NCHS (2012)
Design: Historical life-table estimates
Sample & caveat: US women/men ages 15-44; Table 5 first-marriage survival; old cohort, not a current lifetime forecast.
- S28Population statisticsOnline dating in the US (opens in new tab)
Pew Research Center (2023)
Design: Representative survey
Sample & caveat: 6,034 US adults, July 2022; denominator differs for app use and meeting current partner.
- S29Model or heuristicThe magic ratio and relationship satisfaction (opens in new tab)
The Gottman Institute
Design: Originator explanation of observational finding
Sample & caveat: 5:1 during conflict is a contextual heuristic; not an independently validated personal cutoff.
- S30Model or heuristicDevelopment of the Couples Satisfaction Index (opens in new tab)
Funk & Rogge (2007)
Design: Measurement development
Sample & caveat: 5,315 online participants; investigate current instrument terms and validated translations before deployment.
- S31Official guidanceCouples therapy with an abusive partner (opens in new tab)
National Domestic Violence Hotline
Design: Specialist safety guidance
Sample & caveat: Joint disclosure can increase danger under abuse or coercive control; use individual specialist support.
- S32Measured human associationRisks, rewards and roles for AI in relationship advice (opens in new tab)
Tseng & Liang (CHI 2026)
Design: Qualitative HCI study
Sample & caveat: 25 questionnaire participants, 90 prompts and 17 interviews; explores sycophancy and overreliance, not efficacy.
- S33Official guidanceGetting started with GPT-Live (opens in new tab)
OpenAI documentation (checked Nov. 14, 2026)
Design: Official technical documentation
Sample & caveat: Full-duplex speech plus a separate reasoning/tool backend; browser connection via WebRTC.
- S34Official guidanceGPT-Live-1 model (opens in new tab)
OpenAI documentation (checked Nov. 14, 2026)
Design: Official technical documentation
Sample & caveat: Voice session rate listed at $0.05/minute; backend and tools billed separately; verify before budgeting.
- S35Official guidanceMobile health app developer best practices (opens in new tab)
FTC
Design: Official developer guidance
Sample & caveat: Minimize data, limit permissions, provide clear notice and truthful privacy claims.
- S36Official guidanceMy Health My Data overview and FAQs (opens in new tab)
Washington Attorney General
Design: Official legal guidance
Sample & caveat: Some health-related inferences can be covered data; applicability depends on actual data and product functions.
- S37Product descriptionOfficial features and support (opens in new tab)
Paired
Design: Official product description
Sample & caveat: Daily free conversation; premium question, quiz and game library. Features can change.
- S38Product descriptionOfficial app listing (opens in new tab)
Agapé
Design: Developer app listing
Sample & caveat: Daily prompts for partners, friends and family; optional premium features.
- S39Product descriptionOfficial account and program help (opens in new tab)
Lasting
Design: Official product description
Sample & caveat: Paired assessments, sessions and comparison; research-informed content does not establish app-specific efficacy.
- S40Product descriptionOfficial program details and pricing (opens in new tab)
OurRelationship
Design: Official product description
Sample & caveat: 8-10 hours; guided/self-directed/individual options; $249/$99/$79 listed at review date; eligibility-dependent free access.
- S41Product descriptionGottman Card Decks app (opens in new tab)
The Gottman Institute
Design: Official product description
Sample & caveat: Free conversation and activity prompts; deck count differs between website and app-store listing.
- S42Product descriptionLove Nudge app (opens in new tab)
Love Language Brand
Design: Official product description
Sample & caveat: Love-language quiz, partner pairing, goals and nudges; free setup, current app listing includes in-app purchases.
- S43Product descriptionOfficial couples AI app (opens in new tab)
Maia
Design: Official product description
Sample & caveat: Text and voice for couples, AI suggestions and shared activities. Public claims are not outcome validation.
- S44Product descriptionOfficial FAQ (opens in new tab)
Lovon
Design: Official product description
Sample & caveat: Adult voice AI emotional support; broader than relationships; trial/subscription; privacy promises need independent verification.
- S45Product descriptionCurrent official landing page (opens in new tab)
Meeno
Design: Official product description
Sample & caveat: Current positioning emphasizes spoken dating-conversation practice, feedback and scores; differs from older mentor pages.
- S46Measured human associationSexual frequency and well-being (opens in new tab)
Muise, Schimmack & Impett (2016)
Design: Three observational studies
Sample & caveat: 30,645 participants; average association plateaus around weekly frequency; not a causal or individual prescription.
- S47Measured human associationDemand/withdraw interaction pattern: meta-analysis (opens in new tab)
Schrodt et al. (2014)
Design: Meta-analysis
Sample & caveat: 74 studies, 14,255 people; r=.360 is an aggregate across mixed individual, relational and communication outcomes.
- S48Model or heuristicThe mathematics of marital conflict: dynamic nonlinear modeling (opens in new tab)
Gottman, Swanson & Murray (1999)
Design: Dynamic interaction model
Sample & caveat: Nonlinear difference equations fitted to observed interactions; specialized sample/model, not a universal love calculator.
- B01Model or heuristicProximate and Ultimate Perspectives on Romantic Love (opens in new tab)
Bode & Kushnick (2021)
Design: Integrative theoretical review
Sample & caveat: Organizes love through mechanisms, development, possible functions and evolutionary history; hypotheses and established findings differ.
