How Long Does Limerence Last? Duration, Recurrence, and What Research Shows
Author: Ukrainian Psychological Hub · Published: September 26, 2026 · Editorial Policy
How long does limerence last? The most accurate current answer is that there is no scientifically validated universal timeline. The widely repeated estimate of about 18 months to 3 years comes from Dorothy Tennov’s original qualitative work, not from a modern prospective study that followed a representative sample from the beginning to the end of limerence. A major 2026 study of impairing limerence adds an important newer data point: participants retrospectively described episodes involving about two years of obsessive fixation and preoccupation on average, together with repeated episodes across adulthood. That finding is substantial, but it applies to a selected sample of people with impairing limerence and should not be turned into a countdown clock for everyone.
A person can therefore experience limerence for less than two years, around two years, or much longer. Current science does not establish a point at which limerence “must” end. It also does not establish a reliable number of weeks or months after rejection, no contact, reciprocation, a breakup, or therapy when the experience will resolve.
The duration question is best answered by separating three things that are often collapsed online: how long one episode lasts, whether the same person remains the focus after intensity changes, and whether new limerent episodes recur with different people. Those are different research questions.
The Short Answer: What Research Can Actually Say About Duration
The best-supported summary in 2026 is simple.
The classic 18-month-to-3-year figure is a historical descriptive estimate. A modern review of romantic passion explains that Tennov’s estimate was approximately two years, with 18 months to 3 years as the most typical interval in her qualitative material (Carswell & Impett, 2021). It was not a population-based survival estimate, a clinical prognosis, or a rule that predicts when an individual episode will end.
The strongest newer quantitative evidence comes from Evans and colleagues’ 2026 research on impairing limerence. Study 1 included 1,647 participants and reported episodes involving prolonged preoccupation of about two years, with participants also reporting approximately five episodes across adulthood on average (Evans et al., 2026). The study is currently the largest quantitative investigation of limerence, but its sample was recruited around impairing limerence rather than drawn randomly from the general population.
That distinction changes how the number should be read. “About two years in this sample” is an empirical description. “Limerence normally lasts two years” is a broader claim the study does not establish.
Where Does the “18 Months to 3 Years” Limerence Timeline Come From?
The 18-month-to-3-year number appears so often online that it can look like the result of a large modern longitudinal study. It is not.
Dorothy Tennov introduced the term limerence after qualitative interviews and questionnaire work on the experience of being in love. Later scholarship summarized her estimate as an average of roughly two years, with 18 months to 3 years as the most typical interval (Carswell & Impett, 2021). A later clinical case report also repeated Tennov’s description that individual episodes could be much shorter or much longer than that average (Wyant, 2021).
Tennov’s work was historically important because it identified a recognizable configuration of intrusive romantic preoccupation, longing for reciprocation, uncertainty, and emotional reactivity. It was not designed as a contemporary epidemiological study. There was no representative sampling frame, standardized modern limerence measure, prospective enrollment at episode onset, repeated measurement until remission, or survival analysis estimating the probability that an episode would still be active at a given month.
That means the famous range is useful as history, not as a deadline.
If someone has been intensely preoccupied for six months, the historical estimate does not prove that another year is inevitable. If someone remains preoccupied after three years, it does not prove that their experience cannot be limerence. And if an episode subsides in a few months, that does not make it “too short” to resemble the construct.
What the 2026 Study Changes
Until recently, the limerence literature was extremely thin. A 2025 scoping review described the field as having extraordinarily limited literature, with much of the available work consisting of conceptual discussion, qualitative research, case material, and adjacent evidence on rumination and fixation.
The 2026 Acta Psychologica paper by Evans and colleagues materially changed the evidence base. It included two studies. Study 1 surveyed 1,647 people with impairing limerence and examined core features, onset, recurrence, correlates, and comorbidity. Study 2 used experience sampling with 51 participants over seven days to examine limerent thoughts in real time (Evans et al., 2026).
For duration, the most relevant Study 1 finding is the report of prolonged episodes involving about two years of obsessive fixation and preoccupation. Participants also reported recurrence across adulthood, averaging approximately five episodes.
