Pocketing: What the Word Means, and What the Research on Hidden Relationships Actually Shows

What the search engines answer, and where that answer came from

We put the question in the title to three search engines on 21 October 2026 and wrote down what came back, because what came back is what most readers will have seen before arriving here.

Google returned the unusual-traffic challenge and contributed nothing. Brave and DuckDuckGo both answered with a generated summary placed above any page. Brave's opens with a definition — one partner deliberately hides the other from their friends, family, social circles, and public identity, including social media — then supplies motives (fear of rejection, relationship ambivalence, avoidant attachment styles, or a desire to maintain independence and keep alternative options open) and finishes under a heading of its own, Diagnostic Value: the behaviour reveals a partner's limited conflict tolerance, emotional maturity, and capacity for vulnerability, often signaling that they do not envision a long-term, public future with the partner.

That last sentence is not a synthesis of research. It is a close paraphrase of one page sitting on the same results screen — an article published in June 2026 on a therapy-booking platform, which exists to sell sessions. DuckDuckGo's shorter summary names that same page, and Psychology Today, as its two sources.

Underneath the summaries: Psychology Today at the top of both engines, Medium, PureWow, Marriage.com, Bonobology, NBC News, Quora, Yahoo, MSN, LADbible, four Reddit threads, several therapy practices, a dating-app glossary, six short videos and two Instagram posts. No Healthline, no Cleveland Clinic, no Verywell Mind — the same split this site has measured a dozen times, in which the explanatory questions of this subject are left to lifestyle media and forums. Brave's own follow-up prompts, offered under People also ask, are two invitations to book therapy and one real question: how do I know if my partner is actually hiding me or just a private person?

One page in the set cites studies — Psychology Today's, and a later section works through what those studies measured. The rest cite nothing. That is the gap this page is for, because the behaviour the word names has been measured for thirty years under a different name, and some of what was found runs the other way. The precedent is recent: the same engines recently supplied a causal mechanism for the honeymoon phase ending sourced to a podcast and to TikTok, and a summary of why people stay in unhappy relationships that inverted the conclusion of the study it was paraphrasing.

The word has an origin, and it is a newspaper column

Pocketing is usually introduced as something researchers have identified. It is a coinage, and the coinage has a paper trail.

The sibling term stashing is the one that can be dated. Dictionary.com's slang entry, published on 19 October 2018, credits it to the journalist Ellen Scott, in a column for the UK tabloid Metro in August 2017, and records that Cosmopolitan and others picked it up as a buzzword afterwards. Pocketing itself has no origin that any dictionary has pinned down, so we are not going to invent one. The earliest mainstream American coverage we can find is NBC News, 8 July 2019, by Danielle Page, and it establishes the template every page since has followed: a definition, a list of signs, and two practitioners quoted. No study.

There is one detail in that 2018 dictionary entry worth preserving, because it has been edited out of the 2026 versions. It lists four reasons a person might stash a partner: wanting to keep their options open, embarrassment, viewing the partner as temporary — and concern about the reaction of friends and families based on religious or cultural differences. The fourth is structural rather than characterological. It is about a family, not about a person's capacity for vulnerability. It has disappeared from the generated summaries and from most of the first page, and a later section is about what the research says happens to couples in exactly that position.

None of this makes the word useless. A coined term can name a real thing; ghosting did. It means two separate questions have to be kept apart. Does the behaviour exist and has anything about it been measured — yes, and that is most of this page. Does the word carry a diagnosis that research supports — that is a different question with a different answer.

This site has worked through the same structure twice before, with the three-day rule and the 3-6-9 rule. The pattern repeats: a term with no academic origin acquires an explanatory mechanism on the way round the internet, and nobody goes back to check whether the mechanism was ever measured.

The behaviour does have a literature. It is filed under secrecy

Search the research databases for pocketing and you get surgery and dentistry. Search for romantic secrecy or relationship concealment and a literature appears that runs from 1994 to the present.

The construct is narrower than the dating term and that narrowness is useful. Romantic secrecy, as it is measured, is the extent to which one or both partners conceal the existence of the relationship from third parties: family, friends, colleagues, and more recently the people who would see a social media profile. It is a property of what is shown outward.

