Ch.4 Psychobiology notes
Ch.4 Psychobiology notes Psych 5600
Popular in Psychobiology of Learning and Memory
Popular in Psychlogy
This 4 page Class Notes was uploaded by alvey.15 Notetaker on Saturday February 6, 2016. The Class Notes belongs to Psych 5600 at Ohio State University taught by Derek Lindquist in Spring 2016. Since its upload, it has received 48 views. For similar materials see Psychobiology of Learning and Memory in Psychlogy at Ohio State University.
Reviews for Ch.4 Psychobiology notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/06/16
Ch. 4 Lecture notes 2/04/2016. Notes by Alexandra Alvey. Dr. Lindquist Psych 5600 Classical conditioning is a form of associative learning because two stimuli, sensations or events are linked by contiguity. Remember contiguity is closeness in space and time. Pavlov is credited with the empirical approach of studying classical conditioning with dogs’ and their salivation response to a bell (CS) in prediction of food (US). This predictive response between the CS and US is called psychic secretions. Psychic as in the dogs’ were “psychic” because they did not need to see the food to know that the food was coming and thus secreted saliva in advance of the food. The mechanisms that underlie Pavlovian conditioning are the US, CS, UR and CR. The US is the biologically relevant stimulus that already causes a response (ex. Food, sex, pain). The CS is a neutral stimulus that is presented right before the US for many trials to eventually gain association strength and elicit the behavioral response on its own. The UR is the naturally occurring reaction to the US (ex. Salivation, mounting, running). The CR is the response to the CS which can be the same as the UR or different. Many animals and humans are capable of learning by classical conditioning. An experiment with fruit flies showed that they were even capable of classical conditioning. The flies were presented with two odors. Odor one was the control stimulus and odor two was the experimental stimulus because when the flies were in odor two they would receive a mild shock. When tested in a container with the two odors in opposite chambers all the flies flocked to the side where odor one was present. This is an example of aversive conditioning because the shock (US) associated with odor two (CS) averted the flies away. An example of appetitive conditioning, where the US is positive, is the quail sex experiment. A male quail is placed in a closed chamber with a door that leads to a side chamber that contains a female quail in estrous. A light bulb is placed above the door and when it lights up the door opens to allow the male access to the female. The US is the door leading to the female quail and the UR is sex. When the light (CS) is strongly associated with the opening of the door (US) after many trials the quail spends more time near the area of the door than the rest of the chamber. This is indicative that the quail learned to associate the light with the opening of the door leading to the female. Know how to identify CS, US, UR and CR in appetitive and aversive conditioning. Pavlovian conditioning can also occur in unicellular organisms like E. coli. This finding implies that conditioning is an evolutionarily old mechanism and any organism is capable of learning by this mechanism. In the experiment with E. coli the bacteria were given fructose (CS) first to consume and then glucose (US). The glucose is considered the US in this experiment because the bacteria have to speed up their metabolism to consume it (UR). After many trials of pairing fructose (CS) with glucose(US) the consumption of just fructose initiated a speed up of metabolism (CR) indicating that the E. coli were predicting glucose to come next. Advertising is classical conditioning. Consider an attractive model posing next to a sports car. The US is the model, the UR response is feelings of attraction, the CS is the sports car and the CR is feelings of attraction towards the sports car. These feelings of attraction make you want to get the car in theory. Classical Conditioning is one of the learning mechanisms we have the best understanding of in terms of behavior and brain substrates used. Before eye blink conditioning was used the experimenter would pair a slap (US) after the tone (CS). For ethical reasons an air puff (US) is now presented after the tone in both humans and animals. Rabbits are commonly used in this experiment because they don’t mind sitting still for hours at a time and they blink spontaneously on 2-3 times a minute. An air puff (US) precedes a tone (CS) that makes the rabbit blink (UR) to protect its eyes. The tone and air puff have to overlap otherwise the animal will not make an association between the two stimuli. An electromyogram records the blinks of the rabbit. Blink conditioning is slowly acquired because the rabbit’s have to learn when to blink in order to block the presentation of the air puff (US). In other words, conditioning has a temporal component that is important for a predictive response. The amplitude of the wave from the electromyogram indicates the strength of the blink. On day 1 the rabbit weakly blinks (UR) after the puff of air (US). On day 2 the rabbit starts to blink around the end of the tone and the beginning of the air puff. Also the strength of the blink is much stronger. By day 3 the rabbit begins to blink around the middle of the tone and is protecting its eye from the air puff. The peak(strength) of the blink is timed well with the beginning of the air puff. See slide 11 for diagrams. Humans have a higher CR rate to begin with because they learn faster than rabbits. The shape of the curve is the most important object to pay attention to on the graph. Notice how the percent of CR in the humans goes up at a steady rate while the rabbit’s CR rate jumps and plateaus every once in a while. When setting up a classic conditioning experiment the effects of habituation and sensitization need to be controlled for in order to truly know if learning is based on the CS and US paired together. Control procedures include presenting the CS and US alone to see if behavioral output is being modified without other variables. Explicitly unpaired CS-US never overlap in contiguity and because they don’t overlap the CS should not be predictive of the US. In aversive conditioning the CS becomes a safety signal to the organism because it never is associated with the aversive US. The last control procedure