Conditioned taste aversion is the rebel of classical conditioning, since it breaks some well-established rules. Let's start with a real-life example. For many people, this pizza probably triggers thoughts of yumminess, but for other people, like Sherry, it provokes strong feelings of nausea.
Why does Sherry feel this way about pizza? Many years ago, she had some pizza for lunch, and later on that day suffered from a nasty stomach bug and ended up spending the next 24 hours in the bathroom. Even though it wasn't the pizza that caused this illness, she now has a lifelong aversion to it, and she hasn't eaten pizza since. This experience is also common in people who've experienced gastrointestinal illness from food poisoning. This incredibly strong, learned association between illness and food is known as conditioned taste aversion.
Conditioned taste aversion is a learned avoidance of a particular taste, specifically when nausea occurs after food is eaten. If we compare what happened in Pavlov's original experiments with Sherry's learned aversion to pizza, we can see similarities. In both situations, there's a stimulus that automatically triggers an involuntary response, meat makes dogs drool, and stomach bugs make us burp. In Pavlovian terms, these are the unconditioned stimuli and unconditioned responses.
In both types of learning, there's a stimulus that occurs before the unconditioned stimulus, which on its own shouldn't trigger this response. The pairing of these stimuli, the bell and the pizza, with the unconditioned stimuli results in them taking on new meaning, and they can then trigger the responses on their own. In Pavlovian terms, these newly learned triggers are called conditioned stimuli, and these learned reactions are called conditioned responses. Let's take a moment to review Pavlov's fundamental rules of classical conditioning.
First, multiple pairings of the conditioned stimulus with the unconditioned stimulus are necessary in order for the conditioned stimulus to take on the association. Second, the conditioned stimulus and unconditioned stimulus must be presented post-together in time. These rules are based upon a mountain of laboratory research observed in meticulous experiments under tightly controlled lab conditions by Pavlov and others.
However, Sherry's conditioned taste aversion to pizza did not appear to require multiple pairings. In fact, there was only one time when the pizza was paired with illness. Also, Sherry ate the pizza hours before the nausea set in, so conditioned taste aversion appears not to fit neatly with the research on classical Pavlovian conditioning. Of course, Sherry's learning happened in messy real life with the slights of pizza and a tummy bug, so it's useful to look at this phenomenon in a laboratory setting.
Conditioned taste aversion is also referred to as the Garcia effect, after John Garcia, whose classic research first investigated this. Garcia accidentally stumbled upon his now famous research studying the effects of radiation exposure. He noticed that rats responded to low levels of radiation with reduced feeding as a result of gastrointestinal upset. Therefore, he set out to ask two questions. Can rats be conditioned to avoid a sweet taste, which rats ordinarily love, if it's paired with radiation which makes them sick?
And, if so, how long does this food aversion last? So this is what he did to test his hypothesis. There were six groups of rats, exposed to one of three irradiation conditions, a no radiation control group, low radiation, and high radiation for a period of six hours.
For the six-hour irradiation period, rats were either given access to regular water or water with the artificial sweetener saccharin. Keep in mind that there are two independent variables, the level of radiation, indicated by the colors, and the fluid they were given while exposed to the radiation, which is indicated on the water bottle. After this six-hour conditioning period, rats were given a choice between two drinking bottles, one with water and another with, you guessed it, saccharin water.
The dependent variable was how much of the saccharin water the rats consumed. Like humans, rats like things that taste like sugar, so ordinarily they prefer sweet-tasting saccharin over plain, unflavored water. Since radiation produces gastrointestinal upset, Garcia hypothesized that the rats might associate their illness with a sweet taste and avoid it in the future.
and this is what he found. The main measure was the percentage of saccharin water the rats drank when given a choice, shown here on the y-axis on the left. Rats that got water during the conditioning session drank mostly saccharin water afterwards, close to 100%, regardless of whether they were in the control group or the radiation group. This strong preference for sweet saccharin was also similar in the rats that got saccharin water during training but were not irradiated. Since they didn't get sick, they didn't have an unconditioned stimulus to associate with the sweet taste.
The interesting part of this experiment is that rats who received saccharin with radiation avoided drinking saccharin water later. The more irradiation, the more they avoided the sweet tasting water. Garcia tested saccharin preference later in these same rats and found that there was a strong aversion that lasted 30 days, which didn't go back to pre-irradiation levels until 60 days later. This seems like a tidy experiment, but you might say, hey, wait a minute.
Did the rats given water while irradiated, these ones here, avoid water? Does their saccharin preference indicate that they like sweet taste? Or is it... that they now actually hate regular water.
Garcia thought otherwise, so he did some more experiments about the nature of the relationship between the different types of conditioned stimuli and nausea-inducing and other aversive events. In classical conditioning experiments, animals have a whole host of sights, sounds, smells, and tastes in the environment to choose from. So which ones get chosen for conditioned taste aversion?
In another experiment, Garcia and colleagues exposed the rats to a few different types of conditioned stimuli when they drank. In a pre-test, they either got water with a sweet taste of saccharin or a bottle when licked that turned on a bright light and a noise. They then presented these rats with different types of aversive events during conditioning. Rats were randomly assigned to have all of these stimuli paired with either nausea-inducing x-rays or painful electric shocks.
After conditioning, there was a post-test where rats were given a choice, sweet water or the tasteless but bright and noisy plain water. This way, they separated the conditioned stimuli to see if they were preferentially associated with particular types of aversive unconditioned stimuli. The x-ray experiment is shown on the left in blue, and the shock experiment is shown in red on the right.
In both cases before conditioning, rats drank roughly equal amounts of the sweet saccharine water as the bright noisy water, showing no strong preference for either. However, after conditioning with a different aversive unconditioned stimuli, there was a clear distinction in preference depending upon whether the rat had been conditioned to pain or illness. The rats in the nausea condition clearly avoided the sweet saccharine water, while those in the pain condition avoided the tasteless but bright and noisy water. This shook up the learning world, which firmly supported the idea that conditioned learning is universal, so it shouldn't matter which stimulus is paired with what. Garcia showed that our sense of taste is more easily conditioned to illness while our sight and hearing are more easily conditioned to fear.
Garcia's findings sparked all sorts of debate on Skinner's and Pavlov's perspectives on learning. They maintained that the processes of associative learning were universal and general, and that any arbitrary stimulus could be used as a conditioned stimulus. But these conditioned taste aversion experiments showed otherwise. Garcia's findings fit with an evolutionary perspective on taste aversion, in that it's important to learn which taste might make us sick.
It's valuable to quickly associate the tastes and smells of particular foods with nausea in order to avoid contact with those potentially harmful substances again. This knowledge has been used to deter predators, such as coyotes, from eating livestock by placing carcasses laced with lithium chloride, a non-lethal chemical that produces nausea, near the livestock they want to protect. Many farms have dramatically reduced coyote predation on sheep herds and turkey farms without trapping or shooting the predators.
since the predator's own natural food aversion keeps them away. Hopefully that gives you a clear idea of the phenomenon of conditioned taste aversion.