John Garcia

A General Theory of Aversion Learninga

Infinitely various are the modifications of light and sound, whence they are each capable of supplying an endless variety of signs, and, accordingly, have been each employed to form languages; the one by the arbitrary appointment of mankind, the other by that of God himself. A connection established by the Author of Nature, in the ordinary course of things, may surely be called natural; as that made by men will be called artificial. GEORGE BERKELEY, 1733

Flavor-illness aversions: potentiation of odor by taste in rats

For the Sprague–Dawley rat, the odor of almond-scented water proved to be a weak cue for lithium chloride-induced illness, while the taste of saccharin water was very effective. However, when the weak almond color was combined with the strong saccharin taste into a compound cue and followed by lithium chloride-induced illness in a compound conditioning paradigm, taste strongly potentiated odor. Before illness, rats were hesitant (neophobic) to consume almond-scented saccharin solutions. During extinction tests conducted after the compound solution was paired with illness, rats displayed a stronger aversion to the odor component than to the taste component. Aversions for the odor component increased in direct proportion to the concentration of the taste component that was present only on the conditioning trial. Tests with anosmic rats indicated that almond-scented tap water was primarily an olfactory cue having little if any taste component. In contrast, when the almond odor was combined with saccharin taste in a second-order conditioning paradigm, rats did not use odor to avoid taste previously made aversive by illness; nor did they acquire aversions for taste paired with aversive odor. Synergistic potentiation of aversions for the weak stimulus component by the strong component contrasts with the interference effects usually observed in classical conditioning. Taste may index the memory for distal poison cues when toxicosis is delayed after ingesting poison to circumvent interference by intervening cues. This allows the animal to form selectively an aversion for distal poison cues, thus reducing the risk of tasting poison.

Coyote Predation Control by Aversive Conditioning

Conditioned aversions were induced in coyotes by producing lithium chloride illness in them following a meal, and the effects upon eating and attack behavior were observed. One trial with a given meat and lithium is sufficient to establish a strong aversion which inhibits eating the flesh of that prey. One or two trials with a given flesh (lamb or rabbit) specifically suppresses the attack upon the averted prey but leaves the coyote free to attack the alternative prey. A method of saving both prey and predator is discussed.

NIR‐Triggered Release of Caged Nitric Oxide using Upconverting Nanostructured Materials

In this manuscript, a novel strategy for the therapeutic delivery of nitric oxide to physiological targets is described. Nitric oxide (NO) is a key vasodilator in mammalian cardiovascular systems and has been shown to sensitize tissue to γ -radiation. However, photochemical NO precursors that can be activated by these near-infrared (NIR) wavelengths are extremely limited. Herein, we have addressed this issue by utilizing the known NIR to visible upconversion properties of lanthanide cations doped NaYF 4 nanocrystals. This result is a potential game changer in multiphoton excitation based therapeutic delivery of NO and other small molecule bioregulators. The bioregulatory molecule nitric oxide (NO) plays important roles in cancer biology and has been implicated in both tumor growth and suppression. [ 1 ] Furthermore, NO is a γ -radiation sensitizer that can enhance selective killing of neoplastic tissues. For these reasons, there is considerable interest in developing methods for NO delivery to specifi c physiological targets. In this context, several laboratories have explored photoactivated NO delivery to specifi c tissues upon demand. [ 2‐7 ] A key advantage of photochemical triggering

Preferences and aversions for stimuli paired with ethanol in hungry rats

Two experiments explored the reinforcing effect of ethanol on conditioned location and flavor preferences in hungry rats. In Experiment 1, rats were administered ethanol (.5, 1.0, or 2.0 g/kg, ig) prior to confinement in one side of a shuttlebox with access to a flavored solution. On control trials, H2O was administered prior to confinement to the opposite side with a different flavored solution. Location choice tests revealed an overall aversion for the ethanol-associated side that was largest at the 2.0-g/kg dose. Flavor choice tests revealed an aversion for the ethanol-associated flavor at the 2.0-g/kg dose, no reliable difference at the 1.0-g/kg dose, and, of particular interest, a preference at the .5-g/kg dose. The results of Experiment 2 suggest that caloric restoration served as the reinforcing mechanism for the conditioned flavor preference. An isocaloric glucose solution conditioned a flavor preference of the same magnitude as that obtained with ethanol. Moreover, when ethanol provided no caloric advantage, the associated flavor was less preferred than a flavor associated with an isocaloric glucose solution.

