Classical Conditioning: From Pavlov to Watson (PSYC10003)

Classical Conditioning: From Pavlov to Watson Mind, Brain, & Behaviour 1 PSYC10003 Learning & Cognition Week 2, Lecture 1 Associate Professor Meredith McKague [email protected] The lectures in this series trace a historical journey through the science of psychology from early theories of lear...

Classical Conditioning: From Pavlov to Watson Mind, Brain, & Behaviour 1 PSYC10003 Learning & Cognition Week 2, Lecture 1 Associate Professor Meredith McKague [email protected] The lectures in this series trace a historical journey through the science of psychology from early theories of learning that focussed on the idea that human behaviour is controlled by learned responses to environmental stimuli to an approach that included understanding how internal mental processes are involved in understanding human thoughts, feelings and behaviours. Today we start with one of the most fundamental forms of learning, called Classical Conditioning. We will trace the development of our understanding of classical conditioning from the seminal work of Russian Physiologist, Ivan Pavlov through to his influence on a young experimental psychologist by the name of John Brodus Watson. Watson saw in Pavlov’s work a model for how we might study psychology using the methods of an objective science - focussing on the control of observable stimuli and their effects on human behavioural responses. There was no room for internal mental processes like memory in this approach - as we work through the lectures we’ll see how this changes. 1 Recapping Week 1 Multiple levels of analysis in psychology Biological Psychological Behavioral Socio-cultural Cognition as an active process Cognition produces and acts on mental representations Perceptual-cognitive cycle. Schemas, mental models, situation models, causal models Propositional (symbolic) and analogue (imagery-based) representations Learning and cognitive development begins with sensorimotor exploration, learning approach and avoidance through emotional responses, which forms the basis for higher-order concepts expressed in more abstract representations. 2 2 Learning outcomes for Classical Conditioning • Understand the processes, terminology, and phenomena of classical conditioning • Habituation and sensitisation • NS, UCS, UCR, CS, CR • Stimulus generalisation, stimulus discrimination • Extinction, spontaneous recovery, rapid re-acquisition • Apply this knowledge to analyse new scenarios • Generate hypotheses for scenarios based on classical conditioning • Interpret graphs depicting classically conditioned responses 3 At the beginning of each lecture I will show you the set of learning outcomes. These provide guidance to you about the key areas of focus in the lecture, and the likely targets of examination. 3 “The normal animal must respond not only to stimuli which themselves bring immediate benefit or harm, but also to those that only signal the approach of these stimuli; though it is not the sight and sound of the beast of prey which is in itself harmful…..but its teeth and claws.” Ivan Pavlov 4 I like to start the lecture with this quote from Ivan Pavlov himself. Pavlov was a Russian Physiologist working in the later part of the 19th century and the early 20th C, studying the processes of digestion using dogs. His work took a turn away from digestion towards the processes of learning when he found some unexpected outcomes of his work with the dogs. I will tell you about this shortly. First – this quote sets us up why classical conditioning is such an important learning process. It is so important because of its role in survival. It is a learning mechanism that is shared by all living species, past and present. At the heart of classical conditioning is its ability to allow us to predict the future, and potentially save ourselves, from elimination..... Pavlov put it this way....(quote) 4 Learning: Definition • Learning is the set of biological, cognitive and social processes through which organisms make meaning from their experiences, producing long-lasting changes in their behaviour, abilities, and knowledge. • Learning helps us predict the future from past experience and to use these predictions to guide adaptive behaviours. 5 Before we go on to look at the work that led Pavlov to his understanding of classical conditioning and its importance for learning, let’s start with a general definition of learning. Learning is something that is demonstrated in the behaviour of the learner, we can infer that learning has occurred when we observe an enduring change in the way in which an animal responds to its environment based on its past experiences. In humans, as in other animals, learning is central to adaptation and survival. Knowing how to distinguish edible from inedible foods, or friends from enemies or predators, is essential for survival. The range of possible foods or threats is simply too great to be prewired into the brain. Learning is essentially about prediction — predicting the future from past experience and using these predictions to guide behaviour. 5 Foundations of learning • Two fundamental forms of non-associative learning are shared by all species : • Sensitisation • Habituation • Sensitisation is the temporary state of heightened attention and responsivity that accompanies sudden and surprising events. The learner remains alert to potentially threatening stimuli in the environment and has an increased response to subsequent stimuli. • Habituation is the gradual diminishing of attention and responsivity that occurs when a stimulus persists. 