It is worth separating from a neighbour it is constantly confused with. Keeping secrets from your partner is a different variable with a different literature and different results — Catrin Finkenauer and Harry Hazam found disclosure and secrecy within a marriage contributing to satisfaction independently, which is covered in what the research says about how people open up. That page is about information withheld inside the couple. This one is about the couple itself being withheld from everyone else. The two come apart: a pair can tell each other everything and tell no one else anything.

The other thing to fix before going further is the asymmetry the dating term builds in. Pocketing names a one-sided act — someone does it to someone. The research measures secrecy at the level of the relationship, usually reported by one partner about the relationship as a whole, and in at least one study the partners of concealing people were themselves less visible. That does not mean both people chose it. It means the measurement cannot tell you who chose it, which is a limit worth carrying through everything below.

What the literature can tell you is what concealment co-occurs with, what it appears to do over short stretches of time, and — in one memorable case — that the single most repeated claim about secret relationships was tested directly and did not survive.

The famous finding about secret relationships did not hold up

If you have heard anything about hidden relationships from a research angle, it is probably this: secrecy makes attraction stronger. It comes from one paper and it is a good one.

Daniel Wegner, Julie Lane and Sara Dimitri (1994), in the Journal of Personality and Social Psychology, ran two surveys and a laboratory experiment. The surveys found that past relationships people still thought about were more likely to have been secret, and that respondents who described a past relationship as secret also described continuing obsessive preoccupation with it. The experiment is the part everyone remembers: mixed-sex couples were induced to play footsie under a table in the presence of another couple, and when the contact was secret, participants reported greater attraction to the partner. The authors' account was that secrecy produces a cognitive preoccupation which gets read as attraction.

Then it was tested on actual ongoing relationships. Justin Foster and Keith Campbell (2005), in Personal Relationships, predicted the opposite — that secrecy interferes with the interdependence a relationship runs on — and ran three studies. All three confirmed the prediction. In Study 1, secrecy predicted lower initial relationship quality and decreasing quality over a two-week period. In Study 2, relationship burden partially mediated the negative association. In Study 3, the association was explained primarily by decreased satisfaction.

That leaves an obvious reconciliation available: secret is thrilling at first and corrosive later. Hot, then not. Foster, Foster and Campbell (2010) tested exactly that, in The Journal of Social Psychology, on an internet sample of 564 people, predicting a positive influence in newer relationships and a negative one in older ones. Their own summary of the result: romantic secrecy was a burden to both new and old romantic relationships. The compromise position failed too.

So the one claim in this area that a reader is likely to have encountered as science — that hiding a relationship intensifies it — rests on a laboratory manipulation of concealed foot contact between strangers, and has twice failed to transfer to relationships people were actually in. The shape is familiar from the 7-38-55 figure and from the Romeo and Juliet effect, which did not replicate either.

What concealment does, in the study that modelled the mechanism

The most directly relevant paper is Justin Lehmiller's (2009) in Personality and Social Psychology Bulletin, and it is the one the first page of results should be built on and is not.

Two studies. Study 1 found that greater secrecy was associated with reduced commitment to the relationship, lower self-esteem, and more reported physical health symptoms. Study 2 built a structural equation model of how that happens, and the model fitted. Its two paths are worth stating precisely because both are more specific than the folk account.

First, secrecy undermines commitment by constraining cognitive interdependence — by limiting the degree to which a person's thinking comes to include the partner. That is a mechanical claim rather than a moral one. The mental shift the commitment literature calls inclusion of other in the self is built partly out of talking about the person, planning around them, and being treated by others as half of a pair. A relationship that cannot be mentioned has fewer of the occasions on which that shift is normally assembled.

Second, secrecy threatens personal health by generating negative affect — nervousness and fear specifically. Not sadness, not anger: the affect of possible exposure.

The caution matters as much as the finding. This is self-report, and the modelling establishes that a causal story fits the data, not that it is the true one. The reverse account is entirely live: people may conceal relationships that were already shakier, in which case low commitment produces the secrecy rather than the other way round. Lehmiller says as much about the limits of the design. And none of it licenses a reading of the concealing partner's character, which is the step the search summaries take and the papers do not.

