Chapter Outline

Chapter Outline

1. CONTEMPORARY DIRECTIONS FOR LEARNING THEORIES

   Since the 1960's, theories of learning have focused more on specific aspects of learning than on learning in general. This shift in perspective has occurred because the older global theories (1) concentrated on instrumental behavior, assuming that Pavlovian processes were simple and rare, (2) assumed that learning processes were the same for all species, and (3) attempted to reduce all learning processes to simple, automatic associative mechanisms. Modern theories show that (1) Pavlovian processes are more important and complicated than previously thought, (2) not all species learn in the same fashion, and (3) cognitive explanations of behavior can be more useful than previously believed.

2. THEORIES OF PAVLOVIAN CONDITIONING

   1. The Nature of the Conditioned Response: Recent research has tried to clarify the nature of Pavlovian conditioning and has reexamined the assumptions of Pavlov concerning the conditioning process.

      1. The Stimulus-Substitution Theory

         Pavlov's view that the pairing of the CS and the UCS allows the CS to elicit a CR that was similar in form to the UCR. Theoretically, the CS excites the neural centers for the UCS which then triggers the occurrence of reflexive behavior. This is known as the stimulus-substitution theory.

      2. The Conditioning of an Opponent Response

         The opponent process view attempts to state the conditions when the CR and the UCR are not the same response. A clear example showing that the CR and the UCR are opposite reactions comes from the conditioning of drug reactions.

         The phenomenon of drug tolerance shows that the CR and the UCR can be different in character. Tolerance is a reduction in analgesia with repeated exposure to drugs such as morphine. Siegel suggests that the conditioning of an opponent response, hyperalgesia, contributes to the development of tolerance.

      3. Sometimes Opponent-Process (SOP) Theory

         Wagner's sometimes-opponent process (SOP) theory attempts to explain when the CR is similar to and different from the UCR. According to Wagner, the UCS elicits two unconditioned responses, called A1 and A2. The A1 component has a rapid onset and offset in contrast to A2 which has a slow onset and offset.

      4. The Importance of the Nature of the A2 Response

         Wagner claims that conditioning only occurs to the A2 component. Therefore, when the A1 and A2 components are similar, the CR and the UCR are similar. When the A1 and A2 components are different, the CR and the UCR are different. The conditioned emotional reaction is an example where the A1 and A2 components are different and the A2 component becomes conditioned to the CS. Research on hypoalgesic responses are also relevant.

         Some neural/anatomical evidence for the SOP theory comes from eyeblink conditioning in rabbits.

      5. Backward Conditioning of an Excitatory CR

         Excitatory conditioning may occur with a backward paradigm when the CS immediately precedes the A2 response.

      6. Problems with SOP Theory

         SOP theory predicts that all A2 response measures should follow the same empirical laws of conditioning. This does not always occur which has led to a revision in the SOP theory.

      7. Affective Extension of SOP, or AESOP

         Wagner proposed a revised theory, the affective extension of SOP theory (AESOP) that assumes that the A1 and A2 components each elicit separate sensory and affective unconditioned response sequences. The parameters of the conditioning situation determine whether the sensory or emotive reaction predominates.

   2. The Nature of the Pavlovian Conditioning Process: Several theories attempt to explain the importance of predictiveness of CSs in Pavlovian conditioning.

      1. Rescorla-Wagner Associative Model

         The Rescorla-Wagner associative model of conditioning is based upon four assumptions that refer to the process by which the CS and UCS gain associative strength, V: (1) a particular US can only support a specific level of conditioning, l, (2) associative strength increases with each reinforced trial, but depends upon prior conditioning, (3) particular CSs and UCSs can support different rates of conditioning, K, and (4) when two or more stimuli are paired with the UCS, the stimuli compete for the associative strength available for conditioning. The model is based on an important equation

         ΔVA = K(λ - VAX)

      2. An Evaluation of the Rescorla-Wagner Model

         The Rescorla-Wagner model accurately describes many research findings in Pavlovian conditioning, including for example the UCS preexposure effect. Preexposure to UCS without a CS impairs subsequent CS-UCS conditioning.