6 The animal in this slide is a Nobel prize winning creature called Aplysia. Aplysia is a type of sea slug that has been exploring its particular ocean environment for hundreds of millions of years. Aplysia has two key defensive mechanisms that are hard-wired into its nervous system that allow it to protect itself from predators – its gill withdrawal reflex, and its inking reflex, which you can see displayed dramatically here. As you will see reflexes are the basis for all classically conditioned learning. Aplysia was the subject of the Nobel prize winning work of neuroscientist Eric Kandel. Kandel studied habituation and sensitisation in Aplysia, and then described the neural basis of the process of classical conditioning. Aplysia learned to associate stimuli in its environment with small electric shocks to its tail so that it learned, through classical conditioning, to withdraw its gill pre-emptively, before an electric shock was administered . Kandel mapped the neurophysiology of learning in Aplysias’s nervous system, which consists of about 20,0000 neurons - about 1 millionth of the number of neurons in our brains. Aplysia’s neurons are large and so made an excellent model for Kandel to understand how Aplysia represented its experiences in its nervous system, and how neural connections changed to represent its learning and inform its future experiences. You can read about Kandel’s famous work in his autobiography - In Search of Memory. 6 • It is important to learn associations between environmental events (stimuli) that reliably predict biologically significant events (stimuli), and to learn adaptive behavioural responses to stimuli that predict threat or reward. Conditioning: Learning predictive ('conditional’) relationships. • Biologically significant stimuli relate to survival • Stimuli that naturally cause either defensive or appetitive reflex responses. • Stimuli that are naturally rewarding (appetitive) or punishing (aversive). • Called “unconditioned stimuli’ or “reinforcers” in the language of conditioning. • Learning about the causal structure of the environment • “if X, then Y” • Conditioning is also called associative learning: • learning associations (relationships) between stimuli, and/or between stimuli and behavioural responses. 7 Classical conditioning is a form of associative learning. We will discuss another form of associative learning tomorrow. Learning how events in the environment are related to one another – how they are associated with one another – is a key to all learning. And is especially important for survival. 7 Classical conditioning: Pavlov’s dogs Pavlov was interested in understanding the processes involved in digestion and was using dogs for his studies. He began his studies with the first part of the digestive process, which is the production of saliva. In order to get the dogs to produce saliva, Pavlov would put food on their tongues. After working with the same animals for a number of days the dogs started to produce saliva automatically, before he put the food on their tongues. It was this observation that intrigued Pavlov. The dogs seemed to have learned to predict the presentation of the food, with the salivation reflex response occurring before the food stimulus that causes it. Pavlovs work changed at that moment from an interest in digestion, to an interest in how it was that the dogs had learned to predict the future. His initial working hypothesis was that they had learned to associate certain signals, or cues, from the environment that reliably preceded the presentation of the meat, and that the reflex was occurring as preparation for the meat. He undertook a detailed series of studies to uncover and describe the processes of learning involved – the process we call classical conditioning 8 Classical Conditioning Classical Conditioning involves: • Learning a predictive relationship between an originally neutral environmental event and a biologically significant event that naturally causes a reflex response, so that the neutral event becomes able to produce the reflex response on its own. • A classically conditioned response is a learned reflex response to a stimulus that would not usually cause it. • For example, salivation to the sound of a bell is a learned reflex response to the bell, through its association with food. 9 The type of association that is learned in CC is that between one stimulus, that reliably predicts another stimulus, which is itself naturally associated with a defensive or appetitive reflex response. This definition will take some digesting – pardon the pavlovian pun. Come back to it after the lecture, and the activities we will do in tutorials in Week 3 and see if you can unpack all the elements. Classical conditioning is the process of learning an involuntary association between a neutral stimulus (NS) and an unconditioned stimulus (UCS) so that the neutral stimulus becomes a conditioned stimulus that can cause a conditioned reflex response in the absence of the unconditioned stimulus. 9 Classical Conditioning • Pavlov began to control the stimuli that occurred before the food. • Used the ‘bell’ sound of a metronome as his neutral stimulus • Presented the sound of right before he presented the food. • The food naturally causes a reflex salivation response • Pavlov wanted to see if the bell could could come to cause salivation on its own • Let’s step through this process Ollie. My dog Ollie is going to help us understand the processes involved that Pavlov described with his dogs over 100 years ago. 10 Three-phases of Classical Conditioning 1. The conditions that exist before conditioning (before learning) a. The innate reflex responses of the learner that occur to stimuli that are naturally rewarding (appetitive) or punishing (aversive or threatening) b. The neutrality of stimuli that have not been associated with appetitive or aversive stimuli. 