What it does establish is that concealment is not neutral and not merely hurtful feelings. It is associated with the commitment variable itself, by a route that makes sense, and with the concealed person's reported health.

The base rate the checklist leaves out

Every version of this list contains the same item: they never post you on social media. It is treated as the easiest sign to check, which it is, and as informative, which depends entirely on a number nobody quotes.

Pew Research Center surveyed 4,860 U.S. adults between 16 and 28 October 2019, with a margin of error of ±2.1 points. Among social media users, 28% said they had ever used these platforms to share or discuss anything about their relationship or dating life. Among users under 30, 48%.

Read that back against the sign. On the most generous reading, roughly seven in ten adult social media users have never posted about their romantic life at all. A criterion that flags they do not post about me flags, in the general population, the large majority of people, nearly all of whom are not concealing anything. It is the lowest-specificity item on every list and it leads every list.

The survey also explains why it feels otherwise. 81% of social media users said they at least sometimes see other people posting about their relationships, and 46% said they see it often. Both things are true at once, and they are not in tension: the feed is composed of the minority who post, because the people who do not post are invisible by construction. The impression that everyone documents their relationship is an artefact of the only people you can see being the ones who do.

One qualification in the other direction, from the same survey. Among partnered social media users aged 18 to 29, 48% said social media is very or somewhat important to them for showing how much they care about their partner. For that group the channel does carry meaning, and a partner who knows it carries meaning and still withholds it is doing something more pointed than a partner who never thinks about the platform. The sign is not empty. It is just far weaker on its own than the lists imply, and it is the only one most people can check.

Visibility goes up with insecurity, not down — which breaks the inference

The reading everyone performs is: visible means committed, therefore invisible means uncommitted. There is a study designed to test what relationship visibility actually tracks, and it complicates both halves.

Lydia Emery, Amy Muise, Emily Dix and Benjamin Le (2014), in Personality and Social Psychology Bulletin, ran three studies on what they named relationship visibility — the centrality of the relationship in the self-image a person conveys to others — using Facebook as the setting. Three results.

Attachment avoidance went with low desired visibility; attachment anxiety went with high desired visibility. So the disposition that produces the behaviour is a known dimension, measurable, and not specific to this relationship or this partner. If you want the caution about treating those dimensions as types rather than as two independent continua, the evidence on whether attachment styles change has it at length, and what the research says about anxious attachment and love covers telling the feeling apart from the pattern.

Both avoidant individuals and their partners were less likely to make the relationship visible. Visibility behaves as a property of the pair, not only of the person withholding it.

And the one that breaks the inference. In a daily study, on days when people felt more insecure about their partner's feelings, they made the relationship more visible. Not less. The posting went up with the doubt.

Take that seriously and the sign runs backwards from how it is used. On any given day, a more visible relationship may be one in which somebody needed reassurance, and a quieter one may be a day on which nobody needed anything. Across people, in the same studies, avoidance also predicts low visibility, so the signal is genuinely mixed rather than reversed. What it is not is a clean readout of commitment, and the confidence with which it is read as one is unearned.

There is one more thing in the same paper that the lists cannot accommodate: in Study 1, similar motives drove the actual visibility of both the anxious and the avoidant groups, despite their opposite stated desires. People arrive at the same observable behaviour from different places.

What a visible relationship buys, and what it costs the person posting

The sequel to that work asked a question nobody asks: what do other people take from a visible relationship, and is it free?

Emery, Muise, Emily Alpert and Le (2015), in Personal Relationships, ran two studies on how accurately outsiders read relationship quality off a Facebook profile, and on what those profiles did to their judgements of the people in them.

The first finding supports the folk intuition. Perceived relationship quality corresponded to self-reported relationship quality, and people with more visible relationships were perceived as having greater relationship satisfaction and commitment, and as being more likable. There is real signal in visibility, which is why reading it is not irrational.

The second finding is the one worth the trip. High disclosure about the relationship predicted greater perceived relationship quality but lower likability. Posting more about your partner makes observers rate the relationship better and rate you worse.