is random CS-US training which regulates the degree of association by overlapping the CS and US randomly. The time and the order that the CS and US is presented can effect the learning process and the brain substrates that are being used. In delay eye blink conditioning the tone (CS) is played and the air puff (US) is administered after a short delay but while the tone is still playing. Delay conditioning is easier for animals and humans to learn because it is implicit and reflexive. Delay conditioning uses the cerebellum. In trace eye blink conditioning the tone (CS) is played, there is a delay of no tone and then the air puff (US) is administered. This is a form of explicit learning because it requires the subject to use working memory to create a memory trace. The memory trace will be used on subsequent trials in order to blink at the correct moment. Again, in order to form a memory trace attention to stimuli and working memory have to be used. Trace conditioning uses the hippocampus and cerebellum. Animals do not perform well on trace memory tasks but capable humans can perform well. Slide 16 figure The interstimulus interval (ISI) is the time between the CS and the US. The learning rate depends on the ISI and generally the shorter the ISI the better CR performance. The US needs to be omitted to test the association. The dotted line is where the US would normally occur. The ISI of 280 milliseconds has a strong blink (CR) to the tone (CS) around where the US would occur. As the ISI increases the CR becomes less accurate. It is also possible to have to short of a ISI. The main point is that there is an optimal ISI to produce a predictive CR. The graph of an optimal ISI looks like an inverted U with the peak occurring around where the US used to happen. Associations are not static and extinction can occur by the presentation of the CS without the US following for many trials. Extinction is not unlearning of an association but a learning mechanism that competes with and eventually inhibits the association. Proof that extinction is a form of inhibitory learning is spontaneous recovery. If the association was forgotten or unlearned, then the subject would have to relearn the excitatory association at the same rate as before but this does not happen. After extinction occurs and some time passes the CR recovers spontaneously. Rapid reacquisition, renewal and reinstatement are also proof that extinction is inhibition learning. Reinstatement is US specific. You can remember this by the s in reinstatement and the s in US. Renewal has to do the the context specificity of extinction and rapid reacquisition is the faster learning rate of the association after extinction. We will learn more about the four R’s later. Contiguity is not sufficient in and of itself to produce an association as Aristotle thought. Blocking is an example of how contiguity does not always result in an association. A compound stimulus is when two stimuli are presented together as the CS such as light and a tone. In an experiment the control group of mice were presented with a CS of light and tone together before the US of a shock happened. The experimental group was presented with a tone as the CS and then a shock as the US when training. When the control group was tested on the CR to the tone and light they presented and equally strong CR to both stimuli. When the experimental group was tested on the CR to the tone and light they presented a strong response to the tone but a minimal response to the light. This happened because the light was redundant information and provided no predictive value for the US. These rats learned that the tone was the most predictive CS for the US(shock). The tone blocked the light from becoming a strong CS. Kamin’s Blocking effect was also demonstrated in humans. In the experiment participants were taught to categorize circles in to class A and triangles in to class B based on the shape. When dots were added to the top of the circles and bottoms of the triangle to indicate a different way to categorize the participants continued to use shape. When new shapes were presented with dots on the top or bottom none of the participants categorized correctly indicating they were using shape as the CS the entire time. The shape category strategy blocked the dot category strategy as the CS. The Rescorla-Wagner Model of learning is error correction learning. Error correction means that if the organism did not react as intended then a correction needs to be made. For example, if you were expecting to get an A on the exam that would be your expected US. If you really got a B on the exam that is the actual US. The amount of learning is determined by the difference between the actual US and the expected US. This is called the prediction error which in this case would be a positive error because an unexpected US occurs. Rationally you would want to correct your actions to predict the actual US (your grade) next time. That example was just hypothetical to get the understanding of prediction error. If the RW model the actual US is either assigned the value of 100 or 0 which means either the US occurs or does not. The expected US increases for conditioning or decreases for extinction on every trial and is the sum of all cues (light, tone, food, etc.). The prediction error is multiplied by beta which is the salience of the cue or how strong of a response the US stimulus produces which effects the learning rate. This is why beta controls the shape of the learning curve because if it is high then the curve will be steeper and low will be a more gradual slope. Beta multiplied by the prediction error is called change in v-cue or delta v-cue. Delta v-cue in trial one becomes the expected US in the next trial because learning has occurred. The shape of the curve is what you need to be concerned about instead of the number of trials. In compound conditioning the prediction error is multiplied by beta for each cue. The delta v- cue for each cue are added together and used as the expected US for the next trial. The sum of the cues will always add to 100. The prediction error in compound conditioning drops more rapidly than in classical conditioning. The RW model is based on the delta rule which is the outcome of one trial will feed in to the next. A fall back to this model is that the equation can’t tell you what is happening during the trial only from one trial to the next.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'