Prey-lithium aversions. III: Buteo hawks

While mammalian predators, such as the coyote, follow an olfactory spoor in hunting, hawks rely primarily on visual information. Also, mammalian predators kill with their teeth, whereas hawks kill with taloned feet and so do not taste their prey immediately. In this experiment, captive Buteo hawks were studied to determine (1) if hawks can learn to avoid prey that have been paired with illness as effectively as the coyote, and (2) if distal visual cues are more significant than proximal taste cues in the conditioning of such aversions. Lithium chloride illness followed consumption of “poison” mice that differed from alternative “safe” mice in taste and/or color. Acquisition, generalization, and extinction data indicated that (1) hawks, like the coyote, can acquire pronounced aversions for prey, and (2) while visual cues were sufficient to inhibit attack directed at the prey, taste cues were much more effective in inhibiting consumption.

Conditioned taste aversions: vagal and circulatory mediation of the toxic unconditioned stimulus.

In these experiments we tested the hypothesis that the same afferent systems which mediate vomiting reflexes also participate in conditioned taste aversions which are formed when ingestion of a taste cue is paired with subsequent internal malaise. Borison and Wang have shown that emetic reflexes in dogs and cats are integrated in a brainstem center which responds to local gastric irritation mediated by the vagus or to bloodborne toxins monitored by brain stem chemoreceptors. This same afferent system may mediate taste-illness conditioning since the emetic center is anatomically associated with the relay nucleus for primary taste afferents. We have shown that subdiaphragmatic vagotomy in rats disrupts the acquisition of a saccharin aversion induced by repeated conditioning trials with low doses of intragastric or intraperitoneal copper sulfate. In contrast, there is little effect on the aversion produced by a single, larger dose of copper sulfate injected directly into the blood via the tail vein, although vagotomized rats displayed an accelerated extinction pattern. These results parallel reports on the effect of vagotomy on vomiting reflexes in dogs and suggest a functional relationship between acute vomiting reflexes in dogs and suggest a functional relationship between acute vomiting reflexes which eject poison and chronic conditioned taste aversions which inhibit subsequent ingestion of poison; further, they provide evidence that these responses are subserved, at least in part, by similar neural mechanisms.

The attenuation of a specific cue-to-consequence association by antiemetic agents.

Previous research has shown that rats develop a conditioned taste aversion after a single pairing of a distinct taste and subsequent toxicosis. The experiments reported here test the hypothesis that the expression of a taste aversion may reflect classically conditioned nausea mediated by activation of brainstem emetic centers by taste stimuli. Rats were allowed to drink a saccharin solution (1 g/l) and 10 min later were intubated with LiCl (180 mg/kg) to produce nausea. When control rats were posttested for saccharin preference they consumed less than 50% of their pretest intake. Experimental rats were injected with one of four pharmacologically distinct antiemetic drugs 30 min prior to their posttest with saccharin. Each drug significantly attenuated the aversion to saccharin at one dose level. The antiemetic drugs we used were scopolamine HBr, cyclizine, prochlorperazine dimaleate, and trimethobenzamide. These drugs had no effect on the conditioned fear of a noise that signaled foot shock or on a natural aversion to a bitter fluid (quinine monohydrochloride, 100 mg/l). Our data suggest that pharmacological suppression of the neural mechanisms of emesis selectively disrupts conditioned taste aversions, and that moderate dose levels are critical for obtaining this effect.

Prey-lithium aversions. I: coyotes and wolves.

Captive coyotes were fed rabbit flesh treated with lithium chloride (LiCl) and captive wolves were fed similarly treated sheep flesh. One or two treatments inhibited predatory attack upon the living prey, but left the appetite for alternative prey unaffected. A caged cougar refused to eat deer meat after one meal of venison laced with LiCl. Sheep flavored baits and sheep carcasses laced with LiCl, distributed on a 3000 acre sheep ranch in southeastern Washington, were consumed by feral coyotes. A comparison of this years sheep losses with the ranchers past records suggested a 30–60% reduction in sheep killed by coyotes following this application of taste aversion conditioning in the field.

What the Nose Learns from the Mouth

The mouth is a border zone dividing the feeding domain into two separate regions, each with its own empirical laws. The first region is the external arena, in which terrestrial animals cope with signals precisely located in space and time to obtain food or drink. The second is the internal homeostatic pool, which places demands for food or drink upon the animal. Pavlovian laws of conditioning were formulated in the external environment, where the flavor of food acts like an unconditioned stimulus (UCS) which, if applied promptly, can be used to modify a vast array of coping behaviors in man, rat or pigeon. These same laws do not apply to the internal environment, where the flavor of food acts more like a conditioned stimulus (CS) and a specialized visceral feedback acts as the UCS to modify the specific flavor incentive for which man, rat or pigeon strives.