That is a cost, measured, in the currency the behaviour is performed in. It supplies a reason for low visibility that appears nowhere on the first page of results and is not a character flaw, a hedge, or ambivalence: some people have correctly worked out that heavy relationship posting makes them less liked, and have declined to pay for a reputational benefit that accrues to the relationship rather than to them.

Two limits. These are profile-based judgements by strangers and acquaintances, not partners, and the correspondence between perceived and self-reported quality is a correlation across a sample rather than evidence that any particular reading is right. And it is Facebook in the mid-2010s, which is not the platform most of this behaviour now happens on.

Still, between this and the previous section the picture has changed shape. Relationship visibility is a self-presentation behaviour with its own incentives, its own costs, and its own dispositional predictors. Treating it as a thermometer pointed at somebody's feelings is a category error, and it is the error the entire genre is built on.

The real cost of being hidden, and nobody writes it

There is a measurable thing that concealment takes away, and it is not approval. It is better than approval, and as far as we can find it has never appeared in a page on this subject.

Christopher Agnew, Timothy Loving and Stephen Drigotas (2001), in the Journal of Personality and Social Psychology, ran a three-wave longitudinal study on something awkward: what a couple's friends think about the relationship, measured directly from the friends. Three findings.

The network's view was significantly more negative than the couple's own, consistent with the within-dyad idealization that is one of the better-replicated findings in this field. People inside a relationship rate it higher than everyone around them does.

Both the couple's perceptions and the network's predicted the relationship's fate. So far, unsurprising.

And then: friends of the female partner were particularly successful at predicting dissolution. A specific slice of the surrounding people outperformed, as forecasters, the two people in the relationship.

The mechanism the authors examined and found support for is the part that matters here. The friends' predictive edge was linked to couple-disclosure — to how much they had been told. Their accuracy is not mystical insight; it is information, supplied by the couple, processed without the idealization the couple cannot switch off.

Read that into a concealed relationship and the loss becomes concrete. Being hidden does not merely deprive you of a partner's family liking you. It switches off the observers who were measurably better calibrated than you are about where this is going, and it switches them off by removing their input, which is disclosure. The person most deprived of the signal is the person inside the relationship, on both sides.

This is one study, in a North American sample, on heterosexual couples, twenty-five years old. It is also the only one we can find that measured the network's accuracy rather than its approval, and it reframes the question from do they respect me enough to introduce me to who is currently in a position to tell either of us anything useful.

Being introduced is not the same as being approved of

The network literature proper — whether friends and family liking a relationship keeps it alive — is covered on this site in the piece on signs someone is falling for you, including Diane Felmlee's hazard analysis of 290 daters and the result that complicates the easy version, where approval from a person's own family ran the other way in the full model. This page does not restate it.

What that page does not have, and what bears directly on the timing question, is Susan Sprecher and Felmlee's (2000) five-wave study in Personal Relationships of how network perceptions change as a relationship develops. Four things from it.

Perceived network approval increased over time among participants whose relationships stayed intact, and most clearly from the male partner's friends.

The transitions to engagement and to marriage were associated with an increase in approval from the male's friends — that is, approval tracked the milestone rather than preceding it.

A breakup was significantly less likely the more approval women reported from their friends, and the more women liked their partner's family at Time 1. Both associations were nonsignificant for men.

And the finding nobody quotes: people who broke up during the study reported more network approval than disapproval for the breakup. The network has an opinion about the ending too.

The first two of those do something quiet to the standard reading. If approval rises over time and rises again at formal transitions, then network integration is at least partly downstream of commitment rather than a declaration of it. Michael Parks, Charlotte Stan and Leona Eggert (1983) found the same shape in 193 premarital relationships: romantic involvement was positively associated with communication with the partner's network and with the number of people met in it, among other network variables, and the relationship between social support and romantic involvement was predominantly positive and linear. Involvement and integration grow together.

That is still a real inference — people do tend to integrate partners they expect to keep — but it is an inference about a gradual process with no step in it, not a test a partner either passes or fails on a given weekend.

What the network's opinion predicts, and the moderator that inverts it

One more study belongs here, because it measured the thing the genre is actually claiming and found a condition on it.