2. During conditioning (learning associations) • Experiencing a predictive relationship between a neutral stimulus and a biologically relevant stimulus. 3. After conditioning • The previously neutral stimulus becomes able to produce a learned reflex response in preparation for (or expectancy of) a biologically relevant stimulus. 11 11 How could we train Ollie to salivate to a bell? 12 How can we train Ollie to salivate to the sound of a bell? We’ll follow Pavlov’s process through with Ollie, introducing Pavlov’s formal terminology for each of the component of the learning process as we go. 12 1a. Before conditioning Unconditioned Stimulus US Unconditioned Response UCR 13 There’s two things that exist before we begin the conditioning process. The first is that, before conditioning Ollie has a natural reflex salivation response to the presentation of food. In Pavlovian terms, the food is an unconditioned stimulus (UCS) and salivation response is an unconditioned response (UCR). The unconditioned stimulus is a biologically significant stimulus that naturally causes a reflex response. The reflex response is an unconditioned response. Unconditioned means ‘unlearned’ or innate. We innately salivate in the presence of food - this stimulus-response relationship does not need to be learned. Label the UCS and the UCR on this slide. The UCS + UCR = reflex 13 Before conditioning -1 UCS + UCR = Reflex The UCS and UCR are not learned, they are innate - we say that they are ‘unconditioned’, or unconditional on a learning experience 14 1b. Before Conditioning Initially Neutral Stimulus (NS) No reflex response 15 Before we do any conditioning with Ollie, we also have a pre-existing situation in which the sound of a bell does not cause salivation. In the terminology of CC – we say that the bell is “initially neutral” to Ollie – it might attract his attention briefly, but it does not cause salivation on its own. It is not a biologically significant stimulus. We want to see if we can teach Ollie to salivate to the sound of the bell on its own. 15 Before conditioning - 2 A stimulus that does not produce the reflex = Neutral Stimulus (NS) In this case the NS is the bell 16 NS UCS UCR 2. During Conditioning 17 In this slide we can see the process of conditioning. This involves a series of ‘acquisition trials’ in which we first ring the bell, and then present Ollie with a treat. Presenting the NS before the UCS is repeated over several trials. Why is the temporal sequencing of events important - bell then food? Label each image in terms of classical conditioning 17 During Conditioning… Establishing a conditioned response • Present the sound of the bell just before giving the food • Repeat a few trials in one session, and again after delay (days, a week) 18 3. After conditioning Conditioned Stimulus (CS) Conditioned Response (CR) 19 We can test whether conditioning has occurred by ringing the bell on its own and measuring Ollie’s salivation response. Ollie has established a conditioned salivation response when he salivates to the sound of the bell on its own (in the absences of the UCS) Once this occurs, we call the bell the conditioned stimulus. And, we call the salivation response to the bell, a conditioned (salivation) response, or CR. What do you call the bell after conditioning is complete? What do you call the salivation response to the bell? Why? 19 After conditioning Conditioning has occurred when: • The reflex salivation response occurs in response to the formerly neutral stimulus when presented on its own • The neutral stimulus has become a…..? • The unconditioned response has become a….? The initially neutral stimulus has become a conditioned stimulus The unconditioned response has become a conditioned response. 20 Acquisition of a classically conditioned response 21 Images from Burton text, page 214 This slide shows a graphical depiction of the CC process. On the left you see the sequence of events we just went through. The graph on the right depicts a generic learning curve for a CC response. Each data point on the graph represents the strength of the conditioned salivation response when the bell is presented on it own, as a function of the number of conditioning trials that have occurred. On the Y axis we are measuring mls of saliva. On the X axis is represented the number of conditioning trials prior to each test of the CS on its own. 21 Classical conditioning - Review Learning a predictive relationship between an originally neutral environmental event and a biologically significant event that naturally causes a reflex response, so that the neutral event becomes able to produce the reflex response on its own. Learning a predictive relationship between an originally neutral stimulus and an unconditioned stimulus (UCS) and its unconditioned response, so that the formerly neutral stimulus becomes a conditioned stimulus that can causes a conditioned reflex response. Slide 9 22 In this part of the lecture, I am going to start with a series of questions Chapter 6 of the recommended text is provided for you with the content for this week’s lectures. You will find the answers to these questions there. We will follow up these questions in the practical class in Week 3, and you can seek clarification of any of these processes by posting a question on the discussion board. 