The real-time study contributes a different kind of evidence. In that selected high-severity sample, thoughts about the limerent object occupied about half of sampled waking thought and were often intrusive and immersive. Those findings help explain why a long episode can feel psychologically continuous even though its intensity may fluctuate from hour to hour or week to week.
The paper does not turn two years into a validated prognosis. It does not show that half of all people who experience limerence will remain limerent for two years. It does not establish a median time to recovery in the general population. And because the key history variables were reported retrospectively, the study is not equivalent to following people prospectively from first onset until a predefined endpoint.
Why an Average Duration Is Not a Countdown Clock
An average compresses variation. It does not tell an individual person where they fall in the distribution.
This is especially important for limerence because the field has not yet standardized what counts as the beginning or end of an episode. One person may date onset from the first intrusive thought. Another may date it from the moment attraction became consuming. A researcher could define an endpoint as the disappearance of longing, a major reduction in intrusive thoughts, restored functioning, or a score falling below a research threshold. Those endpoints are not interchangeable.
Current limerence research also contains substantial sampling constraints. People who volunteer for studies about impairing limerence are more likely to have noticed the phenomenon, labeled it, experienced distress, or spent enough time with it to seek information. A severe or persistent sample can tell us a great deal about severe or persistent experiences without telling us how common that duration is among everyone who has ever had a brief limerent-like period.
The 2026 study therefore provides a stronger empirical anchor than the older 18-month-to-3-year estimate, but it does not erase the uncertainty around individual prognosis.
What Counts as the End of Limerence?
There is no formally accepted remission criterion for limerence.
That matters because “it ended” can describe several different changes. Intrusive thoughts may become less frequent while affection remains. The urge to check messages or social media may fade while occasional fantasies continue. Emotional reactions to contact may become smaller before the person stops thinking about the limerent object altogether. A relationship may become reciprocal while idealization and preoccupation gradually decrease. Or a person may still remember the experience vividly without being functionally dominated by it.
A practical research-oriented description of improvement is therefore dimensional rather than binary: the person of focus occupies less involuntary attention, reciprocation feels less necessary for emotional stability, cues produce less intense emotional swing, fantasy consumes less time, and ordinary responsibilities and relationships regain space.
The Limerence Questionnaire-11 introduced in 2025 gives researchers a validated way to measure dimensions of limerence, especially intense need for attachment and neglect of self and others. It does not provide a diagnostic cutoff, a remission definition, or a calculator that predicts how many months an episode will last.
Limerence Is Not a DSM or ICD Diagnosis
Duration is sometimes discussed online as though limerence were a disorder with a required minimum duration and a defined clinical course. It is not.
Limerence is currently a research and descriptive psychological construct. It is not a standalone diagnosis in the current DSM-5-TR framework maintained by the American Psychiatric Association, and it is not a separate disorder in the World Health Organization’s ICD-11 clinical framework.
This means there is no official rule such as “symptoms must last six months” or “the condition remits after three years.” The calendar alone also cannot determine whether someone has another mental health condition. Persistent romantic preoccupation does not automatically mean OCD, an anxiety disorder, depression, trauma, addiction, a personality disorder, or an attachment disorder.
For the broader construct, see Limerence: What It Is, Signs, Causes, and What Research Shows.
Duration, Persistence, and Recurrence Are Different Questions
One reason online discussions become confusing is that they use “lasting” to describe several patterns.
A single episode can remain focused on one person for an extended period. That is episode duration.
A period can decrease substantially and later become more intense again after a cue, contact, change in circumstances, or renewed hope. That may be reactivation of the same episode, but research has not established a standardized rule for distinguishing reactivation from a new episode.
A person can also experience intense limerent preoccupation toward different people at different times. That is recurrence across targets. The phrase “serial limerence” is used for repeated patterns, but it is not a formal diagnosis and has its own separate search intent. Read the dedicated recurrence review: Serial Limerence: Why Intense Romantic Fixation Can Repeat Across Different People.
Evans and colleagues’ 2026 sample is important here because participants reported approximately five episodes across adulthood on average (Evans et al., 2026). The result shows that recurrence can occur in people with impairing limerence. It does not establish that five episodes is typical in the general population, and it does not mean that one long episode and five distinct episodes are psychologically equivalent.