Paul Etcheverry and Agnew (2004), in Personal Relationships, worked with subjective norms: not what the network thinks, but the perceived beliefs of a network member weighted by how motivated you are to comply with that person. It is a sharper instrument than approval, because it builds in the fact that some people's opinions move you and some do not.

In a longitudinal study of college students in dating relationships, subjective norms predicted commitment — on their own, and still when satisfaction, quality of alternatives and investment size were in the regression alongside them. Longitudinally, commitment mediated the effect of subjective norms on still being in the relationship around eight months later.

Then the moderator, and it is the kind of result that gets left out of summaries because it is inconvenient. Level of dependence on the relationship moderated the predictive value of subjective norms, with lower dependence yielding the greatest predictive value. The surrounding people's opinions predicted most strongly where the person needed the relationship least.

Turn that over and it says something specific about the situation a concealed partner is in. The network's view is a strong predictor in the relationships that matter least to the person holding it, and a weaker one in the relationships they most depend on. Somebody deeply invested in a relationship their friends would disapprove of is, in this data, relatively insulated from the disapproval — which is one plausible reason concealment persists even when it is costly, and a reason to be sceptical of the advice that just introducing everyone would resolve things.

Sample limits apply and are the usual ones: college-aged daters, self-report, eight months. What the study establishes is modest and real. The network's opinion is in the commitment equation. It is not a dominant term, and its weight depends on how much the person needs the relationship — not on how good the relationship is.

The timing question has been investigated once

The question underneath nearly every search for this term is a number. How long is too long before I meet their friends? Before I meet their family? Is six months normal?

There is one study. Maryanne Fisher and Catherine Salmon (2013), in the Journal of Family Studies, set out to examine the introduction of a partner to one's parents, and their own abstract states the position plainly: although the introduction is important, there had been no prior investigation into its timing, or into the reasons for it beyond signalling commitment.

What they found supported their hypotheses about motive. People are motivated to bring a partner home to seek parental feedback and approval, and to indicate to the partner that they are serious. In the other direction, people want to meet a partner's parents for cues about how the partner will look when older, their future health, and the family resources that may be available. The frame is evolutionary and the mechanism is ordinary: the introduction is an information-gathering event in both directions, not only a status announcement.

One further result is directly relevant to the generated summaries. They examined attachment style and birth order as influences on the timing of the introduction, and found minimal influence for either.

That sits interestingly beside Emery and colleagues, who did find attachment avoidance predicting low relationship visibility on social media. The reasonable reading is that these are two different behaviours that the dating term has welded together: what you put on a profile and when you take someone to meet your mother are not the same act, and the one study that looked found attachment barely moving the second.

What we cannot give you is the number, and it is worth saying why rather than papering over it. The paper is paywalled, has been cited once, and we could not obtain the timing figures. The top-ranking page on this query cites this same study and also gives no number. So there is no established normal interval, and anyone offering you one — three months, six months, a year — is not citing anything. They are telling you what they would do.

“They are just a private person” — what actually separates the two

This is the one genuine question in Brave's follow-up prompts, and it deserves a straight answer, including the part where the straight answer is nobody has measured this.

No study we can find has compared partner-specific concealment against general low disclosure in the same people. That comparison is the whole question and it has not been run. What exists are three things that bear on it, and they are worth having in the right order.

The measurement that produced the findings on this page is specific. Lehmiller's secrecy scales, and Foster and Campbell's, ask about concealing the relationship, not about a person's general reticence. The commitment and health associations attach to the first thing. Nothing in this literature says that a quiet person is a hazard.

The base rate cuts the same way. Seven in ten adult social media users have never posted about their romantic life, so absence of posting is close to uninformative on its own and the people arguing they are private are, statistically, usually right.

And visibility is partly a couple-level outcome, not one person's decision — Emery and colleagues found the partners of avoidant people also less visible.

