Animal testing has been widely used as a tool for hallucinogen study. Of course, it is impossible to tell when — or if — an animal is hallucinating. But it is possible to train an animal, using standard operant conditioning techniques, to recognize the effects of a given agent. In other words, a rat can be trained to press one lever to get food when under the influence of a particular hallucinogen, and a different lever to get food when it is not.

Two sorts of test are then possible. The trained animal can be given a different agent in a test of stimulus generalization — that is, to test whether the new agent produces effects similar to those of the training drug. Thus, a hallucinogen-trained rat can be given a new substance; if it presses the hallucinogen lever for food, then — for a rat, anyway — the new substance is a hallucinogen; if it presses the non-hallucinogen lever, the new substance is not.

The results can also be quantitative; that is, stimulus generalization studies can be conducted at various doses, in order to determine the potency of the new agent relative to the original agent. It is also possible to use animals to test stimulus antagonism; that is, an animal can be given a substance known to be an antagonist at a particular receptor site, and then tested to see if it still responds to the effects of the hallucinogen to which it has been trained.

There are two problems with such animal studies. First, it is an animal that is doing the discriminating. We have no way of knowing what is salient to a rat about its LSD or psilocin experience. What a rat or monkey considers to be the key features of a hallucinogenic experience may be very different from what a human would think.

Second, there are often significant neuropharmacogical differences among animals.

For example, a rat trained to recognize LSD does not respond if first given the 5-HT2A antagonists ketanserin or pirenperone. This indicates that the effect of LSD in the rat is mediated through the serotonin 5-HT2A receptor, which these antagonists block. But a monkey trained to recognize LSD continues to recognize it even when previously given either ketanserin or pirenperone. And mescaline, a 5-HT2A agonist, fails to substitute for LSD in monkeys, while 5-MeO-DMT, an agonist at both 5-HT1A and 5-HT2A receptors, does substitute for LSD. Does this mean that the experience of LSD is mediated through the 5-HT2A receptor in rats, but through the 5-HT1A receptor in monkeys? Does it mean that the LSD experience is mediated through both receptors in both animals, but what is important to a rat is mediated through one receptor and what is important to a monkey is mediated through the other?

Now, there are structural reasons to group DMT with LSD, psilocin, and other tryptamine hallucinogens. There are also reasons found in animal studies. One widely used training substance has been 2,5-dimethoxy-4-methylamphetamine (DOM). Rats who have been trained to distinguish between DOM and an inactive compound generalize the DOM stimulus to DMT, psilocin, LSD, mescaline, harmaline, N,N–diethyltryptamine, and α-methyltryptamine. There is something rat-salient in common among DMT, psilocin, LSD, and mescaline.

But there is reason to believe that the salient characteristic of ayahuasca and DMT experiences for humans may be opaque to rats. It may well be that rats experience with DMT what the Cashinahua call nixi pae besti, “only vine things” — visual distortions, geometric figures, shifting colors, pinwheels, fireflies — and that rats may perceive these as similar to the visual distortions induced by LSD, psilocin, and mescaline. Yet what is human-salient in ayahuasca and DMT experience is precisely the presence of other-than-human persons and of an extended explorable space within which these persons dwell; perhaps only humans can be — as the Cashinahua put it — blessed by yube, spirit of the boa, the owner of the vine, to see yuxin, spirits, beings with the appearance and agency of human beings.

Indeed, it turns out that the rat 5-HT2A receptor is structurally different from the human 5-HT2A receptor. At a single position in transmembrane 5 of the receptor protein, rats have the amino acid alanine, while humans have the amino acid serine. This difference is apparently enough for human 5-HT2A receptors to have a fifteen-fold higher affinity for psilocin than the rat receptor. As researcher David Nichols puts it, such observations “lend some uncertainty to the generalization of rat data to the human situation.”

We should be as clear as possible: nobody knows how ayahuasca creates hallucinations — or, put another way, how ayahuasca leads into the realm of spirits. Richard Glennon, a leading researcher in hallucinogen pharmacology, states, simply, “It is not yet known how these agents produce their hallucinogenic effects.” Biochemist Bryan Hanson writes: “Unfortunately, even if we know exactly how many receptor types there are, where in the brain they are located, what molecules bind to them, and how strongly, we still won’t know how hallucinations are produced.” The famed hallucinogen chemist Alexander Shulgin puts it this way: current studies may show where psychedelic drugs go, he says, but no one really knows how they work. As leading researcher David Nichols says, the missing pieces of the puzzle are the links between all these biochemical events and the parts of the brain that must be involved in changing consciousness.

And he adds, “It will probably be a long time before this connection can be made.”