Can Limerence Last for Years?
Yes. Existing literature supports the possibility that limerent preoccupation can persist for years.
The older qualitative literature described wide variation, and modern quantitative evidence in an impairing sample found episodes around two years on average. The field also contains case reports of much longer courses. Those reports show possibility, not frequency.
The scientifically important question is therefore not whether a multi-year course can occur. It can. The unresolved question is how the distribution of episode lengths looks in a representative population and which variables predict persistence, remission, or recurrence.
As of September 2026, the field still lacks the kind of prospective longitudinal cohort that would allow researchers to say, for example, what proportion of newly identified episodes remain active at 3, 6, 12, 24, or 36 months under standardized definitions.
Can Limerence End in a Few Weeks or Months?
It can, but current evidence cannot supply a reliable probability.
Historical descriptions included brief episodes as well as very long ones, and there is no formal minimum duration for the construct. A short but intense period of intrusive romantic preoccupation can therefore resemble limerence phenomenologically. At the same time, ordinary attraction, a crush, early-stage passionate love, rejection distress, and situational rumination can all produce temporary preoccupation.
Duration alone cannot distinguish among them. The broader pattern matters: involuntary mental capture, longing for reciprocation, idealization, emotional dependence on perceived cues, neglect of ordinary life, and functional impact.
That is one reason a brief online checklist should not be used as a diagnosis. The LQ-11 is a research measure for the construct, not a diagnostic test with an official duration criterion (Marshall et al., 2025).
Does Uncertainty Make Limerence Last Longer?
Uncertainty about reciprocation is central to classic limerence theory, and it remains one of the most plausible explanations for why some episodes stay cognitively active. The important evidence distinction is that uncertainty is theoretically central and supported by adjacent romantic-passion research, while direct evidence that uncertainty predicts limerence duration is still limited.
Carswell and Impett’s integrative review describes Limerence Theory as proposing that difficulty obtaining affection and uncertainty can increase rumination and perceived desirability, with passion changing when reciprocation becomes clearer (Carswell & Impett, 2021).
That does not establish a universal causal rule. Some people remain preoccupied after clear rejection. Some experience limerence inside an actual relationship. Others become less preoccupied without a decisive conversation. Current research has not demonstrated that one unit of “uncertainty” adds a predictable number of months to an episode.
The scientifically stronger formulation is therefore: uncertainty is a theoretically and relationally relevant candidate maintenance factor, but there is no validated equation linking uncertainty to episode duration.
Does Rejection End Limerence?
Not necessarily.
Clear rejection can change the informational environment because it reduces one kind of ambiguity. It can also produce grief, rumination, counterfactual thinking, or repeated mental review. For some people, those processes may keep the person psychologically salient even after the possibility of a relationship has become clearer.
Direct research has not established a standard post-rejection limerence timeline. A person who still thinks intensely about someone after rejection should not assume that persistence proves a disorder, nor should they assume that rejection must instantly extinguish limerence.
The important distinction is between the status of the relationship and the status of the person’s attention. External certainty can arrive before internal preoccupation has faded.
Does Reciprocation End Limerence?
Reciprocation does not guarantee an immediate end either.
Classic theory treats certainty of reciprocation as one route by which limerent uncertainty can change. In real relationships, however, being chosen by the other person does not automatically remove idealization, intrusive attention, fear of loss, or the emotional significance of the bond. Some intense early romantic states gradually transform as the relationship becomes more familiar and reciprocal; others remain unstable; some relationships end while the preoccupation continues.
Modern limerence research has not produced a prospective study comparing the duration of episodes that become reciprocal with episodes that remain unreciprocated. It is therefore too strong to say that “getting together” reliably cures limerence or that unreciprocated limerence always lasts longer.
What About “No Contact”? Is There a Proven Timeline?
No. Current peer-reviewed limerence research does not establish a validated no-contact timeline.
Online claims such as “limerence fades in 30 days,” “90 days,” “3 to 6 months,” or a specific range by attachment style are not supported by a mature body of controlled limerence trials. The direct 2026 study was descriptive and observational; it did not randomize people to different contact conditions or test how quickly a no-contact intervention ended limerence (Evans et al., 2026).