22 What if we use a different sounding bell after initial conditioned learning? • Pavlov demonstrated that the classically conditioned salivation response would generalise (transfer) to other similar stimuli. • He called this stimulus generalization • How could you test this? • What would the graph look like? 23 The following slides outline a range of phenomena that Pavlov studied. The first is a phenomenon he called stimulus generalization. He wanted to know how strong the conditioned salivation response would be to other similar stimuli that were not experienced during the training trials. 23 What if I want Ollie to only salivate to the sound of a particular bell? • How could I train Ollie salivate to the sound of one specific bell, and not to others? • That is, how could I train stimulus discrimination? 24 Given that stimulus generalization occurs naturally – how could you then train an animal to produce the response only to a specific stimulus – say a particular tone of bell? Answer in the google doc. Your textbook will be helpful for this. 24 What if I want Ollie to stop salivating to the sound of a bell? • How could I extinguish Ollie’s conditioned salivation response? • What would the graph of extinction trials look like? 25 25 Can a classically conditioned response come back after extinction? • Yes, we call this spontaneous recovery • If you rest Ollie for a period of time after a series of extinction trials and then present the bell again, the conditioned response will return. • Extinction over multiple sessions will gradually prevent spontaneous recovery (at least in contexts like the extinction context). 26 26 What if I wanted to teach Ollie the association again after sustained extinction? • Ollie would re-learn the conditioned salivation response more quickly than he did the first time. • We call this rapid reacquisition. 27 27 What do spontaneous recovery and rapid reacquisition suggest about the extinction process? 28 28 Pavlov to Watson: Behaviorism “Give me a dozen healthy infants… and my own specified world to bring them up in and I’ll guarantee to take any one at random and train him to become any type of specialist I might select – doctor, lawyer, artist….even beggar man and thief, regardless of his talents, tendencies, abilities, vocations and race of his ancestors”. J.B. Watson, 1919. 29 Pavlov’s work had a strong impact on a young experimental psychologist in the early part of the 20th Century. This quote from John Brodus Watson articulates his belief in the role of conditioned learning in shaping human development. Watson set out to demonstrate classical conditioning in a human infant. This was quite a radical view at the time, flying in the face of popular beliefs stemming from Darwin’s work on evolution that much of our learning is genetically determined. This is the classic Nature-Nurture debate. Which side of the debate is Watson on – and why? 29 Conditioned Emotional Responses in Humans: The Little Albert Study (Watson & Rayner, 1920) • Watson and Rayner (1920) first demonstrated that Albert was not afraid of a range of stimuli • But Albert showed a natural startle response and distress to the sudden loud sound of a metal bar being struck • The sudden loud sound gave Watson the UCS-UCR he needed to show that Albert could be conditioned to fear a tame white rat that had not previously produced a fearful response. • The next slide describes the process of associating the rat with the loud sound to produce a conditioned fear response to the white rat alone. • This is the first human fear-conditioning study. • See the original footage here: https://youtu.be/PqqRdZ8x9Ws 30 Watson worked with his research assistant, Rosalie Rayner, to demonstrate the acquisition of a classically conditioned fear response in a human infant. This silent film shows the initial stages of Watson and Rayner’s work - demonstrating that Albert was not innately fearful of a range of stimuli. https://youtu.be/PqqRdZ8x9Ws 30 The Little Albert Experiment Watson & Rayner, 1920 • Albert was exposed to two instances in an initial session, and another five instances a week later. • Sufficient to elicit an extreme fear response to the white rat alone • Generalisation also occurred to other furry animals, and Santa’s white beard. • They did not get a chance to extinguish the response. 31 Watson set out to demonstrate classical conditioning in a human infant. Watson wanted to demonstrate that he could classically condition a fearful response in Albert that was not previously present. He filmed his work with his research assistant Rosalie Rayner. You can view the original footage from 1920 via the links I have provided. You can read the original report of their work from The Journal of Experimental Psychology, 1920, in the link provided in the slide. Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. The Journal of Experimental Psychology, 1(1), 1-14. The study is discussed on page 216-217 of Burton chapter on learning. 31 The Little Albert Experiment (Watson & Rayner, 1920) • In the case of Little Albert.. • What is the formerly neutral stimulus that becomes the CS? • What is the UCS? • What is the UCR? • What is the CR? • How does the process of a learned fear response demonstrated in Little Albert relate to phobias and other anxiety and stress disorders? • This content is foundational for your essay topic. 32 Answer these questions for yourself and post any questions you have on the discussion board for the Q&A session in Week 2. I will clarify any misunderstandings in the first part of our live Q&A session in Week 2. Thanks for listening. Our next lecture for this week will be on another form of associative learning, called operant conditioning. 32