Which leaves a practical distinction that follows from the evidence rather than from anyone's opinion: the diagnostic contrast is not visible versus invisible, it is uniform versus patterned. A person whose entire life is low-disclosure — no posts about anyone, few introductions of anyone to anyone, a family nobody in their life has met — is doing one thing. A person whose life is otherwise documented and populated, with one hole in it shaped exactly like you, is doing another. That second pattern is what the word was coined to name.

It still is not a diagnosis, because the hole can be produced by a family, an employer, an immigration status, a custody arrangement or a community, and the next section but one is about what the research says happens to couples in that position. But it is a better question than do they post, it is answerable from information you already have, and it does not require reading anybody's character.

Why they will not simply explain it

The second question people arrive with is why the conversation goes nowhere. Two findings make it less mysterious and considerably less personal.

Leslie Baxter and William Wilmot (1985), in the Journal of Social and Personal Relationships, conducted 90 ethnographic interviews asking what was off limits in a current opposite-sex relationship. They found six types of taboo topic, and the state of the relationship was the most pervasive of them all. The reasons people gave are the useful part: informants held a negative view of relationship talk as destructive, inefficient, futile and risky. Not I have something to hide. This conversation makes things worse.

So the single hardest thing to raise, in the general population of people in relationships, is the thing a concealed partner most needs to raise. The silence is the modal case, not a tell.

Leanne Knobloch and Kathryn Carpenter-Theune (2004), in Communication Research, put a shape on it with 216 people in romantic relationships. Topic avoidance had a convex curvilinear association with intimacy — most avoidance at middling closeness, less at both ends. Relational uncertainty was positively associated with avoidance, and uncertainty mediated the link between intimacy and avoidance.

Read that onto the situation this page is about. Someone being kept out of a partner's life is, almost by definition, at moderate intimacy with high uncertainty — the exact coordinates where the research says topic avoidance peaks. The conversation is hardest to have precisely where it is most needed, and that is a structural feature of relationship development rather than evidence about one person's character. The broader uncertainty literature, including the predictions its own authors made and failed to confirm, is set out in what has actually been measured about undefined relationships.

None of which means the conversation should not happen, and the research has nothing against asking. It means that they got defensive when I brought it up is weak evidence of anything, because defensiveness about the state of the relationship is what the baseline looks like.

The half the genre deletes: when concealment is not about you

This is the section the first page of results does not have, and it is the one with real stakes.

Some relationships are hidden because showing them is punished. Interfaith and interracial couples, same-sex couples, partners whose families have told them which people are acceptable, people in communities or countries where the relationship is a risk rather than an embarrassment. For those couples, pocketing misnames the situation: what is being measured is a family's or a community's behaviour, not a partner's maturity.

And that is measurable. Lehmiller and Agnew (2006), in Personality and Social Psychology Bulletin, tested marginalization inside the investment model. Marginalization was a significant negative predictor of commitment, and people in marginalized relationships invested significantly less than people in non-marginalized ones. Despite investing less, they were significantly more committed — and the compensation ran through reduced perception of alternatives rather than through greater satisfaction. A relationship the world disapproves of narrows the world.

Kristin Rosenthal and Tyrel Starks (2015), in the Journal of Family Psychology, surveyed 480 U.S. adults in interracial and same-sex relationships. Above and beyond individually experienced discrimination and the standard predictors, relationship stigma from friends in particular was associated with lower commitment, trust, love and sexual communication, and with greater odds of intimate partner aggression victimization. Egalitarianism and dyadic coping moderated some associations. In a follow-up with 467 adults, stigma from friends was associated with greater depressive symptoms and, indirectly through them, with poorer self-rated health.

Karen Blair and Diane Holmberg (2008) supply the other side: among 183 people in same-sex and 275 in mixed-sex relationships, support specifically for the relationship predicted relationship well-being, which predicted mental and physical health — with no difference in model fit between the two groups, and holding when general social support and support for one's sexual orientation were controlled. What concealment withholds has a measured value, equally, for both kinds of couple.

So the cost is real on both sides, which is why the next section is about the choice rather than the verdict.

Being out is not unconditionally better, and that has been measured too

The obvious conclusion from the previous section is that the hidden partner should push for visibility, since concealment costs commitment and support helps. The one study that tested whether disclosure is unconditionally good found that it is not.

