Bob Kentridge 1995

Comparative Psychology: Lecture 3.

Basic concepts in Classical Conditioning again!

Today we will continue to learn about classical conditioning, which, you will recall, initially developed out of early work on the physiology of reflexes. I can't emphasise too much how important it is to understand the basic components of the classical conditioning paradigm, so to start this week's lecture I will reiterate the basic concepts we ended with last time. As a quick historical aside (I can't resist!) I'd like to point out that there has recently been some controversy over our example - conditioning salivation to the toll of a bell. The eminent behaviourist Charles Catania suggested that there was no documentary evidence that Pavlov had ever carried out this experiment, that Pavlov had never used a bell as a CS. Luckily for us and our (or at least my) cherished mental images of Pavlov's dogs evidence was unearthed to show that Pavlov described the use of bells as CSs in work described in Science and the British Medical Journal in 1906, moreover, a film, "The Mechanics of the Brain", made in Pavlov's laboratory in the 1920's shows the use of a bell as a CS. We can therefore safely return to our example in which the experimenter tolls a bell (the CS) and presents food (the US) which elicits reflexive salivation (the UR) in a dog. After a number of such paired presentations the toll of the bell alone now elicits salivation (the CR), which may eventually differ slightly from that elicited by the food. So, to reiterate, the basic terms we must have clear are: Now we have defined our terminology we can begin to ask questions about classical conditioning. I have three in mind - what determines whether classical conditioning is effective? What is learned during classical conditioning? Can the processes underlying classical conditioning be explained?

What determines whether classical conditioning is effective?

First of all, as our definition states, we need a US which elicits and innate reflexive response. In addition to food induced salivation other reflexes commonly used include access to an opposite-sex conspecific in order to condition courtship behaviour in birds, eye- blinks (or nictitating membrane closure in animals like rabbits) elicited by puffs of air directed at the eye, leg-withdrawal from electric shock, together with the more complex constellation of 'fear' reactions to shock such as changes in skin-conductivity (galvanic skin response, GSR), changes in heart-rate and suppression of ongoing behaviours (when these responses are conditioned they are known as a conditioned emotional response - CER), the reflexive pecking which food elicits in birds, or the gagging and nausea which the flavour of a poisonous food comes to elicit. It must be emphasised that the these reflexes may exist in some species but not others. Given an effective US, however, there are sill many factors which influence conditioning a particular CS using it.

Timing.

We have not been very precise in our description of the timing of the conditioning procedure so far. I have used phrases like 'paired with' or 'presented at the same time 'approximately' when describing the timing of the presentation of the CS and US. In fact, the precise timing of these events has a great influence on whether the CS can be reliably conditioned to produce the CR. First of all let us look at the development of the CR more precisely. I will be using the same type of diagram to represent a number of situations, it is important that you understand it. In this sort of diagram we represent the occurrence of events over time which flows from left to right. Each sort of event is represented on a separated line. The points or regions over which this line is raised denote the times when the event is occurring. Here is a diagram showing how the salivation response of a dog changes as the presentation of food, the US, is repeatedly paired with the sound of a buzzer, the CS. The development of the CR is shown most clearly in a version of the conditioning procedure called 'Delay Conditioning' - note that the presentation of the CS exactly precedes the presentation of the US - in the normal classical conditioning procedure the CS onset precedes the US but they overlap in time for a little before the CS stops.
You'd see a Development of the CR image here if you were using a graphical web browser like Mosaic or Netscape.
Notice that initially the dog only salivates when presented with the food US, but gradually the buzzer begins to elicit salivation, until, after a number of trials the dog salivates as soon as the buzzer sounds. Now let us look at a number of other arrangements of the relative timing of CS and US.
You'd see a CS-US Timing image here if you were using a graphical web browser like Mosaic or Netscape.
As you can see the relative timing of the CS and US is crucial to the success of conditioning. I've indicated variations between the definitions given in different textbooks with dashed lines. Most of these results appear to have simple explanations. In delay conditioning the salivation elicited by the US gradually extends backwards as it becomes associated with the US. In the standard paradigm the same thing happens although it is a little less obvious because of the overlap between the CS and US. In trace conditioning we might assume that the very recent memory trace of the CS begins to be associated with the US and hence the UR gradually extends back, albeit weakly, to the actual occurrence of the CS. The strength of trace conditioning is generally inversely related to the delay between the CS and US. The same cannot be happening in backwards conditioning which is usually ineffective. Once the CS appears the US has already elicited a response (the UR not the CR) so the effectiveness of conditioning is not simply dependent on the temporal contiguity of the CS and US. If we try to think what is about classical conditioning that makes it so ubiquitous in evolutionary terms this finding might suggest that a crucial feature of successful CS-US pairing in time is that the CS can be used to predict the occurrence of the US. This suggestion may also explain one of the most surprising results - if the CS and US are presented exactly simultaneously, so that both the onset and offset both occur together, then conditioning fails.

