How 747 students read a cannabis ad predicted their intentions better than the ad itself

Interpretation is doing the work in this study, and interpretation is the one thing nobody assigned. The researchers chose which real Instagram ad each participant saw ^[5], but they had no control over whether that participant came away believing the ad promised better sex, and it is that belief which tracked with expectancies and with saying they might use cannabis before sex ^[2]. A student who already intends to use cannabis before sex has every reason to read a couple in bed, or friends around a bonfire, as endorsement ^[6]. The correlation is compatible with the ad producing the belief, the belief coloring the ad, or something else producing both. Which one happened is what this data can't settle.

Even so, the design has real value. The usual approach in this literature asks people to recall cannabis marketing they have already encountered, which tangles exposure with what a person noticed and retained; showing everyone the same stimulus removes that tangle and is what makes comparing two readings of one ad possible at all ^[5].

The gendered results are where the practical content sits. Men who wanted to be like the men in the sexual and romantic ads, described as physically fit, attractive and masculine, reported stronger intentions to use cannabis before sex ^[7]. Women who read consent concerns into the ads reported weaker ones ^[8]. Hust's inference is that messages spelling out how cannabis may affect decision making and consent could lower intentions, particularly among women ^[12]. That points at a message worth testing rather than a clause worth drafting.

On the regulatory side, the asymmetry is plain. Washington keeps cannabis ads away from under-21 audiences and says nothing about claims that cannabis improves sex or romance ^[9], and there is no nationwide cannabis code of the kind alcohol marketers wrote for themselves ^[10]. The spirits comparison also carries a timescale worth doing the arithmetic on: voluntary guidelines arrived shortly after prohibition ended in 1934, and the prohibition on linking alcohol to sexual success is dated to at least the mid-1990s ^[11], roughly sixty years later ^[13]. Willoughby's point is narrower and harder to argue with, that there are no rules saying you cannot tie cannabis to sex ^[4].

What a rulemaking record would want from a study like this is a magnitude. The announcement reports directions of association without effect sizes or per-condition sample counts ^[16], and every outcome is a stated expectation or intention rather than a measured behavior ^[14]. The sample is 747 college students in a single legalized state, recruited online ^[1].

My read: the vacuum is well documented, and the proposed mechanism is plausible. What's still missing is the causal step, because the variable regulators would restrict is the one the design controlled, and it did not consistently move the answers ^[3]. A follow-up that assigns a sexual-success claim and a consent counter-message, then measures what people do, would settle it.