Reducing voluntary exposure to cues can be an evidence-informed practical strategy when repeated checking, rereading, monitoring, or contact is obviously consuming attention. That is different from claiming that no contact has a scientifically established dose-response curve for limerence.
Context also matters. Contact may be unavoidable with a coworker, co-parent, classmate, family connection, or shared community. Safety, employment, caregiving, and legal obligations can matter more than an internet rule. The duration page therefore should not turn “no contact” into a universal protocol.
What Might Influence How Long Limerence Lasts?
The evidence is not strong enough to rank predictors of duration. It is useful to separate what is directly shown from what is plausible.
Direct limerence evidence shows that impairing episodes can be prolonged and recurrent, that intrusive thoughts can occupy a large share of waking attention in severe samples, and that those thoughts can fluctuate with negative affect, especially sadness and loneliness (Evans et al., 2026). The same study found associations with insecure attachment, adverse childhood experiences, obsessive-compulsive cognitive traits, and maladaptive daydreaming. Those are correlates of impairing limerence, not proven causes of longer duration.
The 2025 LQ-11 validation work found associations between limerence scores and several constructs, including attachment anxiety, psychological inflexibility, need to belong, and imaginative fantasy (Marshall et al., 2025). Again, association does not establish that any one of these variables prolongs an episode.
Adjacent relationship science and classic theory make uncertainty about reciprocation a plausible maintenance variable. But current direct evidence does not allow researchers to say that mixed signals, social media use, anxious attachment, or intermittent contact adds a specific amount of time.
A popular-psychology explanation often goes farther and says that limerence lasts because of “dopamine addiction,” “intermittent reinforcement,” “trauma bonding,” or “nervous system dysregulation.” Those phrases can sound mechanistic while hiding large evidentiary gaps. Current limerence research does not establish a specific addiction diagnosis, a universal reinforcement schedule, or one neurobiological pathway that determines episode length.
Why Some Episodes Feel as if They Are “Starting Over”
Intensity can fluctuate without returning literally to day one.
In the 2026 experience-sampling study, limerent cognition varied with momentary emotional states, and sadness and loneliness were linked with more intrusive-distracting fantasy (Evans et al., 2026). This supports a dynamic view: an episode can contain periods of relative quiet and periods of renewed mental capture.
A message, chance encounter, online cue, memory, anniversary, dream, or change in the other person’s availability may coincide with renewed attention. Research has not established a formal “relapse” model for limerence, so it is more precise to describe reactivation than to assume that every spike means a new disorder episode.
For someone tracking progress, trend is more informative than one bad day. Fewer intrusive thoughts over weeks, less checking, less emotional dependence on cues, and better concentration can all represent meaningful improvement even if the person still has occasional intense moments.
Is Two Years “Normal”?
“Normal” is too imprecise for what the evidence shows.
Two years is close to both Tennov’s historical average estimate and the approximate episode duration reported in the 2026 impairing-limerence sample. That makes it a legitimate research-relevant number. It does not establish a normal range for the population.
If an episode has lasted around two years, the duration alone neither confirms nor disproves limerence. It also does not tell you whether you should wait passively for it to end. Functional impact matters more clinically than whether the experience matches a famous timeline.
If sleep, work, study, caregiving, friendships, finances, safety, or daily functioning are being substantially affected, that impact is worth addressing now rather than waiting for a particular anniversary.
Can Limerence Last 5, 10, or More Years?
Long courses are possible, but the scientific literature cannot tell us how common they are.
Historical descriptions and clinical case material include prolonged fixation. Modern research demonstrates multi-year duration in severe samples but has not produced representative prevalence estimates for 5-year or 10-year episodes. Internet anecdotes can demonstrate that people report very long experiences; they cannot estimate the population distribution.
A very long course also raises a useful assessment question: is the person describing one continuous limerent state, repeated reactivations, ongoing contact, unresolved rejection rumination, a broader obsessive-compulsive pattern, depression-related rumination, grief, or another overlapping process? Those possibilities require careful differentiation rather than assuming that duration alone identifies the mechanism.
Can Limerence Recur With the Same Person?
It can appear to re-intensify around the same person, but research has not standardized how to classify that pattern.