Nicole Legate, Richard Ryan and Netta Weinstein (2012), in Social Psychological and Personality Science, asked lesbian, gay and bisexual participants to rate how autonomy-supportive versus controlling five separate contexts were — family, friends, coworkers, school, religious community — along with how out they were in each, and their context-specific self-esteem, depression and anger. Multilevel modelling of the within-person variation found two things.

People were more likely to disclose in autonomy-supportive contexts. And disclosure was associated with better well-being in autonomy-supportive contexts, and was not in controlling ones.

That is as close as this literature comes to adjudicating the question, and the answer is conditional. Being open about a relationship is associated with doing better where the surrounding people are not coercive. Where they are, the association disappears. The benefit is not in the act of disclosing; it is in disclosing somewhere that can hold it.

The other half of the ledger is also measured. John Pachankis (2007), in Psychological Bulletin, assembled the evidence on carrying a concealable stigma into a cognitive-affective-behavioural model, and the summary is that people with a hidden stigma do not escape the difficulties of a visible one: the ambiguity of social situations combined with the threat of discovery is its own sustained stressor. Concealment is not free either.

What this does to the page's subject is specific. A partner who will not introduce you to a hostile family is not failing a test of emotional maturity; they are making a trade between two costed options, and at least one study suggests the option everyone recommends is only beneficial under a condition they may not have. The honest question is not whether they should be braver. It is what the arrangement is costing you, which is a separate matter and is the one you can answer.

The diagnostic claim, and what would have to exist for it to be true

Back to the sentence this page started with: that the behaviour reveals a partner's limited conflict tolerance, emotional maturity, and capacity for vulnerability.

For that to be a research finding, somebody would need to have measured concealment and those three constructs in the same sample and found them associated. As far as we can determine, nobody has. Conflict tolerance has instruments. Emotional maturity, as this site has set out at length, barely survives being operationalised at all. Neither has been put alongside relationship concealment.

What has been measured is weaker, more specific, and more useful: concealment co-occurs with lower commitment and is associated with the concealed partner's reported health; attachment avoidance predicts low relationship visibility while attachment barely moved introduction timing in the one study that looked; network approval and subjective norms sit in the commitment equation with modest weights and conditions attached. None of those is a reading of a person.

It is worth being fair to the one page that cites anything. Psychology Today's May 2024 piece — first result on both engines — links five studies, which puts it ahead of everything else on the query. Two of them are used in ways that outrun what they measured. Its support for the claim that such a lack of trust can exacerbate emotional instability and conflict is a 2021 cross-sectional survey of 329 married people in which lack of trust predicted relationship problems with cell phone snooping as a partial mediator: a different behaviour, a different population, and a mediator that has nothing to do with being hidden. And its support for relationship ambivalence is a 2020 thematic analysis of 35 interviews about how parents influence romantic development — qualitative, about role models, not about concealment.

The Fisher and Salmon citation is the exception and the irony: it is genuinely the right paper, and the page cites it for the uncontroversial half (introductions are a turning point) while passing over the half that would have answered its readers' question, which is the timing.

The pattern across this whole query is not that nobody knows. It is that the relevant studies exist, several of them say something narrower or sideways to the claim being made, and nobody checked.

Keeping options open, and what cannot be read from outside

One of the four motives in the generated summary is a desire to maintain independence and keep alternative options open, and it is the one most people are really asking about. Is there somebody else.

The honest answer is that no study has measured how often a concealed partner has a concurrent one, and there is no obvious way to run it. Concealment is, by construction, hard to sample.

What has been measured is that keeping options in reserve is ordinary rather than exotic. Jayson Dibble and Michelle Drouin (2014), in Computers in Human Behavior, introduced back burners — people kept in contact with in case a current relationship does not work out — as a construct and analysed them through the investment model. The behaviour exists at scale, is conducted through the same channels as everything else, and is not confined to the uncommitted.

The adjacent thing this site has covered in depth is a better fit for most of these situations anyway. What the research on situationships shows sets out asymmetric commitment as it was measured in 315 couples with both partners reporting: around a third of those relationships had a substantial gap between the two people's commitment levels. A hidden relationship is one of the shapes an asymmetric one takes, and asymmetry explains most of the observable features without requiring a second partner to exist.

