What alcohol actually does is narrower and stranger than a simple slide into incompetence. In experiments where people are given controlled doses, sexual decision-making measurably suffers: a systematic review of 43 randomized experiments concluded that acute alcohol causally increases sexual risk decisions, and in one laboratory scenario, intoxicated women were more likely than sober ones to let the partner decide how far things go. The best-known account of why — labeled "alcohol myopia" in 1990 — is that alcohol narrows attention to whatever is most salient in the moment.

That mechanism matters, because it means the effect is situational, not a uniform decline. With strong inhibiting cues present, intoxicated people have even reported more cautious intentions than sober ones; at low doses, some consent communication actually becomes more explicit. The impairment is real — but it is uneven, and it doesn't announce itself. Which is precisely why no drink count can certify "they're still fine," and why the abilities below are the thing to look at, not the glass.

The second catch: the person drinking can't audit this from the inside. In a field study at real drinking events, participants averaging a breath-alcohol level around .075 — already in the range where measurable impairment appears — were asked whether they could consent to sex. 92.6% said yes about themselves, and their self-ratings had no relationship to their measured levels. Under alcohol, "I'm fine" is not data — and "they seem fine to me" is the same sentence in the second person.

Exhaustion does similar work by a different route. Sleep-loss research finds large, reliable impairments in attention and alertness, consistent damage to executive functions — working memory, inhibition, flexible thinking — and specific harm to non-routine decisions and effective communication: exactly the ingredients a live "yes" requires. One honest limit: on familiar, rule-based tasks, tired people often hold up well, so exhaustion doesn't feel like drunkenness. And research hasn't studied fatigue and consent directly — "an exhausted yes is weaker" is an inference from the cognitive evidence, not a finding. It is still a better-supported inference than "tiredness doesn't matter."

None of this is a fringe scenario. About half of sexual assaults involve drinking by at least one of the people involved — a number from assault research, quoted here only to say these questions are common, not to characterize your night out. On the everyday side, students describe plenty of alcohol-involved sex they consider consensual, usually judged through a chain of cues: who initiated leaving the party, what was said earlier, sometimes a sober look back the next day. "We'd both been drinking" is one of the most common states in which real people try to give and read consent — which is why the question deserves a real answer rather than an awkward laugh.

In real situations, the two threats mapped below overlap and feed each other — people drink partly because of pressures; exhaustion is often distributed unequally. The split is about emphasis, not clean borders.