      3. Problems with the Rescorla-Wagner Model

         1. The Potentiation of a Conditioned Response

            When two CSs are presented simultaneously and are followed by a UCS, the more salient CS is more strongly conditioned than the less salient CS, a phenomenon called overshadowing. The Rescorla-Wagner model accounts for overshadowing well. However, in some cases, the opposite outcome is observed with the salient CS actually enhancing the conditioning of the less salient CS, a phenomenon known as potentiation. The Rescorla-Wagner model predicts that the salient cue should reduce the conditioning of the less salient cue.

         2. The CS Preexposure Effect

            Just as preexposure to the UCS retard subsequent CS-UCS conditioning, so also does CS preexposure reduce later conditioning with the CS, a phenomenon known as the CS preexposure effect. The Rescorla-Wagner model does not readily account for this effect.

         3. The Cue Deflation Effect

            The cue deflation effect sometimes occurs in a situation where two simultaneous CSs of different salience are paired with an UCS. The extinction of responding to the more salient (overshadowing) CS sometimes produces increased CR strength to the less salient CS. The Rescorla-Wagner theory cannot account for the cue deflation effect.

      4. The Importance of Within-Compound Associations

         More recently, Rescorla and associates have proposed that within-compound associations are formed between the CS elements. The assumption that within-compound associations are formed during conditioning helps the Rescorla-Wagner model to explain phenomena such as potentiation and the cue-deflation effect.

      5. A Comparator Theory of Pavlovian Conditioning

         The comparator theory maintains that the ability of a particular stimulus to elicit a CR is dependent upon a comparison of the level of conditioning to that stimulus and other stimuli paired in compound with the UCS.

      6. Mackintosh's Attentional View

         Mackintosh's attentional view suggests that animals seek information from their environment that predicts the occurrence of biologically relevant events. Thus animals play an active role in the conditioning process. Therefore, the presentation of a CS without the UCS produces learned irrelevance indicating that the CS carries no informational value. Therefore, attention is removed from the CS.

      7. The Retrospective Processing View

         According to Baker and Mercier's retrospective processing theory, animals constantly assess the contingencies among different events in their environment and compare those experiences with prior experiences. This continual activity means that new information received can change the meaning of previously established CSs. Backward blocking is an example of research supporting the retrospective view. In backward blocking, two CSs are simultaneously paired with an UCS. Then one of those CSs continues to be paired with the UCS. Finally, when the other CS is tested, its associative strength is reduced (backward blocking) leading to a weakening of conditioning to the other CS.

3. THEORIES OF OPERANT/INSTRUMENTAL CONDITIONING

   1. The Nature of Reinforcement: Although Skinner suggested that theories of reinforcement are impediments to scientific progress, several empirical and theoretical explanations have been offered for how reinforcement strengthens the behavior it follows.

      1. Premack's Probability-Differential Theory

         Premack's probability-differential theory claims that an activity will have reinforcing properties when its probability of occurrence is greater than that of the behavior it is intended to reinforce.

      2. Application: The Use of Activity Reinforcers

         The use of activities as reinforcers, such as in educational and business situations, has been successful.

      3. Response Deprivation Theory

         Timberlake and Allison proposed a response deprivation theory that says that when an animal's normal response rate (e.g., eating food) is restricted (e.g., by food deprivation), that behavior becomes more preferred and therefore reinforcing.

      4. Behavioral Allocation

         This proposal claims that when an animal's ability to emit responses is restricted, it will emit the minimal number of contingent responses in order to obtain the maximum number of reinforcing activities, a phenomenon called blisspoint. This behavioral allocation view is studied in situations which allow the animal unrestricted performance of two behaviors.

      5. Choice Behavior

         At times, an individual can choose from two or more simultaneously available reinforcement contingencies. Herrnstein's matching law predicts that responding to each contingency will be proportional to the reinforcement available on each schedule. The matching law is based on economic principles that assume animals act as if they compute behavioral costs with the probability of reward benefits. This "law" assumes that animals will match the proportion of their responses in each of two situations to the probability of reward in those situations. On the other hand, the maximizing law assumes that animals will behave in such a way to achieve the maximum number/rate of rewards. Other hypotheses support a momentary maximization theory that considers how animals perceive their temporary alternatives for reinforcement, and the delay-reduction theory which suggests that animals are sensitive to different delays of reinforcement in choice situations.