Conditioned and Unconditioned Stimuli
Like other forms of pavlovian associative learning, conditoned taste aversions are acquired when a conditioned stimulus (the CS) is paired with an unconditioned stimulus (US). CTA learning occurs optimally within specific domains of CS and US.
- 1 Conditioned Stimuli: Taste and Other Orosensory Stimuli
- 2 Toxins and Unconditioned Stimuli
Conditioned Stimuli: Taste and Other Orosensory Stimuli
CTA is also referred to as conditioned flavor aversion or conditioned food aversion. In most cases, CTA (or its synonyms) is used generically to refer to any pairing of orosensory food stimulus with a toxic effect. Strictly speaking, the food stimulus can be broken down into at least 3 components which engage 3 different sensory modalities:
Mediated by oral taste receptors and relayed by the 3 taste nerves (glossopharyngeal, chorda tympani, and greater superficial petrosal -- some include the laryngeal branch of the vagus as well) to the rostral nucleus of the solitary tract (NTS). The primary taste qualities are sweet, sour, salty and bitter (there are 2 other candidate taste qualities: oligosaccharides (Polycose) and unami (monosodium glutamate).
Mediated by the olfactory epthelium and relayed by olfactory nerves to the olfactory bulb. A pure odor stimulus would stimulate only the olfactory epithelium when smelled or ingested. A flavor stimulus is a mixture of taste and odor – we typically use grape or cherry kool-aid, which combines sweet and sour taste with cherry or grape odor. Note that the oral cavity communicates with the olfactory epithelium via the retonasal passages – so what we informally call “taste” is almost always “flavor” – a combination of oral gustatory and nasal olfactory stimulation.
Mediated by mechanoreceptors, thermoreceptors, and nocioceptors and relayed by the trigeminal nerve to the somatosensory relays of the brainstem (although the trigeminal also projects to the “gustatory” rostral NTS). Trigeminal stimuli include things like temperature, texture (mouthfeel or creaminess), and pain; they can be elicited by non-chemical, mechanical stimulation like brushing the surface of the tongue. It is thought that stimuli such as oil and fat and capsaicin (hot pepper) are detected by trigeminal innervation rather than by taste cells and taste nerves.
We typically use a 5% sucrose, which is a relatively pure taste stimulus acting at sweet taste receptors. Olfactory cues are minimized by using clean vessels and distilled water. Of course, the solution also has physical properties (volume, viscosity, temperature) that inevitably provide trigeminal stimulation. However, by using sucrose we can confidently assert that we are studying conditioned taste aversions; if we used kool-aid, we would say we are studying conditioned flavor aversions.
[Incidently, the animal does not need to swallow the taste, or have the taste enter the GI tract, for the taste to serve as the CS in CTA acquisition. Just stimulating the oral nerves appears to be sufficient.]
Toxins and Unconditioned Stimuli
There are 4 major pathways by which unconditioned stimuli are detected by the brain in CTA learning. As a short-hand, we call these unconditioned stimuli “toxins”, and as a rule of thumb anything that makes a person sick will give a rat a CTA. But rats will often form CTAs when a taste is paired with a non-toxic substance or at a non-toxic dose of a mild toxin, or even form a CTA with a “pleasurable” experience (like a drug of abuse). Jim Smith has put it best: the rat may form a CTA after any treatment that makes it feel “different”.
Gastrointestinal (Visceral) Toxins
The effects of toxins that act within the gut are relayed from the abdominal visceral to the brain either as nervous signals or as humoral signals. Neural input is relayed by the afferent subdiaphragmatic vagus, and perhaps by other nerves (e.g. splanchnic nerve, spinal afferents). An example is intragastric copper sulfate, which is not absorbed into the blood but which does activate vagal afferents to induce CTA. The humoral pathway is mediated by chemical signals released by the gut onto nearby nerves (paracrine signals) or into the blood stream that can affect distant nerves or the brain itself (endocrine signals). Examples: fat in the gut causes CCK release from the small intestine that activates vagal afferents; ionizing radiation (e.g x-rays) causes damage of intestinal mast cells that release histamine into the blood. The vagus nerve terminates in the visceral NTS, immediately caudal to the gustatory NTS.
Toxins which are absorbed into the blood, or which are administered systemically (like intraperitoneal LiCl) can directly activate nerves, endocrine systems, and brain regions. The area postrema (AP), a circumventricular organ which sits on top of the NTS on the floor of the fourth ventricle, is particulary sensitive to blood borne chemicals; lesions of the AP block CTA acqusition to intraperitoneal LiCl or intravenous copper sulfate. The AP and several other circumventricular organs have “leaky” capillaries, and so have no blood brain barrier; this increases their sensitivity to blood-borne chemical and peptides.
Some drugs can cross the blood brain barrier after systemic administration and directly stimulate receptors on brain cells. For example, amphetamine or apopmorphine rapidly activates dopamine receptors in the brain to induce CTA, even in AP-lesioned rats. Many drugs and peptides can induce taste aversions when directly injected into the brain (e.g. GLP-1). Often the receptors for these drugsa re scattered throughout the brain. Without extensive mapping with site-specific injections, it is difficult to tell where these drugs have their “aversive” effect.
[Note that several of centrally acting drugs appear to make rats hungry (NPY, 2DG) or thirsty (angiotensin), so they actually eat or drink more after the injection – only to show a CTA the next day.]
Many (if not all?) species also form a CTA if a taste is paired with vestibular stimulation such as rotation (real or optical), shaking, or damage to the vestibular apparatus of the inner ear – in other words, motion sickness. The detection of motion comes from the inner ear via the vestibular nerve, or from motion of the visual field via the optic nerve; motion sickness is thought to arise when the inner ear and the eyes relay conflicting perceptions of motion. Suggestively, the central vestibular nuclei that process incoming vestibular and ocular motion information sit on top of the gustatory NTS, much as the AP sits on top of the visceral NTS.
Toxins and Satiety
Interestingly, there seems to be an inverse relationship between a drug’s ability to induce a taste aversion and its ability to decrease food intake. For example, compare LiCl and CCK. CCK is an endogenous satiety hormone that makes you feel “full” and satiated; CCK is released when food enters the small intestine. CCK is very potent at decreasing intake of a familiar food in rats, but it requires much higher doses to act as the toxin in CTA acquisiton. Conversely, LiCl is very potent at acting as a CTA unconditioned stimulus, but the same doses given before a meal of a familiar food do not actually decrease intake. So there may be separate pathways for satiety and for sickness that can lead to decreased food intake.
For the majority of stimuli that can induce CTA, the peripheral or central brain sites that are sufficient or necessary for CTA acquistion are not well defined. Furthermore, the time course of the toxin’s effect is very fuzzy. We know the duration of the taste stimulus, from when a rat starts licking until it stops licking at the solution (or when we stop an intraoral infusion of a taste). While we can control the onset of illness (the time we inject the LiCl), we can’t control the termination of the toxic effect,because we can’t remove all the toxin from the blood abruptly, nor can we specify the exact time the rat is no longer feeling sick.