Which leaves the uncomfortable conclusion. From where the hidden person is standing, a partner who is ambivalent, a partner who is avoidant, a partner with a hostile family, a partner protecting a version of themselves they have constructed, and a partner with someone else all look identical. The behaviour does not discriminate between them, no measurement discriminates between them, and the lists that claim to are selling confidence. The only instrument that distinguishes them is the person, which is why the research — here as in every other ambiguity this site has written about — keeps arriving at the same unglamorous recommendation: ask, specifically, about the thing itself, and treat the answer as data rather than as a verdict.

When this is not pocketing at all

If a partner is restricting who you see, monitoring where you are, controlling your money or your transport, going through your phone, continuing after you have said no, or punishing you with silence or anger, that is not the subject of this page and nothing here applies to it. Find A Helpline, in the sources below, lists free and confidential services by country. In the United States the National Domestic Violence Hotline is 1-800-799-7233, or text START to 88788, 24 hours a day.

The reason this needs saying is visible in the search results themselves. The highest-voted Reddit thread for this query is titled “Boyfriend is Keeping Me From Friends Family”, and it is filed under pocketing — but the title describes two opposite things at once, and the distinction is the most important one on this page.

Being excluded from their network is what the word names. You have not met their friends; their family does not know about you; their profile does not have you in it. Everything above is about that.

Being separated from your own network is a different behaviour with a different literature. It is one of the two factors that emerged when Richard Tolman (1989) developed the Psychological Maltreatment of Women Inventory on 407 men and 207 women at intake to a domestic violence programme: factor analysis produced a similar structure in both, and the factors were dominance-isolation and emotional-verbal abuse. Isolation from one's own people is not an unfortunate side effect of a relationship being private. It is a measured dimension of psychological abuse, named as such in the instrument the field uses.

The two can occur together, and when they do the second is the one that matters. A relationship in which you have met nobody of theirs and see less of yours than you used to is not a privacy preference. The piece on jealousy works through where the line sits and why naming the behaviour matters more than measuring the feeling.

One more, briefly, because the research says the period around leaving carries elevated risk: if any of this is familiar, safety planning is work to do with a service rather than from an article.

What none of this can tell you

The limits belong in the article, because the most common use of a page like this is to get an answer it cannot give.

It cannot tell you why your partner is doing it. That is the whole question, and nothing above discriminates between ambivalence, avoidance, a hostile family, a constructed self-image and another relationship. No measurement does.

The causal direction on the central finding is unestablished. Lehmiller's model fits; it is not proven. Concealment may erode commitment, or low commitment may produce concealment, and the available designs cannot separate them. The same applies to every association on this page.

There is no measured normal interval. One study has looked at the timing of partner-to-parent introductions, it is paywalled, and we could not get the figure. Any timetable you are given is somebody's preference.

The partner-specific versus private-person comparison has not been run. The distinction this page offers — uniform low disclosure against a life with one hole in it — follows from the evidence rather than from a study designed to test it, and it is offered as the better question rather than as a finding.

The samples are narrow. North American, heavily student and online-panel, mostly mixed-sex except where noted, and several of the key papers are between fifteen and forty years old. The Facebook studies measure a platform most of this now happens off.

What survives is smaller than a verdict and more use than a checklist. Concealing a relationship is associated with less commitment and with worse reported health in the person concealed, by a route that makes sense. Not posting is nearly uninformative on its own. Visibility is a self-presentation behaviour with its own costs, and it goes up with insecurity as well as with security. Integration tends to follow commitment rather than announce it. The friends who were measurably best at predicting how a relationship would end were the ones who had been told things. And where a relationship is hidden because showing it is punished, the research describes a cost on both sides and takes nobody's part.

What can be answered is the smaller question: what this arrangement is costing you, which is measurable from where you are standing. Our relationship satisfaction test prints satisfaction, alternatives and investment as separate numbers rather than a verdict, and the attachment style test scores two dimensions and says in the page what it is not. More of this under Dating, and the claims this site has had to retract are listed on the corrections page.