If preoccupation fades substantially and later returns after renewed contact or changed circumstances, one researcher might call it reactivation of the original episode while another might classify it as a new episode. Until the field adopts prospective definitions, precise recurrence rates for the same target are not available.
The 2026 study’s report of approximately five episodes across adulthood is evidence that repeated episodes occur in people with impairing limerence. It does not by itself tell us how many were directed toward the same versus different people, how long the gap between episodes was, or what caused recurrence.
Does Limerence Always Fade on Its Own?
The available evidence does not justify a guarantee.
Many intense romantic states change over time, and classic limerence theory assumes that limerent intensity usually declines or transforms. But “usually” is not a clinical promise for an individual person. Long-standing preoccupation has been described, and severe functional impairment can persist.
It is more useful to ask whether the pattern is changing and whether the person’s life is widening again. Waiting becomes a poor strategy when the experience is consuming large amounts of time, driving unsafe behavior, impairing work or relationships, or co-occurring with significant anxiety, depression, compulsions, disordered sleep, or self-harm thoughts.
The absence of a formal limerence diagnosis does not make the distress unreal. A clinician can work with the actual symptoms and behaviors that are present without needing to treat “limerence” as a standalone disorder.
Can a Questionnaire Tell Me How Much Longer I Have?
No.
The LQ-11 is a validated research measure of limerence-related experience, with two main dimensions identified in its validation studies: intense need for attachment and neglect to self and others (Marshall et al., 2025). It can help researchers quantify severity or variation in the construct.
It was not validated as a diagnostic instrument, prognosis calculator, or time-to-remission tool. A high score does not mean that an episode will last longer, and a lower score does not establish that it is nearly over.
The same principle applies to online “limerence tests” that provide countdowns or attachment-style-specific timelines. Without prospective validation, those numbers are interpretations rather than established prognosis.
What Does the Evidence Say About Shortening Limerence?
There is not yet a validated limerence-specific treatment protocol supported by a mature randomized clinical-trial literature.
A 2021 case report described a cognitive-behavioral approach for one individual and illustrates how clinicians may formulate intrusive thoughts and compulsive behavior (Wyant, 2021). A case report cannot establish an average treatment effect or a standard time to remission.
The 2026 study identifies possible clinical targets such as intrusive cognition, immersive fantasy, obsessive-compulsive traits, mood, and functional impairment, but the authors explicitly call for research evaluating psychotherapeutic interventions rather than presenting an already validated protocol (Evans et al., 2026).
Evidence-informed practical steps can still be reasonable: reduce behaviors that repeatedly pull attention back to the person, create clearer boundaries where appropriate, restore sleep and daily routines, invest attention in relationships and activities outside the fixation, and seek professional help when distress or impairment is significant. These are not proven ways to cut an episode from a specific number of months to another specific number.
If symptoms meet criteria for another condition, evidence-based treatment should target that condition. Limerence research should not replace the evidence base for OCD, depression, anxiety disorders, trauma-related disorders, or other diagnosable problems.
When Is It Worth Seeking Professional Help?
The duration itself is not the best threshold. Impact is.
Professional support is reasonable when preoccupation is consuming hours of the day, sleep or appetite is persistently disrupted, work or school performance is deteriorating, relationships or caregiving are being neglected, checking or monitoring feels difficult to control, rejection triggers severe distress, or the person is behaving in ways that violate another person’s boundaries.
It is also important to seek appropriate clinical assessment when the experience includes compulsions, major depressive symptoms, panic, severe anxiety, dissociation, self-harm thoughts, psychotic symptoms, or other problems that extend beyond romantic preoccupation.
A clinician does not need limerence to be a DSM or ICD diagnosis to help with intrusive thoughts, rumination, emotion regulation, compulsive checking, grief, boundaries, sleep, or co-occurring mental health conditions.
What Research Still Needs to Answer
The duration question is one of the clearest examples of where limerence science is developing faster than popular certainty.
The next major advance would be prospective longitudinal research. Researchers would recruit people near the onset of an episode, use a validated measure such as the LQ-11 alongside careful phenomenological assessment, repeatedly measure intensity and functioning, define remission in advance, and follow participants long enough to capture recurrence.