Measuring the strength of conditioning - the suppression ratio.

Before we go on to consider the factors which effect classical conditioning in more detail we need to understand how the effects of variations in conditioning procedures on the strength of conditioning can be measured objectively. We need some method by which we can decide whether an experimental manipulation increased or decreased the strength of conditioning. The most commonly used measurement technique is a procedure called conditioned suppression. As I mentioned above, one of the reflexes elicited by painful stimuli is the suppression of ongoing behaviour. It is therefore possible to measure the strength of association between a neutral CS (e.g. a tone) and a painful US (e.g. an electric shock) by measuring how much an animal's behaviour is reduced in the presence of the CS compared to its absence. If we train an animal to perform some repeated measurable behaviour, such a pressing a bar in order to obtain food rewards, then the strength of a conditioned emotional response to a separately learned tone-shock association can be determined by measuring the reduction in the animals rate of bar-pressing when the tone is sounded. The measure of the extent to which the CS suppresses responding is called the suppression ratio and is normally defined as being the rate of responding in the presence of the CS divided by the sum of the response rate in the presence of the CS and in the absence of the CS. If A is the response rate during CS and B is the response rate in the absence of the CS (usually measured immediately prior to CS presentation) then the suppression ratio is A/(A+B). With this formula a CS which completely suppresses responding will score 0.0, one that has no particular effect will score 0.5, a stimulus which elevates responding for some reason will score between 0.5 and 1.0. It is important to be clear that although the animal has been trained to bar-press for food before the conditioning experiment begins, and although the effect we measure is on the rate of bar-pressing, conditioned suppression is measuring the strength of the classically conditioned CER - usually a tone-shock association, not a change in the things the animal has learned about bar-pressing for food.

Predictability.

The results of the different timing procedures discussed above suggested that an important determinant of the success and strength of conditioning might be the extent to which the CS predicts the US. Is predictability important above and beyond the number of times the CS and US occur together? In order to examine this question we require a procedure in which the correlation between CS and US occurrence is varied while the total number of times the subjects are exposed CS-US pairings are held fixed. An experiment which examined this question was devised by Robert Rescorla in 1968. He took four groups of rats and exposed them all to tone-shock pairings In each test session the animals heard a number of 2 minute tones interspersed by silences. For all groups the probability that they would experience a shock while hearing the tone was 0.4 (on 24 out of the total of 60 tones they heard over the whole of training they also received a shock). The groups differed in the probability that they were also shocked during the silences occurring between tones - one group received no shocks during these intervals (a probability of 0.0), the other groups received shocks during the 'no tone' intervals with probabilities of 0.1, 0.2 and 0.4. So, although all the groups receive the same number of tone-shock pairings the tone becomes a progressively better predictor of shock as the probability of shock occurring during the 'no-tone' intervals decreases. He also tested a control group who received no shocks but heard the same number of tones as the experimental groups. The results clearly show that CS-US predictability is an important factor in determining the efficacy of conditioning - the more the experimental groups were shocked during no-tone intervals the less the tone could predict shock and the less their bar pressing was suppressed by the tone during the conditioned suppression test-phase of the experiment.
You'd see an image of Rescorla's results here if you were using a graphical web browser like Mosaic or Netscape.
This experiment was the forerunner of many others which were used in attempts to discover precisely how animals use information available to them about stimuli to learn about their environment. Two other phenomena concerned with the way a CS can be used to predict the occurrence of a US form the building blocks for this work - overshadowing and blocking. We will discuss them and the development of models of the process of learning in classical conditioning next week.

Sources.

The debate about Pavlov and the toll of the bell is in A.C. Catania and various others in Pavlov-Bell in volume 5 (1994/5) of the electronically published journal 'Psycoloquy' (you can read it with Mosaic by opening the URL (unversal resource locator) http://cogsci.ecs.soton.ac.uk/cgi-bin/psycoloquy). The section on timing draws from Schwarz, B. (1989) Psychology of learning and behavior. (3rd ed) New York: Norton and Davey, G. (1981) Animal learning and conditioning. London: Macmillan. Both are good clear textbooks. The final section on predictability is derived from Dickinson, A. (1980) Contemporary animal learning theory. Cambridge: CUP. This is not really a textbook, but it contains quite readable explanations and summaries of recent(ish) work in animal learning. Its a good last step before taking the plunge and reading the original journal papers it discusses.