Long ago I attended a lecture by a researcher who studied exactly this phenomena. As I recall it, they found that context is very important. If you are in a setting you associate with drinking and you drink something that replicates important elements of the experience, the chances of you subjectively feeling intoxicated increase substantially. People would be shocked to find out that they had been drinking cocktails that only had alcohol on the rim or non-alcoholic beer at the end of the experiment because they had felt it so strongly. Almost everyone they tested with replication of a real drinking experience maximized felt like they had drank alcohol to some extent and you could see their behavior change in very obvious ways as if they had consumed alcohol. The less they replicated the experience the less people felt it, so a nonalcoholic cocktail with alcohol around the rim you smell as you drink was more impactful than the same drink without the smell of alcohol on the rim, and whether the experiment was done in a bar like environment or a lab like environment made a significant difference.