That design could answer questions current studies cannot: median episode length, distribution of short and long courses, probability of persistence at specific time points, differences between reciprocal and unreciprocated contexts, same-person reactivation, recurrence with new people, and which variables prospectively predict change.
The field also needs treatment studies. Without them, statements that a specific intervention “shortens limerence” remain hypotheses, clinical extrapolations, or anecdotal reports rather than quantified treatment effects.
Frequently Asked Questions
How long does limerence usually last?
There is no validated universal duration. Tennov’s historical qualitative estimate was about two years, with 18 months to 3 years as a commonly cited interval. A 2026 study of 1,647 people with impairing limerence also reported episodes involving about two years of prolonged fixation on average. Neither figure establishes a population-wide deadline.
Is 18 months to 3 years scientifically proven?
No. It is a historical estimate derived from Tennov’s qualitative work and later repeated in scholarly and popular literature. It should not be presented as the result of a modern prospective population study.
Can limerence last less than a year?
Yes. There is no official minimum duration, and historical descriptions include shorter episodes. Current research cannot give a reliable percentage of episodes that end within a year.
Can limerence last longer than three years?
Yes. Long courses have been described. The evidence does not establish how common they are in the general population.
Does no contact make limerence end faster?
There is no controlled limerence evidence establishing a universal no-contact timeline. Reducing cue exposure and repetitive checking can be a reasonable evidence-informed strategy when those behaviors are maintaining attention, but it should not be presented as a guaranteed protocol with a fixed number of days or months.
Does rejection make limerence stop?
Sometimes clarity may reduce one source of uncertainty, but rejection does not guarantee immediate relief. Rumination and intrusive preoccupation can continue after the relationship possibility has become clearer.
Can limerence come back after it seems to be over?
Intense attention can reactivate, and repeated episodes occur. Research has not yet standardized the distinction between reactivation of the same episode and a new episode with the same person.
Does recurrent limerence mean I have an attachment disorder, OCD, trauma, or an addiction?
No. Recurrence alone does not diagnose any of those conditions. The 2026 impairing-limerence study found associations with insecure attachment, adverse childhood experiences, obsessive-compulsive cognitive traits, and other symptoms, but association is not diagnosis or proof of causation.
Can the LQ-11 predict how long limerence will last?
No. The LQ-11 is a research measure of limerence-related dimensions. It is not a diagnostic cutoff or a prognosis tool.
When should I stop waiting for limerence to pass on its own?
There is no calendar threshold. If the experience is significantly impairing sleep, work, study, relationships, safety, or daily functioning, or if it co-occurs with serious mental health symptoms, it is reasonable to seek help now rather than waiting for a duration milestone.
Related Articles
References
American Psychiatric Association. (2025). Updates to DSM-5-TR criteria and text. https://www.psychiatry.org/psychiatrists/practice/dsm/updates-to-dsm/updates-to-dsm-5-tr-criteria-text
Bradbury, P., Short, E., & Bleakley, P. (2025). Limerence, hidden obsession, fixation, and rumination: A scoping review of human behaviour. Journal of Police and Criminal Psychology, 40, 417–426. https://doi.org/10.1007/s11896-024-09674-x
Carswell, K. L., & Impett, E. A. (2021). What fuels passion? An integrative review of competing theories of romantic passion. Social and Personality Psychology Compass, 15(8), e12629. https://doi.org/10.1111/spc3.12629
Evans, C., Panton, S. O., Strawson, W. H., Floyd, E., Kellett, S., & Poerio, G. L. (2026). Love, longing and obsession: Features, correlates, comorbidities, and real-time cognitive-affective dynamics of limerence. Acta Psychologica, 267, 107043. https://doi.org/10.1016/j.actpsy.2026.107043
Marshall, L., Waldeck, D., Pancani, L., Churchill, S., & Tyndall, I. (2025). Development and validation of the Limerence Questionnaire (LQ-11). Psychological Reports. https://doi.org/10.1177/00332941251394980
Wyant, B. E. (2021). Treatment of limerence using a cognitive behavioral approach: A case study. Journal of Patient Experience, 8, 23743735211060812. https://doi.org/10.1177/23743735211060812
World Health Organization. (2024). Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. https://www.who.int/publications/i/item/9789240077263