- B02Model or heuristicThe neural mechanisms and circuitry of the pair bond (opens in new tab)
Walum & Young (2018)
Design: Comparative neuroscience review
Sample & caveat: Much mechanistic evidence is from voles and mice; human guidance cannot be inferred directly from animal experiments.
- B03Model or heuristicPair-bonding, romantic love, and evolution: the curious case of Homo sapiens (opens in new tab)
Fletcher et al. (2015)
Design: Evolutionary theoretical review
Sample & caveat: Evaluates pair bonding, care and cooperation accounts; a proposed history does not prescribe marriage or reproduction.
- B04Model or heuristicHumans are not cooperative breeders but practice biocultural reproduction (opens in new tab)
Bogin, Bragg & Kuzawa (2014)
Design: Critical comparative review
Sample & caveat: Challenges a simple cooperative-breeding label and emphasizes culturally flexible care beyond genetic kin.
- B05Model or heuristicTinbergen’s four questions: Two proximate, two evolutionary (opens in new tab)
Nesse (2019)
Design: Conceptual framework
Sample & caveat: Mechanism, development, adaptive significance and phylogeny are distinct questions; an explanation is not a moral rule.
- B06Measured human associationIntimacy as an interpersonal process (opens in new tab)
Laurenceau, Barrett & Pietromonaco (1998)
Design: Human interaction/diary studies
Sample & caveat: Disclosure and perceived responsiveness relate to experienced intimacy; the app’s scripts were not tested in these studies.
- B07Measured human associationCouples’ shared participation in novel and arousing activities and experienced relationship quality (opens in new tab)
Aron et al. (2000)
Design: Human surveys and laboratory experiments
Sample & caveat: Novel shared tasks had short-term effects in included experiments; no guaranteed lasting benefit or prescribed activity dose.
- B08Measured human associationIt’s the little things: Everyday gratitude as a booster shot for romantic relationships (opens in new tab)
Algoe, Gable & Maisel (2010)
Design: Dyadic daily-experience study
Sample & caveat: Gratitude predicted next-day connection; observational patterns do not make compliments a debt or an abuse remedy.
- B09Measured human associationThe Role of Sleep in Interpersonal Conflict: Do Sleepless Nights Mean Worse Fights? (opens in new tab)
Gordon & Chen (2014)
Design: Daily diary and laboratory couple study
Sample & caveat: Poor sleep was associated with conflict patterns; do not infer sleep physiology or prescribe treatment.
- B10Experimental human resultLending a hand: social regulation of the neural response to threat (opens in new tab)
Coan, Schaefer & Davidson (2006)
Design: Small human fMRI experiment
Sample & caveat: 16 married women under laboratory threat; desired supportive contact is not universally safe or beneficial.
- B11Model or heuristicSocial Baseline Theory: The Social Regulation of Risk and Effort (opens in new tab)
Coan & Sbarra (2015)
Design: Theory/review
Sample & caveat: Proposes familiar social resources affect perceived risk and effort; does not imply a romantic partner must be available at all times.
- B12Model or heuristicSocial baseline theory: State of the science and new directions (opens in new tab)
Beckes & Sbarra (2022)
Design: Theory/review
Sample & caveat: Reviews social emotion regulation and needed research; direct application to a specific couple remains uncertain.
- B13Experimental human resultTouch between romantic partners: Being stroked is more pleasant than stroking and decelerates heart rate (opens in new tab)
Triscoli et al. (2017)
Design: Human touch experiment
Sample & caveat: Context-specific physiological findings; not a required hug duration or a treatment for relationship problems.
- B14Experimental human resultSocial support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress (opens in new tab)
Heinrichs et al. (2003)
Design: Small placebo-controlled human experiment
Sample & caveat: 37 healthy men in a laboratory stress task; no recommendation for hormone products, dose or personal hormone inference.
- B15Measured human associationReward, addiction, and emotion regulation systems associated with rejection in love (opens in new tab)
Fisher et al. (2010)
Design: Small selected-sample fMRI study
Sample & caveat: 15 recently rejected people still intensely in love; overlapping neural responses do not establish an addiction diagnosis.
- B16Measured human associationWho am I without you? The influence of romantic breakup on the self-concept (opens in new tab)
Slotter, Gardner & Finkel (2010)
Design: Three human studies
Sample & caveat: Breakup-related self-concept change and reduced clarity related to distress; recovery practices here are proposed applications.
- B17Measured human associationThe role of need fulfillment in relationship functioning and well-being (opens in new tab)
Patrick et al. (2007)
Design: Three human studies
Sample & caveat: Autonomy, competence and relatedness were examined in relationship functioning; not a biological compatibility score.
- B18Measured human associationWhen a Partner Supports Your Sexual Autonomy (opens in new tab)
Shoikhedbrod et al. (2026; online 2024)
Design: Three multimethod human studies
Sample & caveat: 786 participants; associations with perceived autonomy support and satisfaction, not proof that any sexual act is owed.
- B19Model or heuristicThe benefits (and costs) of sexual responsiveness in romantic relationships (opens in new tab)
Muise et al. (2023)
Design: Human relationship research review
Sample & caveat: Highlights responsiveness together with autonomy and possible costs; willingness and consent remain central.
- B20Experimental human resultAn Experimental Test of Jealousy’s Evolved Function: Imagined Partner Infidelity Induces Jealousy, Which Predicts Positive Attitude Towards Mate Retention (opens in new tab)
Jealousy experiment (2024)
Design: Human imagined-scenario experiment
Sample & caveat: 222 partnered adults; priming and intentions do not establish actual infidelity, inevitable behavior or ethical permission.