A study of 743 people on both sides of a camming platform applied a forty-five-year-old relationship model to paid intimacy. Most of it held — and the part that didn’t is the interesting part.
Paying members of a camming site rate their connection to a performer as more satisfying, more intimate, and more invested-in than the performer rates the same connection.
On one measure, though, the two sides are indistinguishable: how committed they say they are to keeping the relationship going.
That result comes from research published on 23 July 2026 in Archives of Sexual Behavior, and it is the first study to measure commitment from both sides of a camming relationship rather than only from the paying side. It matters for anyone who interacts with a performer for money, because it undercuts the two stories usually told about these exchanges — that the viewer is deluded, or that the performer is purely transacting. The data support neither cleanly.
What was measured
The study, led by Rhonda N. Balzarini of Texas State University and published open access, surveyed users of LiveJasmin. The sample was 544 paying members — 512 of them men, average age 44.5 — and 199 cam models, of whom 169 were women, average age 30.7.
The two groups were recruited separately and could not be matched into pairs. Members saw a pop-up after ten seconds on the platform’s listing page; models were invited through a newsletter from the site. Each participant was asked to answer about the person they had interacted with longest.
The framework was the investment model of commitment processes, in use in relationship research since 1980, which holds that people stay in relationships when they are satisfied, when they have invested a lot, and when they rate their alternatives poorly. The researchers swapped that third variable for intimacy, reasoning that on a platform displaying thousands of available alternatives at any moment, scarcity of options could hardly be doing the work.
Where the asymmetry is — and where it stops
Members scored higher than models on satisfaction, on investment, and on intimacy, the last by the widest margin. They also rated their alternatives more highly.
On commitment, the difference was not statistically significant.
A second asymmetry runs the other way. Cam models expected members to spend more on them in future than members themselves projected spending.
The relational framing differs too, but less than the stereotype predicts. Among members, 42.5% described their interactions with a model as constituting a relationship rather than casual encounters. Among models, 31.2% said the same. That is a gap, but it is not the gap between one person in love and one person at work.
The scale each side operates at explains some of it. Members who did not describe their interactions as a relationship reported having interacted with about 14 models on average. Models in the same group reported interacting with about 144 members.
What actually predicts whether someone stays
The quality of a person’s alternatives did not predict commitment for either group. Adding it to the statistical model produced no meaningful improvement, for members or for performers.
What did predict commitment was the combination of satisfaction, investment, and intimacy — together accounting for roughly 68% of the variation among members and about half among models. For members, investment was the strongest single predictor, outweighing satisfaction, which is unusual in this literature.
The authors offer a plausible reading: on these platforms, investment is tracked and visible. Tip histories and member status are concrete in a way that day-to-day satisfaction is not. They also note the pattern is consistent with sunk-cost reasoning, where continued commitment functions partly as justification for what has already been spent.
Commitment, in turn, predicted intended future spending — though it explained under 10% of the variation in that, so it is a real association rather than a strong one.
The paper’s own summary is that the results <cite index=”145-1″>challenge the assumption that intimacy and economic exchange are mutually exclusive</cite>. Earlier work in the same area points the same way: around 60% of cam clients report a current emotional bond with at least one performer, and about half believe the performer cares somewhat about their life beyond the commercial exchange.
The thing camming has that text platforms don’t
There is a structural feature of this research worth isolating, because it does not transfer.
Camming is synchronous and on video. Whatever else is uncertain, the person on screen is the person you are paying. That is not a small thing, and it is precisely what the study was measuring — a relationship with an identifiable individual, sustained over time.
Text-based subscription platforms carry no such guarantee. In April 2025, two OnlyFans subscribers sued the platform’s parent companies, alleging they had been messaging agency staff rather than the creators they had paid. As Vice noted at the time, the plaintiffs offered suspicion rather than evidence; what is well documented is that management agencies handling creator inboxes, using either overseas staff or chatbots, are an established part of the industry.
That case did not go well for them. In a decision filed on 12 December 2025 and reported by The Blast, US District Judge Fred W. Slaughter dismissed every claim, rejecting the racketeering and wiretap arguments and finding no evidence of a coordinated scheme, while allowing the plaintiffs one attempt at an amended complaint. Counsel for one of the cleared agencies said afterwards that <cite index=”147-1″>operating a legitimate business is not misconduct</cite>.
The most useful detail for a paying viewer is the reason the court gave. OnlyFans’ terms already state that creators may use agents to manage their accounts, that transactions run between fan and creator rather than through the platform, and that the platform makes no guarantee about how creators handle messages. The plaintiffs could not claim deception about something disclosed in writing.
Which makes the practical step an unglamorous one: read how a given platform actually structures interaction before spending on it — whether third-party account management is permitted, who transactions are with, what the refund position is. Review sites such as dude-hack.com exist to document those differences, and the terms themselves are public.
What the study cannot tell you
It is cross-sectional, so it cannot establish direction. Whether intimacy produces commitment or commitment produces intimacy is unresolved, and Balzarini has been explicit that the data do not show performers becoming committed because they anticipate payment.
Participants were not matched, so nobody’s answers can be compared against their counterpart’s. Core measures used single items rather than full scales, a trade-off the authors make openly to keep the survey short enough that working performers would finish it. The sample skewed heterosexual and gender-conforming. And 199 performers is a modest base for the statistical modelling involved.
The authors’ next step is longitudinal work, following these connections as they form and end rather than photographing them once. Until that exists, the honest summary is narrow: on one large camming platform, in 2026, the same three factors that hold ordinary relationships together also predicted who stayed — on both sides of the payment.